From 8aa13d5dce2789a7d0bd802e6d62453b3c380496 Mon Sep 17 00:00:00 2001 From: Anthony Fu Date: Mon, 16 Oct 2023 14:12:18 +0800 Subject: Interrupt after current generation --- modules/call_queue.py | 1 + modules/img2img.py | 2 +- modules/processing.py | 2 +- modules/shared_options.py | 1 + modules/shared_state.py | 11 +++++++++-- 5 files changed, 13 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/call_queue.py b/modules/call_queue.py index ddf0d573..01c6d17f 100644 --- a/modules/call_queue.py +++ b/modules/call_queue.py @@ -78,6 +78,7 @@ def wrap_gradio_call(func, extra_outputs=None, add_stats=False): shared.state.skipped = False shared.state.interrupted = False + shared.state.interrupted_next = False shared.state.job_count = 0 if not add_stats: diff --git a/modules/img2img.py b/modules/img2img.py index 52cb577a..31f8c2aa 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -49,7 +49,7 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=Fal if state.skipped: state.skipped = False - if state.interrupted: + if state.interrupted or state.interrupted_next: break try: diff --git a/modules/processing.py b/modules/processing.py index 40598f5c..e7eecd66 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -819,7 +819,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if state.skipped: state.skipped = False - if state.interrupted: + if state.interrupted or state.interrupted_next: break sd_models.reload_model_weights() # model can be changed for example by refiner diff --git a/modules/shared_options.py b/modules/shared_options.py index 32bf7353..4638ef06 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -113,6 +113,7 @@ options_templates.update(options_section(('system', "System"), { "disable_mmap_load_safetensors": OptionInfo(False, "Disable memmapping for loading .safetensors files.").info("fixes very slow loading speed in some cases"), "hide_ldm_prints": OptionInfo(True, "Prevent Stability-AI's ldm/sgm modules from printing noise to console."), "dump_stacks_on_signal": OptionInfo(False, "Print stack traces before exiting the program with ctrl+c."), + "interrupt_after_current": OptionInfo(False, "Interrupt generation after current image is finished on batch processing"), })) options_templates.update(options_section(('API', "API"), { diff --git a/modules/shared_state.py b/modules/shared_state.py index a68789cc..c72c3f63 100644 --- a/modules/shared_state.py +++ b/modules/shared_state.py @@ -12,6 +12,7 @@ log = logging.getLogger(__name__) class State: skipped = False interrupted = False + interrupted_next = False job = "" job_no = 0 job_count = 0 @@ -76,8 +77,12 @@ class State: log.info("Received skip request") def interrupt(self): - self.interrupted = True - log.info("Received interrupt request") + if shared.opts.interrupt_after_current and self.job_count > 1: + self.interrupted_next = True + log.info("Received interrupt request, interrupt after current job") + else: + self.interrupted = True + log.info("Received interrupt request") def nextjob(self): if shared.opts.live_previews_enable and shared.opts.show_progress_every_n_steps == -1: @@ -91,6 +96,7 @@ class State: obj = { "skipped": self.skipped, "interrupted": self.interrupted, + "interrupted_next": self.interrupted_next, "job": self.job, "job_count": self.job_count, "job_timestamp": self.job_timestamp, @@ -114,6 +120,7 @@ class State: self.id_live_preview = 0 self.skipped = False self.interrupted = False + self.interrupted_next = False self.textinfo = None self.job = job devices.torch_gc() -- cgit v1.2.1 From 3d15e58b0a30f2ef1e731f9e429f4d3cf1c259c5 Mon Sep 17 00:00:00 2001 From: Anthony Fu Date: Mon, 16 Oct 2023 15:00:17 +0800 Subject: feat: refactor --- modules/shared_state.py | 12 ++++++------ modules/ui.py | 8 +++++++- 2 files changed, 13 insertions(+), 7 deletions(-) (limited to 'modules') diff --git a/modules/shared_state.py b/modules/shared_state.py index c72c3f63..532fdcd8 100644 --- a/modules/shared_state.py +++ b/modules/shared_state.py @@ -77,12 +77,12 @@ class State: log.info("Received skip request") def interrupt(self): - if shared.opts.interrupt_after_current and self.job_count > 1: - self.interrupted_next = True - log.info("Received interrupt request, interrupt after current job") - else: - self.interrupted = True - log.info("Received interrupt request") + self.interrupted = True + log.info("Received interrupt request") + + def interrupt_next(self): + self.interrupted_next = True + log.info("Received interrupt request, interrupt after current job") def nextjob(self): if shared.opts.live_previews_enable and shared.opts.show_progress_every_n_steps == -1: diff --git a/modules/ui.py b/modules/ui.py index bcf39199..c30093d7 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -216,8 +216,14 @@ class Toprow: outputs=[], ) + def interrupt_fn(): + if shared.state.job_count > 1 and shared.opts.interrupt_after_current: + shared.state.interrupt_next() + else: + shared.state.interrupt() + self.interrupt.click( - fn=lambda: shared.state.interrupt(), + fn=interrupt_fn, inputs=[], outputs=[], ) -- cgit v1.2.1 From 7c128bbdac0da1767c239174e91af6f327845372 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Thu, 19 Oct 2023 13:56:17 +0800 Subject: Add fp8 for sd unet --- modules/cmd_args.py | 1 + modules/sd_models.py | 3 +++ 2 files changed, 4 insertions(+) (limited to 'modules') diff --git a/modules/cmd_args.py b/modules/cmd_args.py index 4e602a84..0f14c71e 100644 --- a/modules/cmd_args.py +++ b/modules/cmd_args.py @@ -118,3 +118,4 @@ parser.add_argument('--timeout-keep-alive', type=int, default=30, help='set time parser.add_argument("--disable-all-extensions", action='store_true', help="prevent all extensions from running regardless of any other settings", default=False) parser.add_argument("--disable-extra-extensions", action='store_true', help="prevent all extensions except built-in from running regardless of any other settings", default=False) parser.add_argument("--skip-load-model-at-start", action='store_true', help="if load a model at web start, only take effect when --nowebui", ) +parser.add_argument("--opt-unet-fp8-storage", action='store_true', help="use fp8 for SD UNet to save vram", default=False) diff --git a/modules/sd_models.py b/modules/sd_models.py index 3b6cdea1..3b8ff820 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -391,6 +391,9 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer devices.dtype_unet = torch.float16 timer.record("apply half()") + if shared.cmd_opts.opt_unet_fp8_storage: + model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) + timer.record("apply fp8 unet") devices.unet_needs_upcast = shared.cmd_opts.upcast_sampling and devices.dtype == torch.float16 and devices.dtype_unet == torch.float16 -- cgit v1.2.1 From 5f9ddfa46f28ca2aa9e0bd832f6bbd67069be63e Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Thu, 19 Oct 2023 23:57:22 +0800 Subject: Add sdxl only arg --- modules/cmd_args.py | 1 + modules/sd_models.py | 3 +++ 2 files changed, 4 insertions(+) (limited to 'modules') diff --git a/modules/cmd_args.py b/modules/cmd_args.py index 0f14c71e..20bfb2c4 100644 --- a/modules/cmd_args.py +++ b/modules/cmd_args.py @@ -119,3 +119,4 @@ parser.add_argument("--disable-all-extensions", action='store_true', help="preve parser.add_argument("--disable-extra-extensions", action='store_true', help="prevent all extensions except built-in from running regardless of any other settings", default=False) parser.add_argument("--skip-load-model-at-start", action='store_true', help="if load a model at web start, only take effect when --nowebui", ) parser.add_argument("--opt-unet-fp8-storage", action='store_true', help="use fp8 for SD UNet to save vram", default=False) +parser.add_argument("--opt-unet-fp8-storage-xl", action='store_true', help="use fp8 for SD UNet to save vram", default=False) diff --git a/modules/sd_models.py b/modules/sd_models.py index 3b8ff820..08af128f 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -394,6 +394,9 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer if shared.cmd_opts.opt_unet_fp8_storage: model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) timer.record("apply fp8 unet") + elif model.is_sdxl and shared.cmd_opts.opt_unet_fp8_storage_xl: + model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) + timer.record("apply fp8 unet for sdxl") devices.unet_needs_upcast = shared.cmd_opts.upcast_sampling and devices.dtype == torch.float16 and devices.dtype_unet == torch.float16 -- cgit v1.2.1 From eaa9f5162fbca2ebcb2682eb861bc7e5510a2b66 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 24 Oct 2023 01:49:05 +0800 Subject: Add CPU fp8 support Since norm layer need fp32, I only convert the linear operation layer(conv2d/linear) And TE have some pytorch function not support bf16 amp in CPU. I add a condition to indicate if the autocast is for unet. --- modules/devices.py | 6 +++++- modules/processing.py | 2 +- modules/sd_models.py | 20 ++++++++++++++++---- 3 files changed, 22 insertions(+), 6 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index 1d4eb563..0cd2b55d 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -71,6 +71,7 @@ def enable_tf32(): errors.run(enable_tf32, "Enabling TF32") cpu: torch.device = torch.device("cpu") +fp8: bool = False device: torch.device = None device_interrogate: torch.device = None device_gfpgan: torch.device = None @@ -93,10 +94,13 @@ def cond_cast_float(input): nv_rng = None -def autocast(disable=False): +def autocast(disable=False, unet=False): if disable: return contextlib.nullcontext() + if unet and fp8 and device==cpu: + return torch.autocast("cpu", dtype=torch.bfloat16, enabled=True) + if dtype == torch.float32 or shared.cmd_opts.precision == "full": return contextlib.nullcontext() diff --git a/modules/processing.py b/modules/processing.py index 40598f5c..2df8a7ea 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -865,7 +865,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.n_iter > 1: shared.state.job = f"Batch {n+1} out of {p.n_iter}" - with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(): + with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(unet=True): samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts) if getattr(samples_ddim, 'already_decoded', False): diff --git a/modules/sd_models.py b/modules/sd_models.py index 08af128f..c5fe57bf 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -391,12 +391,24 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer devices.dtype_unet = torch.float16 timer.record("apply half()") - if shared.cmd_opts.opt_unet_fp8_storage: + + if shared.cmd_opts.opt_unet_fp8_storage: + enable_fp8 = True + elif model.is_sdxl and shared.cmd_opts.opt_unet_fp8_storage_xl: + enable_fp8 = True + + if enable_fp8: + devices.fp8 = True + if devices.device == devices.cpu: + for module in model.model.diffusion_model.modules(): + if isinstance(module, torch.nn.Conv2d): + module.to(torch.float8_e4m3fn) + elif isinstance(module, torch.nn.Linear): + module.to(torch.float8_e4m3fn) + timer.record("apply fp8 unet for cpu") + else: model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) timer.record("apply fp8 unet") - elif model.is_sdxl and shared.cmd_opts.opt_unet_fp8_storage_xl: - model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) - timer.record("apply fp8 unet for sdxl") devices.unet_needs_upcast = shared.cmd_opts.upcast_sampling and devices.dtype == torch.float16 and devices.dtype_unet == torch.float16 -- cgit v1.2.1 From 9c1eba2af3a6f9cd6282b3a367656793cbe70c01 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 24 Oct 2023 02:11:27 +0800 Subject: Fix lint --- modules/sd_models.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index c5fe57bf..44d4038b 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -396,7 +396,7 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer enable_fp8 = True elif model.is_sdxl and shared.cmd_opts.opt_unet_fp8_storage_xl: enable_fp8 = True - + if enable_fp8: devices.fp8 = True if devices.device == devices.cpu: -- cgit v1.2.1 From 1df6c8bfec4715610d64684b6ad2fa38c76c1df6 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Wed, 25 Oct 2023 11:36:43 +0800 Subject: fp8 for TE --- modules/sd_models.py | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 44d4038b..69395294 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -407,6 +407,13 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer module.to(torch.float8_e4m3fn) timer.record("apply fp8 unet for cpu") else: + if model.is_sdxl: + cond_stage = model.conditioner + else: + cond_stage = model.cond_stage_model + for module in cond_stage.modules(): + if isinstance(module, torch.nn.Linear): + module.to(torch.float8_e4m3fn) model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) timer.record("apply fp8 unet") -- cgit v1.2.1 From 4830b251366436ee8499c003fe87e46ddb4a4581 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Wed, 25 Oct 2023 11:53:37 +0800 Subject: Fix alphas_cumprod dtype --- modules/sd_models.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 69395294..23660454 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -416,6 +416,7 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer module.to(torch.float8_e4m3fn) model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) timer.record("apply fp8 unet") + model.alphas_cumprod = model.alphas_cumprod.to(torch.float32) devices.unet_needs_upcast = shared.cmd_opts.upcast_sampling and devices.dtype == torch.float16 and devices.dtype_unet == torch.float16 -- cgit v1.2.1 From bf5067f50ca32cd4764638702e3cc38bca8bfd8b Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Wed, 25 Oct 2023 12:54:28 +0800 Subject: Fix alphas cumprod --- modules/sd_models.py | 3 ++- modules/sd_models_xl.py | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 23660454..7ed89a9c 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -396,6 +396,8 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer enable_fp8 = True elif model.is_sdxl and shared.cmd_opts.opt_unet_fp8_storage_xl: enable_fp8 = True + else: + enable_fp8 = False if enable_fp8: devices.fp8 = True @@ -416,7 +418,6 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer module.to(torch.float8_e4m3fn) model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) timer.record("apply fp8 unet") - model.alphas_cumprod = model.alphas_cumprod.to(torch.float32) devices.unet_needs_upcast = shared.cmd_opts.upcast_sampling and devices.dtype == torch.float16 and devices.dtype_unet == torch.float16 diff --git a/modules/sd_models_xl.py b/modules/sd_models_xl.py index 01123321..11259a36 100644 --- a/modules/sd_models_xl.py +++ b/modules/sd_models_xl.py @@ -93,7 +93,7 @@ def extend_sdxl(model): model.parameterization = "v" if isinstance(model.denoiser.scaling, sgm.modules.diffusionmodules.denoiser_scaling.VScaling) else "eps" discretization = sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization() - model.alphas_cumprod = torch.asarray(discretization.alphas_cumprod, device=devices.device, dtype=dtype) + model.alphas_cumprod = torch.asarray(discretization.alphas_cumprod, device=devices.device, dtype=torch.float32) model.conditioner.wrapped = torch.nn.Module() -- cgit v1.2.1 From dda067f64d3289cee3ffd65767126cb30ae73b13 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Wed, 25 Oct 2023 19:53:22 +0800 Subject: ignore mps for fp8 --- modules/sd_models.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 7ed89a9c..ccb6afd2 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -392,7 +392,9 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer devices.dtype_unet = torch.float16 timer.record("apply half()") - if shared.cmd_opts.opt_unet_fp8_storage: + if devices.get_optimal_device_name() == "mps": + enable_fp8 = False + elif shared.cmd_opts.opt_unet_fp8_storage: enable_fp8 = True elif model.is_sdxl and shared.cmd_opts.opt_unet_fp8_storage_xl: enable_fp8 = True -- cgit v1.2.1 From 0beb131c7ffae6f756a6339206da311232a36970 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Wed, 25 Oct 2023 20:07:37 +0800 Subject: change torch version --- modules/launch_utils.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/launch_utils.py b/modules/launch_utils.py index 8cdbafa5..636da679 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -308,8 +308,8 @@ def requirements_met(requirements_file): def prepare_environment(): - torch_index_url = os.environ.get('TORCH_INDEX_URL', "https://download.pytorch.org/whl/cu118") - torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.0.1 torchvision==0.15.2 --extra-index-url {torch_index_url}") + torch_index_url = os.environ.get('TORCH_INDEX_URL', "https://download.pytorch.org/whl/cu121") + torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.1.0 torchvision==0.16.0 --extra-index-url {torch_index_url}") requirements_file = os.environ.get('REQS_FILE', "requirements_versions.txt") xformers_package = os.environ.get('XFORMERS_PACKAGE', 'xformers==0.0.20') -- cgit v1.2.1 From d4d3134f6d2d232c7bcfa80900a362921e644976 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sat, 28 Oct 2023 15:24:26 +0800 Subject: ManualCast for 10/16 series gpu --- modules/devices.py | 57 +++++++++++++++++++++++++++++++++++++++++++++------ modules/processing.py | 2 +- modules/sd_models.py | 21 +++++++++++-------- 3 files changed, 64 insertions(+), 16 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index 0cd2b55d..c05f2b35 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -16,6 +16,23 @@ def has_mps() -> bool: return mac_specific.has_mps +def cuda_no_autocast(device_id=None) -> bool: + if device_id is None: + device_id = get_cuda_device_id() + return ( + torch.cuda.get_device_capability(device_id) == (7, 5) + and torch.cuda.get_device_name(device_id).startswith("NVIDIA GeForce GTX 16") + ) + + +def get_cuda_device_id(): + return ( + int(shared.cmd_opts.device_id) + if shared.cmd_opts.device_id is not None and shared.cmd_opts.device_id.isdigit() + else 0 + ) or torch.cuda.current_device() + + def get_cuda_device_string(): if shared.cmd_opts.device_id is not None: return f"cuda:{shared.cmd_opts.device_id}" @@ -60,8 +77,7 @@ def enable_tf32(): # enabling benchmark option seems to enable a range of cards to do fp16 when they otherwise can't # see https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/4407 - device_id = (int(shared.cmd_opts.device_id) if shared.cmd_opts.device_id is not None and shared.cmd_opts.device_id.isdigit() else 0) or torch.cuda.current_device() - if torch.cuda.get_device_capability(device_id) == (7, 5) and torch.cuda.get_device_name(device_id).startswith("NVIDIA GeForce GTX 16"): + if cuda_no_autocast(): torch.backends.cudnn.benchmark = True torch.backends.cuda.matmul.allow_tf32 = True @@ -92,15 +108,44 @@ def cond_cast_float(input): nv_rng = None - - -def autocast(disable=False, unet=False): +patch_module_list = [ + torch.nn.Linear, + torch.nn.Conv2d, + torch.nn.MultiheadAttention, + torch.nn.GroupNorm, + torch.nn.LayerNorm, +] + +@contextlib.contextmanager +def manual_autocast(): + def manual_cast_forward(self, *args, **kwargs): + org_dtype = next(self.parameters()).dtype + self.to(dtype) + result = self.org_forward(*args, **kwargs) + self.to(org_dtype) + return result + for module_type in patch_module_list: + org_forward = module_type.forward + module_type.forward = manual_cast_forward + module_type.org_forward = org_forward + try: + yield None + finally: + for module_type in patch_module_list: + module_type.forward = module_type.org_forward + + +def autocast(disable=False): + print(fp8, dtype, shared.cmd_opts.precision, device) if disable: return contextlib.nullcontext() - if unet and fp8 and device==cpu: + if fp8 and device==cpu: return torch.autocast("cpu", dtype=torch.bfloat16, enabled=True) + if fp8 and (dtype == torch.float32 or shared.cmd_opts.precision == "full" or cuda_no_autocast()): + return manual_autocast() + if dtype == torch.float32 or shared.cmd_opts.precision == "full": return contextlib.nullcontext() diff --git a/modules/processing.py b/modules/processing.py index 2df8a7ea..40598f5c 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -865,7 +865,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.n_iter > 1: shared.state.job = f"Batch {n+1} out of {p.n_iter}" - with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(unet=True): + with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(): samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts) if getattr(samples_ddim, 'already_decoded', False): diff --git a/modules/sd_models.py b/modules/sd_models.py index ccb6afd2..31bcb913 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -403,23 +403,26 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer if enable_fp8: devices.fp8 = True + if model.is_sdxl: + cond_stage = model.conditioner + else: + cond_stage = model.cond_stage_model + + for module in cond_stage.modules(): + if isinstance(module, torch.nn.Linear): + module.to(torch.float8_e4m3fn) + if devices.device == devices.cpu: for module in model.model.diffusion_model.modules(): if isinstance(module, torch.nn.Conv2d): module.to(torch.float8_e4m3fn) elif isinstance(module, torch.nn.Linear): module.to(torch.float8_e4m3fn) - timer.record("apply fp8 unet for cpu") else: - if model.is_sdxl: - cond_stage = model.conditioner - else: - cond_stage = model.cond_stage_model - for module in cond_stage.modules(): - if isinstance(module, torch.nn.Linear): - module.to(torch.float8_e4m3fn) model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) - timer.record("apply fp8 unet") + timer.record("apply fp8") + else: + devices.fp8 = False devices.unet_needs_upcast = shared.cmd_opts.upcast_sampling and devices.dtype == torch.float16 and devices.dtype_unet == torch.float16 -- cgit v1.2.1 From ddc2a3499b8cd120b4a42358bcd33137ce1d1e75 Mon Sep 17 00:00:00 2001 From: KohakuBlueleaf Date: Sat, 28 Oct 2023 16:52:35 +0800 Subject: Add MPS manual cast --- modules/devices.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index c05f2b35..d7c905c2 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -121,6 +121,8 @@ def manual_autocast(): def manual_cast_forward(self, *args, **kwargs): org_dtype = next(self.parameters()).dtype self.to(dtype) + args = [arg.to(dtype) if isinstance(arg, torch.Tensor) else arg for arg in args] + kwargs = {k: v.to(dtype) if isinstance(v, torch.Tensor) else v for k, v in kwargs.items()} result = self.org_forward(*args, **kwargs) self.to(org_dtype) return result @@ -136,7 +138,6 @@ def manual_autocast(): def autocast(disable=False): - print(fp8, dtype, shared.cmd_opts.precision, device) if disable: return contextlib.nullcontext() @@ -146,6 +147,9 @@ def autocast(disable=False): if fp8 and (dtype == torch.float32 or shared.cmd_opts.precision == "full" or cuda_no_autocast()): return manual_autocast() + if has_mps() and shared.cmd_opts.precision != "full": + return manual_autocast() + if dtype == torch.float32 or shared.cmd_opts.precision == "full": return contextlib.nullcontext() -- cgit v1.2.1 From f2b83517aa49268219706f9718d734c6f242fd93 Mon Sep 17 00:00:00 2001 From: Nick Harrison <42382362+nickpharrison@users.noreply.github.com> Date: Sun, 29 Oct 2023 15:40:13 +0000 Subject: Add new arguments to known command prompts --- modules/cmd_args.py | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'modules') diff --git a/modules/cmd_args.py b/modules/cmd_args.py index 4e602a84..c6d4b612 100644 --- a/modules/cmd_args.py +++ b/modules/cmd_args.py @@ -76,7 +76,9 @@ parser.add_argument("--port", type=int, help="launch gradio with given server po parser.add_argument("--show-negative-prompt", action='store_true', help="does not do anything", default=False) parser.add_argument("--ui-config-file", type=str, help="filename to use for ui configuration", default=os.path.join(data_path, 'ui-config.json')) parser.add_argument("--hide-ui-dir-config", action='store_true', help="hide directory configuration from webui", default=False) -parser.add_argument("--freeze-settings", action='store_true', help="disable editing settings", default=False) +parser.add_argument("--freeze-settings", action='store_true', help="disable editing of all settings globally", default=False) +parser.add_argument("--freeze-settings-in-sections", type=str, help='disable editing settings in specific sections of the settings page by specifying a comma-delimited list such like "saving-images,upscaling". The list of setting names can be found in the modules/shared_options.py file', default=None) +parser.add_argument("--freeze-specific-settings", type=str, help='disable editing of individual settings by specifying a comma-delimited list like "samples_save,samples_format". The list of setting names can be found in the config.json file', default=None) parser.add_argument("--ui-settings-file", type=str, help="filename to use for ui settings", default=os.path.join(data_path, 'config.json')) parser.add_argument("--gradio-debug", action='store_true', help="launch gradio with --debug option") parser.add_argument("--gradio-auth", type=str, help='set gradio authentication like "username:password"; or comma-delimit multiple like "u1:p1,u2:p2,u3:p3"', default=None) @@ -90,7 +92,7 @@ parser.add_argument("--autolaunch", action='store_true', help="open the webui UR parser.add_argument("--theme", type=str, help="launches the UI with light or dark theme", default=None) parser.add_argument("--use-textbox-seed", action='store_true', help="use textbox for seeds in UI (no up/down, but possible to input long seeds)", default=False) parser.add_argument("--disable-console-progressbars", action='store_true', help="do not output progressbars to console", default=False) -parser.add_argument("--enable-console-prompts", action='store_true', help="does not do anything", default=False) # Legacy compatibility, use as default value shared.opts.enable_console_prompts +parser.add_argument("--enable-console-prompts", action='store_true', help="print prompts to console when generating with txt2img and img2img", default=False) parser.add_argument('--vae-path', type=str, help='Checkpoint to use as VAE; setting this argument disables all settings related to VAE', default=None) parser.add_argument("--disable-safe-unpickle", action='store_true', help="disable checking pytorch models for malicious code", default=False) parser.add_argument("--api", action='store_true', help="use api=True to launch the API together with the webui (use --nowebui instead for only the API)") @@ -112,9 +114,8 @@ parser.add_argument("--skip-version-check", action='store_true', help="Do not ch parser.add_argument("--no-hashing", action='store_true', help="disable sha256 hashing of checkpoints to help loading performance", default=False) parser.add_argument("--no-download-sd-model", action='store_true', help="don't download SD1.5 model even if no model is found in --ckpt-dir", default=False) parser.add_argument('--subpath', type=str, help='customize the subpath for gradio, use with reverse proxy') -parser.add_argument('--add-stop-route', action='store_true', help='does not do anything') +parser.add_argument('--add-stop-route', action='store_true', help='add /_stop route to stop server') parser.add_argument('--api-server-stop', action='store_true', help='enable server stop/restart/kill via api') parser.add_argument('--timeout-keep-alive', type=int, default=30, help='set timeout_keep_alive for uvicorn') parser.add_argument("--disable-all-extensions", action='store_true', help="prevent all extensions from running regardless of any other settings", default=False) -parser.add_argument("--disable-extra-extensions", action='store_true', help="prevent all extensions except built-in from running regardless of any other settings", default=False) -parser.add_argument("--skip-load-model-at-start", action='store_true', help="if load a model at web start, only take effect when --nowebui", ) +parser.add_argument("--disable-extra-extensions", action='store_true', help=" prevent all extensions except built-in from running regardless of any other settings", default=False) -- cgit v1.2.1 From 844c23975f369f85c7179379ec419e1bd067de18 Mon Sep 17 00:00:00 2001 From: Nick Harrison <42382362+nickpharrison@users.noreply.github.com> Date: Sun, 29 Oct 2023 15:40:58 +0000 Subject: Add assertions for checking additional settings freezing parameters --- modules/options.py | 23 +++++++++++++++++++---- 1 file changed, 19 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/options.py b/modules/options.py index ab40aff7..1ac32d95 100644 --- a/modules/options.py +++ b/modules/options.py @@ -85,18 +85,35 @@ class Options: if self.data is not None: if key in self.data or key in self.data_labels: + + # Check that settings aren't globally frozen assert not cmd_opts.freeze_settings, "changing settings is disabled" + # Get the info related to the setting being changed info = self.data_labels.get(key, None) if info.do_not_save: return + # Restrict component arguments comp_args = info.component_args if info else None if isinstance(comp_args, dict) and comp_args.get('visible', True) is False: - raise RuntimeError(f"not possible to set {key} because it is restricted") + raise RuntimeError(f"not possible to set '{key}' because it is restricted") + # Check that this section isn't frozen + if cmd_opts.freeze_settings_in_sections is not None: + frozen_sections = list(map(str.strip, cmd_opts.freeze_settings_in_sections.split(','))) # Trim whitespace from section names + section_key = info.section[0] + section_name = info.section[1] + assert section_key not in frozen_sections, f"not possible to set '{key}' because settings in section '{section_name}' ({section_key}) are frozen with --freeze-settings-in-sections" + + # Check that this section of the settings isn't frozen + if cmd_opts.freeze_specific_settings is not None: + frozen_keys = list(map(str.strip, cmd_opts.freeze_specific_settings.split(','))) # Trim whitespace from setting keys + assert key not in frozen_keys, f"not possible to set '{key}' because this setting is frozen with --freeze-specific-settings" + + # Check shorthand option which disables editing options in "saving-paths" if cmd_opts.hide_ui_dir_config and key in self.restricted_opts: - raise RuntimeError(f"not possible to set {key} because it is restricted") + raise RuntimeError(f"not possible to set '{key}' because it is restricted with --hide_ui_dir_config") self.data[key] = value return @@ -210,8 +227,6 @@ class Options: def add_option(self, key, info): self.data_labels[key] = info - if key not in self.data: - self.data[key] = info.default def reorder(self): """reorder settings so that all items related to section always go together""" -- cgit v1.2.1 From be31e7e71a08dc27543d31aa6e6532463ccbf20f Mon Sep 17 00:00:00 2001 From: Nick Harrison <42382362+nickpharrison@users.noreply.github.com> Date: Sun, 29 Oct 2023 16:05:01 +0000 Subject: Remove blank line whitespace --- modules/options.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/options.py b/modules/options.py index 1ac32d95..e270a42d 100644 --- a/modules/options.py +++ b/modules/options.py @@ -85,7 +85,7 @@ class Options: if self.data is not None: if key in self.data or key in self.data_labels: - + # Check that settings aren't globally frozen assert not cmd_opts.freeze_settings, "changing settings is disabled" -- cgit v1.2.1 From 598da5cd4928618b166886d3485ce30ce3a43490 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sun, 19 Nov 2023 15:50:06 +0800 Subject: Use options instead of cmd_args --- modules/cmd_args.py | 2 -- modules/devices.py | 25 ++++++++++--------- modules/initialize_util.py | 1 + modules/sd_models.py | 61 ++++++++++++++++++++++++---------------------- modules/shared_options.py | 1 + 5 files changed, 48 insertions(+), 42 deletions(-) (limited to 'modules') diff --git a/modules/cmd_args.py b/modules/cmd_args.py index 088d5dea..a9fb9bfa 100644 --- a/modules/cmd_args.py +++ b/modules/cmd_args.py @@ -118,5 +118,3 @@ parser.add_argument('--timeout-keep-alive', type=int, default=30, help='set time parser.add_argument("--disable-all-extensions", action='store_true', help="prevent all extensions from running regardless of any other settings", default=False) parser.add_argument("--disable-extra-extensions", action='store_true', help="prevent all extensions except built-in from running regardless of any other settings", default=False) parser.add_argument("--skip-load-model-at-start", action='store_true', help="if load a model at web start, only take effect when --nowebui", ) -parser.add_argument("--opt-unet-fp8-storage", action='store_true', help="use fp8 for SD UNet to save vram", default=False) -parser.add_argument("--opt-unet-fp8-storage-xl", action='store_true', help="use fp8 for SD UNet to save vram", default=False) diff --git a/modules/devices.py b/modules/devices.py index d7c905c2..03e7bdb7 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -20,15 +20,15 @@ def cuda_no_autocast(device_id=None) -> bool: if device_id is None: device_id = get_cuda_device_id() return ( - torch.cuda.get_device_capability(device_id) == (7, 5) + torch.cuda.get_device_capability(device_id) == (7, 5) and torch.cuda.get_device_name(device_id).startswith("NVIDIA GeForce GTX 16") ) def get_cuda_device_id(): return ( - int(shared.cmd_opts.device_id) - if shared.cmd_opts.device_id is not None and shared.cmd_opts.device_id.isdigit() + int(shared.cmd_opts.device_id) + if shared.cmd_opts.device_id is not None and shared.cmd_opts.device_id.isdigit() else 0 ) or torch.cuda.current_device() @@ -116,16 +116,19 @@ patch_module_list = [ torch.nn.LayerNorm, ] + +def manual_cast_forward(self, *args, **kwargs): + org_dtype = next(self.parameters()).dtype + self.to(dtype) + args = [arg.to(dtype) if isinstance(arg, torch.Tensor) else arg for arg in args] + kwargs = {k: v.to(dtype) if isinstance(v, torch.Tensor) else v for k, v in kwargs.items()} + result = self.org_forward(*args, **kwargs) + self.to(org_dtype) + return result + + @contextlib.contextmanager def manual_autocast(): - def manual_cast_forward(self, *args, **kwargs): - org_dtype = next(self.parameters()).dtype - self.to(dtype) - args = [arg.to(dtype) if isinstance(arg, torch.Tensor) else arg for arg in args] - kwargs = {k: v.to(dtype) if isinstance(v, torch.Tensor) else v for k, v in kwargs.items()} - result = self.org_forward(*args, **kwargs) - self.to(org_dtype) - return result for module_type in patch_module_list: org_forward = module_type.forward module_type.forward = manual_cast_forward diff --git a/modules/initialize_util.py b/modules/initialize_util.py index 2e9b6d89..1b11ead6 100644 --- a/modules/initialize_util.py +++ b/modules/initialize_util.py @@ -177,6 +177,7 @@ def configure_opts_onchange(): shared.opts.onchange("temp_dir", ui_tempdir.on_tmpdir_changed) shared.opts.onchange("gradio_theme", shared.reload_gradio_theme) shared.opts.onchange("cross_attention_optimization", wrap_queued_call(lambda: sd_hijack.model_hijack.redo_hijack(shared.sd_model)), call=False) + shared.opts.onchange("fp8_storage", wrap_queued_call(lambda: sd_models.reload_model_weights()), call=False) startup_timer.record("opts onchange") diff --git a/modules/sd_models.py b/modules/sd_models.py index a6c8b2fa..eb491434 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -339,10 +339,28 @@ class SkipWritingToConfig: SkipWritingToConfig.skip = self.previous +def check_fp8(model): + if model is None: + return None + if devices.get_optimal_device_name() == "mps": + enable_fp8 = False + elif shared.opts.fp8_storage == "Enable": + enable_fp8 = True + elif getattr(model, "is_sdxl", False) and shared.opts.fp8_storage == "Enable for SDXL": + enable_fp8 = True + else: + enable_fp8 = False + return enable_fp8 + + def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer): sd_model_hash = checkpoint_info.calculate_shorthash() timer.record("calculate hash") + if not check_fp8(model) and devices.fp8: + # prevent model to load state dict in fp8 + model.half() + if not SkipWritingToConfig.skip: shared.opts.data["sd_model_checkpoint"] = checkpoint_info.title @@ -395,34 +413,16 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer devices.dtype_unet = torch.float16 timer.record("apply half()") - if devices.get_optimal_device_name() == "mps": - enable_fp8 = False - elif shared.cmd_opts.opt_unet_fp8_storage: - enable_fp8 = True - elif model.is_sdxl and shared.cmd_opts.opt_unet_fp8_storage_xl: - enable_fp8 = True - else: - enable_fp8 = False - - if enable_fp8: + if check_fp8(model): devices.fp8 = True - if model.is_sdxl: - cond_stage = model.conditioner - else: - cond_stage = model.cond_stage_model - - for module in cond_stage.modules(): - if isinstance(module, torch.nn.Linear): + first_stage = model.first_stage_model + model.first_stage_model = None + for module in model.modules(): + if isinstance(module, torch.nn.Conv2d): module.to(torch.float8_e4m3fn) - - if devices.device == devices.cpu: - for module in model.model.diffusion_model.modules(): - if isinstance(module, torch.nn.Conv2d): - module.to(torch.float8_e4m3fn) - elif isinstance(module, torch.nn.Linear): - module.to(torch.float8_e4m3fn) - else: - model.model.diffusion_model = model.model.diffusion_model.to(torch.float8_e4m3fn) + elif isinstance(module, torch.nn.Linear): + module.to(torch.float8_e4m3fn) + model.first_stage_model = first_stage timer.record("apply fp8") else: devices.fp8 = False @@ -769,7 +769,7 @@ def reuse_model_from_already_loaded(sd_model, checkpoint_info, timer): return None -def reload_model_weights(sd_model=None, info=None): +def reload_model_weights(sd_model=None, info=None, forced_reload=False): checkpoint_info = info or select_checkpoint() timer = Timer() @@ -781,11 +781,14 @@ def reload_model_weights(sd_model=None, info=None): current_checkpoint_info = None else: current_checkpoint_info = sd_model.sd_checkpoint_info - if sd_model.sd_model_checkpoint == checkpoint_info.filename: + if check_fp8(sd_model) != devices.fp8: + # load from state dict again to prevent extra numerical errors + forced_reload = True + elif sd_model.sd_model_checkpoint == checkpoint_info.filename: return sd_model sd_model = reuse_model_from_already_loaded(sd_model, checkpoint_info, timer) - if sd_model is not None and sd_model.sd_checkpoint_info.filename == checkpoint_info.filename: + if not forced_reload and sd_model is not None and sd_model.sd_checkpoint_info.filename == checkpoint_info.filename: return sd_model if sd_model is not None: diff --git a/modules/shared_options.py b/modules/shared_options.py index f1003f21..d27f35e9 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -200,6 +200,7 @@ options_templates.update(options_section(('optimizations', "Optimizations"), { "pad_cond_uncond": OptionInfo(False, "Pad prompt/negative prompt to be same length", infotext='Pad conds').info("improves performance when prompt and negative prompt have different lengths; changes seeds"), "persistent_cond_cache": OptionInfo(True, "Persistent cond cache").info("do not recalculate conds from prompts if prompts have not changed since previous calculation"), "batch_cond_uncond": OptionInfo(True, "Batch cond/uncond").info("do both conditional and unconditional denoising in one batch; uses a bit more VRAM during sampling, but improves speed; previously this was controlled by --always-batch-cond-uncond comandline argument"), + "fp8_storage": OptionInfo("Disable", "FP8 weight", gr.Dropdown, {"choices": ["Disable", "Enable for SDXL", "Enable"]}).info("Use FP8 to store Linear/Conv layers' weight. Require pytorch>=2.1.0."), })) options_templates.update(options_section(('compatibility', "Compatibility"), { -- cgit v1.2.1 From 890181e1d456b613bf60f6e8378dc68b39011af9 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sun, 19 Nov 2023 15:54:39 +0800 Subject: Update the xformers/torch versions --- modules/errors.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/errors.py b/modules/errors.py index 8c339464..a3498c11 100644 --- a/modules/errors.py +++ b/modules/errors.py @@ -93,8 +93,8 @@ def check_versions(): import torch import gradio - expected_torch_version = "2.0.0" - expected_xformers_version = "0.0.20" + expected_torch_version = "2.1.0" + expected_xformers_version = "0.0.22.post7" expected_gradio_version = "3.41.2" if version.parse(torch.__version__) < version.parse(expected_torch_version): -- cgit v1.2.1 From f383af2729ec2d1969200218577ab19dd78f7d48 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sun, 19 Nov 2023 15:56:23 +0800 Subject: update xformers/torch versions --- modules/launch_utils.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/launch_utils.py b/modules/launch_utils.py index 636da679..c225bbc1 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -312,7 +312,7 @@ def prepare_environment(): torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.1.0 torchvision==0.16.0 --extra-index-url {torch_index_url}") requirements_file = os.environ.get('REQS_FILE', "requirements_versions.txt") - xformers_package = os.environ.get('XFORMERS_PACKAGE', 'xformers==0.0.20') + xformers_package = os.environ.get('XFORMERS_PACKAGE', 'xformers==0.0.22.post7') clip_package = os.environ.get('CLIP_PACKAGE', "https://github.com/openai/CLIP/archive/d50d76daa670286dd6cacf3bcd80b5e4823fc8e1.zip") openclip_package = os.environ.get('OPENCLIP_PACKAGE', "https://github.com/mlfoundations/open_clip/archive/bb6e834e9c70d9c27d0dc3ecedeebeaeb1ffad6b.zip") -- cgit v1.2.1 From 043d2edcf6a543f236f1f3cb70ac72e7b3b357b6 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sun, 19 Nov 2023 15:56:31 +0800 Subject: Better naming --- modules/devices.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index 03e7bdb7..c19a7f40 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -128,7 +128,7 @@ def manual_cast_forward(self, *args, **kwargs): @contextlib.contextmanager -def manual_autocast(): +def manual_cast(): for module_type in patch_module_list: org_forward = module_type.forward module_type.forward = manual_cast_forward @@ -148,10 +148,10 @@ def autocast(disable=False): return torch.autocast("cpu", dtype=torch.bfloat16, enabled=True) if fp8 and (dtype == torch.float32 or shared.cmd_opts.precision == "full" or cuda_no_autocast()): - return manual_autocast() + return manual_cast() if has_mps() and shared.cmd_opts.precision != "full": - return manual_autocast() + return manual_cast() if dtype == torch.float32 or shared.cmd_opts.precision == "full": return contextlib.nullcontext() -- cgit v1.2.1 From 370a77f8e78e65a8a1339289d684cb43df142f70 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 21 Nov 2023 19:59:34 +0800 Subject: Option for using fp16 weight when apply lora --- modules/initialize_util.py | 1 + modules/sd_models.py | 14 +++++++++++--- modules/shared_options.py | 1 + 3 files changed, 13 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/initialize_util.py b/modules/initialize_util.py index 1b11ead6..7fb1d8d5 100644 --- a/modules/initialize_util.py +++ b/modules/initialize_util.py @@ -178,6 +178,7 @@ def configure_opts_onchange(): shared.opts.onchange("gradio_theme", shared.reload_gradio_theme) shared.opts.onchange("cross_attention_optimization", wrap_queued_call(lambda: sd_hijack.model_hijack.redo_hijack(shared.sd_model)), call=False) shared.opts.onchange("fp8_storage", wrap_queued_call(lambda: sd_models.reload_model_weights()), call=False) + shared.opts.onchange("cache_fp16_weight", wrap_queued_call(lambda: sd_models.reload_model_weights()), call=False) startup_timer.record("opts onchange") diff --git a/modules/sd_models.py b/modules/sd_models.py index eb491434..0a7777f1 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -413,14 +413,22 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer devices.dtype_unet = torch.float16 timer.record("apply half()") + for module in model.modules(): + if hasattr(module, 'fp16_weight'): + del module.fp16_weight + if hasattr(module, 'fp16_bias'): + del module.fp16_bias + if check_fp8(model): devices.fp8 = True first_stage = model.first_stage_model model.first_stage_model = None for module in model.modules(): - if isinstance(module, torch.nn.Conv2d): - module.to(torch.float8_e4m3fn) - elif isinstance(module, torch.nn.Linear): + if isinstance(module, (torch.nn.Conv2d, torch.nn.Linear)): + if shared.opts.cache_fp16_weight: + module.fp16_weight = module.weight.clone().half() + if module.bias is not None: + module.fp16_bias = module.bias.clone().half() module.to(torch.float8_e4m3fn) model.first_stage_model = first_stage timer.record("apply fp8") diff --git a/modules/shared_options.py b/modules/shared_options.py index d27f35e9..eaa9f135 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -201,6 +201,7 @@ options_templates.update(options_section(('optimizations', "Optimizations"), { "persistent_cond_cache": OptionInfo(True, "Persistent cond cache").info("do not recalculate conds from prompts if prompts have not changed since previous calculation"), "batch_cond_uncond": OptionInfo(True, "Batch cond/uncond").info("do both conditional and unconditional denoising in one batch; uses a bit more VRAM during sampling, but improves speed; previously this was controlled by --always-batch-cond-uncond comandline argument"), "fp8_storage": OptionInfo("Disable", "FP8 weight", gr.Dropdown, {"choices": ["Disable", "Enable for SDXL", "Enable"]}).info("Use FP8 to store Linear/Conv layers' weight. Require pytorch>=2.1.0."), + "cache_fp16_weight": OptionInfo(False, "Cache FP16 weight for LoRA").info("Cache fp16 weight when enabling FP8, will increase the quality of LoRA. Use more system ram."), })) options_templates.update(options_section(('compatibility', "Compatibility"), { -- cgit v1.2.1 From f5d719d1f1baa775d838aa75d9af1971bcc78e8f Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Wed, 22 Nov 2023 01:45:56 +0800 Subject: Add forced reload for fp16 cache --- modules/initialize_util.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/initialize_util.py b/modules/initialize_util.py index 7fb1d8d5..b6767138 100644 --- a/modules/initialize_util.py +++ b/modules/initialize_util.py @@ -178,7 +178,7 @@ def configure_opts_onchange(): shared.opts.onchange("gradio_theme", shared.reload_gradio_theme) shared.opts.onchange("cross_attention_optimization", wrap_queued_call(lambda: sd_hijack.model_hijack.redo_hijack(shared.sd_model)), call=False) shared.opts.onchange("fp8_storage", wrap_queued_call(lambda: sd_models.reload_model_weights()), call=False) - shared.opts.onchange("cache_fp16_weight", wrap_queued_call(lambda: sd_models.reload_model_weights()), call=False) + shared.opts.onchange("cache_fp16_weight", wrap_queued_call(lambda: sd_models.reload_model_weights(forced_reload=True)), call=False) startup_timer.record("opts onchange") -- cgit v1.2.1 From 40ac134c553ac824d4a96666bba14d550300daa5 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sat, 25 Nov 2023 12:35:09 +0800 Subject: Fix pre-fp8 --- modules/sd_models.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 0a7777f1..90437c87 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -357,7 +357,7 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer sd_model_hash = checkpoint_info.calculate_shorthash() timer.record("calculate hash") - if not check_fp8(model) and devices.fp8: + if devices.fp8: # prevent model to load state dict in fp8 model.half() -- cgit v1.2.1 From 29f04149b60bcf6e8e2b41a161d6cc7e8981710f Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 26 Nov 2023 12:07:33 +0300 Subject: update torch to 2.1.0 --- modules/errors.py | 4 ++-- modules/launch_utils.py | 6 +++--- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'modules') diff --git a/modules/errors.py b/modules/errors.py index eb234a83..c534a5d6 100644 --- a/modules/errors.py +++ b/modules/errors.py @@ -107,8 +107,8 @@ def check_versions(): import torch import gradio - expected_torch_version = "2.0.0" - expected_xformers_version = "0.0.20" + expected_torch_version = "2.1.0" + expected_xformers_version = "0.0.22.post7" expected_gradio_version = "3.41.2" if version.parse(torch.__version__) < version.parse(expected_torch_version): diff --git a/modules/launch_utils.py b/modules/launch_utils.py index 264ec9ca..1f2b6c5e 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -308,11 +308,11 @@ def requirements_met(requirements_file): def prepare_environment(): - torch_index_url = os.environ.get('TORCH_INDEX_URL', "https://download.pytorch.org/whl/cu118") - torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.0.1 torchvision==0.15.2 --extra-index-url {torch_index_url}") + torch_index_url = os.environ.get('TORCH_INDEX_URL', "https://download.pytorch.org/whl/cu121") + torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.1.0 torchvision==0.16.0 --extra-index-url {torch_index_url}") requirements_file = os.environ.get('REQS_FILE', "requirements_versions.txt") - xformers_package = os.environ.get('XFORMERS_PACKAGE', 'xformers==0.0.20') + xformers_package = os.environ.get('XFORMERS_PACKAGE', 'xformers==0.0.22.post7') clip_package = os.environ.get('CLIP_PACKAGE', "https://github.com/openai/CLIP/archive/d50d76daa670286dd6cacf3bcd80b5e4823fc8e1.zip") openclip_package = os.environ.get('OPENCLIP_PACKAGE', "https://github.com/mlfoundations/open_clip/archive/bb6e834e9c70d9c27d0dc3ecedeebeaeb1ffad6b.zip") -- cgit v1.2.1 From dec791d35ddcd02ca33563d3d0355e05e45de8ad Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Tue, 28 Nov 2023 15:05:01 -0700 Subject: Removed code which forces the inpainting mask to be 0 or 1. Now fractional values (e.g. 0.5) are accepted. --- modules/processing.py | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index e124e7f0..317458f5 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -83,7 +83,7 @@ def apply_overlay(image, paste_loc, index, overlays): def create_binary_mask(image): if image.mode == 'RGBA' and image.getextrema()[-1] != (255, 255): - image = image.split()[-1].convert("L").point(lambda x: 255 if x > 128 else 0) + image = image.split()[-1].convert("L") else: image = image.convert('L') return image @@ -319,9 +319,6 @@ class StableDiffusionProcessing: conditioning_mask = np.array(image_mask.convert("L")) conditioning_mask = conditioning_mask.astype(np.float32) / 255.0 conditioning_mask = torch.from_numpy(conditioning_mask[None, None]) - - # Inpainting model uses a discretized mask as input, so we round to either 1.0 or 0.0 - conditioning_mask = torch.round(conditioning_mask) else: conditioning_mask = source_image.new_ones(1, 1, *source_image.shape[-2:]) @@ -1504,7 +1501,6 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): latmask = init_mask.convert('RGB').resize((self.init_latent.shape[3], self.init_latent.shape[2])) latmask = np.moveaxis(np.array(latmask, dtype=np.float32), 2, 0) / 255 latmask = latmask[0] - latmask = np.around(latmask) latmask = np.tile(latmask[None], (4, 1, 1)) self.mask = torch.asarray(1.0 - latmask).to(shared.device).type(self.sd_model.dtype) -- cgit v1.2.1 From bbba133f054706c3668b7d03b0e6d0afc15705db Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Tue, 28 Nov 2023 15:09:43 -0700 Subject: Removed conflicting step that replaces the softly inpainted latents with a naive blend with the original latents. --- modules/processing.py | 3 --- 1 file changed, 3 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 317458f5..ae894f1a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1523,9 +1523,6 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): samples = self.sampler.sample_img2img(self, self.init_latent, x, conditioning, unconditional_conditioning, image_conditioning=self.image_conditioning) - if self.mask is not None: - samples = samples * self.nmask + self.init_latent * self.mask - del x devices.torch_gc() -- cgit v1.2.1 From e715e46b6aa7f2e5e147cfa1fa2f49b1d926a074 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Tue, 28 Nov 2023 16:10:22 -0700 Subject: Implements "scheduling" for blending of the original latents and a latent blending formula that preserves details in blend transition areas. --- modules/sd_samplers_cfg_denoiser.py | 61 +++++++++++++++++++++++++++++++++++-- 1 file changed, 59 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index b8101d38..c4d6fda6 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -43,6 +43,9 @@ class CFGDenoiser(torch.nn.Module): self.model_wrap = None self.mask = None self.nmask = None + self.mask_blend_power = 1 + self.mask_blend_scale = 1 + self.mask_blend_offset = 0 self.init_latent = None self.steps = None """number of steps as specified by user in UI""" @@ -56,6 +59,9 @@ class CFGDenoiser(torch.nn.Module): self.sampler = sampler self.model_wrap = None self.p = None + + # NOTE: masking before denoising can cause the original latents to be oversmoothed + # as the original latents do not have noise self.mask_before_denoising = False @property @@ -89,6 +95,55 @@ class CFGDenoiser(torch.nn.Module): self.sampler.sampler_extra_args['uncond'] = uc def forward(self, x, sigma, uncond, cond, cond_scale, s_min_uncond, image_cond): + def latent_blend(a, b, t): + """ + Interpolates two latent image representations according to the parameter t, + where the interpolated vectors' magnitudes are also interpolated separately. + The "detail_preservation" factor biases the magnitude interpolation towards + the larger of the two magnitudes. + """ + # Record the original latent vector magnitudes. + # We bring them to a power so that larger magnitudes are favored over smaller ones. + # 64-bit operations are used here to allow large exponents. + detail_preservation = 32 + a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64) ** detail_preservation + b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64) ** detail_preservation + + one_minus_t = 1 - t + + # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). + interp_magnitude = (a_magnitude * one_minus_t + b_magnitude * t) ** (1 / detail_preservation) + + # Linearly interpolate the image vectors. + image_interp = a * one_minus_t + b * t + + # Calculate the magnitude of the interpolated vectors. (We will remove this magnitude.) + # 64-bit operations are used here to allow large exponents. + image_interp_magnitude = torch.norm(image_interp, p=2, dim=1).to(torch.float64) + 0.0001 + + # Change the linearly interpolated image vectors' magnitudes to the value we want. + # This is the last 64-bit operation. + image_interp *= (interp_magnitude / image_interp_magnitude).to(image_interp.dtype) + + return image_interp + + def get_modified_nmask(nmask, _sigma): + """ + Converts a negative mask representing the transparency of the original latent vectors being overlayed + to a mask that is scaled according to the denoising strength for this step. + + Where: + 0 = fully opaque, infinite density, fully masked + 1 = fully transparent, zero density, fully unmasked + + We bring this transparency to a power, as this allows one to simulate N number of blending operations + where N can be any positive real value. Using this one can control the balance of influence between + the denoiser and the original latents according to the sigma value. + + NOTE: "mask" is not used + """ + return torch.pow(nmask, (_sigma ** self.mask_blend_power) * self.mask_blend_scale + self.mask_blend_offset) + if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException @@ -105,8 +160,9 @@ class CFGDenoiser(torch.nn.Module): assert not is_edit_model or all(len(conds) == 1 for conds in conds_list), "AND is not supported for InstructPix2Pix checkpoint (unless using Image CFG scale = 1.0)" + # Blend in the original latents (before) if self.mask_before_denoising and self.mask is not None: - x = self.init_latent * self.mask + self.nmask * x + x = latent_blend(self.init_latent, x, get_modified_nmask(self.nmask, sigma)) batch_size = len(conds_list) repeats = [len(conds_list[i]) for i in range(batch_size)] @@ -207,8 +263,9 @@ class CFGDenoiser(torch.nn.Module): else: denoised = self.combine_denoised(x_out, conds_list, uncond, cond_scale) + # Blend in the original latents (after) if not self.mask_before_denoising and self.mask is not None: - denoised = self.init_latent * self.mask + self.nmask * denoised + denoised = latent_blend(self.init_latent, denoised, get_modified_nmask(self.nmask, sigma)) self.sampler.last_latent = self.get_pred_x0(torch.cat([x_in[i:i + 1] for i in denoised_image_indexes]), torch.cat([x_out[i:i + 1] for i in denoised_image_indexes]), sigma) -- cgit v1.2.1 From a6e584645305c0a91a3d46f73546e191b249210f Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Tue, 28 Nov 2023 16:13:42 -0700 Subject: Nerfs the aggressive post-processing step of overlaying the original image. --- modules/processing.py | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index ae894f1a..12e08e87 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1412,7 +1412,12 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image_mask = Image.fromarray(np_mask) if self.inpaint_full_res: - self.mask_for_overlay = image_mask + np_mask = np.array(image_mask).astype(np.float32) + np_mask /= 255 + np_mask = 1-pow(1-np_mask, 100) + np_mask *= 255 + np_mask = np.clip(np_mask, 0, 255).astype(np.uint8) + self.mask_for_overlay = Image.fromarray(np_mask) mask = image_mask.convert('L') crop_region = masking.get_crop_region(np.array(mask), self.inpaint_full_res_padding) crop_region = masking.expand_crop_region(crop_region, self.width, self.height, mask.width, mask.height) @@ -1423,8 +1428,11 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): self.paste_to = (x1, y1, x2-x1, y2-y1) else: image_mask = images.resize_image(self.resize_mode, image_mask, self.width, self.height) - np_mask = np.array(image_mask) - np_mask = np.clip((np_mask.astype(np.float32)) * 2, 0, 255).astype(np.uint8) + np_mask = np.array(image_mask).astype(np.float32) + np_mask /= 255 + np_mask = 1-pow(1-np_mask, 100) + np_mask *= 255 + np_mask = np.clip(np_mask, 0, 255).astype(np.uint8) self.mask_for_overlay = Image.fromarray(np_mask) self.overlay_images = [] -- cgit v1.2.1 From debf836fcc8d9becc3da8b1a29e33f40b0d9ef3e Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Tue, 28 Nov 2023 16:15:36 -0700 Subject: Added UI elements to control blending parameters. --- modules/img2img.py | 48 ++++++++++++++++++++++++++++++++++++++++++- modules/processing.py | 3 +++ modules/sd_samplers_common.py | 3 +++ modules/ui.py | 9 ++++++++ 4 files changed, 62 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/img2img.py b/modules/img2img.py index 1519e132..240d0588 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -116,7 +116,47 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=Fal process_images(p) -def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_styles, init_img, sketch, init_img_with_mask, inpaint_color_sketch, inpaint_color_sketch_orig, init_img_inpaint, init_mask_inpaint, steps: int, sampler_name: str, mask_blur: int, mask_alpha: float, inpainting_fill: int, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, denoising_strength: float, selected_scale_tab: int, height: int, width: int, scale_by: float, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, override_settings_texts, img2img_batch_use_png_info: bool, img2img_batch_png_info_props: list, img2img_batch_png_info_dir: str, request: gr.Request, *args): +def img2img(id_task: str, + mode: int, + prompt: str, + negative_prompt: str, + prompt_styles, + init_img, + sketch, + init_img_with_mask, + inpaint_color_sketch, + inpaint_color_sketch_orig, + init_img_inpaint, + init_mask_inpaint, + steps: int, + sampler_name: str, + mask_blur: int, + mask_alpha: float, + mask_blend_power: float, + mask_blend_scale: float, + mask_blend_offset: float, + inpainting_fill: int, + n_iter: int, + batch_size: int, + cfg_scale: float, + image_cfg_scale: float, + denoising_strength: float, + selected_scale_tab: int, + height: int, + width: int, + scale_by: float, + resize_mode: int, + inpaint_full_res: bool, + inpaint_full_res_padding: int, + inpainting_mask_invert: int, + img2img_batch_input_dir: str, + img2img_batch_output_dir: str, + img2img_batch_inpaint_mask_dir: str, + override_settings_texts, + img2img_batch_use_png_info: bool, + img2img_batch_png_info_props: list, + img2img_batch_png_info_dir: str, + request: gr.Request, *args): override_settings = create_override_settings_dict(override_settings_texts) is_batch = mode == 5 @@ -174,6 +214,9 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s init_images=[image], mask=mask, mask_blur=mask_blur, + mask_blend_power=mask_blend_power, + mask_blend_scale=mask_blend_scale, + mask_blend_offset=mask_blend_offset, inpainting_fill=inpainting_fill, resize_mode=resize_mode, denoising_strength=denoising_strength, @@ -194,6 +237,9 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s if mask: p.extra_generation_params["Mask blur"] = mask_blur + p.extra_generation_params["Mask blend power"] = mask_blend_power + p.extra_generation_params["Mask blend scale"] = mask_blend_scale + p.extra_generation_params["Mask blend offset"] = mask_blend_offset with closing(p): if is_batch: diff --git a/modules/processing.py b/modules/processing.py index 12e08e87..da4d6fda 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1349,6 +1349,9 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): mask_blur_x: int = 4 mask_blur_y: int = 4 mask_blur: int = None + mask_blend_power: float = 1 + mask_blend_scale: float = 1 + mask_blend_offset: float = 0 inpainting_fill: int = 0 inpaint_full_res: bool = True inpaint_full_res_padding: int = 0 diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 58efcad2..8904da2f 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -277,6 +277,9 @@ class Sampler: self.model_wrap_cfg.p = p self.model_wrap_cfg.mask = p.mask if hasattr(p, 'mask') else None self.model_wrap_cfg.nmask = p.nmask if hasattr(p, 'nmask') else None + self.model_wrap_cfg.mask_blend_power = p.mask_blend_power if hasattr(p, 'mask_blend_power') else None + self.model_wrap_cfg.mask_blend_scale = p.mask_blend_scale if hasattr(p, 'mask_blend_scale') else None + self.model_wrap_cfg.mask_blend_offset = p.mask_blend_offset if hasattr(p, 'mask_blend_offset') else None self.model_wrap_cfg.step = 0 self.model_wrap_cfg.image_cfg_scale = getattr(p, 'image_cfg_scale', None) self.eta = p.eta if p.eta is not None else getattr(opts, self.eta_option_field, 0.0) diff --git a/modules/ui.py b/modules/ui.py index 579bab98..86c13086 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -732,6 +732,9 @@ def create_ui(): with FormRow(): mask_blur = gr.Slider(label='Mask blur', minimum=0, maximum=64, step=1, value=4, elem_id="img2img_mask_blur") mask_alpha = gr.Slider(label="Mask transparency", visible=False, elem_id="img2img_mask_alpha") + mask_blend_power = gr.Slider(label='Mask blend power', minimum=0, maximum=8, step=0.1, value=1, elem_id="img2img_mask_blend_power") + mask_blend_scale = gr.Slider(label='Mask blend scale', minimum=0, maximum=8, step=0.1, value=1, elem_id="img2img_mask_blend_scale") + mask_blend_offset = gr.Slider(label='Mask blend offset', minimum=-4, maximum=4, step=0.1, value=0, elem_id="img2img_mask_blend_offset") with FormRow(): inpainting_mask_invert = gr.Radio(label='Mask mode', choices=['Inpaint masked', 'Inpaint not masked'], value='Inpaint masked', type="index", elem_id="img2img_mask_mode") @@ -781,6 +784,9 @@ def create_ui(): sampler_name, mask_blur, mask_alpha, + mask_blend_power, + mask_blend_scale, + mask_blend_offset, inpainting_fill, batch_count, batch_size, @@ -879,6 +885,9 @@ def create_ui(): (toprow.ui_styles.dropdown, lambda d: d["Styles array"] if isinstance(d.get("Styles array"), list) else gr.update()), (denoising_strength, "Denoising strength"), (mask_blur, "Mask blur"), + (mask_blend_power, "Mask blend power"), + (mask_blend_scale, "Mask blend scale"), + (mask_blend_offset, "Mask blend offset"), *scripts.scripts_img2img.infotext_fields ] parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields, override_settings) -- cgit v1.2.1 From c5c7fa06aae1ae9f8b6d29ae2da3874921d4729b Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Tue, 28 Nov 2023 22:35:07 -0700 Subject: Added slider for detail preservation strength, removed largely needless offset parameter, changed labels in UI and for saving to/pasting data from PNG files. --- modules/img2img.py | 10 +++++----- modules/processing.py | 2 +- modules/sd_samplers_cfg_denoiser.py | 11 +++++------ modules/sd_samplers_common.py | 2 +- modules/ui.py | 14 +++++++------- 5 files changed, 19 insertions(+), 20 deletions(-) (limited to 'modules') diff --git a/modules/img2img.py b/modules/img2img.py index 240d0588..023808d6 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -134,7 +134,7 @@ def img2img(id_task: str, mask_alpha: float, mask_blend_power: float, mask_blend_scale: float, - mask_blend_offset: float, + inpaint_detail_preservation: float, inpainting_fill: int, n_iter: int, batch_size: int, @@ -216,7 +216,7 @@ def img2img(id_task: str, mask_blur=mask_blur, mask_blend_power=mask_blend_power, mask_blend_scale=mask_blend_scale, - mask_blend_offset=mask_blend_offset, + inpaint_detail_preservation=inpaint_detail_preservation, inpainting_fill=inpainting_fill, resize_mode=resize_mode, denoising_strength=denoising_strength, @@ -237,9 +237,9 @@ def img2img(id_task: str, if mask: p.extra_generation_params["Mask blur"] = mask_blur - p.extra_generation_params["Mask blend power"] = mask_blend_power - p.extra_generation_params["Mask blend scale"] = mask_blend_scale - p.extra_generation_params["Mask blend offset"] = mask_blend_offset + p.extra_generation_params["Mask blending bias"] = mask_blend_power + p.extra_generation_params["Mask blending preservation"] = mask_blend_scale + p.extra_generation_params["Mask blending detail boost"] = inpaint_detail_preservation with closing(p): if is_batch: diff --git a/modules/processing.py b/modules/processing.py index da4d6fda..361e8b05 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1351,7 +1351,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): mask_blur: int = None mask_blend_power: float = 1 mask_blend_scale: float = 1 - mask_blend_offset: float = 0 + inpaint_detail_preservation: float = 16 inpainting_fill: int = 0 inpaint_full_res: bool = True inpaint_full_res_padding: int = 0 diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index c4d6fda6..598cd487 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -45,7 +45,7 @@ class CFGDenoiser(torch.nn.Module): self.nmask = None self.mask_blend_power = 1 self.mask_blend_scale = 1 - self.mask_blend_offset = 0 + self.inpaint_detail_preservation = 16 self.init_latent = None self.steps = None """number of steps as specified by user in UI""" @@ -105,14 +105,13 @@ class CFGDenoiser(torch.nn.Module): # Record the original latent vector magnitudes. # We bring them to a power so that larger magnitudes are favored over smaller ones. # 64-bit operations are used here to allow large exponents. - detail_preservation = 32 - a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64) ** detail_preservation - b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64) ** detail_preservation + a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64) ** self.inpaint_detail_preservation + b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64) ** self.inpaint_detail_preservation one_minus_t = 1 - t # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). - interp_magnitude = (a_magnitude * one_minus_t + b_magnitude * t) ** (1 / detail_preservation) + interp_magnitude = (a_magnitude * one_minus_t + b_magnitude * t) ** (1 / self.inpaint_detail_preservation) # Linearly interpolate the image vectors. image_interp = a * one_minus_t + b * t @@ -142,7 +141,7 @@ class CFGDenoiser(torch.nn.Module): NOTE: "mask" is not used """ - return torch.pow(nmask, (_sigma ** self.mask_blend_power) * self.mask_blend_scale + self.mask_blend_offset) + return torch.pow(nmask, (_sigma ** self.mask_blend_power) * self.mask_blend_scale) if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 8904da2f..ecd8ab0a 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -279,7 +279,7 @@ class Sampler: self.model_wrap_cfg.nmask = p.nmask if hasattr(p, 'nmask') else None self.model_wrap_cfg.mask_blend_power = p.mask_blend_power if hasattr(p, 'mask_blend_power') else None self.model_wrap_cfg.mask_blend_scale = p.mask_blend_scale if hasattr(p, 'mask_blend_scale') else None - self.model_wrap_cfg.mask_blend_offset = p.mask_blend_offset if hasattr(p, 'mask_blend_offset') else None + self.model_wrap_cfg.inpaint_detail_preservation = p.inpaint_detail_preservation if hasattr(p, 'inpaint_detail_preservation') else None self.model_wrap_cfg.step = 0 self.model_wrap_cfg.image_cfg_scale = getattr(p, 'image_cfg_scale', None) self.eta = p.eta if p.eta is not None else getattr(opts, self.eta_option_field, 0.0) diff --git a/modules/ui.py b/modules/ui.py index 86c13086..f5e20147 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -732,9 +732,9 @@ def create_ui(): with FormRow(): mask_blur = gr.Slider(label='Mask blur', minimum=0, maximum=64, step=1, value=4, elem_id="img2img_mask_blur") mask_alpha = gr.Slider(label="Mask transparency", visible=False, elem_id="img2img_mask_alpha") - mask_blend_power = gr.Slider(label='Mask blend power', minimum=0, maximum=8, step=0.1, value=1, elem_id="img2img_mask_blend_power") - mask_blend_scale = gr.Slider(label='Mask blend scale', minimum=0, maximum=8, step=0.1, value=1, elem_id="img2img_mask_blend_scale") - mask_blend_offset = gr.Slider(label='Mask blend offset', minimum=-4, maximum=4, step=0.1, value=0, elem_id="img2img_mask_blend_offset") + mask_blend_power = gr.Slider(label='Blending bias', minimum=0, maximum=8, step=0.1, value=1, elem_id="img2img_mask_blend_power") + mask_blend_scale = gr.Slider(label='Blending preservation', minimum=0, maximum=8, step=0.05, value=1, elem_id="img2img_mask_blend_scale") + inpaint_detail_preservation = gr.Slider(label='Blending detail boost', minimum=1, maximum=32, step=0.5, value=16, elem_id="img2img_mask_blend_offset") with FormRow(): inpainting_mask_invert = gr.Radio(label='Mask mode', choices=['Inpaint masked', 'Inpaint not masked'], value='Inpaint masked', type="index", elem_id="img2img_mask_mode") @@ -786,7 +786,7 @@ def create_ui(): mask_alpha, mask_blend_power, mask_blend_scale, - mask_blend_offset, + inpaint_detail_preservation, inpainting_fill, batch_count, batch_size, @@ -885,9 +885,9 @@ def create_ui(): (toprow.ui_styles.dropdown, lambda d: d["Styles array"] if isinstance(d.get("Styles array"), list) else gr.update()), (denoising_strength, "Denoising strength"), (mask_blur, "Mask blur"), - (mask_blend_power, "Mask blend power"), - (mask_blend_scale, "Mask blend scale"), - (mask_blend_offset, "Mask blend offset"), + (mask_blend_power, "Mask blending bias"), + (mask_blend_scale, "Mask blending preservation"), + (inpaint_detail_preservation, "Mask blending detail boost"), *scripts.scripts_img2img.infotext_fields ] parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields, override_settings) -- cgit v1.2.1 From 284fd8f415ec70e14ae5de0b7f5ce738007a6b7f Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Tue, 28 Nov 2023 23:03:50 -0700 Subject: Tweaked UI sliders and labels. --- modules/img2img.py | 2 +- modules/ui.py | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/img2img.py b/modules/img2img.py index 023808d6..0ae16365 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -239,7 +239,7 @@ def img2img(id_task: str, p.extra_generation_params["Mask blur"] = mask_blur p.extra_generation_params["Mask blending bias"] = mask_blend_power p.extra_generation_params["Mask blending preservation"] = mask_blend_scale - p.extra_generation_params["Mask blending detail boost"] = inpaint_detail_preservation + p.extra_generation_params["Mask blending contrast boost"] = inpaint_detail_preservation with closing(p): if is_batch: diff --git a/modules/ui.py b/modules/ui.py index f5e20147..3a9038b2 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -733,8 +733,8 @@ def create_ui(): mask_blur = gr.Slider(label='Mask blur', minimum=0, maximum=64, step=1, value=4, elem_id="img2img_mask_blur") mask_alpha = gr.Slider(label="Mask transparency", visible=False, elem_id="img2img_mask_alpha") mask_blend_power = gr.Slider(label='Blending bias', minimum=0, maximum=8, step=0.1, value=1, elem_id="img2img_mask_blend_power") - mask_blend_scale = gr.Slider(label='Blending preservation', minimum=0, maximum=8, step=0.05, value=1, elem_id="img2img_mask_blend_scale") - inpaint_detail_preservation = gr.Slider(label='Blending detail boost', minimum=1, maximum=32, step=0.5, value=16, elem_id="img2img_mask_blend_offset") + mask_blend_scale = gr.Slider(label='Blending preservation', minimum=0, maximum=8, step=0.05, value=0.5, elem_id="img2img_mask_blend_scale") + inpaint_detail_preservation = gr.Slider(label='Blending contrast boost', minimum=1, maximum=32, step=0.5, value=4, elem_id="img2img_mask_blend_offset") with FormRow(): inpainting_mask_invert = gr.Radio(label='Mask mode', choices=['Inpaint masked', 'Inpaint not masked'], value='Inpaint masked', type="index", elem_id="img2img_mask_mode") @@ -887,7 +887,7 @@ def create_ui(): (mask_blur, "Mask blur"), (mask_blend_power, "Mask blending bias"), (mask_blend_scale, "Mask blending preservation"), - (inpaint_detail_preservation, "Mask blending detail boost"), + (inpaint_detail_preservation, "Mask blending contrast boost"), *scripts.scripts_img2img.infotext_fields ] parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields, override_settings) -- cgit v1.2.1 From c7a1ff87207544dd4bcf3aefffa67a4a38678c16 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Tue, 28 Nov 2023 23:31:10 -0700 Subject: Tweaked default values. --- modules/processing.py | 4 ++-- modules/sd_samplers_cfg_denoiser.py | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 361e8b05..92fdebad 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1350,8 +1350,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): mask_blur_y: int = 4 mask_blur: int = None mask_blend_power: float = 1 - mask_blend_scale: float = 1 - inpaint_detail_preservation: float = 16 + mask_blend_scale: float = 0.5 + inpaint_detail_preservation: float = 4 inpainting_fill: int = 0 inpaint_full_res: bool = True inpaint_full_res_padding: int = 0 diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index 598cd487..ceb612d7 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -44,8 +44,8 @@ class CFGDenoiser(torch.nn.Module): self.mask = None self.nmask = None self.mask_blend_power = 1 - self.mask_blend_scale = 1 - self.inpaint_detail_preservation = 16 + self.mask_blend_scale = 0.5 + self.inpaint_detail_preservation = 4 self.init_latent = None self.steps = None """number of steps as specified by user in UI""" -- cgit v1.2.1 From b25c126ccdbc4da22ade46597a9addf808998989 Mon Sep 17 00:00:00 2001 From: drhead <1313496+drhead@users.noreply.github.com> Date: Wed, 29 Nov 2023 17:38:53 -0500 Subject: Protect alphas_cumprod from downcasting --- modules/sd_models.py | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 841402e8..de80a493 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -387,7 +387,11 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer if shared.cmd_opts.upcast_sampling and depth_model: model.depth_model = None + alphas_cumprod = model.alphas_cumprod + model.alphas_cumprod = None model.half() + model.alphas_cumprod = alphas_cumprod + model.alphas_cumprod_original = alphas_cumprod model.first_stage_model = vae if depth_model: model.depth_model = depth_model @@ -642,6 +646,7 @@ def load_model(checkpoint_info=None, already_loaded_state_dict=None): else: weight_dtype_conversion = { 'first_stage_model': None, + 'alphas_cumprod': None, '': torch.float16, } -- cgit v1.2.1 From 588a52891dca4d030ca7028dd9c0b56022a68b57 Mon Sep 17 00:00:00 2001 From: drhead <1313496+drhead@users.noreply.github.com> Date: Wed, 29 Nov 2023 17:40:23 -0500 Subject: Add options for zero terminal SNR --- modules/shared_options.py | 2 ++ 1 file changed, 2 insertions(+) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index 04e68a71..51596777 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -218,6 +218,7 @@ options_templates.update(options_section(('compatibility', "Compatibility", "sd" "dont_fix_second_order_samplers_schedule": OptionInfo(False, "Do not fix prompt schedule for second order samplers."), "hires_fix_use_firstpass_conds": OptionInfo(False, "For hires fix, calculate conds of second pass using extra networks of first pass."), "use_old_scheduling": OptionInfo(False, "Use old prompt editing timelines.", infotext="Old prompt editing timelines").info("For [red:green:N]; old: If N < 1, it's a fraction of steps (and hires fix uses range from 0 to 1), if N >= 1, it's an absolute number of steps; new: If N has a decimal point in it, it's a fraction of steps (and hires fix uses range from 1 to 2), othewrwise it's an absolute number of steps"), + "use_downcasted_alpha_bar": OptionInfo(False, "Downcast model alphas_cumprod to fp16 before sampling. For reproducing old seeds.") })) options_templates.update(options_section(('interrogate', "Interrogate"), { @@ -335,6 +336,7 @@ options_templates.update(options_section(('sampler-params', "Sampler parameters" 'uni_pc_skip_type': OptionInfo("time_uniform", "UniPC skip type", gr.Radio, {"choices": ["time_uniform", "time_quadratic", "logSNR"]}, infotext='UniPC skip type'), 'uni_pc_order': OptionInfo(3, "UniPC order", gr.Slider, {"minimum": 1, "maximum": 50, "step": 1}, infotext='UniPC order').info("must be < sampling steps"), 'uni_pc_lower_order_final': OptionInfo(True, "UniPC lower order final", infotext='UniPC lower order final'), + 'sd_noise_schedule': OptionInfo("Default", "Noise schedule for sampling", gr.Radio, {"choices": ["Default", "Zero Terminal SNR"]}, infotext="Noise schedule for sampling").info("for use with zero terminal SNR trained models") })) options_templates.update(options_section(('postprocessing', "Postprocessing", "postprocessing"), { -- cgit v1.2.1 From 6d0a8dcd892f7ad9b399fed6edbad6ede13c5f69 Mon Sep 17 00:00:00 2001 From: drhead <1313496+drhead@users.noreply.github.com> Date: Wed, 29 Nov 2023 17:42:07 -0500 Subject: Implement zero terminal SNR schedule option --- modules/processing.py | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index ac58ef86..c88eec70 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -863,6 +863,34 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.n_iter > 1: shared.state.job = f"Batch {n+1} out of {p.n_iter}" + + def rescale_zero_terminal_snr_abar(alphas_cumprod): + alphas_bar_sqrt = alphas_cumprod.sqrt() + + # Store old values. + alphas_bar_sqrt_0 = alphas_bar_sqrt[0].clone() + alphas_bar_sqrt_T = alphas_bar_sqrt[-1].clone() + + # Shift so the last timestep is zero. + alphas_bar_sqrt -= (alphas_bar_sqrt_T) + + # Scale so the first timestep is back to the old value. + alphas_bar_sqrt *= alphas_bar_sqrt_0 / (alphas_bar_sqrt_0 - alphas_bar_sqrt_T) + + # Convert alphas_bar_sqrt to betas + alphas_bar = alphas_bar_sqrt**2 # Revert sqrt + alphas_bar[-1] = 4.8973451890853435e-08 + return alphas_bar + + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) + + if opts.use_downcasted_alpha_bar: + p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device) + if opts.sd_noise_schedule == "Zero Terminal SNR": + p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule + print("rescaling noise schedule for zero snr") + p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device) with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(): samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts) -- cgit v1.2.1 From ec6ee5c13bf3453f8703e225a191333a9bbcf10a Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Wed, 29 Nov 2023 18:10:27 -0500 Subject: Fix infotext for ztSNR --- modules/shared_options.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index 51596777..bc3d56de 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -218,7 +218,7 @@ options_templates.update(options_section(('compatibility', "Compatibility", "sd" "dont_fix_second_order_samplers_schedule": OptionInfo(False, "Do not fix prompt schedule for second order samplers."), "hires_fix_use_firstpass_conds": OptionInfo(False, "For hires fix, calculate conds of second pass using extra networks of first pass."), "use_old_scheduling": OptionInfo(False, "Use old prompt editing timelines.", infotext="Old prompt editing timelines").info("For [red:green:N]; old: If N < 1, it's a fraction of steps (and hires fix uses range from 0 to 1), if N >= 1, it's an absolute number of steps; new: If N has a decimal point in it, it's a fraction of steps (and hires fix uses range from 1 to 2), othewrwise it's an absolute number of steps"), - "use_downcasted_alpha_bar": OptionInfo(False, "Downcast model alphas_cumprod to fp16 before sampling. For reproducing old seeds.") + "use_downcasted_alpha_bar": OptionInfo(False, "Downcast model alphas_cumprod to fp16 before sampling. For reproducing old seeds.", infotext="Downcast alphas_cumprod") })) options_templates.update(options_section(('interrogate', "Interrogate"), { @@ -336,7 +336,7 @@ options_templates.update(options_section(('sampler-params', "Sampler parameters" 'uni_pc_skip_type': OptionInfo("time_uniform", "UniPC skip type", gr.Radio, {"choices": ["time_uniform", "time_quadratic", "logSNR"]}, infotext='UniPC skip type'), 'uni_pc_order': OptionInfo(3, "UniPC order", gr.Slider, {"minimum": 1, "maximum": 50, "step": 1}, infotext='UniPC order').info("must be < sampling steps"), 'uni_pc_lower_order_final': OptionInfo(True, "UniPC lower order final", infotext='UniPC lower order final'), - 'sd_noise_schedule': OptionInfo("Default", "Noise schedule for sampling", gr.Radio, {"choices": ["Default", "Zero Terminal SNR"]}, infotext="Noise schedule for sampling").info("for use with zero terminal SNR trained models") + 'sd_noise_schedule': OptionInfo("Default", "Noise schedule for sampling", gr.Radio, {"choices": ["Default", "Zero Terminal SNR"]}, infotext="Noise Schedule").info("for use with zero terminal SNR trained models") })) options_templates.update(options_section(('postprocessing', "Postprocessing", "postprocessing"), { -- cgit v1.2.1 From ffa7f8201d849636bb327b3b40298e7c169ff204 Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Wed, 29 Nov 2023 18:10:43 -0500 Subject: Lint --- modules/processing.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index c88eec70..f3883d5b 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -863,7 +863,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.n_iter > 1: shared.state.job = f"Batch {n+1} out of {p.n_iter}" - + def rescale_zero_terminal_snr_abar(alphas_cumprod): alphas_bar_sqrt = alphas_cumprod.sqrt() @@ -881,7 +881,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: alphas_bar = alphas_bar_sqrt**2 # Revert sqrt alphas_bar[-1] = 4.8973451890853435e-08 return alphas_bar - + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) if opts.use_downcasted_alpha_bar: -- cgit v1.2.1 From de79597ab9894965e3702939b8536ec3dcc3c859 Mon Sep 17 00:00:00 2001 From: catboxanon <122327233+catboxanon@users.noreply.github.com> Date: Wed, 29 Nov 2023 18:33:32 -0500 Subject: Only apply ztSNR related code if alphas_cumprod exists --- modules/processing.py | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index f3883d5b..7e73d7e2 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -882,15 +882,16 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: alphas_bar[-1] = 4.8973451890853435e-08 return alphas_bar - p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) - - if opts.use_downcasted_alpha_bar: - p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar - p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device) - if opts.sd_noise_schedule == "Zero Terminal SNR": - p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule - print("rescaling noise schedule for zero snr") - p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device) + if hasattr(p.sd_model, 'alphas_cumprod') and hasattr(p.sd_model, 'alphas_cumprod_original'): + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) + + if opts.use_downcasted_alpha_bar: + p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device) + if opts.sd_noise_schedule == "Zero Terminal SNR": + p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule + print("rescaling noise schedule for zero snr") + p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device) with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(): samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts) -- cgit v1.2.1 From 668ae34e21df848ef4909b8b49c4142a3674701b Mon Sep 17 00:00:00 2001 From: drhead <1313496+drhead@users.noreply.github.com> Date: Wed, 29 Nov 2023 22:48:31 -0500 Subject: remove debug print --- modules/processing.py | 1 - 1 file changed, 1 deletion(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 7e73d7e2..d73c8bfc 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -890,7 +890,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device) if opts.sd_noise_schedule == "Zero Terminal SNR": p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule - print("rescaling noise schedule for zero snr") p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device) with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(): -- cgit v1.2.1 From 50a21cb09fe3e9ea2d4fe058e0484e192c8a86e3 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sat, 2 Dec 2023 22:06:47 +0800 Subject: Ensure the cached weight will not be affected --- modules/sd_models.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 4b8a9ae6..dcf816b3 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -435,9 +435,9 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer for module in model.modules(): if isinstance(module, (torch.nn.Conv2d, torch.nn.Linear)): if shared.opts.cache_fp16_weight: - module.fp16_weight = module.weight.clone().half() + module.fp16_weight = module.weight.data.clone().cpu().half() if module.bias is not None: - module.fp16_bias = module.bias.clone().half() + module.fp16_bias = module.bias.data.clone().cpu().half() module.to(torch.float8_e4m3fn) model.first_stage_model = first_stage timer.record("apply fp8") -- cgit v1.2.1 From 309a606c2fa645b6b8623f96ea56117e685a47fb Mon Sep 17 00:00:00 2001 From: drhead <1313496+drhead@users.noreply.github.com> Date: Sat, 2 Dec 2023 13:07:45 -0500 Subject: ensure that original alpha bar always exists --- modules/processing.py | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index d73c8bfc..bfa59038 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -882,15 +882,17 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: alphas_bar[-1] = 4.8973451890853435e-08 return alphas_bar - if hasattr(p.sd_model, 'alphas_cumprod') and hasattr(p.sd_model, 'alphas_cumprod_original'): - p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) - - if opts.use_downcasted_alpha_bar: - p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar - p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device) - if opts.sd_noise_schedule == "Zero Terminal SNR": - p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule - p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device) + if hasattr(p.sd_model, 'alphas_cumprod') and not hasattr(p.sd_model, 'alphas_cumprod_original'): + p.sd_model.alphas_cumprod_original = p.sd_model.alphas_cumprod + + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) + + if opts.use_downcasted_alpha_bar: + p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device) + if opts.sd_noise_schedule == "Zero Terminal SNR": + p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule + p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device) with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(): samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts) -- cgit v1.2.1 From 81c4ddf6ebebe6f18338de3b0391da1d8521a525 Mon Sep 17 00:00:00 2001 From: drhead <1313496+drhead@users.noreply.github.com> Date: Sat, 2 Dec 2023 13:11:00 -0500 Subject: fix linting --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index bfa59038..eeccea74 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -884,7 +884,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if hasattr(p.sd_model, 'alphas_cumprod') and not hasattr(p.sd_model, 'alphas_cumprod_original'): p.sd_model.alphas_cumprod_original = p.sd_model.alphas_cumprod - + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) if opts.use_downcasted_alpha_bar: -- cgit v1.2.1 From 4a43334376d9e116f7a1446f042f9af9c0484fc6 Mon Sep 17 00:00:00 2001 From: drhead Date: Sat, 2 Dec 2023 14:05:42 -0500 Subject: Revert 309a606c --- modules/processing.py | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index eeccea74..d73c8bfc 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -882,17 +882,15 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: alphas_bar[-1] = 4.8973451890853435e-08 return alphas_bar - if hasattr(p.sd_model, 'alphas_cumprod') and not hasattr(p.sd_model, 'alphas_cumprod_original'): - p.sd_model.alphas_cumprod_original = p.sd_model.alphas_cumprod - - p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) - - if opts.use_downcasted_alpha_bar: - p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar - p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device) - if opts.sd_noise_schedule == "Zero Terminal SNR": - p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule - p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device) + if hasattr(p.sd_model, 'alphas_cumprod') and hasattr(p.sd_model, 'alphas_cumprod_original'): + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod_original.to(shared.device) + + if opts.use_downcasted_alpha_bar: + p.extra_generation_params['Downcast alphas_cumprod'] = opts.use_downcasted_alpha_bar + p.sd_model.alphas_cumprod = p.sd_model.alphas_cumprod.half().to(shared.device) + if opts.sd_noise_schedule == "Zero Terminal SNR": + p.extra_generation_params['Noise Schedule'] = opts.sd_noise_schedule + p.sd_model.alphas_cumprod = rescale_zero_terminal_snr_abar(p.sd_model.alphas_cumprod).to(shared.device) with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(): samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts) -- cgit v1.2.1 From dc1adeecdd02f3fb910481e808a6d60a77100fea Mon Sep 17 00:00:00 2001 From: drhead Date: Sat, 2 Dec 2023 14:06:56 -0500 Subject: Create alphas_cumprod_original on full precision path --- modules/sd_models.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index de80a493..976c7d5b 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -374,6 +374,7 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer if shared.cmd_opts.no_half: model.float() + model.alphas_cumprod_original = alphas_cumprod devices.dtype_unet = torch.float32 timer.record("apply float()") else: -- cgit v1.2.1 From 78acdcf677a96894651ff0d7d8287f2a994f3781 Mon Sep 17 00:00:00 2001 From: drhead Date: Sat, 2 Dec 2023 14:09:18 -0500 Subject: fix variable --- modules/sd_models.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 976c7d5b..5a19a00a 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -374,7 +374,7 @@ def load_model_weights(model, checkpoint_info: CheckpointInfo, state_dict, timer if shared.cmd_opts.no_half: model.float() - model.alphas_cumprod_original = alphas_cumprod + model.alphas_cumprod_original = model.alphas_cumprod devices.dtype_unet = torch.float32 timer.record("apply float()") else: -- cgit v1.2.1 From 609dea36ea919aa7db42fd4233c416a45c74578b Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Sat, 2 Dec 2023 18:56:49 -0700 Subject: Added utility functions related to processing masks. --- modules/images.py | 191 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 191 insertions(+) (limited to 'modules') diff --git a/modules/images.py b/modules/images.py index eb644733..b5a0cead 100644 --- a/modules/images.py +++ b/modules/images.py @@ -776,3 +776,194 @@ def flatten(img, bgcolor): img = background return img.convert('RGB') + + +def weighted_histogram_filter(img, kernel, kernel_center, percentile_min=0.0, percentile_max=1.0, min_width=1.0): + """ + Generalization convolution filter capable of applying + weighted mean, median, maximum, and minimum filters + parametrically using an arbitrary kernel. + + Args: + img (nparray): + The image, a 2-D array of floats, to which the filter is being applied. + kernel (nparray): + The kernel, a 2-D array of floats. + kernel_center (nparray): + The kernel center coordinate, a 1-D array with two elements. + percentile_min (float): + The lower bound of the histogram window used by the filter, + from 0 to 1. + percentile_max (float): + The upper bound of the histogram window used by the filter, + from 0 to 1. + min_width (float): + The minimum size of the histogram window bounds, in weight units. + Must be greater than 0. + + Returns: + (nparray): A filtered copy of the input image "img", a 2-D array of floats. + """ + + # Converts an index tuple into a vector. + def vec(x): + return np.array(x) + + kernel_min = -kernel_center + kernel_max = vec(kernel.shape) - kernel_center + + def weighted_histogram_filter_single(idx): + idx = vec(idx) + min_index = np.maximum(0, idx + kernel_min) + max_index = np.minimum(vec(img.shape), idx + kernel_max) + window_shape = max_index - min_index + + class WeightedElement: + """ + An element of the histogram, its weight + and bounds. + """ + def __init__(self, value, weight): + self.value: float = value + self.weight: float = weight + self.window_min: float = 0.0 + self.window_max: float = 1.0 + + # Collect the values in the image as WeightedElements, + # weighted by their corresponding kernel values. + values = [] + for window_tup in np.ndindex(tuple(window_shape)): + window_index = vec(window_tup) + image_index = window_index + min_index + centered_kernel_index = image_index - idx + kernel_index = centered_kernel_index + kernel_center + element = WeightedElement(img[tuple(image_index)], kernel[tuple(kernel_index)]) + values.append(element) + + def sort_key(x: WeightedElement): + return x.value + + values.sort(key=sort_key) + + # Calculate the height of the stack (sum) + # and each sample's range they occupy in the stack + sum = 0 + for i in range(len(values)): + values[i].window_min = sum + sum += values[i].weight + values[i].window_max = sum + + # Calculate what range of this stack ("window") + # we want to get the weighted average across. + window_min = sum * percentile_min + window_max = sum * percentile_max + window_width = window_max - window_min + + # Ensure the window is within the stack and at least a certain size. + if window_width < min_width: + window_center = (window_min + window_max) / 2 + window_min = window_center - min_width / 2 + window_max = window_center + min_width / 2 + + if window_max > sum: + window_max = sum + window_min = sum - min_width + + if window_min < 0: + window_min = 0 + window_max = min_width + + value = 0 + value_weight = 0 + + # Get the weighted average of all the samples + # that overlap with the window, weighted + # by the size of their overlap. + for i in range(len(values)): + if window_min >= values[i].window_max: + continue + if window_max <= values[i].window_min: + break + + s = max(window_min, values[i].window_min) + e = min(window_max, values[i].window_max) + w = e - s + + value += values[i].value * w + value_weight += w + + return value / value_weight if value_weight != 0 else 0 + + img_out = img.copy() + + # Apply the kernel operation over each pixel. + for index in np.ndindex(img.shape): + img_out[index] = weighted_histogram_filter_single(index) + + return img_out + +def smoothstep(x): + """ + The smoothstep function, input should be clamped to 0-1 range. + Turns a diagonal line (f(x) = x) into a sigmoid-like curve. + """ + return x * x * (3 - 2 * x) + +def smootherstep(x): + """ + The smootherstep function, input should be clamped to 0-1 range. + Turns a diagonal line (f(x) = x) into a sigmoid-like curve. + """ + return x * x * x * (x * (6 * x - 15) + 10) + + +def get_gaussian_kernel(stddev_radius=1.0, max_radius=2): + """ + Creates a Gaussian kernel with thresholded edges. + + Args: + stddev_radius (float): + Standard deviation of the gaussian kernel, in pixels. + max_radius (int): + The size of the filter kernel. The number of pixels is (max_radius*2+1) ** 2. + The kernel is thresholded so that any values one pixel beyond this radius + is weighted at 0. + + Returns: + (nparray, nparray): A kernel array (shape: (N, N)), its center coordinate (shape: (2)) + """ + # Evaluates a 0-1 normalized gaussian function for a given square distance from the mean. + def gaussian(sqr_mag): + return math.exp(-sqr_mag / (stddev_radius * stddev_radius)) + + # Helper function for converting a tuple to an array. + def vec(x): + return np.array(x) + + """ + Since a gaussian is unbounded, we need to limit ourselves + to a finite range. + We taper the ends off at the end of that range so they equal zero + while preserving the maximum value of 1 at the mean. + """ + zero_radius = max_radius + 1.0 + gauss_zero = gaussian(zero_radius * zero_radius) + gauss_kernel_scale = 1 / (1 - gauss_zero) + + def gaussian_kernel_func(coordinate): + x = coordinate[0] ** 2.0 + coordinate[1] ** 2.0 + x = gaussian(x) + x -= gauss_zero + x /= gauss_kernel_scale + x = max(0.0, x) + return x + + size = max_radius * 2 + 1 + kernel_center = max_radius + kernel = np.zeros((size, size)) + + for index in np.ndindex(kernel.shape): + kernel[index] = gaussian_kernel_func(vec(index) - kernel_center) + + return kernel, kernel_center + -- cgit v1.2.1 From 73ab982d1b7394574d1cf2e0a151bc457eeed769 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Sat, 2 Dec 2023 21:07:02 -0700 Subject: Blend masks are now produced afterward, based on an estimate of the visual difference between the original and modified latent images. This should remove ghosting and clipping artifacts from masks, while preserving the details of largely unchanged content. --- modules/processing.py | 119 ++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 90 insertions(+), 29 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 92fdebad..ad716e11 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -9,7 +9,7 @@ from dataclasses import dataclass, field import torch import numpy as np -from PIL import Image, ImageOps +from PIL import Image, ImageOps, ImageFilter import random import cv2 from skimage import exposure @@ -62,6 +62,16 @@ def apply_color_correction(correction, original_image): return image.convert('RGB') +def uncrop(image, dest_size, paste_loc): + x, y, w, h = paste_loc + base_image = Image.new('RGBA', dest_size) + image = images.resize_image(1, image, w, h) + base_image.paste(image, (x, y)) + image = base_image + + return image + + def apply_overlay(image, paste_loc, index, overlays): if overlays is None or index >= len(overlays): return image @@ -69,11 +79,7 @@ def apply_overlay(image, paste_loc, index, overlays): overlay = overlays[index] if paste_loc is not None: - x, y, w, h = paste_loc - base_image = Image.new('RGBA', (overlay.width, overlay.height)) - image = images.resize_image(1, image, w, h) - base_image.paste(image, (x, y)) - image = base_image + image = uncrop(image, (overlay.width, overlay.height), paste_loc) image = image.convert('RGBA') image.alpha_composite(overlay) @@ -140,6 +146,7 @@ class StableDiffusionProcessing: do_not_save_grid: bool = False extra_generation_params: dict[str, Any] = None overlay_images: list = None + masks_for_overlay: list = None eta: float = None do_not_reload_embeddings: bool = False denoising_strength: float = 0 @@ -865,11 +872,66 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if getattr(samples_ddim, 'already_decoded', False): x_samples_ddim = samples_ddim + # todo: generate masks the old fashioned way else: if opts.sd_vae_decode_method != 'Full': p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method - x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) + # Generate the mask(s) based on similarity between the original and denoised latent vectors + if getattr(p, "image_mask", None) is not None: + # latent_mask = p.nmask[0].float().cpu() + + # convert the original mask into a form we use to scale distances for thresholding + # mask_scalar = 1-(torch.clamp(latent_mask, min=0, max=1) ** (p.mask_blend_scale / 2)) + # mask_scalar = mask_scalar / (1.00001-mask_scalar) + # mask_scalar = mask_scalar.numpy() + + latent_orig = p.init_latent + latent_proc = samples_ddim + latent_distance = torch.norm(latent_proc - latent_orig, p=2, dim=1) + + kernel, kernel_center = images.get_gaussian_kernel(stddev_radius=1.5, max_radius=2) + + for i, (distance_map, overlay_image) in enumerate(zip(latent_distance, p.overlay_images)): + converted_mask = distance_map.float().cpu().numpy() + converted_mask = images.weighted_histogram_filter(converted_mask, kernel, kernel_center, + percentile_min=0.9, percentile_max=1, min_width=1) + converted_mask = images.weighted_histogram_filter(converted_mask, kernel, kernel_center, + percentile_min=0.25, percentile_max=0.75, min_width=1) + + # The distance at which opacity of original decreases to 50% + # half_weighted_distance = 1 # * mask_scalar + # converted_mask = converted_mask / half_weighted_distance + + converted_mask = 1 / (1 + converted_mask ** 2) + converted_mask = images.smootherstep(converted_mask) + converted_mask = 1 - converted_mask + converted_mask = 255. * converted_mask + converted_mask = converted_mask.astype(np.uint8) + converted_mask = Image.fromarray(converted_mask) + converted_mask = images.resize_image(2, converted_mask, p.width, p.height) + converted_mask = create_binary_mask(converted_mask) + + # Remove aliasing artifacts using a gaussian blur. + converted_mask = converted_mask.filter(ImageFilter.GaussianBlur(radius=4)) + + # Expand the mask to fit the whole image if needed. + if p.paste_to is not None: + converted_mask = uncrop(converted_mask, + (overlay_image.width, overlay_image.height), + p.paste_to) + + p.masks_for_overlay[i] = converted_mask + + image_masked = Image.new('RGBa', (overlay_image.width, overlay_image.height)) + image_masked.paste(overlay_image.convert("RGBA").convert("RGBa"), + mask=ImageOps.invert(converted_mask.convert('L'))) + + p.overlay_images[i] = image_masked.convert('RGBA') + + x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, + target_device=devices.cpu, + check_for_nans=True) x_samples_ddim = torch.stack(x_samples_ddim).float() x_samples_ddim = torch.clamp((x_samples_ddim + 1.0) / 2.0, min=0.0, max=1.0) @@ -892,7 +954,9 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: x_samples_ddim = batch_params.images def infotext(index=0, use_main_prompt=False): - return create_infotext(p, p.prompts, p.seeds, p.subseeds, use_main_prompt=use_main_prompt, index=index, all_negative_prompts=p.negative_prompts) + return create_infotext(p, p.prompts, p.seeds, p.subseeds, + use_main_prompt=use_main_prompt, index=index, + all_negative_prompts=p.negative_prompts) save_samples = p.save_samples() @@ -923,19 +987,27 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: images.save_image(image_without_cc, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-before-color-correction") image = apply_color_correction(p.color_corrections[i], image) + # If the intention is to show the output from the model + # that is being composited over the original image, + # we need to keep the original image around + # and use it in the composite step. + original_denoised_image = image.copy() image = apply_overlay(image, p.paste_to, i, p.overlay_images) if save_samples: - images.save_image(image, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p) + images.save_image(image, p.outpath_samples, "", p.seeds[i], + p.prompts[i], opts.samples_format, info=infotext(i), p=p) text = infotext(i) infotexts.append(text) if opts.enable_pnginfo: image.info["parameters"] = text output_images.append(image) - if save_samples and hasattr(p, 'mask_for_overlay') and p.mask_for_overlay and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]): - image_mask = p.mask_for_overlay.convert('RGB') - image_mask_composite = Image.composite(image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), images.resize_image(2, p.mask_for_overlay, image.width, image.height).convert('L')).convert('RGBA') + if save_samples and hasattr(p, 'masks_for_overlay') and p.masks_for_overlay and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]): + image_mask = p.masks_for_overlay[i].convert('RGB') + image_mask_composite = Image.composite( + original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), + images.resize_image(2, p.masks_for_overlay[i], image.width, image.height).convert('L')).convert('RGBA') if opts.save_mask: images.save_image(image_mask, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-mask") @@ -1364,7 +1436,6 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): nmask: torch.Tensor = field(default=None, init=False) image_conditioning: torch.Tensor = field(default=None, init=False) init_img_hash: str = field(default=None, init=False) - mask_for_overlay: Image = field(default=None, init=False) init_latent: torch.Tensor = field(default=None, init=False) def __post_init__(self): @@ -1415,12 +1486,6 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image_mask = Image.fromarray(np_mask) if self.inpaint_full_res: - np_mask = np.array(image_mask).astype(np.float32) - np_mask /= 255 - np_mask = 1-pow(1-np_mask, 100) - np_mask *= 255 - np_mask = np.clip(np_mask, 0, 255).astype(np.uint8) - self.mask_for_overlay = Image.fromarray(np_mask) mask = image_mask.convert('L') crop_region = masking.get_crop_region(np.array(mask), self.inpaint_full_res_padding) crop_region = masking.expand_crop_region(crop_region, self.width, self.height, mask.width, mask.height) @@ -1431,13 +1496,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): self.paste_to = (x1, y1, x2-x1, y2-y1) else: image_mask = images.resize_image(self.resize_mode, image_mask, self.width, self.height) - np_mask = np.array(image_mask).astype(np.float32) - np_mask /= 255 - np_mask = 1-pow(1-np_mask, 100) - np_mask *= 255 - np_mask = np.clip(np_mask, 0, 255).astype(np.uint8) - self.mask_for_overlay = Image.fromarray(np_mask) + self.masks_for_overlay = [] self.overlay_images = [] latent_mask = self.latent_mask if self.latent_mask is not None else image_mask @@ -1459,10 +1519,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image = images.resize_image(self.resize_mode, image, self.width, self.height) if image_mask is not None: - image_masked = Image.new('RGBa', (image.width, image.height)) - image_masked.paste(image.convert("RGBA").convert("RGBa"), mask=ImageOps.invert(self.mask_for_overlay.convert('L'))) - - self.overlay_images.append(image_masked.convert('RGBA')) + self.overlay_images.append(image) + self.masks_for_overlay.append(image_mask) # crop_region is not None if we are doing inpaint full res if crop_region is not None: @@ -1486,6 +1544,9 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): if self.overlay_images is not None: self.overlay_images = self.overlay_images * self.batch_size + if self.masks_for_overlay is not None: + self.masks_for_overlay = self.masks_for_overlay * self.batch_size + if self.color_corrections is not None and len(self.color_corrections) == 1: self.color_corrections = self.color_corrections * self.batch_size -- cgit v1.2.1 From bb04d400c95df01d191ef6c1a43e66b95425fa33 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Sat, 2 Dec 2023 21:08:26 -0700 Subject: Rewrote latent_blend() to use in-place operations and to aggressively "del" references with the intention of minimizing allocations and easing garbage collection. --- modules/sd_samplers_cfg_denoiser.py | 41 +++++++++++++++++++++++++------------ 1 file changed, 28 insertions(+), 13 deletions(-) (limited to 'modules') diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index ceb612d7..efbe7a40 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -102,29 +102,44 @@ class CFGDenoiser(torch.nn.Module): The "detail_preservation" factor biases the magnitude interpolation towards the larger of the two magnitudes. """ - # Record the original latent vector magnitudes. - # We bring them to a power so that larger magnitudes are favored over smaller ones. - # 64-bit operations are used here to allow large exponents. - a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64) ** self.inpaint_detail_preservation - b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64) ** self.inpaint_detail_preservation + # NOTE: We use inplace operations wherever possible. one_minus_t = 1 - t - # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). - interp_magnitude = (a_magnitude * one_minus_t + b_magnitude * t) ** (1 / self.inpaint_detail_preservation) - # Linearly interpolate the image vectors. - image_interp = a * one_minus_t + b * t + a_scaled = a * one_minus_t + b_scaled = b * t + image_interp = a_scaled + image_interp.add_(b_scaled) + result_type = image_interp.dtype + del a_scaled, b_scaled # Calculate the magnitude of the interpolated vectors. (We will remove this magnitude.) # 64-bit operations are used here to allow large exponents. - image_interp_magnitude = torch.norm(image_interp, p=2, dim=1).to(torch.float64) + 0.0001 + current_magnitude = torch.norm(image_interp, p=2, dim=1).to(torch.float64).add_(0.00001) + + # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). + a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64).pow_(self.inpaint_detail_preservation) * one_minus_t + b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64).pow_(self.inpaint_detail_preservation) * t + desired_magnitude = a_magnitude + desired_magnitude.add_(b_magnitude).pow_(1 / self.inpaint_detail_preservation) + del a_magnitude, b_magnitude, one_minus_t # Change the linearly interpolated image vectors' magnitudes to the value we want. # This is the last 64-bit operation. - image_interp *= (interp_magnitude / image_interp_magnitude).to(image_interp.dtype) - - return image_interp + image_interp_scaling_factor = desired_magnitude + image_interp_scaling_factor.div_(current_magnitude) + image_interp_scaled = image_interp + image_interp_scaled.mul_(image_interp_scaling_factor) + del current_magnitude + del desired_magnitude + del image_interp + del image_interp_scaling_factor + + image_interp_scaled = image_interp_scaled.to(result_type) + del result_type + + return image_interp_scaled def get_modified_nmask(nmask, _sigma): """ -- cgit v1.2.1 From 28a2b5b4aab43424733039c31d910e8b8dd507cd Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Sun, 3 Dec 2023 14:20:20 -0700 Subject: Fixed a math mistake. --- modules/images.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/images.py b/modules/images.py index 6648097e..94953498 100644 --- a/modules/images.py +++ b/modules/images.py @@ -969,7 +969,7 @@ def get_gaussian_kernel(stddev_radius=1.0, max_radius=2): x = coordinate[0] ** 2.0 + coordinate[1] ** 2.0 x = gaussian(x) x -= gauss_zero - x /= gauss_kernel_scale + x *= gauss_kernel_scale x = max(0.0, x) return x -- cgit v1.2.1 From 552f8bc832cd21ee0338e08b6a701687d0d79fad Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Sun, 3 Dec 2023 14:49:41 -0700 Subject: "Uncrop" the original denoised image for the composite step, fixing a "ValueError: Images do not match" *shudder* --- modules/processing.py | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 66aaab83..cd7216f8 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -994,6 +994,10 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: # we need to keep the original image around # and use it in the composite step. original_denoised_image = image.copy() + + if p.paste_to is not None: + original_denoised_image = uncrop(original_denoised_image, (p.overlay_images[i].width, p.overlay_images[i].height), p.paste_to) + image = apply_overlay(image, p.paste_to, i, p.overlay_images) if save_samples: -- cgit v1.2.1 From aaacf4823241450d88315af9d465d6815119fe0d Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 01:27:22 -0700 Subject: Organized the settings and UI of soft inpainting to allow for toggling the feature, and centralizes default values to reduce the amount of copy-pasta. --- modules/img2img.py | 14 ++-- modules/processing.py | 5 +- modules/sd_samplers_cfg_denoiser.py | 35 +++++++--- modules/sd_samplers_common.py | 4 +- modules/soft_inpainting.py | 133 ++++++++++++++++++++++++++++++++++++ modules/ui.py | 17 +++-- 6 files changed, 179 insertions(+), 29 deletions(-) create mode 100644 modules/soft_inpainting.py (limited to 'modules') diff --git a/modules/img2img.py b/modules/img2img.py index 596f741c..3aa8a9ce 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -15,6 +15,7 @@ import modules.shared as shared import modules.processing as processing from modules.ui import plaintext_to_html import modules.scripts +import modules.soft_inpainting as si def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=False, scale_by=1.0, use_png_info=False, png_info_props=None, png_info_dir=None): @@ -162,6 +163,7 @@ def img2img(id_task: str, sampler_name: str, mask_blur: int, mask_alpha: float, + mask_blend_enabled: bool, mask_blend_power: float, mask_blend_scale: float, inpaint_detail_preservation: float, @@ -227,6 +229,9 @@ def img2img(id_task: str, assert 0. <= denoising_strength <= 1., 'can only work with strength in [0.0, 1.0]' + soft_inpainting = si.SoftInpaintingSettings(mask_blend_power, mask_blend_scale, inpaint_detail_preservation) \ + if mask_blend_enabled else None + p = StableDiffusionProcessingImg2Img( sd_model=shared.sd_model, outpath_samples=opts.outdir_samples or opts.outdir_img2img_samples, @@ -244,9 +249,7 @@ def img2img(id_task: str, init_images=[image], mask=mask, mask_blur=mask_blur, - mask_blend_power=mask_blend_power, - mask_blend_scale=mask_blend_scale, - inpaint_detail_preservation=inpaint_detail_preservation, + soft_inpainting=soft_inpainting, inpainting_fill=inpainting_fill, resize_mode=resize_mode, denoising_strength=denoising_strength, @@ -267,9 +270,8 @@ def img2img(id_task: str, if mask: p.extra_generation_params["Mask blur"] = mask_blur - p.extra_generation_params["Mask blending bias"] = mask_blend_power - p.extra_generation_params["Mask blending preservation"] = mask_blend_scale - p.extra_generation_params["Mask blending contrast boost"] = inpaint_detail_preservation + if soft_inpainting is not None: + soft_inpainting.add_generation_params(p.extra_generation_params) with closing(p): if is_batch: diff --git a/modules/processing.py b/modules/processing.py index cd7216f8..b209c84a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -30,6 +30,7 @@ import modules.sd_models as sd_models import modules.sd_vae as sd_vae from ldm.data.util import AddMiDaS from ldm.models.diffusion.ddpm import LatentDepth2ImageDiffusion +import modules.soft_inpainting as si from einops import repeat, rearrange from blendmodes.blend import blendLayers, BlendType @@ -1425,9 +1426,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): mask_blur_x: int = 4 mask_blur_y: int = 4 mask_blur: int = None - mask_blend_power: float = 1 - mask_blend_scale: float = 0.5 - inpaint_detail_preservation: float = 4 + soft_inpainting: si.SoftInpaintingParameters = si.default inpainting_fill: int = 0 inpaint_full_res: bool = True inpaint_full_res_padding: int = 0 diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index efbe7a40..0ee0b7dd 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -6,6 +6,7 @@ import modules.shared as shared from modules.script_callbacks import CFGDenoiserParams, cfg_denoiser_callback from modules.script_callbacks import CFGDenoisedParams, cfg_denoised_callback from modules.script_callbacks import AfterCFGCallbackParams, cfg_after_cfg_callback +import modules.soft_inpainting as si def catenate_conds(conds): @@ -43,9 +44,7 @@ class CFGDenoiser(torch.nn.Module): self.model_wrap = None self.mask = None self.nmask = None - self.mask_blend_power = 1 - self.mask_blend_scale = 0.5 - self.inpaint_detail_preservation = 4 + self.soft_inpainting: si.SoftInpaintingParameters = None self.init_latent = None self.steps = None """number of steps as specified by user in UI""" @@ -95,7 +94,8 @@ class CFGDenoiser(torch.nn.Module): self.sampler.sampler_extra_args['uncond'] = uc def forward(self, x, sigma, uncond, cond, cond_scale, s_min_uncond, image_cond): - def latent_blend(a, b, t): + def latent_blend(a, b, t, one_minus_t=None): + """ Interpolates two latent image representations according to the parameter t, where the interpolated vectors' magnitudes are also interpolated separately. @@ -104,7 +104,11 @@ class CFGDenoiser(torch.nn.Module): """ # NOTE: We use inplace operations wherever possible. - one_minus_t = 1 - t + if one_minus_t is None: + one_minus_t = 1 - t + + if self.soft_inpainting is None: + return a * one_minus_t + b * t # Linearly interpolate the image vectors. a_scaled = a * one_minus_t @@ -119,10 +123,10 @@ class CFGDenoiser(torch.nn.Module): current_magnitude = torch.norm(image_interp, p=2, dim=1).to(torch.float64).add_(0.00001) # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). - a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64).pow_(self.inpaint_detail_preservation) * one_minus_t - b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64).pow_(self.inpaint_detail_preservation) * t + a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64).pow_(self.soft_inpainting.inpaint_detail_preservation) * one_minus_t + b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64).pow_(self.soft_inpainting.inpaint_detail_preservation) * t desired_magnitude = a_magnitude - desired_magnitude.add_(b_magnitude).pow_(1 / self.inpaint_detail_preservation) + desired_magnitude.add_(b_magnitude).pow_(1 / self.soft_inpainting.inpaint_detail_preservation) del a_magnitude, b_magnitude, one_minus_t # Change the linearly interpolated image vectors' magnitudes to the value we want. @@ -156,7 +160,10 @@ class CFGDenoiser(torch.nn.Module): NOTE: "mask" is not used """ - return torch.pow(nmask, (_sigma ** self.mask_blend_power) * self.mask_blend_scale) + if self.soft_inpainting is None: + return nmask + + return torch.pow(nmask, (_sigma ** self.soft_inpainting.mask_blend_power) * self.soft_inpainting.mask_blend_scale) if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException @@ -176,7 +183,10 @@ class CFGDenoiser(torch.nn.Module): # Blend in the original latents (before) if self.mask_before_denoising and self.mask is not None: - x = latent_blend(self.init_latent, x, get_modified_nmask(self.nmask, sigma)) + if self.soft_inpainting is None: + x = latent_blend(self.init_latent, x, self.nmask, self.mask) + else: + x = latent_blend(self.init_latent, x, get_modified_nmask(self.nmask, sigma)) batch_size = len(conds_list) repeats = [len(conds_list[i]) for i in range(batch_size)] @@ -279,7 +289,10 @@ class CFGDenoiser(torch.nn.Module): # Blend in the original latents (after) if not self.mask_before_denoising and self.mask is not None: - denoised = latent_blend(self.init_latent, denoised, get_modified_nmask(self.nmask, sigma)) + if self.soft_inpainting is None: + denoised = latent_blend(self.init_latent, denoised, self.nmask, self.mask) + else: + denoised = latent_blend(self.init_latent, denoised, get_modified_nmask(self.nmask, sigma)) self.sampler.last_latent = self.get_pred_x0(torch.cat([x_in[i:i + 1] for i in denoised_image_indexes]), torch.cat([x_out[i:i + 1] for i in denoised_image_indexes]), sigma) diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index ecd8ab0a..9682bee3 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -277,9 +277,7 @@ class Sampler: self.model_wrap_cfg.p = p self.model_wrap_cfg.mask = p.mask if hasattr(p, 'mask') else None self.model_wrap_cfg.nmask = p.nmask if hasattr(p, 'nmask') else None - self.model_wrap_cfg.mask_blend_power = p.mask_blend_power if hasattr(p, 'mask_blend_power') else None - self.model_wrap_cfg.mask_blend_scale = p.mask_blend_scale if hasattr(p, 'mask_blend_scale') else None - self.model_wrap_cfg.inpaint_detail_preservation = p.inpaint_detail_preservation if hasattr(p, 'inpaint_detail_preservation') else None + self.model_wrap_cfg.soft_inpainting = p.soft_inpainting if hasattr(p, 'soft_inpainting') else None self.model_wrap_cfg.step = 0 self.model_wrap_cfg.image_cfg_scale = getattr(p, 'image_cfg_scale', None) self.eta = p.eta if p.eta is not None else getattr(opts, self.eta_option_field, 0.0) diff --git a/modules/soft_inpainting.py b/modules/soft_inpainting.py new file mode 100644 index 00000000..259c36ec --- /dev/null +++ b/modules/soft_inpainting.py @@ -0,0 +1,133 @@ +class SoftInpaintingSettings: + def __init__(self, mask_blend_power, mask_blend_scale, inpaint_detail_preservation): + self.mask_blend_power = mask_blend_power + self.mask_blend_scale = mask_blend_scale + self.inpaint_detail_preservation = inpaint_detail_preservation + + def get_paste_fields(self): + return [ + (self.mask_blend_power, gen_param_labels.mask_blend_power), + (self.mask_blend_scale, gen_param_labels.mask_blend_scale), + (self.inpaint_detail_preservation, gen_param_labels.inpaint_detail_preservation), + ] + + def add_generation_params(self, dest): + dest[enabled_gen_param_label] = True + dest[gen_param_labels.mask_blend_power] = self.mask_blend_power + dest[gen_param_labels.mask_blend_scale] = self.mask_blend_scale + dest[gen_param_labels.inpaint_detail_preservation] = self.inpaint_detail_preservation + + +enabled_ui_label = "Soft inpainting" +enabled_gen_param_label = "Soft inpainting enabled" +enabled_el_id = "soft_inpainting_enabled" + +default = SoftInpaintingSettings(1, 0.5, 4) +ui_labels = SoftInpaintingSettings("Schedule bias", "Preservation strength", "Transition contrast boost") + +ui_info = SoftInpaintingSettings( + mask_blend_power="Shifts when preservation of original content occurs during denoising.", + # "Below 1: Stronger preservation near the end (with low sigma)\n" + # "1: Balanced (proportional to sigma)\n" + # "Above 1: Stronger preservation in the beginning (with high sigma)", + mask_blend_scale="How strongly partially masked content should be preserved.", + # "Low values: Favors generated content.\n" + # "High values: Favors original content.", + inpaint_detail_preservation="Amplifies the contrast that may be lost in partially masked regions.") + +gen_param_labels = SoftInpaintingSettings("Soft inpainting schedule bias", "Soft inpainting preservation strength", "Soft inpainting transition contrast boost") +el_ids = SoftInpaintingSettings("mask_blend_power", "mask_blend_scale", "inpaint_detail_preservation") + + +def gradio_ui(): + import gradio as gr + from modules.ui_components import InputAccordion + """ + with InputAccordion(False, label="Refiner", elem_id=self.elem_id("enable")) as enable_refiner: + with gr.Row(): + refiner_checkpoint = gr.Dropdown(label='Checkpoint', elem_id=self.elem_id("checkpoint"), choices=sd_models.checkpoint_tiles(), value='', tooltip="switch to another model in the middle of generation") + create_refresh_button(refiner_checkpoint, sd_models.list_models, lambda: {"choices": sd_models.checkpoint_tiles()}, self.elem_id("checkpoint_refresh")) + + refiner_switch_at = gr.Slider(value=0.8, label="Switch at", minimum=0.01, maximum=1.0, step=0.01, elem_id=self.elem_id("switch_at"), tooltip="fraction of sampling steps when the switch to refiner model should happen; 1=never, 0.5=switch in the middle of generation") + + """ + with InputAccordion(False, label=enabled_ui_label, elem_id=enabled_el_id) as soft_inpainting_enabled: + with gr.Group(): + gr.Markdown( + """ + Soft inpainting allows you to **seamlessly blend original content with inpainted content** according to the mask opacity. + **High _Mask blur_** values are recommended! + """) + + result = SoftInpaintingSettings( + gr.Slider(label=ui_labels.mask_blend_power, + info=ui_info.mask_blend_power, + minimum=0, + maximum=8, + step=0.1, + value=default.mask_blend_power, + elem_id=el_ids.mask_blend_power), + gr.Slider(label=ui_labels.mask_blend_scale, + info=ui_info.mask_blend_scale, + minimum=0, + maximum=8, + step=0.05, + value=default.mask_blend_scale, + elem_id=el_ids.mask_blend_scale), + gr.Slider(label=ui_labels.inpaint_detail_preservation, + info=ui_info.inpaint_detail_preservation, + minimum=1, + maximum=32, + step=0.5, + value=default.inpaint_detail_preservation, + elem_id=el_ids.inpaint_detail_preservation)) + + with gr.Accordion("Help", open=False): + gr.Markdown( + f""" + ### {ui_labels.mask_blend_power} + + The blending strength of original content is scaled proportionally with the decreasing noise level values at each step (sigmas). + This ensures that the influence of the denoiser and original content preservation is roughly balanced at each step. + This balance can be shifted using this parameter, controlling whether earlier or later steps have stronger preservation. + + - **Below 1**: Stronger preservation near the end (with low sigma) + - **1**: Balanced (proportional to sigma) + - **Above 1**: Stronger preservation in the beginning (with high sigma) + """) + gr.Markdown( + f""" + ### {ui_labels.mask_blend_scale} + + Skews whether partially masked image regions should be more likely to preserve the original content or favor inpainted content. + This may need to be adjusted depending on the {ui_labels.mask_blend_power}, CFG Scale, prompt and Denoising strength. + + - **Low values**: Favors generated content. + - **High values**: Favors original content. + """) + gr.Markdown( + f""" + ### {ui_labels.inpaint_detail_preservation} + + This parameter controls how the original latent vectors and denoised latent vectors are interpolated. + With higher values, the magnitude of the resulting blended vector will be closer to the maximum of the two interpolated vectors. + This can prevent the loss of contrast that occurs with linear interpolation. + + - **Low values**: Softer blending, details may fade. + - **High values**: Stronger contrast, may over-saturate colors. + """) + + return ( + [ + soft_inpainting_enabled, + result.mask_blend_power, + result.mask_blend_scale, + result.inpaint_detail_preservation + ], + [ + (soft_inpainting_enabled, enabled_gen_param_label), + (result.mask_blend_power, gen_param_labels.mask_blend_power), + (result.mask_blend_scale, gen_param_labels.mask_blend_scale), + (result.inpaint_detail_preservation, gen_param_labels.inpaint_detail_preservation) + ] + ) diff --git a/modules/ui.py b/modules/ui.py index b13ed66c..0e4fb17a 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -29,6 +29,7 @@ import modules.shared as shared from modules import prompt_parser from modules.sd_hijack import model_hijack from modules.generation_parameters_copypaste import image_from_url_text +import modules.soft_inpainting as si create_setting_component = ui_settings.create_setting_component @@ -678,9 +679,16 @@ def create_ui(): with FormRow(): mask_blur = gr.Slider(label='Mask blur', minimum=0, maximum=64, step=1, value=4, elem_id="img2img_mask_blur") mask_alpha = gr.Slider(label="Mask transparency", visible=False, elem_id="img2img_mask_alpha") + + with FormRow(): + soft_inpainting = si.gradio_ui() + + + """ mask_blend_power = gr.Slider(label='Blending bias', minimum=0, maximum=8, step=0.1, value=1, elem_id="img2img_mask_blend_power") mask_blend_scale = gr.Slider(label='Blending preservation', minimum=0, maximum=8, step=0.05, value=0.5, elem_id="img2img_mask_blend_scale") inpaint_detail_preservation = gr.Slider(label='Blending contrast boost', minimum=1, maximum=32, step=0.5, value=4, elem_id="img2img_mask_blend_offset") + """ with FormRow(): inpainting_mask_invert = gr.Radio(label='Mask mode', choices=['Inpaint masked', 'Inpaint not masked'], value='Inpaint masked', type="index", elem_id="img2img_mask_mode") @@ -736,9 +744,7 @@ def create_ui(): sampler_name, mask_blur, mask_alpha, - mask_blend_power, - mask_blend_scale, - inpaint_detail_preservation, + *(soft_inpainting[0]), inpainting_fill, batch_count, batch_size, @@ -837,11 +843,10 @@ def create_ui(): (toprow.ui_styles.dropdown, lambda d: d["Styles array"] if isinstance(d.get("Styles array"), list) else gr.update()), (denoising_strength, "Denoising strength"), (mask_blur, "Mask blur"), - (mask_blend_power, "Mask blending bias"), - (mask_blend_scale, "Mask blending preservation"), - (inpaint_detail_preservation, "Mask blending contrast boost"), + *(soft_inpainting[1]), *scripts.scripts_img2img.infotext_fields ] + parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields, override_settings) parameters_copypaste.add_paste_fields("inpaint", init_img_with_mask, img2img_paste_fields, override_settings) parameters_copypaste.register_paste_params_button(parameters_copypaste.ParamBinding( -- cgit v1.2.1 From 259d33c3c8e27557cb9bab9b3a1dd7fc7450d16c Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 01:57:21 -0700 Subject: Enables the original functionality to be toggled on and off. --- modules/processing.py | 99 ++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 70 insertions(+), 29 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index b209c84a..b40b1a40 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -88,9 +88,12 @@ def apply_overlay(image, paste_loc, index, overlays): return image -def create_binary_mask(image): +def create_binary_mask(image, round=True): if image.mode == 'RGBA' and image.getextrema()[-1] != (255, 255): - image = image.split()[-1].convert("L") + if round: + image = image.split()[-1].convert("L").point(lambda x: 255 if x > 128 else 0) + else: + image = image.split()[-1].convert("L") else: image = image.convert('L') return image @@ -316,7 +319,7 @@ class StableDiffusionProcessing: c_adm = torch.cat((c_adm, noise_level_emb), 1) return c_adm - def inpainting_image_conditioning(self, source_image, latent_image, image_mask=None): + def inpainting_image_conditioning(self, source_image, latent_image, image_mask=None, round_image_mask=True): self.is_using_inpainting_conditioning = True # Handle the different mask inputs @@ -327,6 +330,11 @@ class StableDiffusionProcessing: conditioning_mask = np.array(image_mask.convert("L")) conditioning_mask = conditioning_mask.astype(np.float32) / 255.0 conditioning_mask = torch.from_numpy(conditioning_mask[None, None]) + + if round_image_mask: + # Caller is requesting a discretized mask as input, so we round to either 1.0 or 0.0 + conditioning_mask = torch.round(conditioning_mask) + else: conditioning_mask = source_image.new_ones(1, 1, *source_image.shape[-2:]) @@ -350,7 +358,7 @@ class StableDiffusionProcessing: return image_conditioning - def img2img_image_conditioning(self, source_image, latent_image, image_mask=None): + def img2img_image_conditioning(self, source_image, latent_image, image_mask=None, round_image_mask=True): source_image = devices.cond_cast_float(source_image) # HACK: Using introspection as the Depth2Image model doesn't appear to uniquely @@ -362,7 +370,10 @@ class StableDiffusionProcessing: return self.edit_image_conditioning(source_image) if self.sampler.conditioning_key in {'hybrid', 'concat'}: - return self.inpainting_image_conditioning(source_image, latent_image, image_mask=image_mask) + return self.inpainting_image_conditioning(source_image, + latent_image, + image_mask=image_mask, + round_image_mask=round_image_mask) if self.sampler.conditioning_key == "crossattn-adm": return self.unclip_image_conditioning(source_image) @@ -878,8 +889,9 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: else: if opts.sd_vae_decode_method != 'Full': p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method + # Generate the mask(s) based on similarity between the original and denoised latent vectors - if getattr(p, "image_mask", None) is not None: + if getattr(p, "image_mask", None) is not None and getattr(p, "soft_inpainting", None) is not None: # latent_mask = p.nmask[0].float().cpu() # convert the original mask into a form we use to scale distances for thresholding @@ -911,7 +923,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: converted_mask = converted_mask.astype(np.uint8) converted_mask = Image.fromarray(converted_mask) converted_mask = images.resize_image(2, converted_mask, p.width, p.height) - converted_mask = create_binary_mask(converted_mask) + converted_mask = create_binary_mask(converted_mask, round=False) # Remove aliasing artifacts using a gaussian blur. converted_mask = converted_mask.filter(ImageFilter.GaussianBlur(radius=4)) @@ -1010,23 +1022,33 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if opts.enable_pnginfo: image.info["parameters"] = text output_images.append(image) - if save_samples and hasattr(p, 'masks_for_overlay') and p.masks_for_overlay and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]): - image_mask = p.masks_for_overlay[i].convert('RGB') - image_mask_composite = Image.composite( - original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), - images.resize_image(2, p.masks_for_overlay[i], image.width, image.height).convert('L')).convert('RGBA') - - if opts.save_mask: - images.save_image(image_mask, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-mask") - - if opts.save_mask_composite: - images.save_image(image_mask_composite, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-mask-composite") - - if opts.return_mask: - output_images.append(image_mask) - - if opts.return_mask_composite: - output_images.append(image_mask_composite) + if save_samples and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]): + if hasattr(p, 'masks_for_overlay') and p.masks_for_overlay: + image_mask = p.masks_for_overlay[i].convert('RGB') + image_mask_composite = Image.composite( + original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), + images.resize_image(2, p.masks_for_overlay[i], image.width, image.height).convert('L')).convert('RGBA') + elif hasattr(p, 'mask_for_overlay') and p.mask_for_overlay: + image_mask = p.mask_for_overlay.convert('RGB') + image_mask_composite = Image.composite( + original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), + images.resize_image(2, p.mask_for_overlay, image.width, image.height).convert('L')).convert('RGBA') + else: + image_mask = None + image_mask_composite = None + + if image_mask is not None and image_mask_composite is not None: + if opts.save_mask: + images.save_image(image_mask, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-mask") + + if opts.save_mask_composite: + images.save_image(image_mask_composite, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-mask-composite") + + if opts.return_mask: + output_images.append(image_mask) + + if opts.return_mask_composite: + output_images.append(image_mask_composite) del x_samples_ddim @@ -1439,6 +1461,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): nmask: torch.Tensor = field(default=None, init=False) image_conditioning: torch.Tensor = field(default=None, init=False) init_img_hash: str = field(default=None, init=False) + mask_for_overlay: Image = field(default=None, init=False) init_latent: torch.Tensor = field(default=None, init=False) def __post_init__(self): @@ -1471,7 +1494,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): if image_mask is not None: # image_mask is passed in as RGBA by Gradio to support alpha masks, # but we still want to support binary masks. - image_mask = create_binary_mask(image_mask) + image_mask = create_binary_mask(image_mask, round=(self.soft_inpainting is None)) if self.inpainting_mask_invert: image_mask = ImageOps.invert(image_mask) @@ -1489,6 +1512,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image_mask = Image.fromarray(np_mask) if self.inpaint_full_res: + self.mask_for_overlay = image_mask if self.soft_inpainting is None else None mask = image_mask.convert('L') crop_region = masking.get_crop_region(np.array(mask), self.inpaint_full_res_padding) crop_region = masking.expand_crop_region(crop_region, self.width, self.height, mask.width, mask.height) @@ -1500,7 +1524,12 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): else: image_mask = images.resize_image(self.resize_mode, image_mask, self.width, self.height) - self.masks_for_overlay = [] + if self.soft_inpainting is None: + np_mask = np.array(image_mask) + np_mask = np.clip((np_mask.astype(np.float32)) * 2, 0, 255).astype(np.uint8) + self.mask_for_overlay = Image.fromarray(np_mask) + + self.masks_for_overlay = [] if self.soft_inpainting is not None else None self.overlay_images = [] latent_mask = self.latent_mask if self.latent_mask is not None else image_mask @@ -1522,8 +1551,15 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image = images.resize_image(self.resize_mode, image, self.width, self.height) if image_mask is not None: - self.overlay_images.append(image) - self.masks_for_overlay.append(image_mask) + if self.soft_inpainting is not None: + # We apply the masks AFTER to adjust mask based on changed content. + self.overlay_images.append(image) + self.masks_for_overlay.append(image_mask) + else: + image_masked = Image.new('RGBa', (image.width, image.height)) + image_masked.paste(image.convert("RGBA").convert("RGBa"), + mask=ImageOps.invert(self.mask_for_overlay.convert('L'))) + self.overlay_images.append(image_masked.convert('RGBA')) # crop_region is not None if we are doing inpaint full res if crop_region is not None: @@ -1576,6 +1612,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): latmask = init_mask.convert('RGB').resize((self.init_latent.shape[3], self.init_latent.shape[2])) latmask = np.moveaxis(np.array(latmask, dtype=np.float32), 2, 0) / 255 latmask = latmask[0] + if self.soft_inpainting is None: + latmask = np.around(latmask) latmask = np.tile(latmask[None], (4, 1, 1)) self.mask = torch.asarray(1.0 - latmask).to(shared.device).type(self.sd_model.dtype) @@ -1587,7 +1625,10 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): elif self.inpainting_fill == 3: self.init_latent = self.init_latent * self.mask - self.image_conditioning = self.img2img_image_conditioning(image * 2 - 1, self.init_latent, image_mask) + self.image_conditioning = self.img2img_image_conditioning(image * 2 - 1, + self.init_latent, + image_mask, + self.soft_inpainting is None) def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts): x = self.rng.next() -- cgit v1.2.1 From 15322e1b1a9e31edcc2f7d72a32d02365058737d Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 4 Dec 2023 12:36:41 +0300 Subject: repair old handler for postprocessing API --- modules/postprocessing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/postprocessing.py b/modules/postprocessing.py index 3c85a74c..d166f859 100644 --- a/modules/postprocessing.py +++ b/modules/postprocessing.py @@ -153,4 +153,4 @@ def run_extras(extras_mode, resize_mode, image, image_folder, input_dir, output_ }, }) - return run_postprocessing(extras_mode, image, image_folder, input_dir, output_dir, show_extras_results, *args, save_output=save_output) + return run_postprocessing("", extras_mode, image, image_folder, input_dir, output_dir, show_extras_results, *args, save_output=save_output) -- cgit v1.2.1 From 883d6a2b34a2817304d23c2481a6f9fc56687a53 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 4 Dec 2023 13:11:00 +0300 Subject: repair old handler for postprocessing API in a way that doesn't break interface --- modules/postprocessing.py | 8 ++++++-- modules/ui_postprocessing.py | 2 +- 2 files changed, 7 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/postprocessing.py b/modules/postprocessing.py index d166f859..0c59fad4 100644 --- a/modules/postprocessing.py +++ b/modules/postprocessing.py @@ -6,7 +6,7 @@ from modules import shared, images, devices, scripts, scripts_postprocessing, ui from modules.shared import opts -def run_postprocessing(id_task, extras_mode, image, image_folder, input_dir, output_dir, show_extras_results, *args, save_output: bool = True): +def run_postprocessing(extras_mode, image, image_folder, input_dir, output_dir, show_extras_results, *args, save_output: bool = True): devices.torch_gc() shared.state.begin(job="extras") @@ -128,6 +128,10 @@ def run_postprocessing(id_task, extras_mode, image, image_folder, input_dir, out return outputs, ui_common.plaintext_to_html(infotext), '' +def run_postprocessing_webui(id_task, *args, **kwargs): + return run_postprocessing(*args, **kwargs) + + def run_extras(extras_mode, resize_mode, image, image_folder, input_dir, output_dir, show_extras_results, gfpgan_visibility, codeformer_visibility, codeformer_weight, upscaling_resize, upscaling_resize_w, upscaling_resize_h, upscaling_crop, extras_upscaler_1, extras_upscaler_2, extras_upscaler_2_visibility, upscale_first: bool, save_output: bool = True): """old handler for API""" @@ -153,4 +157,4 @@ def run_extras(extras_mode, resize_mode, image, image_folder, input_dir, output_ }, }) - return run_postprocessing("", extras_mode, image, image_folder, input_dir, output_dir, show_extras_results, *args, save_output=save_output) + return run_postprocessing(extras_mode, image, image_folder, input_dir, output_dir, show_extras_results, *args, save_output=save_output) diff --git a/modules/ui_postprocessing.py b/modules/ui_postprocessing.py index fbad0800..13d888e4 100644 --- a/modules/ui_postprocessing.py +++ b/modules/ui_postprocessing.py @@ -35,7 +35,7 @@ def create_ui(): tab_batch_dir.select(fn=lambda: 2, inputs=[], outputs=[tab_index]) submit.click( - fn=call_queue.wrap_gradio_gpu_call(postprocessing.run_postprocessing, extra_outputs=[None, '']), + fn=call_queue.wrap_gradio_gpu_call(postprocessing.run_postprocessing_webui, extra_outputs=[None, '']), _js="submit_extras", inputs=[ dummy_component, -- cgit v1.2.1 From 854f8c318c2610c76259056ab02739176aa849e8 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Tue, 5 Dec 2023 04:40:12 +0900 Subject: remove clean_text() --- modules/styles.py | 23 +++-------------------- 1 file changed, 3 insertions(+), 20 deletions(-) (limited to 'modules') diff --git a/modules/styles.py b/modules/styles.py index 4d218cd7..7fb6c2e1 100644 --- a/modules/styles.py +++ b/modules/styles.py @@ -2,7 +2,6 @@ import csv import fnmatch import os import os.path -import re import typing import shutil @@ -14,22 +13,6 @@ class PromptStyle(typing.NamedTuple): path: str = None -def clean_text(text: str) -> str: - """ - Iterating through a list of regular expressions and replacement strings, we - clean up the prompt and style text to make it easier to match against each - other. - """ - re_list = [ - ("multiple commas", re.compile("(,+\s+)+,?"), ", "), - ("multiple spaces", re.compile("\s{2,}"), " "), - ] - for _, regex, replace in re_list: - text = regex.sub(replace, text) - - return text.strip(", ") - - def merge_prompts(style_prompt: str, prompt: str) -> str: if "{prompt}" in style_prompt: res = style_prompt.replace("{prompt}", prompt) @@ -44,7 +27,7 @@ def apply_styles_to_prompt(prompt, styles): for style in styles: prompt = merge_prompts(style, prompt) - return clean_text(prompt) + return prompt def unwrap_style_text_from_prompt(style_text, prompt): @@ -56,8 +39,8 @@ def unwrap_style_text_from_prompt(style_text, prompt): Note that the "cleaned" version of the style text is only used for matching purposes here. It isn't returned; the original style text is not modified. """ - stripped_prompt = clean_text(prompt) - stripped_style_text = clean_text(style_text) + stripped_prompt = prompt + stripped_style_text = style_text if "{prompt}" in stripped_style_text: # Work out whether the prompt is wrapped in the style text. If so, we # return True and the "inner" prompt text that isn't part of the style. -- cgit v1.2.1 From 976c1053efeb5054692ed3cfa294cf79196f3946 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 16:06:58 -0700 Subject: Cleaned up code, moved main code contributions into soft_inpainting.py --- modules/processing.py | 56 ++---------- modules/sd_samplers_cfg_denoiser.py | 84 ++--------------- modules/soft_inpainting.py | 177 ++++++++++++++++++++++++++++++++---- modules/ui.py | 7 -- 4 files changed, 174 insertions(+), 150 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index b40b1a40..0b360387 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -892,55 +892,13 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: # Generate the mask(s) based on similarity between the original and denoised latent vectors if getattr(p, "image_mask", None) is not None and getattr(p, "soft_inpainting", None) is not None: - # latent_mask = p.nmask[0].float().cpu() - - # convert the original mask into a form we use to scale distances for thresholding - # mask_scalar = 1-(torch.clamp(latent_mask, min=0, max=1) ** (p.mask_blend_scale / 2)) - # mask_scalar = mask_scalar / (1.00001-mask_scalar) - # mask_scalar = mask_scalar.numpy() - - latent_orig = p.init_latent - latent_proc = samples_ddim - latent_distance = torch.norm(latent_proc - latent_orig, p=2, dim=1) - - kernel, kernel_center = images.get_gaussian_kernel(stddev_radius=1.5, max_radius=2) - - for i, (distance_map, overlay_image) in enumerate(zip(latent_distance, p.overlay_images)): - converted_mask = distance_map.float().cpu().numpy() - converted_mask = images.weighted_histogram_filter(converted_mask, kernel, kernel_center, - percentile_min=0.9, percentile_max=1, min_width=1) - converted_mask = images.weighted_histogram_filter(converted_mask, kernel, kernel_center, - percentile_min=0.25, percentile_max=0.75, min_width=1) - - # The distance at which opacity of original decreases to 50% - # half_weighted_distance = 1 # * mask_scalar - # converted_mask = converted_mask / half_weighted_distance - - converted_mask = 1 / (1 + converted_mask ** 2) - converted_mask = images.smootherstep(converted_mask) - converted_mask = 1 - converted_mask - converted_mask = 255. * converted_mask - converted_mask = converted_mask.astype(np.uint8) - converted_mask = Image.fromarray(converted_mask) - converted_mask = images.resize_image(2, converted_mask, p.width, p.height) - converted_mask = create_binary_mask(converted_mask, round=False) - - # Remove aliasing artifacts using a gaussian blur. - converted_mask = converted_mask.filter(ImageFilter.GaussianBlur(radius=4)) - - # Expand the mask to fit the whole image if needed. - if p.paste_to is not None: - converted_mask = uncrop(converted_mask, - (overlay_image.width, overlay_image.height), - p.paste_to) - - p.masks_for_overlay[i] = converted_mask - - image_masked = Image.new('RGBa', (overlay_image.width, overlay_image.height)) - image_masked.paste(overlay_image.convert("RGBA").convert("RGBa"), - mask=ImageOps.invert(converted_mask.convert('L'))) - - p.overlay_images[i] = image_masked.convert('RGBA') + si.generate_adaptive_masks(latent_orig=p.init_latent, + latent_processed=samples_ddim, + overlay_images=p.overlay_images, + masks_for_overlay=p.masks_for_overlay, + width=p.width, + height=p.height, + paste_to=p.paste_to) x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index 0ee0b7dd..a700e692 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -94,76 +94,6 @@ class CFGDenoiser(torch.nn.Module): self.sampler.sampler_extra_args['uncond'] = uc def forward(self, x, sigma, uncond, cond, cond_scale, s_min_uncond, image_cond): - def latent_blend(a, b, t, one_minus_t=None): - - """ - Interpolates two latent image representations according to the parameter t, - where the interpolated vectors' magnitudes are also interpolated separately. - The "detail_preservation" factor biases the magnitude interpolation towards - the larger of the two magnitudes. - """ - # NOTE: We use inplace operations wherever possible. - - if one_minus_t is None: - one_minus_t = 1 - t - - if self.soft_inpainting is None: - return a * one_minus_t + b * t - - # Linearly interpolate the image vectors. - a_scaled = a * one_minus_t - b_scaled = b * t - image_interp = a_scaled - image_interp.add_(b_scaled) - result_type = image_interp.dtype - del a_scaled, b_scaled - - # Calculate the magnitude of the interpolated vectors. (We will remove this magnitude.) - # 64-bit operations are used here to allow large exponents. - current_magnitude = torch.norm(image_interp, p=2, dim=1).to(torch.float64).add_(0.00001) - - # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). - a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64).pow_(self.soft_inpainting.inpaint_detail_preservation) * one_minus_t - b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64).pow_(self.soft_inpainting.inpaint_detail_preservation) * t - desired_magnitude = a_magnitude - desired_magnitude.add_(b_magnitude).pow_(1 / self.soft_inpainting.inpaint_detail_preservation) - del a_magnitude, b_magnitude, one_minus_t - - # Change the linearly interpolated image vectors' magnitudes to the value we want. - # This is the last 64-bit operation. - image_interp_scaling_factor = desired_magnitude - image_interp_scaling_factor.div_(current_magnitude) - image_interp_scaled = image_interp - image_interp_scaled.mul_(image_interp_scaling_factor) - del current_magnitude - del desired_magnitude - del image_interp - del image_interp_scaling_factor - - image_interp_scaled = image_interp_scaled.to(result_type) - del result_type - - return image_interp_scaled - - def get_modified_nmask(nmask, _sigma): - """ - Converts a negative mask representing the transparency of the original latent vectors being overlayed - to a mask that is scaled according to the denoising strength for this step. - - Where: - 0 = fully opaque, infinite density, fully masked - 1 = fully transparent, zero density, fully unmasked - - We bring this transparency to a power, as this allows one to simulate N number of blending operations - where N can be any positive real value. Using this one can control the balance of influence between - the denoiser and the original latents according to the sigma value. - - NOTE: "mask" is not used - """ - if self.soft_inpainting is None: - return nmask - - return torch.pow(nmask, (_sigma ** self.soft_inpainting.mask_blend_power) * self.soft_inpainting.mask_blend_scale) if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException @@ -184,9 +114,12 @@ class CFGDenoiser(torch.nn.Module): # Blend in the original latents (before) if self.mask_before_denoising and self.mask is not None: if self.soft_inpainting is None: - x = latent_blend(self.init_latent, x, self.nmask, self.mask) + x = self.init_latent * self.mask + self.nmask * x else: - x = latent_blend(self.init_latent, x, get_modified_nmask(self.nmask, sigma)) + x = si.latent_blend(self.soft_inpainting, + self.init_latent, + x, + si.get_modified_nmask(self.soft_inpainting, self.nmask, sigma)) batch_size = len(conds_list) repeats = [len(conds_list[i]) for i in range(batch_size)] @@ -290,9 +223,12 @@ class CFGDenoiser(torch.nn.Module): # Blend in the original latents (after) if not self.mask_before_denoising and self.mask is not None: if self.soft_inpainting is None: - denoised = latent_blend(self.init_latent, denoised, self.nmask, self.mask) + denoised = self.init_latent * self.mask + self.nmask * denoised else: - denoised = latent_blend(self.init_latent, denoised, get_modified_nmask(self.nmask, sigma)) + denoised = si.latent_blend(self.soft_inpainting, + self.init_latent, + denoised, + si.get_modified_nmask(self.soft_inpainting, self.nmask, sigma)) self.sampler.last_latent = self.get_pred_x0(torch.cat([x_in[i:i + 1] for i in denoised_image_indexes]), torch.cat([x_out[i:i + 1] for i in denoised_image_indexes]), sigma) diff --git a/modules/soft_inpainting.py b/modules/soft_inpainting.py index 259c36ec..b81c8dd9 100644 --- a/modules/soft_inpainting.py +++ b/modules/soft_inpainting.py @@ -4,13 +4,6 @@ class SoftInpaintingSettings: self.mask_blend_scale = mask_blend_scale self.inpaint_detail_preservation = inpaint_detail_preservation - def get_paste_fields(self): - return [ - (self.mask_blend_power, gen_param_labels.mask_blend_power), - (self.mask_blend_scale, gen_param_labels.mask_blend_scale), - (self.inpaint_detail_preservation, gen_param_labels.inpaint_detail_preservation), - ] - def add_generation_params(self, dest): dest[enabled_gen_param_label] = True dest[gen_param_labels.mask_blend_power] = self.mask_blend_power @@ -18,25 +11,169 @@ class SoftInpaintingSettings: dest[gen_param_labels.inpaint_detail_preservation] = self.inpaint_detail_preservation +# ------------------- Methods ------------------- + + +def latent_blend(soft_inpainting, a, b, t): + """ + Interpolates two latent image representations according to the parameter t, + where the interpolated vectors' magnitudes are also interpolated separately. + The "detail_preservation" factor biases the magnitude interpolation towards + the larger of the two magnitudes. + """ + import torch + + # NOTE: We use inplace operations wherever possible. + + one_minus_t = 1 - t + + # Linearly interpolate the image vectors. + a_scaled = a * one_minus_t + b_scaled = b * t + image_interp = a_scaled + image_interp.add_(b_scaled) + result_type = image_interp.dtype + del a_scaled, b_scaled + + # Calculate the magnitude of the interpolated vectors. (We will remove this magnitude.) + # 64-bit operations are used here to allow large exponents. + current_magnitude = torch.norm(image_interp, p=2, dim=1).to(torch.float64).add_(0.00001) + + # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). + a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64).pow_(soft_inpainting.inpaint_detail_preservation) * one_minus_t + b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64).pow_(soft_inpainting.inpaint_detail_preservation) * t + desired_magnitude = a_magnitude + desired_magnitude.add_(b_magnitude).pow_(1 / soft_inpainting.inpaint_detail_preservation) + del a_magnitude, b_magnitude, one_minus_t + + # Change the linearly interpolated image vectors' magnitudes to the value we want. + # This is the last 64-bit operation. + image_interp_scaling_factor = desired_magnitude + image_interp_scaling_factor.div_(current_magnitude) + image_interp_scaling_factor = image_interp_scaling_factor.to(result_type) + image_interp_scaled = image_interp + image_interp_scaled.mul_(image_interp_scaling_factor) + del current_magnitude + del desired_magnitude + del image_interp + del image_interp_scaling_factor + del result_type + + return image_interp_scaled + + +def get_modified_nmask(soft_inpainting, nmask, sigma): + """ + Converts a negative mask representing the transparency of the original latent vectors being overlayed + to a mask that is scaled according to the denoising strength for this step. + + Where: + 0 = fully opaque, infinite density, fully masked + 1 = fully transparent, zero density, fully unmasked + + We bring this transparency to a power, as this allows one to simulate N number of blending operations + where N can be any positive real value. Using this one can control the balance of influence between + the denoiser and the original latents according to the sigma value. + + NOTE: "mask" is not used + """ + import torch + return torch.pow(nmask, (sigma ** soft_inpainting.mask_blend_power) * soft_inpainting.mask_blend_scale) + + +def generate_adaptive_masks( + latent_orig, + latent_processed, + overlay_images, + masks_for_overlay, + width, height, + paste_to): + import torch + import numpy as np + import modules.processing as proc + import modules.images as images + from PIL import Image, ImageOps, ImageFilter + + # TODO: Bias the blending according to the latent mask, add adjustable parameter for bias control. + # latent_mask = p.nmask[0].float().cpu() + # convert the original mask into a form we use to scale distances for thresholding + # mask_scalar = 1-(torch.clamp(latent_mask, min=0, max=1) ** (p.mask_blend_scale / 2)) + # mask_scalar = mask_scalar / (1.00001-mask_scalar) + # mask_scalar = mask_scalar.numpy() + + latent_distance = torch.norm(latent_processed - latent_orig, p=2, dim=1) + + kernel, kernel_center = images.get_gaussian_kernel(stddev_radius=1.5, max_radius=2) + + for i, (distance_map, overlay_image) in enumerate(zip(latent_distance, overlay_images)): + converted_mask = distance_map.float().cpu().numpy() + converted_mask = images.weighted_histogram_filter(converted_mask, kernel, kernel_center, + percentile_min=0.9, percentile_max=1, min_width=1) + converted_mask = images.weighted_histogram_filter(converted_mask, kernel, kernel_center, + percentile_min=0.25, percentile_max=0.75, min_width=1) + + # The distance at which opacity of original decreases to 50% + # half_weighted_distance = 1 # * mask_scalar + # converted_mask = converted_mask / half_weighted_distance + + converted_mask = 1 / (1 + converted_mask ** 2) + converted_mask = images.smootherstep(converted_mask) + converted_mask = 1 - converted_mask + converted_mask = 255. * converted_mask + converted_mask = converted_mask.astype(np.uint8) + converted_mask = Image.fromarray(converted_mask) + converted_mask = images.resize_image(2, converted_mask, width, height) + converted_mask = proc.create_binary_mask(converted_mask, round=False) + + # Remove aliasing artifacts using a gaussian blur. + converted_mask = converted_mask.filter(ImageFilter.GaussianBlur(radius=4)) + + # Expand the mask to fit the whole image if needed. + if paste_to is not None: + converted_mask = proc. uncrop(converted_mask, + (overlay_image.width, overlay_image.height), + paste_to) + + masks_for_overlay[i] = converted_mask + + image_masked = Image.new('RGBa', (overlay_image.width, overlay_image.height)) + image_masked.paste(overlay_image.convert("RGBA").convert("RGBa"), + mask=ImageOps.invert(converted_mask.convert('L'))) + + overlay_images[i] = image_masked.convert('RGBA') + + +# ------------------- Constants ------------------- + + +default = SoftInpaintingSettings(1, 0.5, 4) + enabled_ui_label = "Soft inpainting" enabled_gen_param_label = "Soft inpainting enabled" enabled_el_id = "soft_inpainting_enabled" -default = SoftInpaintingSettings(1, 0.5, 4) -ui_labels = SoftInpaintingSettings("Schedule bias", "Preservation strength", "Transition contrast boost") +ui_labels = SoftInpaintingSettings( + "Schedule bias", + "Preservation strength", + "Transition contrast boost") ui_info = SoftInpaintingSettings( - mask_blend_power="Shifts when preservation of original content occurs during denoising.", - # "Below 1: Stronger preservation near the end (with low sigma)\n" - # "1: Balanced (proportional to sigma)\n" - # "Above 1: Stronger preservation in the beginning (with high sigma)", - mask_blend_scale="How strongly partially masked content should be preserved.", - # "Low values: Favors generated content.\n" - # "High values: Favors original content.", - inpaint_detail_preservation="Amplifies the contrast that may be lost in partially masked regions.") - -gen_param_labels = SoftInpaintingSettings("Soft inpainting schedule bias", "Soft inpainting preservation strength", "Soft inpainting transition contrast boost") -el_ids = SoftInpaintingSettings("mask_blend_power", "mask_blend_scale", "inpaint_detail_preservation") + "Shifts when preservation of original content occurs during denoising.", + "How strongly partially masked content should be preserved.", + "Amplifies the contrast that may be lost in partially masked regions.") + +gen_param_labels = SoftInpaintingSettings( + "Soft inpainting schedule bias", + "Soft inpainting preservation strength", + "Soft inpainting transition contrast boost") + +el_ids = SoftInpaintingSettings( + "mask_blend_power", + "mask_blend_scale", + "inpaint_detail_preservation") + + +# ------------------- UI ------------------- def gradio_ui(): diff --git a/modules/ui.py b/modules/ui.py index 0e4fb17a..4f1265a3 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -683,13 +683,6 @@ def create_ui(): with FormRow(): soft_inpainting = si.gradio_ui() - - """ - mask_blend_power = gr.Slider(label='Blending bias', minimum=0, maximum=8, step=0.1, value=1, elem_id="img2img_mask_blend_power") - mask_blend_scale = gr.Slider(label='Blending preservation', minimum=0, maximum=8, step=0.05, value=0.5, elem_id="img2img_mask_blend_scale") - inpaint_detail_preservation = gr.Slider(label='Blending contrast boost', minimum=1, maximum=32, step=0.5, value=4, elem_id="img2img_mask_blend_offset") - """ - with FormRow(): inpainting_mask_invert = gr.Radio(label='Mask mode', choices=['Inpaint masked', 'Inpaint not masked'], value='Inpaint masked', type="index", elem_id="img2img_mask_mode") -- cgit v1.2.1 From 1455159cf44cd8c21656818463f6095eae887540 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 16:43:57 -0700 Subject: Fixed issue with whitespace, removed commented out code that was meant to be used as a reference. --- modules/soft_inpainting.py | 20 ++++++-------------- 1 file changed, 6 insertions(+), 14 deletions(-) (limited to 'modules') diff --git a/modules/soft_inpainting.py b/modules/soft_inpainting.py index b81c8dd9..56a87774 100644 --- a/modules/soft_inpainting.py +++ b/modules/soft_inpainting.py @@ -179,15 +179,7 @@ el_ids = SoftInpaintingSettings( def gradio_ui(): import gradio as gr from modules.ui_components import InputAccordion - """ - with InputAccordion(False, label="Refiner", elem_id=self.elem_id("enable")) as enable_refiner: - with gr.Row(): - refiner_checkpoint = gr.Dropdown(label='Checkpoint', elem_id=self.elem_id("checkpoint"), choices=sd_models.checkpoint_tiles(), value='', tooltip="switch to another model in the middle of generation") - create_refresh_button(refiner_checkpoint, sd_models.list_models, lambda: {"choices": sd_models.checkpoint_tiles()}, self.elem_id("checkpoint_refresh")) - - refiner_switch_at = gr.Slider(value=0.8, label="Switch at", minimum=0.01, maximum=1.0, step=0.01, elem_id=self.elem_id("switch_at"), tooltip="fraction of sampling steps when the switch to refiner model should happen; 1=never, 0.5=switch in the middle of generation") - """ with InputAccordion(False, label=enabled_ui_label, elem_id=enabled_el_id) as soft_inpainting_enabled: with gr.Group(): gr.Markdown( @@ -223,11 +215,11 @@ def gradio_ui(): gr.Markdown( f""" ### {ui_labels.mask_blend_power} - + The blending strength of original content is scaled proportionally with the decreasing noise level values at each step (sigmas). This ensures that the influence of the denoiser and original content preservation is roughly balanced at each step. This balance can be shifted using this parameter, controlling whether earlier or later steps have stronger preservation. - + - **Below 1**: Stronger preservation near the end (with low sigma) - **1**: Balanced (proportional to sigma) - **Above 1**: Stronger preservation in the beginning (with high sigma) @@ -235,21 +227,21 @@ def gradio_ui(): gr.Markdown( f""" ### {ui_labels.mask_blend_scale} - + Skews whether partially masked image regions should be more likely to preserve the original content or favor inpainted content. This may need to be adjusted depending on the {ui_labels.mask_blend_power}, CFG Scale, prompt and Denoising strength. - + - **Low values**: Favors generated content. - **High values**: Favors original content. """) gr.Markdown( f""" ### {ui_labels.inpaint_detail_preservation} - + This parameter controls how the original latent vectors and denoised latent vectors are interpolated. With higher values, the magnitude of the resulting blended vector will be closer to the maximum of the two interpolated vectors. This can prevent the loss of contrast that occurs with linear interpolation. - + - **Low values**: Softer blending, details may fade. - **High values**: Stronger contrast, may over-saturate colors. """) -- cgit v1.2.1 From 57f29bd61dc30f1a8c94ead9b780f4655f7d7d6d Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 17:41:18 -0700 Subject: Re-introduce latent blending step from the vanilla inpainting procedure. --- modules/processing.py | 3 +++ 1 file changed, 3 insertions(+) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 0b360387..c8dc4d93 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1597,6 +1597,9 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): samples = self.sampler.sample_img2img(self, self.init_latent, x, conditioning, unconditional_conditioning, image_conditioning=self.image_conditioning) + if self.mask is not None and self.soft_inpainting is None: + samples = samples * self.nmask + self.init_latent * self.mask + del x devices.torch_gc() -- cgit v1.2.1 From 60c602232fd760fb548fb0b3d18b5297f8823c2a Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 17:41:51 -0700 Subject: Restored original formatting. --- modules/processing.py | 36 +++++++++++------------------------- 1 file changed, 11 insertions(+), 25 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index c8dc4d93..90ae249a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -370,10 +370,7 @@ class StableDiffusionProcessing: return self.edit_image_conditioning(source_image) if self.sampler.conditioning_key in {'hybrid', 'concat'}: - return self.inpainting_image_conditioning(source_image, - latent_image, - image_mask=image_mask, - round_image_mask=round_image_mask) + return self.inpainting_image_conditioning(source_image, latent_image, image_mask=image_mask, round_image_mask=round_image_mask) if self.sampler.conditioning_key == "crossattn-adm": return self.unclip_image_conditioning(source_image) @@ -885,7 +882,8 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if getattr(samples_ddim, 'already_decoded', False): x_samples_ddim = samples_ddim - # todo: generate masks the old fashioned way + # todo: generate adaptive masks based on pixel differences. + # if p.masks_for_overlay is used, it will already be populated with masks else: if opts.sd_vae_decode_method != 'Full': p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method @@ -900,9 +898,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: height=p.height, paste_to=p.paste_to) - x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, - target_device=devices.cpu, - check_for_nans=True) + x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) x_samples_ddim = torch.stack(x_samples_ddim).float() x_samples_ddim = torch.clamp((x_samples_ddim + 1.0) / 2.0, min=0.0, max=1.0) @@ -927,9 +923,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: x_samples_ddim = batch_params.images def infotext(index=0, use_main_prompt=False): - return create_infotext(p, p.prompts, p.seeds, p.subseeds, - use_main_prompt=use_main_prompt, index=index, - all_negative_prompts=p.negative_prompts) + return create_infotext(p, p.prompts, p.seeds, p.subseeds, use_main_prompt=use_main_prompt, index=index, all_negative_prompts=p.negative_prompts) save_samples = p.save_samples() @@ -972,8 +966,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: image = apply_overlay(image, p.paste_to, i, p.overlay_images) if save_samples: - images.save_image(image, p.outpath_samples, "", p.seeds[i], - p.prompts[i], opts.samples_format, info=infotext(i), p=p) + images.save_image(image, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p) text = infotext(i) infotexts.append(text) @@ -983,14 +976,10 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if save_samples and any([opts.save_mask, opts.save_mask_composite, opts.return_mask, opts.return_mask_composite]): if hasattr(p, 'masks_for_overlay') and p.masks_for_overlay: image_mask = p.masks_for_overlay[i].convert('RGB') - image_mask_composite = Image.composite( - original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), - images.resize_image(2, p.masks_for_overlay[i], image.width, image.height).convert('L')).convert('RGBA') + image_mask_composite = Image.composite(original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), images.resize_image(2, p.masks_for_overlay[i], image.width, image.height).convert('L')).convert('RGBA') elif hasattr(p, 'mask_for_overlay') and p.mask_for_overlay: image_mask = p.mask_for_overlay.convert('RGB') - image_mask_composite = Image.composite( - original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), - images.resize_image(2, p.mask_for_overlay, image.width, image.height).convert('L')).convert('RGBA') + image_mask_composite = Image.composite(original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), images.resize_image(2, p.mask_for_overlay, image.width, image.height).convert('L')).convert('RGBA') else: image_mask = None image_mask_composite = None @@ -1515,8 +1504,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): self.masks_for_overlay.append(image_mask) else: image_masked = Image.new('RGBa', (image.width, image.height)) - image_masked.paste(image.convert("RGBA").convert("RGBa"), - mask=ImageOps.invert(self.mask_for_overlay.convert('L'))) + image_masked.paste(image.convert("RGBA").convert("RGBa"), mask=ImageOps.invert(self.mask_for_overlay.convert('L'))) + self.overlay_images.append(image_masked.convert('RGBA')) # crop_region is not None if we are doing inpaint full res @@ -1583,10 +1572,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): elif self.inpainting_fill == 3: self.init_latent = self.init_latent * self.mask - self.image_conditioning = self.img2img_image_conditioning(image * 2 - 1, - self.init_latent, - image_mask, - self.soft_inpainting is None) + self.image_conditioning = self.img2img_image_conditioning(image * 2 - 1, self.init_latent, image_mask, self.soft_inpainting is None) def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts): x = self.rng.next() -- cgit v1.2.1 From b32a334e3da7b06d82441beaa08a673b4f55bca1 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 17:57:10 -0700 Subject: Applies a convert('RGBA') operation early to mimic previous behaviour. --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 90ae249a..7fc282cf 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1500,7 +1500,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): if image_mask is not None: if self.soft_inpainting is not None: # We apply the masks AFTER to adjust mask based on changed content. - self.overlay_images.append(image) + self.overlay_images.append(image.convert('RGBA')) self.masks_for_overlay.append(image_mask) else: image_masked = Image.new('RGBa', (image.width, image.height)) -- cgit v1.2.1 From 6fc12428e3c5f903584ca7986e0c441f80fa2807 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 19:42:59 -0700 Subject: Fixed issue where batched inpainting (batch size > 1) wouldn't work because of mismatched tensor sizes. The 'already_decoded' decoded case should also be handled correctly (tested indirectly). --- modules/processing.py | 23 ++++++++++------ modules/soft_inpainting.py | 66 +++++++++++++++++++++++++++++++++++++++------- 2 files changed, 71 insertions(+), 18 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 7fc282cf..71bb056a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -883,20 +883,27 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if getattr(samples_ddim, 'already_decoded', False): x_samples_ddim = samples_ddim # todo: generate adaptive masks based on pixel differences. - # if p.masks_for_overlay is used, it will already be populated with masks + if getattr(p, "image_mask", None) is not None and getattr(p, "soft_inpainting", None) is not None: + si.apply_masks(soft_inpainting=p.soft_inpainting, + nmask=p.nmask, + overlay_images=p.overlay_images, + masks_for_overlay=p.masks_for_overlay, + width=p.width, + height=p.height, + paste_to=p.paste_to) else: if opts.sd_vae_decode_method != 'Full': p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method # Generate the mask(s) based on similarity between the original and denoised latent vectors if getattr(p, "image_mask", None) is not None and getattr(p, "soft_inpainting", None) is not None: - si.generate_adaptive_masks(latent_orig=p.init_latent, - latent_processed=samples_ddim, - overlay_images=p.overlay_images, - masks_for_overlay=p.masks_for_overlay, - width=p.width, - height=p.height, - paste_to=p.paste_to) + si.apply_adaptive_masks(latent_orig=p.init_latent, + latent_processed=samples_ddim, + overlay_images=p.overlay_images, + masks_for_overlay=p.masks_for_overlay, + width=p.width, + height=p.height, + paste_to=p.paste_to) x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) diff --git a/modules/soft_inpainting.py b/modules/soft_inpainting.py index 56a87774..b36ac8fa 100644 --- a/modules/soft_inpainting.py +++ b/modules/soft_inpainting.py @@ -25,26 +25,32 @@ def latent_blend(soft_inpainting, a, b, t): # NOTE: We use inplace operations wherever possible. - one_minus_t = 1 - t + # [4][w][h] to [1][4][w][h] + t2 = t.unsqueeze(0) + # [4][w][h] to [1][1][w][h] - the [4] seem redundant. + t3 = t[0].unsqueeze(0).unsqueeze(0) + + one_minus_t2 = 1 - t2 + one_minus_t3 = 1 - t3 # Linearly interpolate the image vectors. - a_scaled = a * one_minus_t - b_scaled = b * t + a_scaled = a * one_minus_t2 + b_scaled = b * t2 image_interp = a_scaled image_interp.add_(b_scaled) result_type = image_interp.dtype - del a_scaled, b_scaled + del a_scaled, b_scaled, t2, one_minus_t2 # Calculate the magnitude of the interpolated vectors. (We will remove this magnitude.) # 64-bit operations are used here to allow large exponents. - current_magnitude = torch.norm(image_interp, p=2, dim=1).to(torch.float64).add_(0.00001) + current_magnitude = torch.norm(image_interp, p=2, dim=1, keepdim=True).to(torch.float64).add_(0.00001) # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). - a_magnitude = torch.norm(a, p=2, dim=1).to(torch.float64).pow_(soft_inpainting.inpaint_detail_preservation) * one_minus_t - b_magnitude = torch.norm(b, p=2, dim=1).to(torch.float64).pow_(soft_inpainting.inpaint_detail_preservation) * t + a_magnitude = torch.norm(a, p=2, dim=1, keepdim=True).to(torch.float64).pow_(soft_inpainting.inpaint_detail_preservation) * one_minus_t3 + b_magnitude = torch.norm(b, p=2, dim=1, keepdim=True).to(torch.float64).pow_(soft_inpainting.inpaint_detail_preservation) * t3 desired_magnitude = a_magnitude desired_magnitude.add_(b_magnitude).pow_(1 / soft_inpainting.inpaint_detail_preservation) - del a_magnitude, b_magnitude, one_minus_t + del a_magnitude, b_magnitude, t3, one_minus_t3 # Change the linearly interpolated image vectors' magnitudes to the value we want. # This is the last 64-bit operation. @@ -78,10 +84,11 @@ def get_modified_nmask(soft_inpainting, nmask, sigma): NOTE: "mask" is not used """ import torch - return torch.pow(nmask, (sigma ** soft_inpainting.mask_blend_power) * soft_inpainting.mask_blend_scale) + # todo: Why is sigma 2D? Both values are the same. + return torch.pow(nmask, (sigma[0] ** soft_inpainting.mask_blend_power) * soft_inpainting.mask_blend_scale) -def generate_adaptive_masks( +def apply_adaptive_masks( latent_orig, latent_processed, overlay_images, @@ -142,6 +149,45 @@ def generate_adaptive_masks( overlay_images[i] = image_masked.convert('RGBA') +def apply_masks( + soft_inpainting, + nmask, + overlay_images, + masks_for_overlay, + width, height, + paste_to): + import torch + import numpy as np + import modules.processing as proc + import modules.images as images + from PIL import Image, ImageOps, ImageFilter + + converted_mask = nmask[0].float() + converted_mask = torch.clamp(converted_mask, min=0, max=1).pow_(soft_inpainting.mask_blend_scale / 2) + converted_mask = 255. * converted_mask + converted_mask = converted_mask.cpu().numpy().astype(np.uint8) + converted_mask = Image.fromarray(converted_mask) + converted_mask = images.resize_image(2, converted_mask, width, height) + converted_mask = proc.create_binary_mask(converted_mask, round=False) + + # Remove aliasing artifacts using a gaussian blur. + converted_mask = converted_mask.filter(ImageFilter.GaussianBlur(radius=4)) + + # Expand the mask to fit the whole image if needed. + if paste_to is not None: + converted_mask = proc.uncrop(converted_mask, + (width, height), + paste_to) + + for i, overlay_image in enumerate(overlay_images): + masks_for_overlay[i] = converted_mask + + image_masked = Image.new('RGBa', (overlay_image.width, overlay_image.height)) + image_masked.paste(overlay_image.convert("RGBA").convert("RGBa"), + mask=ImageOps.invert(converted_mask.convert('L'))) + + overlay_images[i] = image_masked.convert('RGBA') + # ------------------- Constants ------------------- -- cgit v1.2.1 From 49bbf1140731036875573bb7c44aa7e74623c856 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Mon, 4 Dec 2023 19:47:40 -0700 Subject: Fixed unused import. --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 71bb056a..e1823ac3 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -9,7 +9,7 @@ from dataclasses import dataclass, field import torch import numpy as np -from PIL import Image, ImageOps, ImageFilter +from PIL import Image, ImageOps import random import cv2 from skimage import exposure -- cgit v1.2.1 From 895456c4a2e87f5fe3ee23b4482e68fce317a1ca Mon Sep 17 00:00:00 2001 From: Jabasukuriputo Wang Date: Tue, 5 Dec 2023 18:00:48 -0600 Subject: change state dict comparison to ref compare --- modules/sd_disable_initialization.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/sd_disable_initialization.py b/modules/sd_disable_initialization.py index 8863107a..273a7edd 100644 --- a/modules/sd_disable_initialization.py +++ b/modules/sd_disable_initialization.py @@ -215,7 +215,7 @@ class LoadStateDictOnMeta(ReplaceHelper): would be on the meta device. """ - if state_dict == sd: + if state_dict is sd: state_dict = {k: v.to(device="meta", dtype=v.dtype) for k, v in state_dict.items()} original(module, state_dict, strict=strict) -- cgit v1.2.1 From 672dc4efa8e0da38426b121e7c7216d0a8e465fd Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Wed, 6 Dec 2023 15:16:10 +0800 Subject: Fix forced reload --- modules/sd_models.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index dcf816b3..d0046f88 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -801,7 +801,7 @@ def reload_model_weights(sd_model=None, info=None, forced_reload=False): if check_fp8(sd_model) != devices.fp8: # load from state dict again to prevent extra numerical errors forced_reload = True - elif sd_model.sd_model_checkpoint == checkpoint_info.filename: + elif sd_model.sd_model_checkpoint == checkpoint_info.filename and not forced_reload: return sd_model sd_model = reuse_model_from_already_loaded(sd_model, checkpoint_info, timer) -- cgit v1.2.1 From 746783f7a47f38f728f221cc26fe04035d3ca66b Mon Sep 17 00:00:00 2001 From: Nuullll Date: Wed, 6 Dec 2023 20:55:42 +0800 Subject: [IPEX] Fix embedding Cast `torch.bmm` args into same `dtype`. Fixes the following error when using Text Inversion embedding (#14224): ``` RuntimeError: could not create a primitive descriptor for a matmul primitive ``` --- modules/xpu_specific.py | 3 +++ 1 file changed, 3 insertions(+) (limited to 'modules') diff --git a/modules/xpu_specific.py b/modules/xpu_specific.py index d933c790..ec1ad100 100644 --- a/modules/xpu_specific.py +++ b/modules/xpu_specific.py @@ -48,3 +48,6 @@ if has_xpu: CondFunc('torch.nn.modules.conv.Conv2d.forward', lambda orig_func, self, input: orig_func(self, input.to(self.weight.data.dtype)), lambda orig_func, self, input: input.dtype != self.weight.data.dtype) + CondFunc('torch.bmm', + lambda orig_func, input, mat2, out=None: orig_func(input.to(mat2.dtype), mat2, out=out), + lambda orig_func, input, mat2, out=None: input.dtype != mat2.dtype) -- cgit v1.2.1 From 9d2cbf8e97832662e446145d3961c39e78919d3d Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Wed, 6 Dec 2023 23:06:32 +0900 Subject: add option: Live preview in full page image viewer make #13459 "show the preview image in the modal view if available" optional --- modules/shared_options.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index e5de0d01..88cfdded 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -330,6 +330,7 @@ options_templates.update(options_section(('ui', "Live previews", "ui"), { "live_preview_content": OptionInfo("Prompt", "Live preview subject", gr.Radio, {"choices": ["Combined", "Prompt", "Negative prompt"]}), "live_preview_refresh_period": OptionInfo(1000, "Progressbar and preview update period").info("in milliseconds"), "live_preview_fast_interrupt": OptionInfo(False, "Return image with chosen live preview method on interrupt").info("makes interrupts faster"), + "js_live_preview_in_modal_lightbox": OptionInfo(True, "Show Live preview in full page image viewer"), })) options_templates.update(options_section(('sampler-params', "Sampler parameters", "sd"), { -- cgit v1.2.1 From e90d4334ad37024a802f4ef27069b625a6508f72 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Wed, 6 Dec 2023 16:54:42 -0700 Subject: A custom blending function can be provided by p, replacing the use of soft_inpainting. --- modules/sd_samplers_cfg_denoiser.py | 34 +++++++++++++++++----------------- modules/sd_samplers_common.py | 1 - 2 files changed, 17 insertions(+), 18 deletions(-) (limited to 'modules') diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index a700e692..f13e8dcc 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -6,7 +6,6 @@ import modules.shared as shared from modules.script_callbacks import CFGDenoiserParams, cfg_denoiser_callback from modules.script_callbacks import CFGDenoisedParams, cfg_denoised_callback from modules.script_callbacks import AfterCFGCallbackParams, cfg_after_cfg_callback -import modules.soft_inpainting as si def catenate_conds(conds): @@ -44,7 +43,6 @@ class CFGDenoiser(torch.nn.Module): self.model_wrap = None self.mask = None self.nmask = None - self.soft_inpainting: si.SoftInpaintingParameters = None self.init_latent = None self.steps = None """number of steps as specified by user in UI""" @@ -94,7 +92,6 @@ class CFGDenoiser(torch.nn.Module): self.sampler.sampler_extra_args['uncond'] = uc def forward(self, x, sigma, uncond, cond, cond_scale, s_min_uncond, image_cond): - if state.interrupted or state.skipped: raise sd_samplers_common.InterruptedException @@ -111,15 +108,24 @@ class CFGDenoiser(torch.nn.Module): assert not is_edit_model or all(len(conds) == 1 for conds in conds_list), "AND is not supported for InstructPix2Pix checkpoint (unless using Image CFG scale = 1.0)" + # If we use masks, blending between the denoised and original latent images occurs here. + def apply_blend(latent): + if hasattr(self.p, "denoiser_masked_blend_function") and callable(self.p.denoiser_masked_blend_function): + return self.p.denoiser_masked_blend_function( + self, + # Using an argument dictionary so that arguments can be added without breaking extensions. + args= + { + "denoiser": self, + "current_latent": latent, + "sigma": sigma + }) + else: + return self.init_latent * self.mask + self.nmask * latent + # Blend in the original latents (before) if self.mask_before_denoising and self.mask is not None: - if self.soft_inpainting is None: - x = self.init_latent * self.mask + self.nmask * x - else: - x = si.latent_blend(self.soft_inpainting, - self.init_latent, - x, - si.get_modified_nmask(self.soft_inpainting, self.nmask, sigma)) + x = apply_blend(x) batch_size = len(conds_list) repeats = [len(conds_list[i]) for i in range(batch_size)] @@ -222,13 +228,7 @@ class CFGDenoiser(torch.nn.Module): # Blend in the original latents (after) if not self.mask_before_denoising and self.mask is not None: - if self.soft_inpainting is None: - denoised = self.init_latent * self.mask + self.nmask * denoised - else: - denoised = si.latent_blend(self.soft_inpainting, - self.init_latent, - denoised, - si.get_modified_nmask(self.soft_inpainting, self.nmask, sigma)) + denoised = apply_blend(denoised) self.sampler.last_latent = self.get_pred_x0(torch.cat([x_in[i:i + 1] for i in denoised_image_indexes]), torch.cat([x_out[i:i + 1] for i in denoised_image_indexes]), sigma) diff --git a/modules/sd_samplers_common.py b/modules/sd_samplers_common.py index 9682bee3..58efcad2 100644 --- a/modules/sd_samplers_common.py +++ b/modules/sd_samplers_common.py @@ -277,7 +277,6 @@ class Sampler: self.model_wrap_cfg.p = p self.model_wrap_cfg.mask = p.mask if hasattr(p, 'mask') else None self.model_wrap_cfg.nmask = p.nmask if hasattr(p, 'nmask') else None - self.model_wrap_cfg.soft_inpainting = p.soft_inpainting if hasattr(p, 'soft_inpainting') else None self.model_wrap_cfg.step = 0 self.model_wrap_cfg.image_cfg_scale = getattr(p, 'image_cfg_scale', None) self.eta = p.eta if p.eta is not None else getattr(opts, self.eta_option_field, 0.0) -- cgit v1.2.1 From 4608f6236fc24d937f89500b2c9bf48484537cf9 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Wed, 6 Dec 2023 18:11:17 -0700 Subject: Removed changes in some scripts since the arguments for soft painting are no longer passed through the same path as "mask_blur". --- modules/img2img.py | 50 +------------------------------------------------- modules/ui.py | 7 ------- 2 files changed, 1 insertion(+), 56 deletions(-) (limited to 'modules') diff --git a/modules/img2img.py b/modules/img2img.py index 3aa8a9ce..c583290a 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -15,7 +15,6 @@ import modules.shared as shared import modules.processing as processing from modules.ui import plaintext_to_html import modules.scripts -import modules.soft_inpainting as si def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=False, scale_by=1.0, use_png_info=False, png_info_props=None, png_info_dir=None): @@ -147,48 +146,7 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=Fal return batch_results -def img2img(id_task: str, - mode: int, - prompt: str, - negative_prompt: str, - prompt_styles, - init_img, - sketch, - init_img_with_mask, - inpaint_color_sketch, - inpaint_color_sketch_orig, - init_img_inpaint, - init_mask_inpaint, - steps: int, - sampler_name: str, - mask_blur: int, - mask_alpha: float, - mask_blend_enabled: bool, - mask_blend_power: float, - mask_blend_scale: float, - inpaint_detail_preservation: float, - inpainting_fill: int, - n_iter: int, - batch_size: int, - cfg_scale: float, - image_cfg_scale: float, - denoising_strength: float, - selected_scale_tab: int, - height: int, - width: int, - scale_by: float, - resize_mode: int, - inpaint_full_res: bool, - inpaint_full_res_padding: int, - inpainting_mask_invert: int, - img2img_batch_input_dir: str, - img2img_batch_output_dir: str, - img2img_batch_inpaint_mask_dir: str, - override_settings_texts, - img2img_batch_use_png_info: bool, - img2img_batch_png_info_props: list, - img2img_batch_png_info_dir: str, - request: gr.Request, *args): +def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_styles, init_img, sketch, init_img_with_mask, inpaint_color_sketch, inpaint_color_sketch_orig, init_img_inpaint, init_mask_inpaint, steps: int, sampler_name: str, mask_blur: int, mask_alpha: float, inpainting_fill: int, n_iter: int, batch_size: int, cfg_scale: float, image_cfg_scale: float, denoising_strength: float, selected_scale_tab: int, height: int, width: int, scale_by: float, resize_mode: int, inpaint_full_res: bool, inpaint_full_res_padding: int, inpainting_mask_invert: int, img2img_batch_input_dir: str, img2img_batch_output_dir: str, img2img_batch_inpaint_mask_dir: str, override_settings_texts, img2img_batch_use_png_info: bool, img2img_batch_png_info_props: list, img2img_batch_png_info_dir: str, request: gr.Request, *args): override_settings = create_override_settings_dict(override_settings_texts) is_batch = mode == 5 @@ -229,9 +187,6 @@ def img2img(id_task: str, assert 0. <= denoising_strength <= 1., 'can only work with strength in [0.0, 1.0]' - soft_inpainting = si.SoftInpaintingSettings(mask_blend_power, mask_blend_scale, inpaint_detail_preservation) \ - if mask_blend_enabled else None - p = StableDiffusionProcessingImg2Img( sd_model=shared.sd_model, outpath_samples=opts.outdir_samples or opts.outdir_img2img_samples, @@ -249,7 +204,6 @@ def img2img(id_task: str, init_images=[image], mask=mask, mask_blur=mask_blur, - soft_inpainting=soft_inpainting, inpainting_fill=inpainting_fill, resize_mode=resize_mode, denoising_strength=denoising_strength, @@ -270,8 +224,6 @@ def img2img(id_task: str, if mask: p.extra_generation_params["Mask blur"] = mask_blur - if soft_inpainting is not None: - soft_inpainting.add_generation_params(p.extra_generation_params) with closing(p): if is_batch: diff --git a/modules/ui.py b/modules/ui.py index bd2091e1..d80486dd 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -29,7 +29,6 @@ import modules.shared as shared from modules import prompt_parser from modules.sd_hijack import model_hijack from modules.generation_parameters_copypaste import image_from_url_text -import modules.soft_inpainting as si create_setting_component = ui_settings.create_setting_component @@ -680,9 +679,6 @@ def create_ui(): mask_blur = gr.Slider(label='Mask blur', minimum=0, maximum=64, step=1, value=4, elem_id="img2img_mask_blur") mask_alpha = gr.Slider(label="Mask transparency", visible=False, elem_id="img2img_mask_alpha") - with FormRow(): - soft_inpainting = si.gradio_ui() - with FormRow(): inpainting_mask_invert = gr.Radio(label='Mask mode', choices=['Inpaint masked', 'Inpaint not masked'], value='Inpaint masked', type="index", elem_id="img2img_mask_mode") @@ -737,7 +733,6 @@ def create_ui(): sampler_name, mask_blur, mask_alpha, - *(soft_inpainting[0]), inpainting_fill, batch_count, batch_size, @@ -836,10 +831,8 @@ def create_ui(): (toprow.ui_styles.dropdown, lambda d: d["Styles array"] if isinstance(d.get("Styles array"), list) else gr.update()), (denoising_strength, "Denoising strength"), (mask_blur, "Mask blur"), - *(soft_inpainting[1]), *scripts.scripts_img2img.infotext_fields ] - parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields, override_settings) parameters_copypaste.add_paste_fields("inpaint", init_img_with_mask, img2img_paste_fields, override_settings) parameters_copypaste.register_paste_params_button(parameters_copypaste.ParamBinding( -- cgit v1.2.1 From ac4578912395627731f2cd8529f87a95df1f7644 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Wed, 6 Dec 2023 21:16:27 -0700 Subject: Removed soft inpainting, added hooks for softpainting to work instead. --- modules/processing.py | 94 +++++++++++++++---------------------- modules/scripts.py | 70 +++++++++++++++++++++++++++ modules/sd_samplers_cfg_denoiser.py | 23 ++++----- 3 files changed, 118 insertions(+), 69 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 7d46949f..5a1a90af 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -30,7 +30,6 @@ import modules.sd_models as sd_models import modules.sd_vae as sd_vae from ldm.data.util import AddMiDaS from ldm.models.diffusion.ddpm import LatentDepth2ImageDiffusion -import modules.soft_inpainting as si from einops import repeat, rearrange from blendmodes.blend import blendLayers, BlendType @@ -73,12 +72,10 @@ def uncrop(image, dest_size, paste_loc): return image -def apply_overlay(image, paste_loc, index, overlays): - if overlays is None or index >= len(overlays): +def apply_overlay(image, paste_loc, overlay): + if overlay is None: return image - overlay = overlays[index] - if paste_loc is not None: image = uncrop(image, (overlay.width, overlay.height), paste_loc) @@ -150,7 +147,6 @@ class StableDiffusionProcessing: do_not_save_grid: bool = False extra_generation_params: dict[str, Any] = None overlay_images: list = None - masks_for_overlay: list = None eta: float = None do_not_reload_embeddings: bool = False denoising_strength: float = None @@ -880,31 +876,17 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: with devices.without_autocast() if devices.unet_needs_upcast else devices.autocast(): samples_ddim = p.sample(conditioning=p.c, unconditional_conditioning=p.uc, seeds=p.seeds, subseeds=p.subseeds, subseed_strength=p.subseed_strength, prompts=p.prompts) + if p.scripts is not None: + ps = scripts.PostSampleArgs(samples_ddim) + p.scripts.post_sample(p, ps) + samples_ddim = pp.samples + if getattr(samples_ddim, 'already_decoded', False): x_samples_ddim = samples_ddim - # todo: generate adaptive masks based on pixel differences. - if getattr(p, "image_mask", None) is not None and getattr(p, "soft_inpainting", None) is not None: - si.apply_masks(soft_inpainting=p.soft_inpainting, - nmask=p.nmask, - overlay_images=p.overlay_images, - masks_for_overlay=p.masks_for_overlay, - width=p.width, - height=p.height, - paste_to=p.paste_to) else: if opts.sd_vae_decode_method != 'Full': p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method - # Generate the mask(s) based on similarity between the original and denoised latent vectors - if getattr(p, "image_mask", None) is not None and getattr(p, "soft_inpainting", None) is not None: - si.apply_adaptive_masks(latent_orig=p.init_latent, - latent_processed=samples_ddim, - overlay_images=p.overlay_images, - masks_for_overlay=p.masks_for_overlay, - width=p.width, - height=p.height, - paste_to=p.paste_to) - x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) x_samples_ddim = torch.stack(x_samples_ddim).float() @@ -955,9 +937,18 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: pp = scripts.PostprocessImageArgs(image) p.scripts.postprocess_image(p, pp) image = pp.image + + mask_for_overlay = p.mask_for_overlay + overlay_image = p.overlay_images[i] if p.overlay_images is not None and i < len(p.overlay_images) else None + + if p.scripts is not None: + ppmo = scripts.PostProcessMaskOverlayArgs(i, mask_for_overlay, overlay_image) + p.scripts.postprocess_maskoverlay(p, ppmo) + mask_for_overlay, overlay_image = pp.mask_for_overlay, pp.overlay_image + if p.color_corrections is not None and i < len(p.color_corrections): if save_samples and opts.save_images_before_color_correction: - image_without_cc = apply_overlay(image, p.paste_to, i, p.overlay_images) + image_without_cc = apply_overlay(image, p.paste_to, overlay_image) images.save_image(image_without_cc, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p, suffix="-before-color-correction") image = apply_color_correction(p.color_corrections[i], image) @@ -968,9 +959,9 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: original_denoised_image = image.copy() if p.paste_to is not None: - original_denoised_image = uncrop(original_denoised_image, (p.overlay_images[i].width, p.overlay_images[i].height), p.paste_to) + original_denoised_image = uncrop(original_denoised_image, (p.overlay_image.width, p.overlay_image.height), p.paste_to) - image = apply_overlay(image, p.paste_to, i, p.overlay_images) + image = apply_overlay(image, p.paste_to, overlay_image) if save_samples: images.save_image(image, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p) @@ -981,13 +972,6 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: image.info["parameters"] = text output_images.append(image) - if hasattr(p, 'mask_for_overlay') and p.mask_for_overlay: - mask_for_overlay = p.mask_for_overlay - elif hasattr(p, 'masks_for_overlay') and p.masks_for_overlay and p.masks_for_overlay[i]: - mask_for_overlay = p.masks_for_overlay[i] - else: - mask_for_overlay = None - if mask_for_overlay is not None: if opts.return_mask or opts.save_mask: image_mask = mask_for_overlay.convert('RGB') @@ -1401,7 +1385,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): mask_blur_x: int = 4 mask_blur_y: int = 4 mask_blur: int = None - soft_inpainting: si.SoftInpaintingParameters = si.default + mask_round: bool = True inpainting_fill: int = 0 inpaint_full_res: bool = True inpaint_full_res_padding: int = 0 @@ -1447,7 +1431,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): if image_mask is not None: # image_mask is passed in as RGBA by Gradio to support alpha masks, # but we still want to support binary masks. - image_mask = create_binary_mask(image_mask, round=(self.soft_inpainting is None)) + image_mask = create_binary_mask(image_mask, round=self.mask_round) if self.inpainting_mask_invert: image_mask = ImageOps.invert(image_mask) @@ -1465,7 +1449,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image_mask = Image.fromarray(np_mask) if self.inpaint_full_res: - self.mask_for_overlay = image_mask if self.soft_inpainting is None else None + self.mask_for_overlay = image_mask mask = image_mask.convert('L') crop_region = masking.get_crop_region(np.array(mask), self.inpaint_full_res_padding) crop_region = masking.expand_crop_region(crop_region, self.width, self.height, mask.width, mask.height) @@ -1476,13 +1460,10 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): self.paste_to = (x1, y1, x2-x1, y2-y1) else: image_mask = images.resize_image(self.resize_mode, image_mask, self.width, self.height) + np_mask = np.array(image_mask) + np_mask = np.clip((np_mask.astype(np.float32)) * 2, 0, 255).astype(np.uint8) + self.mask_for_overlay = Image.fromarray(np_mask) - if self.soft_inpainting is None: - np_mask = np.array(image_mask) - np_mask = np.clip((np_mask.astype(np.float32)) * 2, 0, 255).astype(np.uint8) - self.mask_for_overlay = Image.fromarray(np_mask) - - self.masks_for_overlay = [] if self.soft_inpainting is not None else None self.overlay_images = [] latent_mask = self.latent_mask if self.latent_mask is not None else image_mask @@ -1504,15 +1485,10 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): image = images.resize_image(self.resize_mode, image, self.width, self.height) if image_mask is not None: - if self.soft_inpainting is not None: - # We apply the masks AFTER to adjust mask based on changed content. - self.overlay_images.append(image.convert('RGBA')) - self.masks_for_overlay.append(image_mask) - else: - image_masked = Image.new('RGBa', (image.width, image.height)) - image_masked.paste(image.convert("RGBA").convert("RGBa"), mask=ImageOps.invert(self.mask_for_overlay.convert('L'))) + image_masked = Image.new('RGBa', (image.width, image.height)) + image_masked.paste(image.convert("RGBA").convert("RGBa"), mask=ImageOps.invert(self.mask_for_overlay.convert('L'))) - self.overlay_images.append(image_masked.convert('RGBA')) + self.overlay_images.append(image_masked.convert('RGBA')) # crop_region is not None if we are doing inpaint full res if crop_region is not None: @@ -1565,7 +1541,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): latmask = init_mask.convert('RGB').resize((self.init_latent.shape[3], self.init_latent.shape[2])) latmask = np.moveaxis(np.array(latmask, dtype=np.float32), 2, 0) / 255 latmask = latmask[0] - if self.soft_inpainting is None: + if self.mask_round: latmask = np.around(latmask) latmask = np.tile(latmask[None], (4, 1, 1)) @@ -1578,7 +1554,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): elif self.inpainting_fill == 3: self.init_latent = self.init_latent * self.mask - self.image_conditioning = self.img2img_image_conditioning(image * 2 - 1, self.init_latent, image_mask, self.soft_inpainting is None) + self.image_conditioning = self.img2img_image_conditioning(image * 2 - 1, self.init_latent, image_mask, self.mask_round) def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts): x = self.rng.next() @@ -1589,8 +1565,14 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): samples = self.sampler.sample_img2img(self, self.init_latent, x, conditioning, unconditional_conditioning, image_conditioning=self.image_conditioning) - if self.mask is not None and self.soft_inpainting is None: - samples = samples * self.nmask + self.init_latent * self.mask + blended_samples = samples * self.nmask + self.init_latent * self.mask + + if self.scripts is not None: + mba = scripts.MaskBlendArgs(self, samples, self.nmask, self.init_latent, self.mask, blended_samples, sigma=None, is_final_blend=True) + self.scripts.on_mask_blend(self, mba) + blended_samples = mba.blended_latent + + samples = blended_samples del x devices.torch_gc() diff --git a/modules/scripts.py b/modules/scripts.py index 7f9454eb..92a07c56 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -11,11 +11,31 @@ from modules import shared, paths, script_callbacks, extensions, script_loading, AlwaysVisible = object() +class MaskBlendArgs: + def __init__(self, current_latent, nmask, init_latent, mask, blended_samples, denoiser=None, sigma=None): + self.current_latent = current_latent + self.nmask = nmask + self.init_latent = init_latent + self.mask = mask + self.blended_samples = blended_samples + + self.denoiser = denoiser + self.is_final_blend = denoiser is None + self.sigma = sigma + +class PostSampleArgs: + def __init__(self, samples): + self.samples = samples class PostprocessImageArgs: def __init__(self, image): self.image = image +class PostProcessMaskOverlayArgs: + def __init__(self, index, mask_for_overlay, overlay_image): + self.index = index + self.mask_for_overlay = mask_for_overlay + self.overlay_image = overlay_image class PostprocessBatchListArgs: def __init__(self, images): @@ -206,6 +226,25 @@ class Script: pass + def on_mask_blend(self, p, mba: MaskBlendArgs, *args): + """ + Called in inpainting mode when the original content is blended with the inpainted content. + This is called at every step in the denoising process and once at the end. + If is_final_blend is true, this is called for the final blending stage. + Otherwise, denoiser and sigma are defined and may be used to inform the procedure. + """ + + pass + + def post_sample(self, p, ps: PostSampleArgs, *args): + """ + Called after the samples have been generated, + but before they have been decoded by the VAE, if applicable. + Check getattr(samples, 'already_decoded', False) to test if the images are decoded. + """ + + pass + def postprocess_image(self, p, pp: PostprocessImageArgs, *args): """ Called for every image after it has been generated. @@ -213,6 +252,13 @@ class Script: pass + def postprocess_maskoverlay(self, p, ppmo: PostProcessMaskOverlayArgs, *args): + """ + Called for every image after it has been generated. + """ + + pass + def postprocess(self, p, processed, *args): """ This function is called after processing ends for AlwaysVisible scripts. @@ -767,6 +813,22 @@ class ScriptRunner: except Exception: errors.report(f"Error running postprocess_batch_list: {script.filename}", exc_info=True) + def post_sample(self, p, ps: PostSampleArgs): + for script in self.alwayson_scripts: + try: + script_args = p.script_args[script.args_from:script.args_to] + script.post_sample(p, ps, *script_args) + except Exception: + errors.report(f"Error running post_sample: {script.filename}", exc_info=True) + + def on_mask_blend(self, p, mba: MaskBlendArgs): + for script in self.alwayson_scripts: + try: + script_args = p.script_args[script.args_from:script.args_to] + script.on_mask_blend(p, mba, *script_args) + except Exception: + errors.report(f"Error running post_sample: {script.filename}", exc_info=True) + def postprocess_image(self, p, pp: PostprocessImageArgs): for script in self.alwayson_scripts: try: @@ -775,6 +837,14 @@ class ScriptRunner: except Exception: errors.report(f"Error running postprocess_image: {script.filename}", exc_info=True) + def postprocess_maskoverlay(self, p, ppmo: PostProcessMaskOverlayArgs): + for script in self.alwayson_scripts: + try: + script_args = p.script_args[script.args_from:script.args_to] + script.postprocess_maskoverlay(p, ppmo, *script_args) + except Exception: + errors.report(f"Error running postprocess_image: {script.filename}", exc_info=True) + def before_component(self, component, **kwargs): for callback, script in self.on_before_component_elem_id.get(kwargs.get("elem_id"), []): try: diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index f13e8dcc..eb9d5daf 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -109,19 +109,16 @@ class CFGDenoiser(torch.nn.Module): assert not is_edit_model or all(len(conds) == 1 for conds in conds_list), "AND is not supported for InstructPix2Pix checkpoint (unless using Image CFG scale = 1.0)" # If we use masks, blending between the denoised and original latent images occurs here. - def apply_blend(latent): - if hasattr(self.p, "denoiser_masked_blend_function") and callable(self.p.denoiser_masked_blend_function): - return self.p.denoiser_masked_blend_function( - self, - # Using an argument dictionary so that arguments can be added without breaking extensions. - args= - { - "denoiser": self, - "current_latent": latent, - "sigma": sigma - }) - else: - return self.init_latent * self.mask + self.nmask * latent + def apply_blend(current_latent): + blended_latent = current_latent * self.nmask + self.init_latent * self.mask + + if self.p.scripts is not None: + from modules import scripts + mba = scripts.MaskBlendArgs(current_latent, self.nmask, self.init_latent, self.mask, blended_latent, denoiser=self, sigma=sigma) + self.p.scripts.on_mask_blend(self.p, mba) + blended_latent = mba.blended_latent + + return blended_latent # Blend in the original latents (before) if self.mask_before_denoising and self.mask is not None: -- cgit v1.2.1 From 2abc417834d752e43a283f8603bfddfb1c80b30f Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Wed, 6 Dec 2023 22:25:53 -0700 Subject: Re-implemented soft inpainting via a script. Also fixed some mistakes with the previous hooks, removed unnecessary formatting changes, removed code that I had forgotten to. --- modules/processing.py | 23 ++-- modules/scripts.py | 4 +- modules/soft_inpainting.py | 308 --------------------------------------------- 3 files changed, 12 insertions(+), 323 deletions(-) delete mode 100644 modules/soft_inpainting.py (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 5a1a90af..f8d85bdf 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -879,14 +879,13 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.scripts is not None: ps = scripts.PostSampleArgs(samples_ddim) p.scripts.post_sample(p, ps) - samples_ddim = pp.samples + samples_ddim = ps.samples if getattr(samples_ddim, 'already_decoded', False): x_samples_ddim = samples_ddim else: if opts.sd_vae_decode_method != 'Full': p.extra_generation_params['VAE Decoder'] = opts.sd_vae_decode_method - x_samples_ddim = decode_latent_batch(p.sd_model, samples_ddim, target_device=devices.cpu, check_for_nans=True) x_samples_ddim = torch.stack(x_samples_ddim).float() @@ -944,7 +943,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if p.scripts is not None: ppmo = scripts.PostProcessMaskOverlayArgs(i, mask_for_overlay, overlay_image) p.scripts.postprocess_maskoverlay(p, ppmo) - mask_for_overlay, overlay_image = pp.mask_for_overlay, pp.overlay_image + mask_for_overlay, overlay_image = ppmo.mask_for_overlay, ppmo.overlay_image if p.color_corrections is not None and i < len(p.color_corrections): if save_samples and opts.save_images_before_color_correction: @@ -959,7 +958,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: original_denoised_image = image.copy() if p.paste_to is not None: - original_denoised_image = uncrop(original_denoised_image, (p.overlay_image.width, p.overlay_image.height), p.paste_to) + original_denoised_image = uncrop(original_denoised_image, (overlay_image.width, overlay_image.height), p.paste_to) image = apply_overlay(image, p.paste_to, overlay_image) @@ -1512,9 +1511,6 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): if self.overlay_images is not None: self.overlay_images = self.overlay_images * self.batch_size - if self.masks_for_overlay is not None: - self.masks_for_overlay = self.masks_for_overlay * self.batch_size - if self.color_corrections is not None and len(self.color_corrections) == 1: self.color_corrections = self.color_corrections * self.batch_size @@ -1565,14 +1561,15 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): samples = self.sampler.sample_img2img(self, self.init_latent, x, conditioning, unconditional_conditioning, image_conditioning=self.image_conditioning) - blended_samples = samples * self.nmask + self.init_latent * self.mask + if self.mask is not None: + blended_samples = samples * self.nmask + self.init_latent * self.mask - if self.scripts is not None: - mba = scripts.MaskBlendArgs(self, samples, self.nmask, self.init_latent, self.mask, blended_samples, sigma=None, is_final_blend=True) - self.scripts.on_mask_blend(self, mba) - blended_samples = mba.blended_latent + if self.scripts is not None: + mba = scripts.MaskBlendArgs(samples, self.nmask, self.init_latent, self.mask, blended_samples) + self.scripts.on_mask_blend(self, mba) + blended_samples = mba.blended_latent - samples = blended_samples + samples = blended_samples del x devices.torch_gc() diff --git a/modules/scripts.py b/modules/scripts.py index 92a07c56..b6fcf96e 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -12,12 +12,12 @@ from modules import shared, paths, script_callbacks, extensions, script_loading, AlwaysVisible = object() class MaskBlendArgs: - def __init__(self, current_latent, nmask, init_latent, mask, blended_samples, denoiser=None, sigma=None): + def __init__(self, current_latent, nmask, init_latent, mask, blended_latent, denoiser=None, sigma=None): self.current_latent = current_latent self.nmask = nmask self.init_latent = init_latent self.mask = mask - self.blended_samples = blended_samples + self.blended_latent = blended_latent self.denoiser = denoiser self.is_final_blend = denoiser is None diff --git a/modules/soft_inpainting.py b/modules/soft_inpainting.py deleted file mode 100644 index b36ac8fa..00000000 --- a/modules/soft_inpainting.py +++ /dev/null @@ -1,308 +0,0 @@ -class SoftInpaintingSettings: - def __init__(self, mask_blend_power, mask_blend_scale, inpaint_detail_preservation): - self.mask_blend_power = mask_blend_power - self.mask_blend_scale = mask_blend_scale - self.inpaint_detail_preservation = inpaint_detail_preservation - - def add_generation_params(self, dest): - dest[enabled_gen_param_label] = True - dest[gen_param_labels.mask_blend_power] = self.mask_blend_power - dest[gen_param_labels.mask_blend_scale] = self.mask_blend_scale - dest[gen_param_labels.inpaint_detail_preservation] = self.inpaint_detail_preservation - - -# ------------------- Methods ------------------- - - -def latent_blend(soft_inpainting, a, b, t): - """ - Interpolates two latent image representations according to the parameter t, - where the interpolated vectors' magnitudes are also interpolated separately. - The "detail_preservation" factor biases the magnitude interpolation towards - the larger of the two magnitudes. - """ - import torch - - # NOTE: We use inplace operations wherever possible. - - # [4][w][h] to [1][4][w][h] - t2 = t.unsqueeze(0) - # [4][w][h] to [1][1][w][h] - the [4] seem redundant. - t3 = t[0].unsqueeze(0).unsqueeze(0) - - one_minus_t2 = 1 - t2 - one_minus_t3 = 1 - t3 - - # Linearly interpolate the image vectors. - a_scaled = a * one_minus_t2 - b_scaled = b * t2 - image_interp = a_scaled - image_interp.add_(b_scaled) - result_type = image_interp.dtype - del a_scaled, b_scaled, t2, one_minus_t2 - - # Calculate the magnitude of the interpolated vectors. (We will remove this magnitude.) - # 64-bit operations are used here to allow large exponents. - current_magnitude = torch.norm(image_interp, p=2, dim=1, keepdim=True).to(torch.float64).add_(0.00001) - - # Interpolate the powered magnitudes, then un-power them (bring them back to a power of 1). - a_magnitude = torch.norm(a, p=2, dim=1, keepdim=True).to(torch.float64).pow_(soft_inpainting.inpaint_detail_preservation) * one_minus_t3 - b_magnitude = torch.norm(b, p=2, dim=1, keepdim=True).to(torch.float64).pow_(soft_inpainting.inpaint_detail_preservation) * t3 - desired_magnitude = a_magnitude - desired_magnitude.add_(b_magnitude).pow_(1 / soft_inpainting.inpaint_detail_preservation) - del a_magnitude, b_magnitude, t3, one_minus_t3 - - # Change the linearly interpolated image vectors' magnitudes to the value we want. - # This is the last 64-bit operation. - image_interp_scaling_factor = desired_magnitude - image_interp_scaling_factor.div_(current_magnitude) - image_interp_scaling_factor = image_interp_scaling_factor.to(result_type) - image_interp_scaled = image_interp - image_interp_scaled.mul_(image_interp_scaling_factor) - del current_magnitude - del desired_magnitude - del image_interp - del image_interp_scaling_factor - del result_type - - return image_interp_scaled - - -def get_modified_nmask(soft_inpainting, nmask, sigma): - """ - Converts a negative mask representing the transparency of the original latent vectors being overlayed - to a mask that is scaled according to the denoising strength for this step. - - Where: - 0 = fully opaque, infinite density, fully masked - 1 = fully transparent, zero density, fully unmasked - - We bring this transparency to a power, as this allows one to simulate N number of blending operations - where N can be any positive real value. Using this one can control the balance of influence between - the denoiser and the original latents according to the sigma value. - - NOTE: "mask" is not used - """ - import torch - # todo: Why is sigma 2D? Both values are the same. - return torch.pow(nmask, (sigma[0] ** soft_inpainting.mask_blend_power) * soft_inpainting.mask_blend_scale) - - -def apply_adaptive_masks( - latent_orig, - latent_processed, - overlay_images, - masks_for_overlay, - width, height, - paste_to): - import torch - import numpy as np - import modules.processing as proc - import modules.images as images - from PIL import Image, ImageOps, ImageFilter - - # TODO: Bias the blending according to the latent mask, add adjustable parameter for bias control. - # latent_mask = p.nmask[0].float().cpu() - # convert the original mask into a form we use to scale distances for thresholding - # mask_scalar = 1-(torch.clamp(latent_mask, min=0, max=1) ** (p.mask_blend_scale / 2)) - # mask_scalar = mask_scalar / (1.00001-mask_scalar) - # mask_scalar = mask_scalar.numpy() - - latent_distance = torch.norm(latent_processed - latent_orig, p=2, dim=1) - - kernel, kernel_center = images.get_gaussian_kernel(stddev_radius=1.5, max_radius=2) - - for i, (distance_map, overlay_image) in enumerate(zip(latent_distance, overlay_images)): - converted_mask = distance_map.float().cpu().numpy() - converted_mask = images.weighted_histogram_filter(converted_mask, kernel, kernel_center, - percentile_min=0.9, percentile_max=1, min_width=1) - converted_mask = images.weighted_histogram_filter(converted_mask, kernel, kernel_center, - percentile_min=0.25, percentile_max=0.75, min_width=1) - - # The distance at which opacity of original decreases to 50% - # half_weighted_distance = 1 # * mask_scalar - # converted_mask = converted_mask / half_weighted_distance - - converted_mask = 1 / (1 + converted_mask ** 2) - converted_mask = images.smootherstep(converted_mask) - converted_mask = 1 - converted_mask - converted_mask = 255. * converted_mask - converted_mask = converted_mask.astype(np.uint8) - converted_mask = Image.fromarray(converted_mask) - converted_mask = images.resize_image(2, converted_mask, width, height) - converted_mask = proc.create_binary_mask(converted_mask, round=False) - - # Remove aliasing artifacts using a gaussian blur. - converted_mask = converted_mask.filter(ImageFilter.GaussianBlur(radius=4)) - - # Expand the mask to fit the whole image if needed. - if paste_to is not None: - converted_mask = proc. uncrop(converted_mask, - (overlay_image.width, overlay_image.height), - paste_to) - - masks_for_overlay[i] = converted_mask - - image_masked = Image.new('RGBa', (overlay_image.width, overlay_image.height)) - image_masked.paste(overlay_image.convert("RGBA").convert("RGBa"), - mask=ImageOps.invert(converted_mask.convert('L'))) - - overlay_images[i] = image_masked.convert('RGBA') - -def apply_masks( - soft_inpainting, - nmask, - overlay_images, - masks_for_overlay, - width, height, - paste_to): - import torch - import numpy as np - import modules.processing as proc - import modules.images as images - from PIL import Image, ImageOps, ImageFilter - - converted_mask = nmask[0].float() - converted_mask = torch.clamp(converted_mask, min=0, max=1).pow_(soft_inpainting.mask_blend_scale / 2) - converted_mask = 255. * converted_mask - converted_mask = converted_mask.cpu().numpy().astype(np.uint8) - converted_mask = Image.fromarray(converted_mask) - converted_mask = images.resize_image(2, converted_mask, width, height) - converted_mask = proc.create_binary_mask(converted_mask, round=False) - - # Remove aliasing artifacts using a gaussian blur. - converted_mask = converted_mask.filter(ImageFilter.GaussianBlur(radius=4)) - - # Expand the mask to fit the whole image if needed. - if paste_to is not None: - converted_mask = proc.uncrop(converted_mask, - (width, height), - paste_to) - - for i, overlay_image in enumerate(overlay_images): - masks_for_overlay[i] = converted_mask - - image_masked = Image.new('RGBa', (overlay_image.width, overlay_image.height)) - image_masked.paste(overlay_image.convert("RGBA").convert("RGBa"), - mask=ImageOps.invert(converted_mask.convert('L'))) - - overlay_images[i] = image_masked.convert('RGBA') - - -# ------------------- Constants ------------------- - - -default = SoftInpaintingSettings(1, 0.5, 4) - -enabled_ui_label = "Soft inpainting" -enabled_gen_param_label = "Soft inpainting enabled" -enabled_el_id = "soft_inpainting_enabled" - -ui_labels = SoftInpaintingSettings( - "Schedule bias", - "Preservation strength", - "Transition contrast boost") - -ui_info = SoftInpaintingSettings( - "Shifts when preservation of original content occurs during denoising.", - "How strongly partially masked content should be preserved.", - "Amplifies the contrast that may be lost in partially masked regions.") - -gen_param_labels = SoftInpaintingSettings( - "Soft inpainting schedule bias", - "Soft inpainting preservation strength", - "Soft inpainting transition contrast boost") - -el_ids = SoftInpaintingSettings( - "mask_blend_power", - "mask_blend_scale", - "inpaint_detail_preservation") - - -# ------------------- UI ------------------- - - -def gradio_ui(): - import gradio as gr - from modules.ui_components import InputAccordion - - with InputAccordion(False, label=enabled_ui_label, elem_id=enabled_el_id) as soft_inpainting_enabled: - with gr.Group(): - gr.Markdown( - """ - Soft inpainting allows you to **seamlessly blend original content with inpainted content** according to the mask opacity. - **High _Mask blur_** values are recommended! - """) - - result = SoftInpaintingSettings( - gr.Slider(label=ui_labels.mask_blend_power, - info=ui_info.mask_blend_power, - minimum=0, - maximum=8, - step=0.1, - value=default.mask_blend_power, - elem_id=el_ids.mask_blend_power), - gr.Slider(label=ui_labels.mask_blend_scale, - info=ui_info.mask_blend_scale, - minimum=0, - maximum=8, - step=0.05, - value=default.mask_blend_scale, - elem_id=el_ids.mask_blend_scale), - gr.Slider(label=ui_labels.inpaint_detail_preservation, - info=ui_info.inpaint_detail_preservation, - minimum=1, - maximum=32, - step=0.5, - value=default.inpaint_detail_preservation, - elem_id=el_ids.inpaint_detail_preservation)) - - with gr.Accordion("Help", open=False): - gr.Markdown( - f""" - ### {ui_labels.mask_blend_power} - - The blending strength of original content is scaled proportionally with the decreasing noise level values at each step (sigmas). - This ensures that the influence of the denoiser and original content preservation is roughly balanced at each step. - This balance can be shifted using this parameter, controlling whether earlier or later steps have stronger preservation. - - - **Below 1**: Stronger preservation near the end (with low sigma) - - **1**: Balanced (proportional to sigma) - - **Above 1**: Stronger preservation in the beginning (with high sigma) - """) - gr.Markdown( - f""" - ### {ui_labels.mask_blend_scale} - - Skews whether partially masked image regions should be more likely to preserve the original content or favor inpainted content. - This may need to be adjusted depending on the {ui_labels.mask_blend_power}, CFG Scale, prompt and Denoising strength. - - - **Low values**: Favors generated content. - - **High values**: Favors original content. - """) - gr.Markdown( - f""" - ### {ui_labels.inpaint_detail_preservation} - - This parameter controls how the original latent vectors and denoised latent vectors are interpolated. - With higher values, the magnitude of the resulting blended vector will be closer to the maximum of the two interpolated vectors. - This can prevent the loss of contrast that occurs with linear interpolation. - - - **Low values**: Softer blending, details may fade. - - **High values**: Stronger contrast, may over-saturate colors. - """) - - return ( - [ - soft_inpainting_enabled, - result.mask_blend_power, - result.mask_blend_scale, - result.inpaint_detail_preservation - ], - [ - (soft_inpainting_enabled, enabled_gen_param_label), - (result.mask_blend_power, gen_param_labels.mask_blend_power), - (result.mask_blend_scale, gen_param_labels.mask_blend_scale), - (result.inpaint_detail_preservation, gen_param_labels.inpaint_detail_preservation) - ] - ) -- cgit v1.2.1 From 56604f08a18588e8e6b57d7c3f9c61d6624846f8 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Thu, 7 Dec 2023 14:53:44 -0700 Subject: Moved image filters used by soft inpainting into soft_inpainting.py from images.py --- modules/images.py | 190 ------------------------------------------------------ 1 file changed, 190 deletions(-) (limited to 'modules') diff --git a/modules/images.py b/modules/images.py index 94953498..16f9ae7c 100644 --- a/modules/images.py +++ b/modules/images.py @@ -792,193 +792,3 @@ def flatten(img, bgcolor): return img.convert('RGB') - -def weighted_histogram_filter(img, kernel, kernel_center, percentile_min=0.0, percentile_max=1.0, min_width=1.0): - """ - Generalization convolution filter capable of applying - weighted mean, median, maximum, and minimum filters - parametrically using an arbitrary kernel. - - Args: - img (nparray): - The image, a 2-D array of floats, to which the filter is being applied. - kernel (nparray): - The kernel, a 2-D array of floats. - kernel_center (nparray): - The kernel center coordinate, a 1-D array with two elements. - percentile_min (float): - The lower bound of the histogram window used by the filter, - from 0 to 1. - percentile_max (float): - The upper bound of the histogram window used by the filter, - from 0 to 1. - min_width (float): - The minimum size of the histogram window bounds, in weight units. - Must be greater than 0. - - Returns: - (nparray): A filtered copy of the input image "img", a 2-D array of floats. - """ - - # Converts an index tuple into a vector. - def vec(x): - return np.array(x) - - kernel_min = -kernel_center - kernel_max = vec(kernel.shape) - kernel_center - - def weighted_histogram_filter_single(idx): - idx = vec(idx) - min_index = np.maximum(0, idx + kernel_min) - max_index = np.minimum(vec(img.shape), idx + kernel_max) - window_shape = max_index - min_index - - class WeightedElement: - """ - An element of the histogram, its weight - and bounds. - """ - def __init__(self, value, weight): - self.value: float = value - self.weight: float = weight - self.window_min: float = 0.0 - self.window_max: float = 1.0 - - # Collect the values in the image as WeightedElements, - # weighted by their corresponding kernel values. - values = [] - for window_tup in np.ndindex(tuple(window_shape)): - window_index = vec(window_tup) - image_index = window_index + min_index - centered_kernel_index = image_index - idx - kernel_index = centered_kernel_index + kernel_center - element = WeightedElement(img[tuple(image_index)], kernel[tuple(kernel_index)]) - values.append(element) - - def sort_key(x: WeightedElement): - return x.value - - values.sort(key=sort_key) - - # Calculate the height of the stack (sum) - # and each sample's range they occupy in the stack - sum = 0 - for i in range(len(values)): - values[i].window_min = sum - sum += values[i].weight - values[i].window_max = sum - - # Calculate what range of this stack ("window") - # we want to get the weighted average across. - window_min = sum * percentile_min - window_max = sum * percentile_max - window_width = window_max - window_min - - # Ensure the window is within the stack and at least a certain size. - if window_width < min_width: - window_center = (window_min + window_max) / 2 - window_min = window_center - min_width / 2 - window_max = window_center + min_width / 2 - - if window_max > sum: - window_max = sum - window_min = sum - min_width - - if window_min < 0: - window_min = 0 - window_max = min_width - - value = 0 - value_weight = 0 - - # Get the weighted average of all the samples - # that overlap with the window, weighted - # by the size of their overlap. - for i in range(len(values)): - if window_min >= values[i].window_max: - continue - if window_max <= values[i].window_min: - break - - s = max(window_min, values[i].window_min) - e = min(window_max, values[i].window_max) - w = e - s - - value += values[i].value * w - value_weight += w - - return value / value_weight if value_weight != 0 else 0 - - img_out = img.copy() - - # Apply the kernel operation over each pixel. - for index in np.ndindex(img.shape): - img_out[index] = weighted_histogram_filter_single(index) - - return img_out - -def smoothstep(x): - """ - The smoothstep function, input should be clamped to 0-1 range. - Turns a diagonal line (f(x) = x) into a sigmoid-like curve. - """ - return x * x * (3 - 2 * x) - -def smootherstep(x): - """ - The smootherstep function, input should be clamped to 0-1 range. - Turns a diagonal line (f(x) = x) into a sigmoid-like curve. - """ - return x * x * x * (x * (6 * x - 15) + 10) - - -def get_gaussian_kernel(stddev_radius=1.0, max_radius=2): - """ - Creates a Gaussian kernel with thresholded edges. - - Args: - stddev_radius (float): - Standard deviation of the gaussian kernel, in pixels. - max_radius (int): - The size of the filter kernel. The number of pixels is (max_radius*2+1) ** 2. - The kernel is thresholded so that any values one pixel beyond this radius - is weighted at 0. - - Returns: - (nparray, nparray): A kernel array (shape: (N, N)), its center coordinate (shape: (2)) - """ - # Evaluates a 0-1 normalized gaussian function for a given square distance from the mean. - def gaussian(sqr_mag): - return math.exp(-sqr_mag / (stddev_radius * stddev_radius)) - - # Helper function for converting a tuple to an array. - def vec(x): - return np.array(x) - - """ - Since a gaussian is unbounded, we need to limit ourselves - to a finite range. - We taper the ends off at the end of that range so they equal zero - while preserving the maximum value of 1 at the mean. - """ - zero_radius = max_radius + 1.0 - gauss_zero = gaussian(zero_radius * zero_radius) - gauss_kernel_scale = 1 / (1 - gauss_zero) - - def gaussian_kernel_func(coordinate): - x = coordinate[0] ** 2.0 + coordinate[1] ** 2.0 - x = gaussian(x) - x -= gauss_zero - x *= gauss_kernel_scale - x = max(0.0, x) - return x - - size = max_radius * 2 + 1 - kernel_center = max_radius - kernel = np.zeros((size, size)) - - for index in np.ndindex(kernel.shape): - kernel[index] = gaussian_kernel_func(vec(index) - kernel_center) - - return kernel, kernel_center - -- cgit v1.2.1 From 0ef4a4cb2365051b1e308f0136a0d8c01d071569 Mon Sep 17 00:00:00 2001 From: CodeHatchling Date: Thu, 7 Dec 2023 14:54:26 -0700 Subject: Fixed error that occurs when using vanilla samplers (somehow). --- modules/processing.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index f8d85bdf..bea01ec6 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -937,8 +937,8 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: p.scripts.postprocess_image(p, pp) image = pp.image - mask_for_overlay = p.mask_for_overlay - overlay_image = p.overlay_images[i] if p.overlay_images is not None and i < len(p.overlay_images) else None + mask_for_overlay = getattr(p, "mask_for_overlay", None) + overlay_image = p.overlay_images[i] if getattr(p, "overlay_images", None) is not None and i < len(p.overlay_images) else None if p.scripts is not None: ppmo = scripts.PostProcessMaskOverlayArgs(i, mask_for_overlay, overlay_image) -- cgit v1.2.1 From 59429793440fb3cb1624ddcc702c6f9807373203 Mon Sep 17 00:00:00 2001 From: Nuullll Date: Sat, 9 Dec 2023 18:09:45 +0800 Subject: Fix ControlNet --- modules/xpu_specific.py | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'modules') diff --git a/modules/xpu_specific.py b/modules/xpu_specific.py index ec1ad100..9bb0a561 100644 --- a/modules/xpu_specific.py +++ b/modules/xpu_specific.py @@ -51,3 +51,9 @@ if has_xpu: CondFunc('torch.bmm', lambda orig_func, input, mat2, out=None: orig_func(input.to(mat2.dtype), mat2, out=out), lambda orig_func, input, mat2, out=None: input.dtype != mat2.dtype) + CondFunc('torch.cat', + lambda orig_func, tensors, dim=0, out=None: orig_func([t.to(tensors[0].dtype) for t in tensors], dim=dim, out=out), + lambda orig_func, tensors, dim=0, out=None: not all(t.dtype == tensors[0].dtype for t in tensors)) + CondFunc('torch.nn.functional.scaled_dot_product_attention', + lambda orig_func, query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False: orig_func(query, key.to(query.dtype), value.to(query.dtype), attn_mask, dropout_p, is_causal), + lambda orig_func, query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False: query.dtype != key.dtype or query.dtype != value.dtype) \ No newline at end of file -- cgit v1.2.1 From 049d5642e58d572ee8657ac754e72d019eea0e6c Mon Sep 17 00:00:00 2001 From: Nuullll Date: Sat, 9 Dec 2023 18:11:26 +0800 Subject: Fix format --- modules/xpu_specific.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/xpu_specific.py b/modules/xpu_specific.py index 9bb0a561..d8da94a0 100644 --- a/modules/xpu_specific.py +++ b/modules/xpu_specific.py @@ -56,4 +56,4 @@ if has_xpu: lambda orig_func, tensors, dim=0, out=None: not all(t.dtype == tensors[0].dtype for t in tensors)) CondFunc('torch.nn.functional.scaled_dot_product_attention', lambda orig_func, query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False: orig_func(query, key.to(query.dtype), value.to(query.dtype), attn_mask, dropout_p, is_causal), - lambda orig_func, query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False: query.dtype != key.dtype or query.dtype != value.dtype) \ No newline at end of file + lambda orig_func, query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False: query.dtype != key.dtype or query.dtype != value.dtype) -- cgit v1.2.1 From 39ec4cfea9040bc94e639eb4aa8ab8ed37a68f01 Mon Sep 17 00:00:00 2001 From: kaalibro Date: Sat, 9 Dec 2023 19:12:59 +0600 Subject: Re-add setting lost as part of e294e46 --- modules/shared_options.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index e5de0d01..acb6e2d4 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -256,6 +256,7 @@ options_templates.update(options_section(('ui_prompt_editing', "Prompt editing", "keyedit_precision_extra": OptionInfo(0.05, "Precision for when editing the prompt with Ctrl+up/down", gr.Slider, {"minimum": 0.01, "maximum": 0.2, "step": 0.001}), "keyedit_delimiters": OptionInfo(r".,\/!?%^*;:{}=`~() ", "Word delimiters when editing the prompt with Ctrl+up/down"), "keyedit_delimiters_whitespace": OptionInfo(["Tab", "Carriage Return", "Line Feed"], "Ctrl+up/down whitespace delimiters", gr.CheckboxGroup, lambda: {"choices": ["Tab", "Carriage Return", "Line Feed"]}), + "keyedit_move": OptionInfo(True, "Alt+left/right moves prompt elements"), "disable_token_counters": OptionInfo(False, "Disable prompt token counters").needs_reload_ui(), })) -- cgit v1.2.1 From 9c201550ddae0b33367adfb99bcbb57ba9b207a9 Mon Sep 17 00:00:00 2001 From: kaalibro Date: Sat, 9 Dec 2023 21:04:45 +0600 Subject: Add keyboard shortcuts for generation (Removed Alt+Enter) Ctrl+Enter to start/restart generation (New) Alt/Option+Enter to skip generation (New) Ctrl+Alt/Option+Enter to interrupt generation --- modules/ui_toprow.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/ui_toprow.py b/modules/ui_toprow.py index 88838f97..c3865e3d 100644 --- a/modules/ui_toprow.py +++ b/modules/ui_toprow.py @@ -79,11 +79,11 @@ class Toprow: def create_prompts(self): with gr.Column(elem_id=f"{self.id_part}_prompt_container", elem_classes=["prompt-container-compact"] if self.is_compact else [], scale=6): with gr.Row(elem_id=f"{self.id_part}_prompt_row", elem_classes=["prompt-row"]): - self.prompt = gr.Textbox(label="Prompt", elem_id=f"{self.id_part}_prompt", show_label=False, lines=3, placeholder="Prompt (press Ctrl+Enter or Alt+Enter to generate)", elem_classes=["prompt"]) + self.prompt = gr.Textbox(label="Prompt", elem_id=f"{self.id_part}_prompt", show_label=False, lines=3, placeholder="Prompt\n(Press Ctrl+Enter to generate, Alt+Enter to skip, Ctrl+Alt+Enter to interrupt)", elem_classes=["prompt"]) self.prompt_img = gr.File(label="", elem_id=f"{self.id_part}_prompt_image", file_count="single", type="binary", visible=False) with gr.Row(elem_id=f"{self.id_part}_neg_prompt_row", elem_classes=["prompt-row"]): - self.negative_prompt = gr.Textbox(label="Negative prompt", elem_id=f"{self.id_part}_neg_prompt", show_label=False, lines=3, placeholder="Negative prompt (press Ctrl+Enter or Alt+Enter to generate)", elem_classes=["prompt"]) + self.negative_prompt = gr.Textbox(label="Negative prompt", elem_id=f"{self.id_part}_neg_prompt", show_label=False, lines=3, placeholder="Negative prompt\n(Press Ctrl+Enter to generate, Alt+Enter to skip, Ctrl+Alt+Enter to interrupt)", elem_classes=["prompt"]) self.prompt_img.change( fn=modules.images.image_data, -- cgit v1.2.1 From 5381405eaa1e809e5cfb97522bd4c19d3c946079 Mon Sep 17 00:00:00 2001 From: drhead <1313496+drhead@users.noreply.github.com> Date: Sat, 9 Dec 2023 14:09:28 -0500 Subject: re-derive sqrt alpha bar and sqrt one minus alphabar This is the only place these values are ever referenced outside of training code so this change is very justifiable and more consistent. --- modules/sd_samplers_timesteps.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/sd_samplers_timesteps.py b/modules/sd_samplers_timesteps.py index b17a8f93..c4bd5c12 100644 --- a/modules/sd_samplers_timesteps.py +++ b/modules/sd_samplers_timesteps.py @@ -36,7 +36,7 @@ class CompVisTimestepsVDenoiser(torch.nn.Module): self.inner_model = model def predict_eps_from_z_and_v(self, x_t, t, v): - return self.inner_model.sqrt_alphas_cumprod[t.to(torch.int), None, None, None] * v + self.inner_model.sqrt_one_minus_alphas_cumprod[t.to(torch.int), None, None, None] * x_t + return torch.sqrt(self.inner_model.alphas_cumprod)[t.to(torch.int), None, None, None] * v + torch.sqrt(1 - self.inner_model.alphas_cumprod)[t.to(torch.int), None, None, None] * x_t def forward(self, input, timesteps, **kwargs): model_output = self.inner_model.apply_model(input, timesteps, **kwargs) -- cgit v1.2.1 From 23a0e60b9bf90a80f8af9732cc6495fbfce2ea21 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sun, 10 Dec 2023 14:03:41 +0900 Subject: fix save styles --- modules/styles.py | 25 +++++-------------------- 1 file changed, 5 insertions(+), 20 deletions(-) (limited to 'modules') diff --git a/modules/styles.py b/modules/styles.py index 7fb6c2e1..07588945 100644 --- a/modules/styles.py +++ b/modules/styles.py @@ -155,10 +155,8 @@ class StyleDatabase: row["name"], prompt, negative_prompt, path ) - def get_style_paths(self) -> list(): - """ - Returns a list of all distinct paths, including the default path, of - files that styles are loaded from.""" + def get_style_paths(self) -> set: + """Returns a set of all distinct paths of files that styles are loaded from.""" # Update any styles without a path to the default path for style in list(self.styles.values()): if not style.path: @@ -172,9 +170,9 @@ class StyleDatabase: style_paths.add(style.path) # Remove any paths for styles that are just list dividers - style_paths.remove("do_not_save") + style_paths.discard("do_not_save") - return list(style_paths) + return style_paths def get_style_prompts(self, styles): return [self.styles.get(x, self.no_style).prompt for x in styles] @@ -196,20 +194,7 @@ class StyleDatabase: # The path argument is deprecated, but kept for backwards compatibility _ = path - # Update any styles without a path to the default path - for style in list(self.styles.values()): - if not style.path: - self.styles[style.name] = style._replace(path=self.default_path) - - # Create a list of all distinct paths, including the default path - style_paths = set() - style_paths.add(self.default_path) - for _, style in self.styles.items(): - if style.path: - style_paths.add(style.path) - - # Remove any paths for styles that are just list dividers - style_paths.remove("do_not_save") + style_paths = self.get_style_paths() csv_names = [os.path.split(path)[1].lower() for path in style_paths] -- cgit v1.2.1 From 8b74389e76a7678e972583ef16100e90e1519e55 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sun, 10 Dec 2023 15:48:16 +0900 Subject: fix styles.csv filename --- modules/styles.py | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/styles.py b/modules/styles.py index 07588945..81d9800d 100644 --- a/modules/styles.py +++ b/modules/styles.py @@ -98,10 +98,8 @@ class StyleDatabase: self.path = path folder, file = os.path.split(self.path) - self.default_file = file.split("*")[0] + ".csv" - if self.default_file == ".csv": - self.default_file = "styles.csv" - self.default_path = os.path.join(folder, self.default_file) + filename, _, ext = file.partition('*') + self.default_path = os.path.join(folder, filename + ext) self.prompt_fields = [field for field in PromptStyle._fields if field != "path"] -- cgit v1.2.1 From 1d42babd324b933bae317cb427fe0513138954f4 Mon Sep 17 00:00:00 2001 From: kaalibro Date: Sun, 10 Dec 2023 16:28:56 +0600 Subject: Replace Ctrl+Alt+Enter with Esc --- modules/ui_toprow.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/ui_toprow.py b/modules/ui_toprow.py index c3865e3d..9caf8faa 100644 --- a/modules/ui_toprow.py +++ b/modules/ui_toprow.py @@ -79,11 +79,11 @@ class Toprow: def create_prompts(self): with gr.Column(elem_id=f"{self.id_part}_prompt_container", elem_classes=["prompt-container-compact"] if self.is_compact else [], scale=6): with gr.Row(elem_id=f"{self.id_part}_prompt_row", elem_classes=["prompt-row"]): - self.prompt = gr.Textbox(label="Prompt", elem_id=f"{self.id_part}_prompt", show_label=False, lines=3, placeholder="Prompt\n(Press Ctrl+Enter to generate, Alt+Enter to skip, Ctrl+Alt+Enter to interrupt)", elem_classes=["prompt"]) + self.prompt = gr.Textbox(label="Prompt", elem_id=f"{self.id_part}_prompt", show_label=False, lines=3, placeholder="Prompt\n(Press Ctrl+Enter to generate, Alt+Enter to skip, Esc to interrupt)", elem_classes=["prompt"]) self.prompt_img = gr.File(label="", elem_id=f"{self.id_part}_prompt_image", file_count="single", type="binary", visible=False) with gr.Row(elem_id=f"{self.id_part}_neg_prompt_row", elem_classes=["prompt-row"]): - self.negative_prompt = gr.Textbox(label="Negative prompt", elem_id=f"{self.id_part}_neg_prompt", show_label=False, lines=3, placeholder="Negative prompt\n(Press Ctrl+Enter to generate, Alt+Enter to skip, Ctrl+Alt+Enter to interrupt)", elem_classes=["prompt"]) + self.negative_prompt = gr.Textbox(label="Negative prompt", elem_id=f"{self.id_part}_neg_prompt", show_label=False, lines=3, placeholder="Negative prompt\n(Press Ctrl+Enter to generate, Alt+Enter to skip, Esc to interrupt)", elem_classes=["prompt"]) self.prompt_img.change( fn=modules.images.image_data, -- cgit v1.2.1 From 3c0c27757944ae17a7fa4c2323ee9ae2d434dbce Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Thu, 14 Dec 2023 19:36:17 +0900 Subject: default False js_live_preview_in_modal_lightbox --- modules/shared_options.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index 41097d8e..d2e86ff1 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -331,7 +331,7 @@ options_templates.update(options_section(('ui', "Live previews", "ui"), { "live_preview_content": OptionInfo("Prompt", "Live preview subject", gr.Radio, {"choices": ["Combined", "Prompt", "Negative prompt"]}), "live_preview_refresh_period": OptionInfo(1000, "Progressbar and preview update period").info("in milliseconds"), "live_preview_fast_interrupt": OptionInfo(False, "Return image with chosen live preview method on interrupt").info("makes interrupts faster"), - "js_live_preview_in_modal_lightbox": OptionInfo(True, "Show Live preview in full page image viewer"), + "js_live_preview_in_modal_lightbox": OptionInfo(False, "Show Live preview in full page image viewer"), })) options_templates.update(options_section(('sampler-params', "Sampler parameters", "sd"), { -- cgit v1.2.1 From 0c5427960b3a4ffe6d673c28e8e135b26f015717 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Fri, 15 Dec 2023 17:11:59 +0900 Subject: make modal toolbar and icon opacity adjustable --- modules/shared_gradio_themes.py | 4 ++++ modules/shared_options.py | 2 ++ 2 files changed, 6 insertions(+) (limited to 'modules') diff --git a/modules/shared_gradio_themes.py b/modules/shared_gradio_themes.py index 822db0a9..b6dc3145 100644 --- a/modules/shared_gradio_themes.py +++ b/modules/shared_gradio_themes.py @@ -65,3 +65,7 @@ def reload_gradio_theme(theme_name=None): except Exception as e: errors.display(e, "changing gradio theme") shared.gradio_theme = gr.themes.Default(**default_theme_args) + + # append additional values gradio_theme + shared.gradio_theme.sd_webui_modal_lightbox_toolbar_opacity = shared.opts.sd_webui_modal_lightbox_toolbar_opacity + shared.gradio_theme.sd_webui_modal_lightbox_icon_opacity = shared.opts.sd_webui_modal_lightbox_icon_opacity diff --git a/modules/shared_options.py b/modules/shared_options.py index e5de0d01..86e7636c 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -266,6 +266,8 @@ options_templates.update(options_section(('ui_gallery', "Gallery", "ui"), { "js_modal_lightbox_initially_zoomed": OptionInfo(True, "Full page image viewer: show images zoomed in by default"), "js_modal_lightbox_gamepad": OptionInfo(False, "Full page image viewer: navigate with gamepad"), "js_modal_lightbox_gamepad_repeat": OptionInfo(250, "Full page image viewer: gamepad repeat period").info("in milliseconds"), + "sd_webui_modal_lightbox_icon_opacity": OptionInfo(1, "Full page image viewer: control icon unfocused opacity", gr.Slider, {"minimum": 0.0, "maximum": 1, "step": 0.01}, onchange=shared.reload_gradio_theme).info('for mouse only').needs_reload_ui(), + "sd_webui_modal_lightbox_toolbar_opacity": OptionInfo(0.9, "Full page image viewer: tool bar opacity", gr.Slider, {"minimum": 0.0, "maximum": 1, "step": 0.01}, onchange=shared.reload_gradio_theme).info('for mouse only').needs_reload_ui(), "gallery_height": OptionInfo("", "Gallery height", gr.Textbox).info("can be any valid CSS value, for example 768px or 20em").needs_reload_ui(), })) -- cgit v1.2.1 From 1242ba08e19f3d317bdc5924db2b73d0c9569a7f Mon Sep 17 00:00:00 2001 From: gayshub Date: Fri, 15 Dec 2023 16:57:17 +0800 Subject: add allow specify the task id and get the location of task in the queue of pending task --- modules/api/api.py | 20 ++++++++++++++++++-- modules/api/models.py | 2 ++ modules/processing.py | 2 ++ modules/progress.py | 21 +++++++++++++++++++-- 4 files changed, 41 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index e6edffe7..5d000ae8 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -33,7 +33,7 @@ from typing import Dict, List, Any import piexif import piexif.helper from contextlib import closing - +from modules.progress import create_task_id, add_task_to_queue, start_task, finish_task, current_task def script_name_to_index(name, scripts): try: @@ -337,6 +337,10 @@ class Api: return script_args def text2imgapi(self, txt2imgreq: models.StableDiffusionTxt2ImgProcessingAPI): + task_id = create_task_id("text2img") + if txt2imgreq.force_task_id != None: + task_id = txt2imgreq.force_task_id + script_runner = scripts.scripts_txt2img if not script_runner.scripts: script_runner.initialize_scripts(False) @@ -363,6 +367,8 @@ class Api: send_images = args.pop('send_images', True) args.pop('save_images', None) + add_task_to_queue(task_id) + with self.queue_lock: with closing(StableDiffusionProcessingTxt2Img(sd_model=shared.sd_model, **args)) as p: p.is_api = True @@ -372,12 +378,14 @@ class Api: try: shared.state.begin(job="scripts_txt2img") + start_task(task_id) if selectable_scripts is not None: p.script_args = script_args processed = scripts.scripts_txt2img.run(p, *p.script_args) # Need to pass args as list here else: p.script_args = tuple(script_args) # Need to pass args as tuple here processed = process_images(p) + finish_task(task_id) finally: shared.state.end() shared.total_tqdm.clear() @@ -387,6 +395,10 @@ class Api: return models.TextToImageResponse(images=b64images, parameters=vars(txt2imgreq), info=processed.js()) def img2imgapi(self, img2imgreq: models.StableDiffusionImg2ImgProcessingAPI): + task_id = create_task_id("img2img") + if img2imgreq.force_task_id != None: + task_id = img2imgreq.force_task_id + init_images = img2imgreq.init_images if init_images is None: raise HTTPException(status_code=404, detail="Init image not found") @@ -423,6 +435,8 @@ class Api: send_images = args.pop('send_images', True) args.pop('save_images', None) + add_task_to_queue(task_id) + with self.queue_lock: with closing(StableDiffusionProcessingImg2Img(sd_model=shared.sd_model, **args)) as p: p.init_images = [decode_base64_to_image(x) for x in init_images] @@ -433,12 +447,14 @@ class Api: try: shared.state.begin(job="scripts_img2img") + start_task(task_id) if selectable_scripts is not None: p.script_args = script_args processed = scripts.scripts_img2img.run(p, *p.script_args) # Need to pass args as list here else: p.script_args = tuple(script_args) # Need to pass args as tuple here processed = process_images(p) + finish_task(task_id) finally: shared.state.end() shared.total_tqdm.clear() @@ -514,7 +530,7 @@ class Api: if shared.state.current_image and not req.skip_current_image: current_image = encode_pil_to_base64(shared.state.current_image) - return models.ProgressResponse(progress=progress, eta_relative=eta_relative, state=shared.state.dict(), current_image=current_image, textinfo=shared.state.textinfo) + return models.ProgressResponse(progress=progress, eta_relative=eta_relative, state=shared.state.dict(), current_image=current_image, textinfo=shared.state.textinfo, current_task=current_task) def interrogateapi(self, interrogatereq: models.InterrogateRequest): image_b64 = interrogatereq.image diff --git a/modules/api/models.py b/modules/api/models.py index 6a574771..7b7f1773 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -109,6 +109,7 @@ StableDiffusionTxt2ImgProcessingAPI = PydanticModelGenerator( {"key": "send_images", "type": bool, "default": True}, {"key": "save_images", "type": bool, "default": False}, {"key": "alwayson_scripts", "type": dict, "default": {}}, + {"key": "force_task_id", "type": str, "default": None}, ] ).generate_model() @@ -126,6 +127,7 @@ StableDiffusionImg2ImgProcessingAPI = PydanticModelGenerator( {"key": "send_images", "type": bool, "default": True}, {"key": "save_images", "type": bool, "default": False}, {"key": "alwayson_scripts", "type": dict, "default": {}}, + {"key": "force_task_id", "type": str, "default": None}, ] ).generate_model() diff --git a/modules/processing.py b/modules/processing.py index e124e7f0..657cacfc 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1023,6 +1023,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): hr_sampler_name: str = None hr_prompt: str = '' hr_negative_prompt: str = '' + force_task_id: str = None cached_hr_uc = [None, None] cached_hr_c = [None, None] @@ -1358,6 +1359,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): inpainting_mask_invert: int = 0 initial_noise_multiplier: float = None latent_mask: Image = None + force_task_id: string = None image_mask: Any = field(default=None, init=False) diff --git a/modules/progress.py b/modules/progress.py index 69921de7..553866db 100644 --- a/modules/progress.py +++ b/modules/progress.py @@ -8,10 +8,13 @@ from pydantic import BaseModel, Field from modules.shared import opts import modules.shared as shared - +from collections import OrderedDict +import string +import random +from typing import List current_task = None -pending_tasks = {} +pending_tasks = OrderedDict() finished_tasks = [] recorded_results = [] recorded_results_limit = 2 @@ -34,6 +37,11 @@ def finish_task(id_task): if len(finished_tasks) > 16: finished_tasks.pop(0) +def create_task_id(task_type): + N = 7 + res = ''.join(random.choices(string.ascii_uppercase + + string.digits, k=N)) + return f"task({task_type}-{res})" def record_results(id_task, res): recorded_results.append((id_task, res)) @@ -44,6 +52,9 @@ def record_results(id_task, res): def add_task_to_queue(id_job): pending_tasks[id_job] = time.time() +class PendingTasksResponse(BaseModel): + size: int = Field(title="Pending task size") + tasks: List[str] = Field(title="Pending task ids") class ProgressRequest(BaseModel): id_task: str = Field(default=None, title="Task ID", description="id of the task to get progress for") @@ -63,8 +74,14 @@ class ProgressResponse(BaseModel): def setup_progress_api(app): + app.add_api_route("/internal/pendingTasks", get_pending_tasks, methods=["GET"]) return app.add_api_route("/internal/progress", progressapi, methods=["POST"], response_model=ProgressResponse) +def get_pending_tasks(): + pending_tasks_ids = [x for x in pending_tasks] + pending_len = len(pending_tasks_ids) + return PendingTasksResponse(size=pending_len, tasks=pending_tasks_ids) + def progressapi(req: ProgressRequest): active = req.id_task == current_task -- cgit v1.2.1 From d859de37d9ec10cb6c804226328a11c87c444852 Mon Sep 17 00:00:00 2001 From: gayshub Date: Fri, 15 Dec 2023 17:48:20 +0800 Subject: fix the problem of ruff of github --- modules/api/api.py | 4 ++-- modules/processing.py | 2 +- modules/progress.py | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 5d000ae8..1f464806 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -338,7 +338,7 @@ class Api: def text2imgapi(self, txt2imgreq: models.StableDiffusionTxt2ImgProcessingAPI): task_id = create_task_id("text2img") - if txt2imgreq.force_task_id != None: + if txt2imgreq.force_task_id is None: task_id = txt2imgreq.force_task_id script_runner = scripts.scripts_txt2img @@ -396,7 +396,7 @@ class Api: def img2imgapi(self, img2imgreq: models.StableDiffusionImg2ImgProcessingAPI): task_id = create_task_id("img2img") - if img2imgreq.force_task_id != None: + if img2imgreq.force_task_id is None: task_id = img2imgreq.force_task_id init_images = img2imgreq.init_images diff --git a/modules/processing.py b/modules/processing.py index 657cacfc..5added65 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1359,7 +1359,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): inpainting_mask_invert: int = 0 initial_noise_multiplier: float = None latent_mask: Image = None - force_task_id: string = None + force_task_id: str = None image_mask: Any = field(default=None, init=False) diff --git a/modules/progress.py b/modules/progress.py index 553866db..6946fb1b 100644 --- a/modules/progress.py +++ b/modules/progress.py @@ -78,7 +78,7 @@ def setup_progress_api(app): return app.add_api_route("/internal/progress", progressapi, methods=["POST"], response_model=ProgressResponse) def get_pending_tasks(): - pending_tasks_ids = [x for x in pending_tasks] + pending_tasks_ids = list(pending_tasks) pending_len = len(pending_tasks_ids) return PendingTasksResponse(size=pending_len, tasks=pending_tasks_ids) -- cgit v1.2.1 From da45e73b4ffde2e2a85b64a3e3258a0625bd307e Mon Sep 17 00:00:00 2001 From: gayshub Date: Fri, 15 Dec 2023 17:57:58 +0800 Subject: fix the problem of ruff of github --- modules/api/api.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 1f464806..9fac7e60 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -340,7 +340,7 @@ class Api: task_id = create_task_id("text2img") if txt2imgreq.force_task_id is None: task_id = txt2imgreq.force_task_id - + script_runner = scripts.scripts_txt2img if not script_runner.scripts: script_runner.initialize_scripts(False) -- cgit v1.2.1 From 6d7e57ba6a4d686d515518b5f90e91b32fa01caf Mon Sep 17 00:00:00 2001 From: gayshub Date: Fri, 15 Dec 2023 18:03:14 +0800 Subject: fix the problem of ruff of github --- modules/api/api.py | 1 - 1 file changed, 1 deletion(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 9fac7e60..8d8e70a4 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -340,7 +340,6 @@ class Api: task_id = create_task_id("text2img") if txt2imgreq.force_task_id is None: task_id = txt2imgreq.force_task_id - script_runner = scripts.scripts_txt2img if not script_runner.scripts: script_runner.initialize_scripts(False) -- cgit v1.2.1 From ea272152e0b50dbb2bd675ec020607f3d50c37d0 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sat, 16 Dec 2023 15:08:08 +0800 Subject: Add FP8 settings into PNG info --- modules/generation_parameters_copypaste.py | 6 ++++++ modules/processing.py | 2 ++ 2 files changed, 8 insertions(+) (limited to 'modules') diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index 4efe53e0..dbffe494 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -314,6 +314,12 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model if "VAE Decoder" not in res: res["VAE Decoder"] = "Full" + if "FP8 weight" not in res: + res["FP8 weight"] = "Disable" + + if "Cache FP16 weight for LoRA" not in res and res["FP8 weight"] != "Disable": + res["Cache FP16 weight for LoRA"] = False + skip = set(shared.opts.infotext_skip_pasting) res = {k: v for k, v in res.items() if k not in skip} diff --git a/modules/processing.py b/modules/processing.py index bea01ec6..179f2c0f 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -688,6 +688,8 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "Size": f"{p.width}x{p.height}", "Model hash": p.sd_model_hash if opts.add_model_hash_to_info else None, "Model": p.sd_model_name if opts.add_model_name_to_info else None, + "FP8 weight": opts.fp8_storage if devices.fp8 else None, + "Cache FP16 weight for LoRA": opts.cache_fp16_weight if devices.fp8 else None, "VAE hash": p.sd_vae_hash if opts.add_vae_hash_to_info else None, "VAE": p.sd_vae_name if opts.add_vae_name_to_info else None, "Variation seed": (None if p.subseed_strength == 0 else (p.all_subseeds[0] if use_main_prompt else all_subseeds[index])), -- cgit v1.2.1 From 7745db6fc02faf19117838c1e7bcc8a60b5f5e90 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 16 Dec 2023 10:15:08 +0300 Subject: torch 2.1.2 --- modules/errors.py | 4 ++-- modules/launch_utils.py | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/errors.py b/modules/errors.py index c534a5d6..48aa13a1 100644 --- a/modules/errors.py +++ b/modules/errors.py @@ -107,8 +107,8 @@ def check_versions(): import torch import gradio - expected_torch_version = "2.1.0" - expected_xformers_version = "0.0.22.post7" + expected_torch_version = "2.1.2" + expected_xformers_version = "0.0.23.post1" expected_gradio_version = "3.41.2" if version.parse(torch.__version__) < version.parse(expected_torch_version): diff --git a/modules/launch_utils.py b/modules/launch_utils.py index 2c54e2a0..dabef0f5 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -315,7 +315,7 @@ def requirements_met(requirements_file): def prepare_environment(): torch_index_url = os.environ.get('TORCH_INDEX_URL', "https://download.pytorch.org/whl/cu121") - torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.1.0 torchvision==0.16.0 --extra-index-url {torch_index_url}") + torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.1.2 torchvision==0.16.2 --extra-index-url {torch_index_url}") if args.use_ipex: if platform.system() == "Windows": # The "Nuullll/intel-extension-for-pytorch" wheels were built from IPEX source for Intel Arc GPU: https://github.com/intel/intel-extension-for-pytorch/tree/xpu-main @@ -338,7 +338,7 @@ def prepare_environment(): torch_command = os.environ.get('TORCH_COMMAND', f"pip install torch==2.0.0a0 intel-extension-for-pytorch==2.0.110+gitba7f6c1 --extra-index-url {torch_index_url}") requirements_file = os.environ.get('REQS_FILE', "requirements_versions.txt") - xformers_package = os.environ.get('XFORMERS_PACKAGE', 'xformers==0.0.22.post7') + xformers_package = os.environ.get('XFORMERS_PACKAGE', 'xformers==0.0.23.post1') clip_package = os.environ.get('CLIP_PACKAGE', "https://github.com/openai/CLIP/archive/d50d76daa670286dd6cacf3bcd80b5e4823fc8e1.zip") openclip_package = os.environ.get('OPENCLIP_PACKAGE', "https://github.com/mlfoundations/open_clip/archive/bb6e834e9c70d9c27d0dc3ecedeebeaeb1ffad6b.zip") -- cgit v1.2.1 From cd9ce2e31c4a264d7cde17c54d24f8ad94c9cf2c Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 16 Dec 2023 10:40:20 +0300 Subject: Use radio for FP8 mode selection --- modules/shared_options.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index d470eb8f..fa542ba8 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -206,7 +206,7 @@ options_templates.update(options_section(('optimizations', "Optimizations", "sd" "pad_cond_uncond": OptionInfo(False, "Pad prompt/negative prompt to be same length", infotext='Pad conds').info("improves performance when prompt and negative prompt have different lengths; changes seeds"), "persistent_cond_cache": OptionInfo(True, "Persistent cond cache").info("do not recalculate conds from prompts if prompts have not changed since previous calculation"), "batch_cond_uncond": OptionInfo(True, "Batch cond/uncond").info("do both conditional and unconditional denoising in one batch; uses a bit more VRAM during sampling, but improves speed; previously this was controlled by --always-batch-cond-uncond comandline argument"), - "fp8_storage": OptionInfo("Disable", "FP8 weight", gr.Dropdown, {"choices": ["Disable", "Enable for SDXL", "Enable"]}).info("Use FP8 to store Linear/Conv layers' weight. Require pytorch>=2.1.0."), + "fp8_storage": OptionInfo("Disable", "FP8 weight", gr.Radio, {"choices": ["Disable", "Enable for SDXL", "Enable"]}).info("Use FP8 to store Linear/Conv layers' weight. Require pytorch>=2.1.0."), "cache_fp16_weight": OptionInfo(False, "Cache FP16 weight for LoRA").info("Cache fp16 weight when enabling FP8, will increase the quality of LoRA. Use more system ram."), })) -- cgit v1.2.1 From 86b3aa94e2d36a4f9d5ef1bb7c6ec995ff8eb517 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 16 Dec 2023 11:04:59 +0300 Subject: rename pending tasks api endpoint to be more in line with others --- modules/progress.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/progress.py b/modules/progress.py index 6946fb1b..85255e82 100644 --- a/modules/progress.py +++ b/modules/progress.py @@ -74,9 +74,10 @@ class ProgressResponse(BaseModel): def setup_progress_api(app): - app.add_api_route("/internal/pendingTasks", get_pending_tasks, methods=["GET"]) + app.add_api_route("/internal/pending-tasks", get_pending_tasks, methods=["GET"]) return app.add_api_route("/internal/progress", progressapi, methods=["POST"], response_model=ProgressResponse) + def get_pending_tasks(): pending_tasks_ids = list(pending_tasks) pending_len = len(pending_tasks_ids) -- cgit v1.2.1 From a97832033427096072d5ea914adac3662cda4fd1 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Sat, 16 Dec 2023 19:39:43 +0800 Subject: Let fp8-related settings to invalidate cond_cache --- modules/processing.py | 2 ++ 1 file changed, 2 insertions(+) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index dd97b4ee..9351e3fb 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -431,6 +431,8 @@ class StableDiffusionProcessing: opts.sdxl_crop_top, self.width, self.height, + opts.fp8_storage, + opts.cache_fp16_weight, ) def get_conds_with_caching(self, function, required_prompts, steps, caches, extra_network_data, hires_steps=None): -- cgit v1.2.1 From de03882d6ca56bc81058f5120f028678a6a54aaa Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 17 Dec 2023 08:55:35 +0300 Subject: make task ids for API work without force_task_id --- modules/api/api.py | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 9637cb81..7154c9d5 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -336,9 +336,8 @@ class Api: return script_args def text2imgapi(self, txt2imgreq: models.StableDiffusionTxt2ImgProcessingAPI): - task_id = create_task_id("text2img") - if txt2imgreq.force_task_id is None: - task_id = txt2imgreq.force_task_id + task_id = txt2imgreq.force_task_id or create_task_id("txt2img") + script_runner = scripts.scripts_txt2img if not script_runner.scripts: script_runner.initialize_scripts(False) @@ -393,9 +392,7 @@ class Api: return models.TextToImageResponse(images=b64images, parameters=vars(txt2imgreq), info=processed.js()) def img2imgapi(self, img2imgreq: models.StableDiffusionImg2ImgProcessingAPI): - task_id = create_task_id("img2img") - if img2imgreq.force_task_id is None: - task_id = img2imgreq.force_task_id + task_id = img2imgreq.force_task_id or create_task_id("img2img") init_images = img2imgreq.init_images if init_images is None: -- cgit v1.2.1 From 10945aa41a158ee03727c5ea77d4ffff6b5370f0 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Mon, 18 Dec 2023 15:27:41 +0900 Subject: only rewrite ui-config when there is change and a typo --- modules/ui.py | 4 +++- modules/ui_loadsave.py | 2 +- 2 files changed, 4 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index d80486dd..f02b5511 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -1086,6 +1086,7 @@ def create_ui(): ) loadsave = ui_loadsave.UiLoadsave(cmd_opts.ui_config_file) + ui_settings_from_file = loadsave.ui_settings.copy() settings = ui_settings.UiSettings() settings.create_ui(loadsave, dummy_component) @@ -1146,7 +1147,8 @@ def create_ui(): modelmerger_ui.setup_ui(dummy_component=dummy_component, sd_model_checkpoint_component=settings.component_dict['sd_model_checkpoint']) - loadsave.dump_defaults() + if ui_settings_from_file != loadsave.ui_settings: + loadsave.dump_defaults() demo.ui_loadsave = loadsave return demo diff --git a/modules/ui_loadsave.py b/modules/ui_loadsave.py index 7826786c..693ff75c 100644 --- a/modules/ui_loadsave.py +++ b/modules/ui_loadsave.py @@ -144,7 +144,7 @@ class UiLoadsave: json.dump(current_ui_settings, file, indent=4, ensure_ascii=False) def dump_defaults(self): - """saves default values to a file unless tjhe file is present and there was an error loading default values at start""" + """saves default values to a file unless the file is present and there was an error loading default values at start""" if self.error_loading and os.path.exists(self.filename): return -- cgit v1.2.1 From e4b4a9c4acf0ca375a8603f7f52fde8467b2d266 Mon Sep 17 00:00:00 2001 From: Nuullll Date: Mon, 18 Dec 2023 18:00:01 +0800 Subject: [IPEX] Slice SDPA into smaller chunks --- modules/xpu_specific.py | 66 +++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 64 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/xpu_specific.py b/modules/xpu_specific.py index d8da94a0..0ebdd596 100644 --- a/modules/xpu_specific.py +++ b/modules/xpu_specific.py @@ -27,6 +27,68 @@ def torch_xpu_gc(): has_xpu = check_for_xpu() + +# Arc GPU cannot allocate a single block larger than 4GB: https://github.com/intel/compute-runtime/issues/627 +# Here we implement a slicing algorithm to split large batch size into smaller chunks, +# so that SDPA of each chunk wouldn't require any allocation larger than ARC_SINGLE_ALLOCATION_LIMIT. +# The heuristic limit (TOTAL_VRAM // 8) is tuned for Intel Arc A770 16G and Arc A750 8G, +# which is the best trade-off between VRAM usage and performance. +ARC_SINGLE_ALLOCATION_LIMIT = min(torch.xpu.get_device_properties(shared.cmd_opts.device_id).total_memory // 8, 4 * 1024 * 1024 * 1024) +orig_sdp_attn_func = torch.nn.functional.scaled_dot_product_attention +def torch_xpu_scaled_dot_product_attention( + query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False, *args, **kwargs +): + # cast to same dtype first + key = key.to(query.dtype) + value = value.to(query.dtype) + + N = query.shape[:-2] # Batch size + L = query.size(-2) # Target sequence length + E = query.size(-1) # Embedding dimension of the query and key + S = key.size(-2) # Source sequence length + Ev = value.size(-1) # Embedding dimension of the value + + total_batch_size = torch.numel(torch.empty(N)) + batch_size_limit = max(1, ARC_SINGLE_ALLOCATION_LIMIT // (L * S * query.element_size())) + + if total_batch_size <= batch_size_limit: + return orig_sdp_attn_func( + query, + key, + value, + attn_mask, + dropout_p, + is_causal, + *args, **kwargs + ) + + query = torch.reshape(query, (-1, L, E)) + key = torch.reshape(key, (-1, S, E)) + value = torch.reshape(value, (-1, S, Ev)) + if attn_mask is not None: + attn_mask = attn_mask.view(-1, L, S) + chunk_count = (total_batch_size + batch_size_limit - 1) // batch_size_limit + outputs = [] + for i in range(chunk_count): + attn_mask_chunk = ( + None + if attn_mask is None + else attn_mask[i * batch_size_limit : (i + 1) * batch_size_limit, :, :] + ) + chunk_output = orig_sdp_attn_func( + query[i * batch_size_limit : (i + 1) * batch_size_limit, :, :], + key[i * batch_size_limit : (i + 1) * batch_size_limit, :, :], + value[i * batch_size_limit : (i + 1) * batch_size_limit, :, :], + attn_mask_chunk, + dropout_p, + is_causal, + *args, **kwargs + ) + outputs.append(chunk_output) + result = torch.cat(outputs, dim=0) + return torch.reshape(result, (*N, L, Ev)) + + if has_xpu: # W/A for https://github.com/intel/intel-extension-for-pytorch/issues/452: torch.Generator API doesn't support XPU device CondFunc('torch.Generator', @@ -55,5 +117,5 @@ if has_xpu: lambda orig_func, tensors, dim=0, out=None: orig_func([t.to(tensors[0].dtype) for t in tensors], dim=dim, out=out), lambda orig_func, tensors, dim=0, out=None: not all(t.dtype == tensors[0].dtype for t in tensors)) CondFunc('torch.nn.functional.scaled_dot_product_attention', - lambda orig_func, query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False: orig_func(query, key.to(query.dtype), value.to(query.dtype), attn_mask, dropout_p, is_causal), - lambda orig_func, query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False: query.dtype != key.dtype or query.dtype != value.dtype) + lambda orig_func, *args, **kwargs: torch_xpu_scaled_dot_product_attention(*args, **kwargs), + lambda orig_func, query, *args, **kwargs: query.is_xpu) -- cgit v1.2.1 From f586f4973a0f715e30b42242bb0e6b3f88c37d90 Mon Sep 17 00:00:00 2001 From: Nuullll Date: Mon, 18 Dec 2023 19:44:52 +0800 Subject: Fix device id --- modules/xpu_specific.py | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/xpu_specific.py b/modules/xpu_specific.py index 0ebdd596..f7687a66 100644 --- a/modules/xpu_specific.py +++ b/modules/xpu_specific.py @@ -33,7 +33,7 @@ has_xpu = check_for_xpu() # so that SDPA of each chunk wouldn't require any allocation larger than ARC_SINGLE_ALLOCATION_LIMIT. # The heuristic limit (TOTAL_VRAM // 8) is tuned for Intel Arc A770 16G and Arc A750 8G, # which is the best trade-off between VRAM usage and performance. -ARC_SINGLE_ALLOCATION_LIMIT = min(torch.xpu.get_device_properties(shared.cmd_opts.device_id).total_memory // 8, 4 * 1024 * 1024 * 1024) +ARC_SINGLE_ALLOCATION_LIMIT = {} orig_sdp_attn_func = torch.nn.functional.scaled_dot_product_attention def torch_xpu_scaled_dot_product_attention( query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False, *args, **kwargs @@ -49,7 +49,10 @@ def torch_xpu_scaled_dot_product_attention( Ev = value.size(-1) # Embedding dimension of the value total_batch_size = torch.numel(torch.empty(N)) - batch_size_limit = max(1, ARC_SINGLE_ALLOCATION_LIMIT // (L * S * query.element_size())) + device_id = query.device.index + if device_id not in ARC_SINGLE_ALLOCATION_LIMIT: + ARC_SINGLE_ALLOCATION_LIMIT[device_id] = min(torch.xpu.get_device_properties(device_id).total_memory // 8, 4 * 1024 * 1024 * 1024) + batch_size_limit = max(1, ARC_SINGLE_ALLOCATION_LIMIT[device_id] // (L * S * query.element_size())) if total_batch_size <= batch_size_limit: return orig_sdp_attn_func( -- cgit v1.2.1 From 9feb034e343d6d7ef63395821658fb3774b30a24 Mon Sep 17 00:00:00 2001 From: wangqyqq Date: Thu, 21 Dec 2023 20:15:51 +0800 Subject: support for sdxl-inpaint model --- modules/processing.py | 19 +++++++++++++++++++ modules/sd_models_config.py | 6 +++++- modules/sd_models_xl.py | 5 +++++ 3 files changed, 29 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 6f01c95f..159548db 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -106,6 +106,20 @@ def txt2img_image_conditioning(sd_model, x, width, height): return x.new_zeros(x.shape[0], 2*sd_model.noise_augmentor.time_embed.dim, dtype=x.dtype, device=x.device) else: + sd = sd_model.model.state_dict() + diffusion_model_input = sd.get('diffusion_model.input_blocks.0.0.weight', None) + if diffusion_model_input.shape[1] == 9: + # The "masked-image" in this case will just be all 0.5 since the entire image is masked. + image_conditioning = torch.ones(x.shape[0], 3, height, width, device=x.device) * 0.5 + image_conditioning = images_tensor_to_samples(image_conditioning, + approximation_indexes.get(opts.sd_vae_encode_method)) + + # Add the fake full 1s mask to the first dimension. + image_conditioning = torch.nn.functional.pad(image_conditioning, (0, 0, 0, 0, 1, 0), value=1.0) + image_conditioning = image_conditioning.to(x.dtype) + + return image_conditioning + # Dummy zero conditioning if we're not using inpainting or unclip models. # Still takes up a bit of memory, but no encoder call. # Pretty sure we can just make this a 1x1 image since its not going to be used besides its batch size. @@ -362,6 +376,11 @@ class StableDiffusionProcessing: if self.sampler.conditioning_key == "crossattn-adm": return self.unclip_image_conditioning(source_image) + sd = self.sampler.model_wrap.inner_model.model.state_dict() + diffusion_model_input = sd.get('diffusion_model.input_blocks.0.0.weight', None) + if diffusion_model_input.shape[1] == 9: + return self.inpainting_image_conditioning(source_image, latent_image, image_mask=image_mask) + # Dummy zero conditioning if we're not using inpainting or depth model. return latent_image.new_zeros(latent_image.shape[0], 5, 1, 1) diff --git a/modules/sd_models_config.py b/modules/sd_models_config.py index deab2f6e..b38137eb 100644 --- a/modules/sd_models_config.py +++ b/modules/sd_models_config.py @@ -15,6 +15,7 @@ config_sd2v = os.path.join(sd_repo_configs_path, "v2-inference-v.yaml") config_sd2_inpainting = os.path.join(sd_repo_configs_path, "v2-inpainting-inference.yaml") config_sdxl = os.path.join(sd_xl_repo_configs_path, "sd_xl_base.yaml") config_sdxl_refiner = os.path.join(sd_xl_repo_configs_path, "sd_xl_refiner.yaml") +config_sdxl_inpainting = os.path.join(sd_configs_path, "sd_xl_inpaint.yaml") config_depth_model = os.path.join(sd_repo_configs_path, "v2-midas-inference.yaml") config_unclip = os.path.join(sd_repo_configs_path, "v2-1-stable-unclip-l-inference.yaml") config_unopenclip = os.path.join(sd_repo_configs_path, "v2-1-stable-unclip-h-inference.yaml") @@ -71,7 +72,10 @@ def guess_model_config_from_state_dict(sd, filename): sd2_variations_weight = sd.get('embedder.model.ln_final.weight', None) if sd.get('conditioner.embedders.1.model.ln_final.weight', None) is not None: - return config_sdxl + if diffusion_model_input.shape[1] == 9: + return config_sdxl_inpainting + else: + return config_sdxl if sd.get('conditioner.embedders.0.model.ln_final.weight', None) is not None: return config_sdxl_refiner elif sd.get('depth_model.model.pretrained.act_postprocess3.0.project.0.bias', None) is not None: diff --git a/modules/sd_models_xl.py b/modules/sd_models_xl.py index 01123321..d8a9a73b 100644 --- a/modules/sd_models_xl.py +++ b/modules/sd_models_xl.py @@ -34,6 +34,11 @@ def get_learned_conditioning(self: sgm.models.diffusion.DiffusionEngine, batch: def apply_model(self: sgm.models.diffusion.DiffusionEngine, x, t, cond): + sd = self.model.state_dict() + diffusion_model_input = sd.get('diffusion_model.input_blocks.0.0.weight', None) + if diffusion_model_input.shape[1] == 9: + x = torch.cat([x] + cond['c_concat'], dim=1) + return self.model(x, t, cond) -- cgit v1.2.1 From edfae95d90a49ea95394b772817a59dde4175222 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sat, 23 Dec 2023 01:21:00 +0900 Subject: prevent crash due to Script __init__ exception --- modules/scripts.py | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/scripts.py b/modules/scripts.py index b6fcf96e..3a766911 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -566,7 +566,12 @@ class ScriptRunner: auto_processing_scripts = scripts_auto_postprocessing.create_auto_preprocessing_script_data() for script_data in auto_processing_scripts + scripts_data: - script = script_data.script_class() + try: + script = script_data.script_class() + except Exception: + errors.report(f"Error # failed to initialize Script {script_data.module}: ", exc_info=True) + continue + script.filename = script_data.path script.is_txt2img = not is_img2img script.is_img2img = is_img2img -- cgit v1.2.1 From 00d4a4d4ac75903d8224e9beb1136584dd66fcd8 Mon Sep 17 00:00:00 2001 From: lanyeeee <1210347077@qq.com> Date: Tue, 26 Dec 2023 14:46:29 +0800 Subject: move thread-unsafe code to __init__ --- modules/api/api.py | 19 +++++++++---------- 1 file changed, 9 insertions(+), 10 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 7154c9d5..f0a68c67 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -251,6 +251,15 @@ class Api: self.default_script_arg_txt2img = [] self.default_script_arg_img2img = [] + script_runner = scripts.scripts_img2img + if not script_runner.scripts: + script_runner.initialize_scripts(True) + ui.create_ui() + if not self.default_script_arg_txt2img: + self.default_script_arg_txt2img = self.init_default_script_args(script_runner) + if not self.default_script_arg_img2img: + self.default_script_arg_img2img = self.init_default_script_args(script_runner) + def add_api_route(self, path: str, endpoint, **kwargs): if shared.cmd_opts.api_auth: return self.app.add_api_route(path, endpoint, dependencies=[Depends(self.auth)], **kwargs) @@ -339,11 +348,6 @@ class Api: task_id = txt2imgreq.force_task_id or create_task_id("txt2img") script_runner = scripts.scripts_txt2img - if not script_runner.scripts: - script_runner.initialize_scripts(False) - ui.create_ui() - if not self.default_script_arg_txt2img: - self.default_script_arg_txt2img = self.init_default_script_args(script_runner) selectable_scripts, selectable_script_idx = self.get_selectable_script(txt2imgreq.script_name, script_runner) populate = txt2imgreq.copy(update={ # Override __init__ params @@ -403,11 +407,6 @@ class Api: mask = decode_base64_to_image(mask) script_runner = scripts.scripts_img2img - if not script_runner.scripts: - script_runner.initialize_scripts(True) - ui.create_ui() - if not self.default_script_arg_img2img: - self.default_script_arg_img2img = self.init_default_script_args(script_runner) selectable_scripts, selectable_script_idx = self.get_selectable_script(img2imgreq.script_name, script_runner) populate = img2imgreq.copy(update={ # Override __init__ params -- cgit v1.2.1 From bfe418a58d39c69ca2672e7d8a1fd7ad2b34869b Mon Sep 17 00:00:00 2001 From: wangqyqq Date: Wed, 27 Dec 2023 10:20:56 +0800 Subject: add some codes for robust --- modules/processing.py | 24 +++++++++++++----------- modules/sd_models_xl.py | 5 +++-- 2 files changed, 16 insertions(+), 13 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 159548db..c05e608a 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -108,17 +108,18 @@ def txt2img_image_conditioning(sd_model, x, width, height): else: sd = sd_model.model.state_dict() diffusion_model_input = sd.get('diffusion_model.input_blocks.0.0.weight', None) - if diffusion_model_input.shape[1] == 9: - # The "masked-image" in this case will just be all 0.5 since the entire image is masked. - image_conditioning = torch.ones(x.shape[0], 3, height, width, device=x.device) * 0.5 - image_conditioning = images_tensor_to_samples(image_conditioning, - approximation_indexes.get(opts.sd_vae_encode_method)) + if diffusion_model_input is not None: + if diffusion_model_input.shape[1] == 9: + # The "masked-image" in this case will just be all 0.5 since the entire image is masked. + image_conditioning = torch.ones(x.shape[0], 3, height, width, device=x.device) * 0.5 + image_conditioning = images_tensor_to_samples(image_conditioning, + approximation_indexes.get(opts.sd_vae_encode_method)) - # Add the fake full 1s mask to the first dimension. - image_conditioning = torch.nn.functional.pad(image_conditioning, (0, 0, 0, 0, 1, 0), value=1.0) - image_conditioning = image_conditioning.to(x.dtype) + # Add the fake full 1s mask to the first dimension. + image_conditioning = torch.nn.functional.pad(image_conditioning, (0, 0, 0, 0, 1, 0), value=1.0) + image_conditioning = image_conditioning.to(x.dtype) - return image_conditioning + return image_conditioning # Dummy zero conditioning if we're not using inpainting or unclip models. # Still takes up a bit of memory, but no encoder call. @@ -378,8 +379,9 @@ class StableDiffusionProcessing: sd = self.sampler.model_wrap.inner_model.model.state_dict() diffusion_model_input = sd.get('diffusion_model.input_blocks.0.0.weight', None) - if diffusion_model_input.shape[1] == 9: - return self.inpainting_image_conditioning(source_image, latent_image, image_mask=image_mask) + if diffusion_model_input is not None: + if diffusion_model_input.shape[1] == 9: + return self.inpainting_image_conditioning(source_image, latent_image, image_mask=image_mask) # Dummy zero conditioning if we're not using inpainting or depth model. return latent_image.new_zeros(latent_image.shape[0], 5, 1, 1) diff --git a/modules/sd_models_xl.py b/modules/sd_models_xl.py index d8a9a73b..162d0fee 100644 --- a/modules/sd_models_xl.py +++ b/modules/sd_models_xl.py @@ -36,8 +36,9 @@ def get_learned_conditioning(self: sgm.models.diffusion.DiffusionEngine, batch: def apply_model(self: sgm.models.diffusion.DiffusionEngine, x, t, cond): sd = self.model.state_dict() diffusion_model_input = sd.get('diffusion_model.input_blocks.0.0.weight', None) - if diffusion_model_input.shape[1] == 9: - x = torch.cat([x] + cond['c_concat'], dim=1) + if diffusion_model_input is not None: + if diffusion_model_input.shape[1] == 9: + x = torch.cat([x] + cond['c_concat'], dim=1) return self.model(x, t, cond) -- cgit v1.2.1 From de04573438bc111f137359b8f4998780bf315275 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Thu, 28 Dec 2023 06:22:51 +0900 Subject: create utility truncate_path utli.truncate_path(target_path, base_path) return the target_path relative to base_path if target_path is a sub path of base_path else return the absolute path --- modules/util.py | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/util.py b/modules/util.py index 60afc067..4861bcb0 100644 --- a/modules/util.py +++ b/modules/util.py @@ -2,7 +2,7 @@ import os import re from modules import shared -from modules.paths_internal import script_path +from modules.paths_internal import script_path, cwd def natural_sort_key(s, regex=re.compile('([0-9]+)')): @@ -56,3 +56,13 @@ def ldm_print(*args, **kwargs): return print(*args, **kwargs) + + +def truncate_path(target_path, base_path=cwd): + abs_target, abs_base = os.path.abspath(target_path), os.path.abspath(base_path) + try: + if os.path.commonpath([abs_target, abs_base]) == abs_base: + return os.path.relpath(abs_target, abs_base) + except ValueError: + pass + return abs_target -- cgit v1.2.1 From af2951ed53da6d357aea9232538f9ea7e1cdc648 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Thu, 28 Dec 2023 06:52:33 +0900 Subject: base default image output on data_path Co-Authored-By: Alberto Cano <34340962+canoalberto@users.noreply.github.com> --- modules/paths_internal.py | 1 + modules/shared_options.py | 19 ++++++++++--------- 2 files changed, 11 insertions(+), 9 deletions(-) (limited to 'modules') diff --git a/modules/paths_internal.py b/modules/paths_internal.py index 89131a54..b86ecd7f 100644 --- a/modules/paths_internal.py +++ b/modules/paths_internal.py @@ -28,5 +28,6 @@ models_path = os.path.join(data_path, "models") extensions_dir = os.path.join(data_path, "extensions") extensions_builtin_dir = os.path.join(script_path, "extensions-builtin") config_states_dir = os.path.join(script_path, "config_states") +default_output_dir = os.path.join(data_path, "output") roboto_ttf_file = os.path.join(modules_path, 'Roboto-Regular.ttf') diff --git a/modules/shared_options.py b/modules/shared_options.py index fa542ba8..752a4f12 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -1,7 +1,8 @@ +import os import gradio as gr -from modules import localization, ui_components, shared_items, shared, interrogate, shared_gradio_themes -from modules.paths_internal import models_path, script_path, data_path, sd_configs_path, sd_default_config, sd_model_file, default_sd_model_file, extensions_dir, extensions_builtin_dir # noqa: F401 +from modules import localization, ui_components, shared_items, shared, interrogate, shared_gradio_themes, util +from modules.paths_internal import models_path, script_path, data_path, sd_configs_path, sd_default_config, sd_model_file, default_sd_model_file, extensions_dir, extensions_builtin_dir, default_output_dir # noqa: F401 from modules.shared_cmd_options import cmd_opts from modules.options import options_section, OptionInfo, OptionHTML, categories @@ -74,14 +75,14 @@ options_templates.update(options_section(('saving-images', "Saving images/grids" options_templates.update(options_section(('saving-paths', "Paths for saving", "saving"), { "outdir_samples": OptionInfo("", "Output directory for images; if empty, defaults to three directories below", component_args=hide_dirs), - "outdir_txt2img_samples": OptionInfo("outputs/txt2img-images", 'Output directory for txt2img images', component_args=hide_dirs), - "outdir_img2img_samples": OptionInfo("outputs/img2img-images", 'Output directory for img2img images', component_args=hide_dirs), - "outdir_extras_samples": OptionInfo("outputs/extras-images", 'Output directory for images from extras tab', component_args=hide_dirs), + "outdir_txt2img_samples": OptionInfo(util.truncate_path(os.path.join(default_output_dir, 'txt2img-images')), 'Output directory for txt2img images', component_args=hide_dirs), + "outdir_img2img_samples": OptionInfo(util.truncate_path(os.path.join(default_output_dir, 'img2img-images')), 'Output directory for img2img images', component_args=hide_dirs), + "outdir_extras_samples": OptionInfo(util.truncate_path(os.path.join(default_output_dir, 'extras-images')), 'Output directory for images from extras tab', component_args=hide_dirs), "outdir_grids": OptionInfo("", "Output directory for grids; if empty, defaults to two directories below", component_args=hide_dirs), - "outdir_txt2img_grids": OptionInfo("outputs/txt2img-grids", 'Output directory for txt2img grids', component_args=hide_dirs), - "outdir_img2img_grids": OptionInfo("outputs/img2img-grids", 'Output directory for img2img grids', component_args=hide_dirs), - "outdir_save": OptionInfo("log/images", "Directory for saving images using the Save button", component_args=hide_dirs), - "outdir_init_images": OptionInfo("outputs/init-images", "Directory for saving init images when using img2img", component_args=hide_dirs), + "outdir_txt2img_grids": OptionInfo(util.truncate_path(os.path.join(default_output_dir, 'txt2img-grids')), 'Output directory for txt2img grids', component_args=hide_dirs), + "outdir_img2img_grids": OptionInfo(util.truncate_path(os.path.join(default_output_dir, 'img2img-grids')), 'Output directory for img2img grids', component_args=hide_dirs), + "outdir_save": OptionInfo(util.truncate_path(os.path.join(data_path, 'log', 'images')), "Directory for saving images using the Save button", component_args=hide_dirs), + "outdir_init_images": OptionInfo(util.truncate_path(os.path.join(default_output_dir, 'init-images')), "Directory for saving init images when using img2img", component_args=hide_dirs), })) options_templates.update(options_section(('saving-to-dirs', "Saving to a directory", "saving"), { -- cgit v1.2.1 From 892e703b59b2f867d8a202a52fab1db89882ef86 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Thu, 28 Dec 2023 06:52:41 +0900 Subject: webpath use truncate_path --- modules/ui_gradio_extensions.py | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) (limited to 'modules') diff --git a/modules/ui_gradio_extensions.py b/modules/ui_gradio_extensions.py index 0d368f8b..a86c368e 100644 --- a/modules/ui_gradio_extensions.py +++ b/modules/ui_gradio_extensions.py @@ -1,17 +1,12 @@ import os import gradio as gr -from modules import localization, shared, scripts -from modules.paths import script_path, data_path, cwd +from modules import localization, shared, scripts, util +from modules.paths import script_path, data_path def webpath(fn): - if fn.startswith(cwd): - web_path = os.path.relpath(fn, cwd) - else: - web_path = os.path.abspath(fn) - - return f'file={web_path}?{os.path.getmtime(fn)}' + return f'file={util.truncate_path(fn)}?{os.path.getmtime(fn)}' def javascript_html(): -- cgit v1.2.1 From dc57ec0296e768ee91290e16ab262404837c566d Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Fri, 29 Dec 2023 01:56:48 +0900 Subject: save info of init image --- modules/processing.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 9351e3fb..141f2f11 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1482,7 +1482,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): # Save init image if opts.save_init_img: self.init_img_hash = hashlib.md5(img.tobytes()).hexdigest() - images.save_image(img, path=opts.outdir_init_images, basename=None, forced_filename=self.init_img_hash, save_to_dirs=False) + images.save_image(img, path=opts.outdir_init_images, basename=None, forced_filename=self.init_img_hash, save_to_dirs=False, existing_info=img.info) image = images.flatten(img, opts.img2img_background_color) -- cgit v1.2.1 From bb07cb6a0df60a96827125ffc09ea182a1ed272c Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 17 Dec 2023 10:22:03 +0300 Subject: a --- modules/api/api.py | 27 ++++++++++++++++++ modules/api/models.py | 2 ++ modules/generation_parameters_copypaste.py | 19 ++++++++++++ modules/processing.py | 2 +- modules/processing_scripts/refiner.py | 7 +++-- modules/processing_scripts/seed.py | 13 +++++---- modules/ui.py | 46 +++++++++++++++--------------- 7 files changed, 83 insertions(+), 33 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 7154c9d5..b3d70940 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -335,6 +335,29 @@ class Api: script_args[alwayson_script.args_from + idx] = request.alwayson_scripts[alwayson_script_name]["args"][idx] return script_args + def apply_infotext(self, request, tabname): + if not request.infotext: + return {} + + params = generation_parameters_copypaste.parse_generation_parameters(request.infotext) + + for field in generation_parameters_copypaste.paste_fields[tabname]["fields"]: + if not field.api: + continue + + value = field.function(params) if field.function else params.get(field.label) + target_type = request.__fields__[field.api].type_ + + if value is None: + continue + + if not isinstance(value, target_type): + value = target_type(value) + + setattr(request, field.api, value) + + return params + def text2imgapi(self, txt2imgreq: models.StableDiffusionTxt2ImgProcessingAPI): task_id = txt2imgreq.force_task_id or create_task_id("txt2img") @@ -342,6 +365,9 @@ class Api: if not script_runner.scripts: script_runner.initialize_scripts(False) ui.create_ui() + + infotext_params = self.apply_infotext(txt2imgreq, "txt2img") + if not self.default_script_arg_txt2img: self.default_script_arg_txt2img = self.init_default_script_args(script_runner) selectable_scripts, selectable_script_idx = self.get_selectable_script(txt2imgreq.script_name, script_runner) @@ -358,6 +384,7 @@ class Api: args.pop('script_name', None) args.pop('script_args', None) # will refeed them to the pipeline directly after initializing them args.pop('alwayson_scripts', None) + args.pop('infotext', None) script_args = self.init_script_args(txt2imgreq, self.default_script_arg_txt2img, selectable_scripts, selectable_script_idx, script_runner) diff --git a/modules/api/models.py b/modules/api/models.py index 58083a34..16edf11c 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -108,6 +108,7 @@ StableDiffusionTxt2ImgProcessingAPI = PydanticModelGenerator( {"key": "save_images", "type": bool, "default": False}, {"key": "alwayson_scripts", "type": dict, "default": {}}, {"key": "force_task_id", "type": str, "default": None}, + {"key": "infotext", "type": str, "default": None}, ] ).generate_model() @@ -126,6 +127,7 @@ StableDiffusionImg2ImgProcessingAPI = PydanticModelGenerator( {"key": "save_images", "type": bool, "default": False}, {"key": "alwayson_scripts", "type": dict, "default": {}}, {"key": "force_task_id", "type": str, "default": None}, + {"key": "infotext", "type": str, "default": None}, ] ).generate_model() diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index dbffe494..4b4727c4 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -28,6 +28,19 @@ class ParamBinding: self.paste_field_names = paste_field_names or [] +class PasteField(tuple): + def __new__(cls, component, target, *, api=None): + return super().__new__(cls, (component, target)) + + def __init__(self, component, target, *, api=None): + super().__init__() + + self.api = api + self.component = component + self.label = target if isinstance(target, str) else None + self.function = target if callable(target) else None + + paste_fields: dict[str, dict] = {} registered_param_bindings: list[ParamBinding] = [] @@ -84,6 +97,12 @@ def image_from_url_text(filedata): def add_paste_fields(tabname, init_img, fields, override_settings_component=None): + + if fields: + for i in range(len(fields)): + if not isinstance(fields[i], PasteField): + fields[i] = PasteField(*fields[i]) + paste_fields[tabname] = {"init_img": init_img, "fields": fields, "override_settings_component": override_settings_component} # backwards compatibility for existing extensions diff --git a/modules/processing.py b/modules/processing.py index 9351e3fb..ee2ccf46 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1135,7 +1135,7 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def init(self, all_prompts, all_seeds, all_subseeds): if self.enable_hr: - if self.hr_checkpoint_name: + if self.hr_checkpoint_name and self.hr_checkpoint_name != 'Use same checkpoint': self.hr_checkpoint_info = sd_models.get_closet_checkpoint_match(self.hr_checkpoint_name) if self.hr_checkpoint_info is None: diff --git a/modules/processing_scripts/refiner.py b/modules/processing_scripts/refiner.py index 29ccb78f..cefad32b 100644 --- a/modules/processing_scripts/refiner.py +++ b/modules/processing_scripts/refiner.py @@ -1,6 +1,7 @@ import gradio as gr from modules import scripts, sd_models +from modules.generation_parameters_copypaste import PasteField from modules.ui_common import create_refresh_button from modules.ui_components import InputAccordion @@ -31,9 +32,9 @@ class ScriptRefiner(scripts.ScriptBuiltinUI): return None if info is None else info.title self.infotext_fields = [ - (enable_refiner, lambda d: 'Refiner' in d), - (refiner_checkpoint, lambda d: lookup_checkpoint(d.get('Refiner'))), - (refiner_switch_at, 'Refiner switch at'), + PasteField(enable_refiner, lambda d: 'Refiner' in d), + PasteField(refiner_checkpoint, lambda d: lookup_checkpoint(d.get('Refiner')), api="refiner_checkpoint"), + PasteField(refiner_switch_at, 'Refiner switch at', api="refiner_switch_at"), ] return enable_refiner, refiner_checkpoint, refiner_switch_at diff --git a/modules/processing_scripts/seed.py b/modules/processing_scripts/seed.py index dc9c2da5..a3e16a12 100644 --- a/modules/processing_scripts/seed.py +++ b/modules/processing_scripts/seed.py @@ -3,6 +3,7 @@ import json import gradio as gr from modules import scripts, ui, errors +from modules.generation_parameters_copypaste import PasteField from modules.shared import cmd_opts from modules.ui_components import ToolButton @@ -51,12 +52,12 @@ class ScriptSeed(scripts.ScriptBuiltinUI): seed_checkbox.change(lambda x: gr.update(visible=x), show_progress=False, inputs=[seed_checkbox], outputs=[seed_extras]) self.infotext_fields = [ - (self.seed, "Seed"), - (seed_checkbox, lambda d: "Variation seed" in d or "Seed resize from-1" in d), - (subseed, "Variation seed"), - (subseed_strength, "Variation seed strength"), - (seed_resize_from_w, "Seed resize from-1"), - (seed_resize_from_h, "Seed resize from-2"), + PasteField(self.seed, "Seed", api="seed"), + PasteField(seed_checkbox, lambda d: "Variation seed" in d or "Seed resize from-1" in d), + PasteField(subseed, "Variation seed", api="subseed"), + PasteField(subseed_strength, "Variation seed strength", api="subseed_strength"), + PasteField(seed_resize_from_w, "Seed resize from-1", api="seed_resize_from_h"), + PasteField(seed_resize_from_h, "Seed resize from-2", api="seed_resize_from_w"), ] self.on_after_component(lambda x: connect_reuse_seed(self.seed, reuse_seed, x.component, False), elem_id=f'generation_info_{self.tabname}') diff --git a/modules/ui.py b/modules/ui.py index d80486dd..9db2407e 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -28,7 +28,7 @@ import modules.textual_inversion.textual_inversion as textual_inversion import modules.shared as shared from modules import prompt_parser from modules.sd_hijack import model_hijack -from modules.generation_parameters_copypaste import image_from_url_text +from modules.generation_parameters_copypaste import image_from_url_text, PasteField create_setting_component = ui_settings.create_setting_component @@ -436,28 +436,28 @@ def create_ui(): ) txt2img_paste_fields = [ - (toprow.prompt, "Prompt"), - (toprow.negative_prompt, "Negative prompt"), - (steps, "Steps"), - (sampler_name, "Sampler"), - (cfg_scale, "CFG scale"), - (width, "Size-1"), - (height, "Size-2"), - (batch_size, "Batch size"), - (toprow.ui_styles.dropdown, lambda d: d["Styles array"] if isinstance(d.get("Styles array"), list) else gr.update()), - (denoising_strength, "Denoising strength"), - (enable_hr, lambda d: "Denoising strength" in d and ("Hires upscale" in d or "Hires upscaler" in d or "Hires resize-1" in d)), - (hr_scale, "Hires upscale"), - (hr_upscaler, "Hires upscaler"), - (hr_second_pass_steps, "Hires steps"), - (hr_resize_x, "Hires resize-1"), - (hr_resize_y, "Hires resize-2"), - (hr_checkpoint_name, "Hires checkpoint"), - (hr_sampler_name, "Hires sampler"), - (hr_sampler_container, lambda d: gr.update(visible=True) if d.get("Hires sampler", "Use same sampler") != "Use same sampler" or d.get("Hires checkpoint", "Use same checkpoint") != "Use same checkpoint" else gr.update()), - (hr_prompt, "Hires prompt"), - (hr_negative_prompt, "Hires negative prompt"), - (hr_prompts_container, lambda d: gr.update(visible=True) if d.get("Hires prompt", "") != "" or d.get("Hires negative prompt", "") != "" else gr.update()), + PasteField(toprow.prompt, "Prompt", api="prompt"), + PasteField(toprow.negative_prompt, "Negative prompt", api="negative_prompt"), + PasteField(steps, "Steps", api="steps"), + PasteField(sampler_name, "Sampler", api="sampler_name"), + PasteField(cfg_scale, "CFG scale", api="cfg_scale"), + PasteField(width, "Size-1", api="width"), + PasteField(height, "Size-2", api="height"), + PasteField(batch_size, "Batch size", api="batch_size"), + PasteField(toprow.ui_styles.dropdown, lambda d: d["Styles array"] if isinstance(d.get("Styles array"), list) else gr.update(), api="styles"), + PasteField(denoising_strength, "Denoising strength", api="denoising_strength"), + PasteField(enable_hr, lambda d: "Denoising strength" in d and ("Hires upscale" in d or "Hires upscaler" in d or "Hires resize-1" in d), api="enable_hr"), + PasteField(hr_scale, "Hires upscale", api="hr_scale"), + PasteField(hr_upscaler, "Hires upscaler", api="hr_upscaler"), + PasteField(hr_second_pass_steps, "Hires steps", api="hr_second_pass_steps"), + PasteField(hr_resize_x, "Hires resize-1", api="hr_resize_x"), + PasteField(hr_resize_y, "Hires resize-2", api="hr_resize_y"), + PasteField(hr_checkpoint_name, "Hires checkpoint", api="hr_checkpoint_name"), + PasteField(hr_sampler_name, "Hires sampler", api="hr_sampler_name"), + PasteField(hr_sampler_container, lambda d: gr.update(visible=True) if d.get("Hires sampler", "Use same sampler") != "Use same sampler" or d.get("Hires checkpoint", "Use same checkpoint") != "Use same checkpoint" else gr.update()), + PasteField(hr_prompt, "Hires prompt", api="hr_prompt"), + PasteField(hr_negative_prompt, "Hires negative prompt", api="hr_negative_prompt"), + PasteField(hr_prompts_container, lambda d: gr.update(visible=True) if d.get("Hires prompt", "") != "" or d.get("Hires negative prompt", "") != "" else gr.update()), *scripts.scripts_txt2img.infotext_fields ] parameters_copypaste.add_paste_fields("txt2img", None, txt2img_paste_fields, override_settings) -- cgit v1.2.1 From ba92135a2ba9e210ce5370715e2defcb43df70d1 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 30 Dec 2023 12:11:09 +0300 Subject: add override_settings support for infotext API --- modules/api/api.py | 10 +++++ modules/generation_parameters_copypaste.py | 66 ++++++++++++++++++++---------- 2 files changed, 54 insertions(+), 22 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index b3d70940..fb108486 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -341,6 +341,7 @@ class Api: params = generation_parameters_copypaste.parse_generation_parameters(request.infotext) + handled_fields = {} for field in generation_parameters_copypaste.paste_fields[tabname]["fields"]: if not field.api: continue @@ -355,6 +356,15 @@ class Api: value = target_type(value) setattr(request, field.api, value) + handled_fields[field.label] = 1 + + if request.override_settings is None: + request.override_settings = {} + + overriden_settings = generation_parameters_copypaste.get_override_settings(params, skip_fields=handled_fields) + for infotext_text, setting_name, value in overriden_settings: + if setting_name not in request.override_settings: + request.override_settings[setting_name] = value return params diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index 4b4727c4..86a36c32 100644 --- a/modules/generation_parameters_copypaste.py +++ b/modules/generation_parameters_copypaste.py @@ -390,6 +390,48 @@ def create_override_settings_dict(text_pairs): return res +def get_override_settings(params, *, skip_fields=None): + """Returns a list of settings overrides from the infotext parameters dictionary. + + This function checks the `params` dictionary for any keys that correspond to settings in `shared.opts` and returns + a list of tuples containing the parameter name, setting name, and new value cast to correct type. + + It checks for conditions before adding an override: + - ignores settings that match the current value + - ignores parameter keys present in skip_fields argument. + + Example input: + {"Clip skip": "2"} + + Example output: + [("Clip skip", "CLIP_stop_at_last_layers", 2)] + """ + + res = [] + + mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] + for param_name, setting_name in mapping + infotext_to_setting_name_mapping: + if param_name in (skip_fields or {}): + continue + + v = params.get(param_name, None) + if v is None: + continue + + if setting_name == "sd_model_checkpoint" and shared.opts.disable_weights_auto_swap: + continue + + v = shared.opts.cast_value(setting_name, v) + current_value = getattr(shared.opts, setting_name, None) + + if v == current_value: + continue + + res.append((param_name, setting_name, v)) + + return res + + def connect_paste(button, paste_fields, input_comp, override_settings_component, tabname): def paste_func(prompt): if not prompt and not shared.cmd_opts.hide_ui_dir_config: @@ -431,29 +473,9 @@ def connect_paste(button, paste_fields, input_comp, override_settings_component, already_handled_fields = {key: 1 for _, key in paste_fields} def paste_settings(params): - vals = {} - - mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] - for param_name, setting_name in mapping + infotext_to_setting_name_mapping: - if param_name in already_handled_fields: - continue - - v = params.get(param_name, None) - if v is None: - continue - - if setting_name == "sd_model_checkpoint" and shared.opts.disable_weights_auto_swap: - continue - - v = shared.opts.cast_value(setting_name, v) - current_value = getattr(shared.opts, setting_name, None) - - if v == current_value: - continue - - vals[param_name] = v + vals = get_override_settings(params, skip_fields=already_handled_fields) - vals_pairs = [f"{k}: {v}" for k, v in vals.items()] + vals_pairs = [f"{infotext_text}: {value}" for infotext_text, setting_name, value in vals] return gr.Dropdown.update(value=vals_pairs, choices=vals_pairs, visible=bool(vals_pairs)) -- cgit v1.2.1 From 8b08b78c03f09898455d54cf099225ed5f8de1ee Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 30 Dec 2023 12:27:23 +0300 Subject: make it so that if an option from infotext conflicts with an argument from API, the latter overrides the former --- modules/api/api.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index fb108486..cabccb4c 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -339,6 +339,7 @@ class Api: if not request.infotext: return {} + set_fields = request.model_dump(exclude_unset=True) if hasattr(request, "request") else request.dict(exclude_unset=True) # pydantic v1/v2 have differenrt names for this params = generation_parameters_copypaste.parse_generation_parameters(request.infotext) handled_fields = {} @@ -346,6 +347,9 @@ class Api: if not field.api: continue + if field.api in set_fields: + continue + value = field.function(params) if field.function else params.get(field.label) target_type = request.__fields__[field.api].type_ @@ -376,7 +380,7 @@ class Api: script_runner.initialize_scripts(False) ui.create_ui() - infotext_params = self.apply_infotext(txt2imgreq, "txt2img") + self.apply_infotext(txt2imgreq, "txt2img") if not self.default_script_arg_txt2img: self.default_script_arg_txt2img = self.init_default_script_args(script_runner) -- cgit v1.2.1 From 0aacd4c72b4008d7153e747301fe8c5ffca57f85 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 30 Dec 2023 13:33:18 +0300 Subject: add support for alwayson scripts for infotext API --- modules/api/api.py | 61 +++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 42 insertions(+), 19 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index cabccb4c..946cfe4a 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -312,8 +312,13 @@ class Api: script_args[script.args_from:script.args_to] = ui_default_values return script_args - def init_script_args(self, request, default_script_args, selectable_scripts, selectable_idx, script_runner): + def init_script_args(self, request, default_script_args, selectable_scripts, selectable_idx, script_runner, *, input_script_args=None): script_args = default_script_args.copy() + + if input_script_args is not None: + for index, value in input_script_args.items(): + script_args[index] = value + # position 0 in script_arg is the idx+1 of the selectable script that is going to be run when using scripts.scripts_*2img.run() if selectable_scripts: script_args[selectable_scripts.args_from:selectable_scripts.args_to] = request.script_args @@ -335,41 +340,58 @@ class Api: script_args[alwayson_script.args_from + idx] = request.alwayson_scripts[alwayson_script_name]["args"][idx] return script_args - def apply_infotext(self, request, tabname): + def apply_infotext(self, request, tabname, *, script_runner=None, mentioned_script_args=None): if not request.infotext: return {} + possible_fields = generation_parameters_copypaste.paste_fields[tabname]["fields"] set_fields = request.model_dump(exclude_unset=True) if hasattr(request, "request") else request.dict(exclude_unset=True) # pydantic v1/v2 have differenrt names for this params = generation_parameters_copypaste.parse_generation_parameters(request.infotext) - handled_fields = {} - for field in generation_parameters_copypaste.paste_fields[tabname]["fields"]: - if not field.api: - continue - - if field.api in set_fields: - continue - + def get_field_value(field, params): value = field.function(params) if field.function else params.get(field.label) - target_type = request.__fields__[field.api].type_ - if value is None: - continue + return None + + if field.api in request.__fields__: + target_type = request.__fields__[field.api].type_ + else: + target_type = type(field.component.value) + + if target_type == type(None): + return None if not isinstance(value, target_type): value = target_type(value) - setattr(request, field.api, value) - handled_fields[field.label] = 1 + return value + + for field in possible_fields: + if not field.api: + continue + + if field.api in set_fields: + continue + + value = get_field_value(field, params) + if value is not None: + setattr(request, field.api, value) if request.override_settings is None: request.override_settings = {} - overriden_settings = generation_parameters_copypaste.get_override_settings(params, skip_fields=handled_fields) - for infotext_text, setting_name, value in overriden_settings: + overriden_settings = generation_parameters_copypaste.get_override_settings(params) + for _, setting_name, value in overriden_settings: if setting_name not in request.override_settings: request.override_settings[setting_name] = value + if script_runner is not None and mentioned_script_args is not None: + indexes = {v: i for i, v in enumerate(script_runner.inputs)} + script_fields = ((field, indexes[field.component]) for field in possible_fields if field.component in indexes) + + for field, index in script_fields: + mentioned_script_args[index] = get_field_value(field, params) + return params def text2imgapi(self, txt2imgreq: models.StableDiffusionTxt2ImgProcessingAPI): @@ -380,7 +402,8 @@ class Api: script_runner.initialize_scripts(False) ui.create_ui() - self.apply_infotext(txt2imgreq, "txt2img") + infotext_script_args = {} + self.apply_infotext(txt2imgreq, "txt2img", script_runner=script_runner, mentioned_script_args=infotext_script_args) if not self.default_script_arg_txt2img: self.default_script_arg_txt2img = self.init_default_script_args(script_runner) @@ -400,7 +423,7 @@ class Api: args.pop('alwayson_scripts', None) args.pop('infotext', None) - script_args = self.init_script_args(txt2imgreq, self.default_script_arg_txt2img, selectable_scripts, selectable_script_idx, script_runner) + script_args = self.init_script_args(txt2imgreq, self.default_script_arg_txt2img, selectable_scripts, selectable_script_idx, script_runner, input_script_args=infotext_script_args) send_images = args.pop('send_images', True) args.pop('save_images', None) -- cgit v1.2.1 From 11a435b4697c2d735a117f31944c4ebe59c2504c Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 30 Dec 2023 13:34:46 +0300 Subject: img2img support for infotext API --- modules/api/api.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 946cfe4a..2c8dc2a0 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -470,6 +470,10 @@ class Api: if not script_runner.scripts: script_runner.initialize_scripts(True) ui.create_ui() + + infotext_script_args = {} + self.apply_infotext(img2imgreq, "img2img", script_runner=script_runner, mentioned_script_args=infotext_script_args) + if not self.default_script_arg_img2img: self.default_script_arg_img2img = self.init_default_script_args(script_runner) selectable_scripts, selectable_script_idx = self.get_selectable_script(img2imgreq.script_name, script_runner) @@ -489,7 +493,7 @@ class Api: args.pop('script_args', None) # will refeed them to the pipeline directly after initializing them args.pop('alwayson_scripts', None) - script_args = self.init_script_args(img2imgreq, self.default_script_arg_img2img, selectable_scripts, selectable_script_idx, script_runner) + script_args = self.init_script_args(img2imgreq, self.default_script_arg_img2img, selectable_scripts, selectable_script_idx, script_runner, input_script_args=infotext_script_args) send_images = args.pop('send_images', True) args.pop('save_images', None) -- cgit v1.2.1 From 8f1826375943718463cec3af97a37886249bdb44 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 30 Dec 2023 13:48:25 +0300 Subject: fix bad values read from infotext for API, add comment --- modules/api/api.py | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 2c8dc2a0..2918f785 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -341,6 +341,13 @@ class Api: return script_args def apply_infotext(self, request, tabname, *, script_runner=None, mentioned_script_args=None): + """Processes `infotext` field from the `request`, and sets other fields of the `request` accoring to what's in infotext. + + If request already has a field set, and that field is encountered in infotext too, the value from infotext is ignored. + + Additionally, fills `mentioned_script_args` dict with index: value pairs for script arguments read from infotext. + """ + if not request.infotext: return {} @@ -361,7 +368,10 @@ class Api: if target_type == type(None): return None - if not isinstance(value, target_type): + if isinstance(value, dict) and value.get('__type__') == 'generic_update': # this is a gradio.update rather than a value + value = value.get('value') + + if value is not None and not isinstance(value, target_type): value = target_type(value) return value @@ -390,7 +400,12 @@ class Api: script_fields = ((field, indexes[field.component]) for field in possible_fields if field.component in indexes) for field, index in script_fields: - mentioned_script_args[index] = get_field_value(field, params) + value = get_field_value(field, params) + + if value is None: + continue + + mentioned_script_args[index] = value return params -- cgit v1.2.1 From c069c2c5628728c9506dd034ef98e6335fd5bb34 Mon Sep 17 00:00:00 2001 From: lanyeeee <1210347077@qq.com> Date: Sat, 30 Dec 2023 21:32:22 +0800 Subject: add locks to ensure init args are thread-safe --- modules/api/api.py | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index f0a68c67..45c5c507 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -251,14 +251,10 @@ class Api: self.default_script_arg_txt2img = [] self.default_script_arg_img2img = [] - script_runner = scripts.scripts_img2img - if not script_runner.scripts: - script_runner.initialize_scripts(True) - ui.create_ui() - if not self.default_script_arg_txt2img: - self.default_script_arg_txt2img = self.init_default_script_args(script_runner) - if not self.default_script_arg_img2img: - self.default_script_arg_img2img = self.init_default_script_args(script_runner) + self.txt2img_script_arg_init_lock = Lock() + self.img2img_script_arg_init_lock = Lock() + + def add_api_route(self, path: str, endpoint, **kwargs): if shared.cmd_opts.api_auth: @@ -348,6 +344,12 @@ class Api: task_id = txt2imgreq.force_task_id or create_task_id("txt2img") script_runner = scripts.scripts_txt2img + with self.txt2img_script_arg_init_lock: + if not script_runner.scripts: + script_runner.initialize_scripts(False) + ui.create_ui() + if not self.default_script_arg_txt2img: + self.default_script_arg_txt2img = self.init_default_script_args(script_runner) selectable_scripts, selectable_script_idx = self.get_selectable_script(txt2imgreq.script_name, script_runner) populate = txt2imgreq.copy(update={ # Override __init__ params @@ -407,6 +409,12 @@ class Api: mask = decode_base64_to_image(mask) script_runner = scripts.scripts_img2img + with self.img2img_script_arg_init_lock: + if not script_runner.scripts: + script_runner.initialize_scripts(True) + ui.create_ui() + if not self.default_script_arg_img2img: + self.default_script_arg_img2img = self.init_default_script_args(script_runner) selectable_scripts, selectable_script_idx = self.get_selectable_script(img2imgreq.script_name, script_runner) populate = img2imgreq.copy(update={ # Override __init__ params -- cgit v1.2.1 From 31992eff9b9714c158b12cec16dfe66c76270dfa Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 30 Dec 2023 16:51:02 +0300 Subject: make it possible again to extract styles that have whitespace at the end. --- modules/styles.py | 47 +++++++++++++++++------------------------------ 1 file changed, 17 insertions(+), 30 deletions(-) (limited to 'modules') diff --git a/modules/styles.py b/modules/styles.py index 81d9800d..026c4300 100644 --- a/modules/styles.py +++ b/modules/styles.py @@ -30,38 +30,29 @@ def apply_styles_to_prompt(prompt, styles): return prompt -def unwrap_style_text_from_prompt(style_text, prompt): - """ - Checks the prompt to see if the style text is wrapped around it. If so, - returns True plus the prompt text without the style text. Otherwise, returns - False with the original prompt. +def extract_style_text_from_prompt(style_text, prompt): + """This function extracts the text from a given prompt based on a provided style text. It checks if the style text contains the placeholder {prompt} or if it appears at the end of the prompt. If a match is found, it returns True along with the extracted text. Otherwise, it returns False and the original prompt. - Note that the "cleaned" version of the style text is only used for matching - purposes here. It isn't returned; the original style text is not modified. + extract_style_text_from_prompt("masterpiece", "1girl, art by greg, masterpiece") outputs (True, "1girl, art by greg") + extract_style_text_from_prompt("masterpiece, {prompt}", "masterpiece, 1girl, art by greg") outputs (True, "1girl, art by greg") + extract_style_text_from_prompt("masterpiece, {prompt}", "exquisite, 1girl, art by greg") outputs (False, "exquisite, 1girl, art by greg") """ - stripped_prompt = prompt - stripped_style_text = style_text + + stripped_prompt = prompt.strip() + stripped_style_text = style_text.strip() + if "{prompt}" in stripped_style_text: - # Work out whether the prompt is wrapped in the style text. If so, we - # return True and the "inner" prompt text that isn't part of the style. - try: - left, right = stripped_style_text.split("{prompt}", 2) - except ValueError as e: - # If the style text has multple "{prompt}"s, we can't split it into - # two parts. This is an error, but we can't do anything about it. - print(f"Unable to compare style text to prompt:\n{style_text}") - print(f"Error: {e}") - return False, prompt + left, right = stripped_style_text.split("{prompt}", 2) if stripped_prompt.startswith(left) and stripped_prompt.endswith(right): - prompt = stripped_prompt[len(left) : len(stripped_prompt) - len(right)] + prompt = stripped_prompt[len(left):len(stripped_prompt)-len(right)] return True, prompt else: - # Work out whether the given prompt ends with the style text. If so, we - # return True and the prompt text up to where the style text starts. if stripped_prompt.endswith(stripped_style_text): - prompt = stripped_prompt[: len(stripped_prompt) - len(stripped_style_text)] - if prompt.endswith(", "): + prompt = stripped_prompt[:len(stripped_prompt)-len(stripped_style_text)] + + if prompt.endswith(', '): prompt = prompt[:-2] + return True, prompt return False, prompt @@ -76,15 +67,11 @@ def extract_original_prompts(style: PromptStyle, prompt, negative_prompt): if not style.prompt and not style.negative_prompt: return False, prompt, negative_prompt - match_positive, extracted_positive = unwrap_style_text_from_prompt( - style.prompt, prompt - ) + match_positive, extracted_positive = extract_style_text_from_prompt(style.prompt, prompt) if not match_positive: return False, prompt, negative_prompt - match_negative, extracted_negative = unwrap_style_text_from_prompt( - style.negative_prompt, negative_prompt - ) + match_negative, extracted_negative = extract_style_text_from_prompt(style.negative_prompt, negative_prompt) if not match_negative: return False, prompt, negative_prompt -- cgit v1.2.1 From 7aa27b000a3087dcb5cc7254600064bf70cacd3e Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Mon, 25 Dec 2023 14:44:15 +0200 Subject: Add types to split_grid --- modules/images.py | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/images.py b/modules/images.py index 16f9ae7c..d30e8865 100644 --- a/modules/images.py +++ b/modules/images.py @@ -64,9 +64,8 @@ def image_grid(imgs, batch_size=1, rows=None): Grid = namedtuple("Grid", ["tiles", "tile_w", "tile_h", "image_w", "image_h", "overlap"]) -def split_grid(image, tile_w=512, tile_h=512, overlap=64): - w = image.width - h = image.height +def split_grid(image: Image.Image, tile_w: int = 512, tile_h: int = 512, overlap: int = 64) -> Grid: + w, h = image.size non_overlap_width = tile_w - overlap non_overlap_height = tile_h - overlap -- cgit v1.2.1 From 12c6f37f8e4b1d1d643c9d8d5dfc763c3203c728 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Wed, 27 Dec 2023 11:01:45 +0200 Subject: Add tile_count property to Grid --- modules/images.py | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/images.py b/modules/images.py index d30e8865..87a7bf22 100644 --- a/modules/images.py +++ b/modules/images.py @@ -61,7 +61,13 @@ def image_grid(imgs, batch_size=1, rows=None): return grid -Grid = namedtuple("Grid", ["tiles", "tile_w", "tile_h", "image_w", "image_h", "overlap"]) +class Grid(namedtuple("_Grid", ["tiles", "tile_w", "tile_h", "image_w", "image_h", "overlap"])): + @property + def tile_count(self) -> int: + """ + The total number of tiles in the grid. + """ + return sum(len(row[2]) for row in self.tiles) def split_grid(image: Image.Image, tile_w: int = 512, tile_h: int = 512, overlap: int = 64) -> Grid: -- cgit v1.2.1 From e472383acbb9e07dca311abe5fb16ee2675e410a Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Wed, 27 Dec 2023 11:04:33 +0200 Subject: Refactor esrgan_upscale to more generic upscale_with_model --- modules/esrgan_model.py | 47 ++++++--------------------------- modules/upscaler_utils.py | 66 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 74 insertions(+), 39 deletions(-) create mode 100644 modules/upscaler_utils.py (limited to 'modules') diff --git a/modules/esrgan_model.py b/modules/esrgan_model.py index 02a1727d..c0d22a99 100644 --- a/modules/esrgan_model.py +++ b/modules/esrgan_model.py @@ -1,13 +1,12 @@ import sys -import numpy as np import torch -from PIL import Image import modules.esrgan_model_arch as arch -from modules import modelloader, images, devices +from modules import modelloader, devices from modules.shared import opts from modules.upscaler import Upscaler, UpscalerData +from modules.upscaler_utils import upscale_with_model def mod2normal(state_dict): @@ -190,40 +189,10 @@ class UpscalerESRGAN(Upscaler): return model -def upscale_without_tiling(model, img): - img = np.array(img) - img = img[:, :, ::-1] - img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255 - img = torch.from_numpy(img).float() - img = img.unsqueeze(0).to(devices.device_esrgan) - with torch.no_grad(): - output = model(img) - output = output.squeeze().float().cpu().clamp_(0, 1).numpy() - output = 255. * np.moveaxis(output, 0, 2) - output = output.astype(np.uint8) - output = output[:, :, ::-1] - return Image.fromarray(output, 'RGB') - - def esrgan_upscale(model, img): - if opts.ESRGAN_tile == 0: - return upscale_without_tiling(model, img) - - grid = images.split_grid(img, opts.ESRGAN_tile, opts.ESRGAN_tile, opts.ESRGAN_tile_overlap) - newtiles = [] - scale_factor = 1 - - for y, h, row in grid.tiles: - newrow = [] - for tiledata in row: - x, w, tile = tiledata - - output = upscale_without_tiling(model, tile) - scale_factor = output.width // tile.width - - newrow.append([x * scale_factor, w * scale_factor, output]) - newtiles.append([y * scale_factor, h * scale_factor, newrow]) - - newgrid = images.Grid(newtiles, grid.tile_w * scale_factor, grid.tile_h * scale_factor, grid.image_w * scale_factor, grid.image_h * scale_factor, grid.overlap * scale_factor) - output = images.combine_grid(newgrid) - return output + return upscale_with_model( + model, + img, + tile_size=opts.ESRGAN_tile, + tile_overlap=opts.ESRGAN_tile_overlap, + ) diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py new file mode 100644 index 00000000..8bdda51c --- /dev/null +++ b/modules/upscaler_utils.py @@ -0,0 +1,66 @@ +import logging +from typing import Callable + +import numpy as np +import torch +import tqdm +from PIL import Image + +from modules import devices, images + +logger = logging.getLogger(__name__) + + +def upscale_without_tiling(model, img: Image.Image): + img = np.array(img) + img = img[:, :, ::-1] + img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255 + img = torch.from_numpy(img).float() + img = img.unsqueeze(0).to(devices.device_esrgan) + with torch.no_grad(): + output = model(img) + output = output.squeeze().float().cpu().clamp_(0, 1).numpy() + output = 255. * np.moveaxis(output, 0, 2) + output = output.astype(np.uint8) + output = output[:, :, ::-1] + return Image.fromarray(output, 'RGB') + + +def upscale_with_model( + model: Callable[[torch.Tensor], torch.Tensor], + img: Image.Image, + *, + tile_size: int, + tile_overlap: int = 0, + desc="tiled upscale", +) -> Image.Image: + if tile_size <= 0: + logger.debug("Upscaling %s without tiling", img) + output = upscale_without_tiling(model, img) + logger.debug("=> %s", output) + return output + + grid = images.split_grid(img, tile_size, tile_size, tile_overlap) + newtiles = [] + + with tqdm.tqdm(total=grid.tile_count, desc=desc) as p: + for y, h, row in grid.tiles: + newrow = [] + for x, w, tile in row: + logger.debug("Tile (%d, %d) %s...", x, y, tile) + output = upscale_without_tiling(model, tile) + scale_factor = output.width // tile.width + logger.debug("=> %s (scale factor %s)", output, scale_factor) + newrow.append([x * scale_factor, w * scale_factor, output]) + p.update(1) + newtiles.append([y * scale_factor, h * scale_factor, newrow]) + + newgrid = images.Grid( + newtiles, + tile_w=grid.tile_w * scale_factor, + tile_h=grid.tile_h * scale_factor, + image_w=grid.image_w * scale_factor, + image_h=grid.image_h * scale_factor, + overlap=grid.overlap * scale_factor, + ) + return images.combine_grid(newgrid) -- cgit v1.2.1 From b0f59342346b1c8b405f97c0e0bb01c6ae05c601 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Mon, 25 Dec 2023 14:43:51 +0200 Subject: Use Spandrel for upscaling and face restoration architectures (aside from GFPGAN and LDSR) --- modules/codeformer/codeformer_arch.py | 276 -------------------- modules/codeformer/vqgan_arch.py | 435 ------------------------------- modules/codeformer_model.py | 195 +++++++------- modules/esrgan_model.py | 153 +---------- modules/esrgan_model_arch.py | 465 ---------------------------------- modules/gfpgan_model.py | 13 +- modules/launch_utils.py | 7 - modules/modelloader.py | 16 ++ modules/paths.py | 1 - modules/realesrgan_model.py | 153 +++++------ modules/sysinfo.py | 2 - modules/upscaler.py | 3 + 12 files changed, 198 insertions(+), 1521 deletions(-) delete mode 100644 modules/codeformer/codeformer_arch.py delete mode 100644 modules/codeformer/vqgan_arch.py delete mode 100644 modules/esrgan_model_arch.py (limited to 'modules') diff --git a/modules/codeformer/codeformer_arch.py b/modules/codeformer/codeformer_arch.py deleted file mode 100644 index 12db6814..00000000 --- a/modules/codeformer/codeformer_arch.py +++ /dev/null @@ -1,276 +0,0 @@ -# this file is copied from CodeFormer repository. Please see comment in modules/codeformer_model.py - -import math -import torch -from torch import nn, Tensor -import torch.nn.functional as F -from typing import Optional - -from modules.codeformer.vqgan_arch import VQAutoEncoder, ResBlock -from basicsr.utils.registry import ARCH_REGISTRY - -def calc_mean_std(feat, eps=1e-5): - """Calculate mean and std for adaptive_instance_normalization. - - Args: - feat (Tensor): 4D tensor. - eps (float): A small value added to the variance to avoid - divide-by-zero. Default: 1e-5. - """ - size = feat.size() - assert len(size) == 4, 'The input feature should be 4D tensor.' - b, c = size[:2] - feat_var = feat.view(b, c, -1).var(dim=2) + eps - feat_std = feat_var.sqrt().view(b, c, 1, 1) - feat_mean = feat.view(b, c, -1).mean(dim=2).view(b, c, 1, 1) - return feat_mean, feat_std - - -def adaptive_instance_normalization(content_feat, style_feat): - """Adaptive instance normalization. - - Adjust the reference features to have the similar color and illuminations - as those in the degradate features. - - Args: - content_feat (Tensor): The reference feature. - style_feat (Tensor): The degradate features. - """ - size = content_feat.size() - style_mean, style_std = calc_mean_std(style_feat) - content_mean, content_std = calc_mean_std(content_feat) - normalized_feat = (content_feat - content_mean.expand(size)) / content_std.expand(size) - return normalized_feat * style_std.expand(size) + style_mean.expand(size) - - -class PositionEmbeddingSine(nn.Module): - """ - This is a more standard version of the position embedding, very similar to the one - used by the Attention is all you need paper, generalized to work on images. - """ - - def __init__(self, num_pos_feats=64, temperature=10000, normalize=False, scale=None): - super().__init__() - self.num_pos_feats = num_pos_feats - self.temperature = temperature - self.normalize = normalize - if scale is not None and normalize is False: - raise ValueError("normalize should be True if scale is passed") - if scale is None: - scale = 2 * math.pi - self.scale = scale - - def forward(self, x, mask=None): - if mask is None: - mask = torch.zeros((x.size(0), x.size(2), x.size(3)), device=x.device, dtype=torch.bool) - not_mask = ~mask - y_embed = not_mask.cumsum(1, dtype=torch.float32) - x_embed = not_mask.cumsum(2, dtype=torch.float32) - if self.normalize: - eps = 1e-6 - y_embed = y_embed / (y_embed[:, -1:, :] + eps) * self.scale - x_embed = x_embed / (x_embed[:, :, -1:] + eps) * self.scale - - dim_t = torch.arange(self.num_pos_feats, dtype=torch.float32, device=x.device) - dim_t = self.temperature ** (2 * (dim_t // 2) / self.num_pos_feats) - - pos_x = x_embed[:, :, :, None] / dim_t - pos_y = y_embed[:, :, :, None] / dim_t - pos_x = torch.stack( - (pos_x[:, :, :, 0::2].sin(), pos_x[:, :, :, 1::2].cos()), dim=4 - ).flatten(3) - pos_y = torch.stack( - (pos_y[:, :, :, 0::2].sin(), pos_y[:, :, :, 1::2].cos()), dim=4 - ).flatten(3) - pos = torch.cat((pos_y, pos_x), dim=3).permute(0, 3, 1, 2) - return pos - -def _get_activation_fn(activation): - """Return an activation function given a string""" - if activation == "relu": - return F.relu - if activation == "gelu": - return F.gelu - if activation == "glu": - return F.glu - raise RuntimeError(F"activation should be relu/gelu, not {activation}.") - - -class TransformerSALayer(nn.Module): - def __init__(self, embed_dim, nhead=8, dim_mlp=2048, dropout=0.0, activation="gelu"): - super().__init__() - self.self_attn = nn.MultiheadAttention(embed_dim, nhead, dropout=dropout) - # Implementation of Feedforward model - MLP - self.linear1 = nn.Linear(embed_dim, dim_mlp) - self.dropout = nn.Dropout(dropout) - self.linear2 = nn.Linear(dim_mlp, embed_dim) - - self.norm1 = nn.LayerNorm(embed_dim) - self.norm2 = nn.LayerNorm(embed_dim) - self.dropout1 = nn.Dropout(dropout) - self.dropout2 = nn.Dropout(dropout) - - self.activation = _get_activation_fn(activation) - - def with_pos_embed(self, tensor, pos: Optional[Tensor]): - return tensor if pos is None else tensor + pos - - def forward(self, tgt, - tgt_mask: Optional[Tensor] = None, - tgt_key_padding_mask: Optional[Tensor] = None, - query_pos: Optional[Tensor] = None): - - # self attention - tgt2 = self.norm1(tgt) - q = k = self.with_pos_embed(tgt2, query_pos) - tgt2 = self.self_attn(q, k, value=tgt2, attn_mask=tgt_mask, - key_padding_mask=tgt_key_padding_mask)[0] - tgt = tgt + self.dropout1(tgt2) - - # ffn - tgt2 = self.norm2(tgt) - tgt2 = self.linear2(self.dropout(self.activation(self.linear1(tgt2)))) - tgt = tgt + self.dropout2(tgt2) - return tgt - -class Fuse_sft_block(nn.Module): - def __init__(self, in_ch, out_ch): - super().__init__() - self.encode_enc = ResBlock(2*in_ch, out_ch) - - self.scale = nn.Sequential( - nn.Conv2d(in_ch, out_ch, kernel_size=3, padding=1), - nn.LeakyReLU(0.2, True), - nn.Conv2d(out_ch, out_ch, kernel_size=3, padding=1)) - - self.shift = nn.Sequential( - nn.Conv2d(in_ch, out_ch, kernel_size=3, padding=1), - nn.LeakyReLU(0.2, True), - nn.Conv2d(out_ch, out_ch, kernel_size=3, padding=1)) - - def forward(self, enc_feat, dec_feat, w=1): - enc_feat = self.encode_enc(torch.cat([enc_feat, dec_feat], dim=1)) - scale = self.scale(enc_feat) - shift = self.shift(enc_feat) - residual = w * (dec_feat * scale + shift) - out = dec_feat + residual - return out - - -@ARCH_REGISTRY.register() -class CodeFormer(VQAutoEncoder): - def __init__(self, dim_embd=512, n_head=8, n_layers=9, - codebook_size=1024, latent_size=256, - connect_list=('32', '64', '128', '256'), - fix_modules=('quantize', 'generator')): - super(CodeFormer, self).__init__(512, 64, [1, 2, 2, 4, 4, 8], 'nearest',2, [16], codebook_size) - - if fix_modules is not None: - for module in fix_modules: - for param in getattr(self, module).parameters(): - param.requires_grad = False - - self.connect_list = connect_list - self.n_layers = n_layers - self.dim_embd = dim_embd - self.dim_mlp = dim_embd*2 - - self.position_emb = nn.Parameter(torch.zeros(latent_size, self.dim_embd)) - self.feat_emb = nn.Linear(256, self.dim_embd) - - # transformer - self.ft_layers = nn.Sequential(*[TransformerSALayer(embed_dim=dim_embd, nhead=n_head, dim_mlp=self.dim_mlp, dropout=0.0) - for _ in range(self.n_layers)]) - - # logits_predict head - self.idx_pred_layer = nn.Sequential( - nn.LayerNorm(dim_embd), - nn.Linear(dim_embd, codebook_size, bias=False)) - - self.channels = { - '16': 512, - '32': 256, - '64': 256, - '128': 128, - '256': 128, - '512': 64, - } - - # after second residual block for > 16, before attn layer for ==16 - self.fuse_encoder_block = {'512':2, '256':5, '128':8, '64':11, '32':14, '16':18} - # after first residual block for > 16, before attn layer for ==16 - self.fuse_generator_block = {'16':6, '32': 9, '64':12, '128':15, '256':18, '512':21} - - # fuse_convs_dict - self.fuse_convs_dict = nn.ModuleDict() - for f_size in self.connect_list: - in_ch = self.channels[f_size] - self.fuse_convs_dict[f_size] = Fuse_sft_block(in_ch, in_ch) - - def _init_weights(self, module): - if isinstance(module, (nn.Linear, nn.Embedding)): - module.weight.data.normal_(mean=0.0, std=0.02) - if isinstance(module, nn.Linear) and module.bias is not None: - module.bias.data.zero_() - elif isinstance(module, nn.LayerNorm): - module.bias.data.zero_() - module.weight.data.fill_(1.0) - - def forward(self, x, w=0, detach_16=True, code_only=False, adain=False): - # ################### Encoder ##################### - enc_feat_dict = {} - out_list = [self.fuse_encoder_block[f_size] for f_size in self.connect_list] - for i, block in enumerate(self.encoder.blocks): - x = block(x) - if i in out_list: - enc_feat_dict[str(x.shape[-1])] = x.clone() - - lq_feat = x - # ################# Transformer ################### - # quant_feat, codebook_loss, quant_stats = self.quantize(lq_feat) - pos_emb = self.position_emb.unsqueeze(1).repeat(1,x.shape[0],1) - # BCHW -> BC(HW) -> (HW)BC - feat_emb = self.feat_emb(lq_feat.flatten(2).permute(2,0,1)) - query_emb = feat_emb - # Transformer encoder - for layer in self.ft_layers: - query_emb = layer(query_emb, query_pos=pos_emb) - - # output logits - logits = self.idx_pred_layer(query_emb) # (hw)bn - logits = logits.permute(1,0,2) # (hw)bn -> b(hw)n - - if code_only: # for training stage II - # logits doesn't need softmax before cross_entropy loss - return logits, lq_feat - - # ################# Quantization ################### - # if self.training: - # quant_feat = torch.einsum('btn,nc->btc', [soft_one_hot, self.quantize.embedding.weight]) - # # b(hw)c -> bc(hw) -> bchw - # quant_feat = quant_feat.permute(0,2,1).view(lq_feat.shape) - # ------------ - soft_one_hot = F.softmax(logits, dim=2) - _, top_idx = torch.topk(soft_one_hot, 1, dim=2) - quant_feat = self.quantize.get_codebook_feat(top_idx, shape=[x.shape[0],16,16,256]) - # preserve gradients - # quant_feat = lq_feat + (quant_feat - lq_feat).detach() - - if detach_16: - quant_feat = quant_feat.detach() # for training stage III - if adain: - quant_feat = adaptive_instance_normalization(quant_feat, lq_feat) - - # ################## Generator #################### - x = quant_feat - fuse_list = [self.fuse_generator_block[f_size] for f_size in self.connect_list] - - for i, block in enumerate(self.generator.blocks): - x = block(x) - if i in fuse_list: # fuse after i-th block - f_size = str(x.shape[-1]) - if w>0: - x = self.fuse_convs_dict[f_size](enc_feat_dict[f_size].detach(), x, w) - out = x - # logits doesn't need softmax before cross_entropy loss - return out, logits, lq_feat diff --git a/modules/codeformer/vqgan_arch.py b/modules/codeformer/vqgan_arch.py deleted file mode 100644 index 09ee6660..00000000 --- a/modules/codeformer/vqgan_arch.py +++ /dev/null @@ -1,435 +0,0 @@ -# this file is copied from CodeFormer repository. Please see comment in modules/codeformer_model.py - -''' -VQGAN code, adapted from the original created by the Unleashing Transformers authors: -https://github.com/samb-t/unleashing-transformers/blob/master/models/vqgan.py - -''' -import torch -import torch.nn as nn -import torch.nn.functional as F -from basicsr.utils import get_root_logger -from basicsr.utils.registry import ARCH_REGISTRY - -def normalize(in_channels): - return torch.nn.GroupNorm(num_groups=32, num_channels=in_channels, eps=1e-6, affine=True) - - -@torch.jit.script -def swish(x): - return x*torch.sigmoid(x) - - -# Define VQVAE classes -class VectorQuantizer(nn.Module): - def __init__(self, codebook_size, emb_dim, beta): - super(VectorQuantizer, self).__init__() - self.codebook_size = codebook_size # number of embeddings - self.emb_dim = emb_dim # dimension of embedding - self.beta = beta # commitment cost used in loss term, beta * ||z_e(x)-sg[e]||^2 - self.embedding = nn.Embedding(self.codebook_size, self.emb_dim) - self.embedding.weight.data.uniform_(-1.0 / self.codebook_size, 1.0 / self.codebook_size) - - def forward(self, z): - # reshape z -> (batch, height, width, channel) and flatten - z = z.permute(0, 2, 3, 1).contiguous() - z_flattened = z.view(-1, self.emb_dim) - - # distances from z to embeddings e_j (z - e)^2 = z^2 + e^2 - 2 e * z - d = (z_flattened ** 2).sum(dim=1, keepdim=True) + (self.embedding.weight**2).sum(1) - \ - 2 * torch.matmul(z_flattened, self.embedding.weight.t()) - - mean_distance = torch.mean(d) - # find closest encodings - # min_encoding_indices = torch.argmin(d, dim=1).unsqueeze(1) - min_encoding_scores, min_encoding_indices = torch.topk(d, 1, dim=1, largest=False) - # [0-1], higher score, higher confidence - min_encoding_scores = torch.exp(-min_encoding_scores/10) - - min_encodings = torch.zeros(min_encoding_indices.shape[0], self.codebook_size).to(z) - min_encodings.scatter_(1, min_encoding_indices, 1) - - # get quantized latent vectors - z_q = torch.matmul(min_encodings, self.embedding.weight).view(z.shape) - # compute loss for embedding - loss = torch.mean((z_q.detach()-z)**2) + self.beta * torch.mean((z_q - z.detach()) ** 2) - # preserve gradients - z_q = z + (z_q - z).detach() - - # perplexity - e_mean = torch.mean(min_encodings, dim=0) - perplexity = torch.exp(-torch.sum(e_mean * torch.log(e_mean + 1e-10))) - # reshape back to match original input shape - z_q = z_q.permute(0, 3, 1, 2).contiguous() - - return z_q, loss, { - "perplexity": perplexity, - "min_encodings": min_encodings, - "min_encoding_indices": min_encoding_indices, - "min_encoding_scores": min_encoding_scores, - "mean_distance": mean_distance - } - - def get_codebook_feat(self, indices, shape): - # input indices: batch*token_num -> (batch*token_num)*1 - # shape: batch, height, width, channel - indices = indices.view(-1,1) - min_encodings = torch.zeros(indices.shape[0], self.codebook_size).to(indices) - min_encodings.scatter_(1, indices, 1) - # get quantized latent vectors - z_q = torch.matmul(min_encodings.float(), self.embedding.weight) - - if shape is not None: # reshape back to match original input shape - z_q = z_q.view(shape).permute(0, 3, 1, 2).contiguous() - - return z_q - - -class GumbelQuantizer(nn.Module): - def __init__(self, codebook_size, emb_dim, num_hiddens, straight_through=False, kl_weight=5e-4, temp_init=1.0): - super().__init__() - self.codebook_size = codebook_size # number of embeddings - self.emb_dim = emb_dim # dimension of embedding - self.straight_through = straight_through - self.temperature = temp_init - self.kl_weight = kl_weight - self.proj = nn.Conv2d(num_hiddens, codebook_size, 1) # projects last encoder layer to quantized logits - self.embed = nn.Embedding(codebook_size, emb_dim) - - def forward(self, z): - hard = self.straight_through if self.training else True - - logits = self.proj(z) - - soft_one_hot = F.gumbel_softmax(logits, tau=self.temperature, dim=1, hard=hard) - - z_q = torch.einsum("b n h w, n d -> b d h w", soft_one_hot, self.embed.weight) - - # + kl divergence to the prior loss - qy = F.softmax(logits, dim=1) - diff = self.kl_weight * torch.sum(qy * torch.log(qy * self.codebook_size + 1e-10), dim=1).mean() - min_encoding_indices = soft_one_hot.argmax(dim=1) - - return z_q, diff, { - "min_encoding_indices": min_encoding_indices - } - - -class Downsample(nn.Module): - def __init__(self, in_channels): - super().__init__() - self.conv = torch.nn.Conv2d(in_channels, in_channels, kernel_size=3, stride=2, padding=0) - - def forward(self, x): - pad = (0, 1, 0, 1) - x = torch.nn.functional.pad(x, pad, mode="constant", value=0) - x = self.conv(x) - return x - - -class Upsample(nn.Module): - def __init__(self, in_channels): - super().__init__() - self.conv = nn.Conv2d(in_channels, in_channels, kernel_size=3, stride=1, padding=1) - - def forward(self, x): - x = F.interpolate(x, scale_factor=2.0, mode="nearest") - x = self.conv(x) - - return x - - -class ResBlock(nn.Module): - def __init__(self, in_channels, out_channels=None): - super(ResBlock, self).__init__() - self.in_channels = in_channels - self.out_channels = in_channels if out_channels is None else out_channels - self.norm1 = normalize(in_channels) - self.conv1 = nn.Conv2d(in_channels, out_channels, kernel_size=3, stride=1, padding=1) - self.norm2 = normalize(out_channels) - self.conv2 = nn.Conv2d(out_channels, out_channels, kernel_size=3, stride=1, padding=1) - if self.in_channels != self.out_channels: - self.conv_out = nn.Conv2d(in_channels, out_channels, kernel_size=1, stride=1, padding=0) - - def forward(self, x_in): - x = x_in - x = self.norm1(x) - x = swish(x) - x = self.conv1(x) - x = self.norm2(x) - x = swish(x) - x = self.conv2(x) - if self.in_channels != self.out_channels: - x_in = self.conv_out(x_in) - - return x + x_in - - -class AttnBlock(nn.Module): - def __init__(self, in_channels): - super().__init__() - self.in_channels = in_channels - - self.norm = normalize(in_channels) - self.q = torch.nn.Conv2d( - in_channels, - in_channels, - kernel_size=1, - stride=1, - padding=0 - ) - self.k = torch.nn.Conv2d( - in_channels, - in_channels, - kernel_size=1, - stride=1, - padding=0 - ) - self.v = torch.nn.Conv2d( - in_channels, - in_channels, - kernel_size=1, - stride=1, - padding=0 - ) - self.proj_out = torch.nn.Conv2d( - in_channels, - in_channels, - kernel_size=1, - stride=1, - padding=0 - ) - - def forward(self, x): - h_ = x - h_ = self.norm(h_) - q = self.q(h_) - k = self.k(h_) - v = self.v(h_) - - # compute attention - b, c, h, w = q.shape - q = q.reshape(b, c, h*w) - q = q.permute(0, 2, 1) - k = k.reshape(b, c, h*w) - w_ = torch.bmm(q, k) - w_ = w_ * (int(c)**(-0.5)) - w_ = F.softmax(w_, dim=2) - - # attend to values - v = v.reshape(b, c, h*w) - w_ = w_.permute(0, 2, 1) - h_ = torch.bmm(v, w_) - h_ = h_.reshape(b, c, h, w) - - h_ = self.proj_out(h_) - - return x+h_ - - -class Encoder(nn.Module): - def __init__(self, in_channels, nf, emb_dim, ch_mult, num_res_blocks, resolution, attn_resolutions): - super().__init__() - self.nf = nf - self.num_resolutions = len(ch_mult) - self.num_res_blocks = num_res_blocks - self.resolution = resolution - self.attn_resolutions = attn_resolutions - - curr_res = self.resolution - in_ch_mult = (1,)+tuple(ch_mult) - - blocks = [] - # initial convultion - blocks.append(nn.Conv2d(in_channels, nf, kernel_size=3, stride=1, padding=1)) - - # residual and downsampling blocks, with attention on smaller res (16x16) - for i in range(self.num_resolutions): - block_in_ch = nf * in_ch_mult[i] - block_out_ch = nf * ch_mult[i] - for _ in range(self.num_res_blocks): - blocks.append(ResBlock(block_in_ch, block_out_ch)) - block_in_ch = block_out_ch - if curr_res in attn_resolutions: - blocks.append(AttnBlock(block_in_ch)) - - if i != self.num_resolutions - 1: - blocks.append(Downsample(block_in_ch)) - curr_res = curr_res // 2 - - # non-local attention block - blocks.append(ResBlock(block_in_ch, block_in_ch)) - blocks.append(AttnBlock(block_in_ch)) - blocks.append(ResBlock(block_in_ch, block_in_ch)) - - # normalise and convert to latent size - blocks.append(normalize(block_in_ch)) - blocks.append(nn.Conv2d(block_in_ch, emb_dim, kernel_size=3, stride=1, padding=1)) - self.blocks = nn.ModuleList(blocks) - - def forward(self, x): - for block in self.blocks: - x = block(x) - - return x - - -class Generator(nn.Module): - def __init__(self, nf, emb_dim, ch_mult, res_blocks, img_size, attn_resolutions): - super().__init__() - self.nf = nf - self.ch_mult = ch_mult - self.num_resolutions = len(self.ch_mult) - self.num_res_blocks = res_blocks - self.resolution = img_size - self.attn_resolutions = attn_resolutions - self.in_channels = emb_dim - self.out_channels = 3 - block_in_ch = self.nf * self.ch_mult[-1] - curr_res = self.resolution // 2 ** (self.num_resolutions-1) - - blocks = [] - # initial conv - blocks.append(nn.Conv2d(self.in_channels, block_in_ch, kernel_size=3, stride=1, padding=1)) - - # non-local attention block - blocks.append(ResBlock(block_in_ch, block_in_ch)) - blocks.append(AttnBlock(block_in_ch)) - blocks.append(ResBlock(block_in_ch, block_in_ch)) - - for i in reversed(range(self.num_resolutions)): - block_out_ch = self.nf * self.ch_mult[i] - - for _ in range(self.num_res_blocks): - blocks.append(ResBlock(block_in_ch, block_out_ch)) - block_in_ch = block_out_ch - - if curr_res in self.attn_resolutions: - blocks.append(AttnBlock(block_in_ch)) - - if i != 0: - blocks.append(Upsample(block_in_ch)) - curr_res = curr_res * 2 - - blocks.append(normalize(block_in_ch)) - blocks.append(nn.Conv2d(block_in_ch, self.out_channels, kernel_size=3, stride=1, padding=1)) - - self.blocks = nn.ModuleList(blocks) - - - def forward(self, x): - for block in self.blocks: - x = block(x) - - return x - - -@ARCH_REGISTRY.register() -class VQAutoEncoder(nn.Module): - def __init__(self, img_size, nf, ch_mult, quantizer="nearest", res_blocks=2, attn_resolutions=None, codebook_size=1024, emb_dim=256, - beta=0.25, gumbel_straight_through=False, gumbel_kl_weight=1e-8, model_path=None): - super().__init__() - logger = get_root_logger() - self.in_channels = 3 - self.nf = nf - self.n_blocks = res_blocks - self.codebook_size = codebook_size - self.embed_dim = emb_dim - self.ch_mult = ch_mult - self.resolution = img_size - self.attn_resolutions = attn_resolutions or [16] - self.quantizer_type = quantizer - self.encoder = Encoder( - self.in_channels, - self.nf, - self.embed_dim, - self.ch_mult, - self.n_blocks, - self.resolution, - self.attn_resolutions - ) - if self.quantizer_type == "nearest": - self.beta = beta #0.25 - self.quantize = VectorQuantizer(self.codebook_size, self.embed_dim, self.beta) - elif self.quantizer_type == "gumbel": - self.gumbel_num_hiddens = emb_dim - self.straight_through = gumbel_straight_through - self.kl_weight = gumbel_kl_weight - self.quantize = GumbelQuantizer( - self.codebook_size, - self.embed_dim, - self.gumbel_num_hiddens, - self.straight_through, - self.kl_weight - ) - self.generator = Generator( - self.nf, - self.embed_dim, - self.ch_mult, - self.n_blocks, - self.resolution, - self.attn_resolutions - ) - - if model_path is not None: - chkpt = torch.load(model_path, map_location='cpu') - if 'params_ema' in chkpt: - self.load_state_dict(torch.load(model_path, map_location='cpu')['params_ema']) - logger.info(f'vqgan is loaded from: {model_path} [params_ema]') - elif 'params' in chkpt: - self.load_state_dict(torch.load(model_path, map_location='cpu')['params']) - logger.info(f'vqgan is loaded from: {model_path} [params]') - else: - raise ValueError('Wrong params!') - - - def forward(self, x): - x = self.encoder(x) - quant, codebook_loss, quant_stats = self.quantize(x) - x = self.generator(quant) - return x, codebook_loss, quant_stats - - - -# patch based discriminator -@ARCH_REGISTRY.register() -class VQGANDiscriminator(nn.Module): - def __init__(self, nc=3, ndf=64, n_layers=4, model_path=None): - super().__init__() - - layers = [nn.Conv2d(nc, ndf, kernel_size=4, stride=2, padding=1), nn.LeakyReLU(0.2, True)] - ndf_mult = 1 - ndf_mult_prev = 1 - for n in range(1, n_layers): # gradually increase the number of filters - ndf_mult_prev = ndf_mult - ndf_mult = min(2 ** n, 8) - layers += [ - nn.Conv2d(ndf * ndf_mult_prev, ndf * ndf_mult, kernel_size=4, stride=2, padding=1, bias=False), - nn.BatchNorm2d(ndf * ndf_mult), - nn.LeakyReLU(0.2, True) - ] - - ndf_mult_prev = ndf_mult - ndf_mult = min(2 ** n_layers, 8) - - layers += [ - nn.Conv2d(ndf * ndf_mult_prev, ndf * ndf_mult, kernel_size=4, stride=1, padding=1, bias=False), - nn.BatchNorm2d(ndf * ndf_mult), - nn.LeakyReLU(0.2, True) - ] - - layers += [ - nn.Conv2d(ndf * ndf_mult, 1, kernel_size=4, stride=1, padding=1)] # output 1 channel prediction map - self.main = nn.Sequential(*layers) - - if model_path is not None: - chkpt = torch.load(model_path, map_location='cpu') - if 'params_d' in chkpt: - self.load_state_dict(torch.load(model_path, map_location='cpu')['params_d']) - elif 'params' in chkpt: - self.load_state_dict(torch.load(model_path, map_location='cpu')['params']) - else: - raise ValueError('Wrong params!') - - def forward(self, x): - return self.main(x) diff --git a/modules/codeformer_model.py b/modules/codeformer_model.py index da42b5e9..517eadfd 100644 --- a/modules/codeformer_model.py +++ b/modules/codeformer_model.py @@ -8,9 +8,6 @@ import modules.shared from modules import shared, devices, modelloader, errors from modules.paths import models_path -# codeformer people made a choice to include modified basicsr library to their project which makes -# it utterly impossible to use it alongside with other libraries that also use basicsr, like GFPGAN. -# I am making a choice to include some files from codeformer to work around this issue. model_dir = "Codeformer" model_path = os.path.join(models_path, model_dir) model_url = 'https://github.com/sczhou/CodeFormer/releases/download/v0.1.0/codeformer.pth' @@ -18,115 +15,127 @@ model_url = 'https://github.com/sczhou/CodeFormer/releases/download/v0.1.0/codef codeformer = None -def setup_model(dirname): - os.makedirs(model_path, exist_ok=True) - - path = modules.paths.paths.get("CodeFormer", None) - if path is None: - return - - try: +class FaceRestorerCodeFormer(modules.face_restoration.FaceRestoration): + def name(self): + return "CodeFormer" + + def __init__(self, dirname): + self.net = None + self.face_helper = None + self.cmd_dir = dirname + + def create_models(self): + from facexlib.detection import retinaface + from facexlib.utils.face_restoration_helper import FaceRestoreHelper + + if self.net is not None and self.face_helper is not None: + self.net.to(devices.device_codeformer) + return self.net, self.face_helper + model_paths = modelloader.load_models( + model_path, + model_url, + self.cmd_dir, + download_name='codeformer-v0.1.0.pth', + ext_filter=['.pth'], + ) + + if len(model_paths) != 0: + ckpt_path = model_paths[0] + else: + print("Unable to load codeformer model.") + return None, None + net = modelloader.load_spandrel_model(ckpt_path, device=devices.device_codeformer) + + if hasattr(retinaface, 'device'): + retinaface.device = devices.device_codeformer + + face_helper = FaceRestoreHelper( + upscale_factor=1, + face_size=512, + crop_ratio=(1, 1), + det_model='retinaface_resnet50', + save_ext='png', + use_parse=True, + device=devices.device_codeformer, + ) + + self.net = net + self.face_helper = face_helper + + def send_model_to(self, device): + self.net.to(device) + self.face_helper.face_det.to(device) + self.face_helper.face_parse.to(device) + + def restore(self, np_image, w=None): from torchvision.transforms.functional import normalize - from modules.codeformer.codeformer_arch import CodeFormer from basicsr.utils import img2tensor, tensor2img - from facelib.utils.face_restoration_helper import FaceRestoreHelper - from facelib.detection.retinaface import retinaface - - net_class = CodeFormer - - class FaceRestorerCodeFormer(modules.face_restoration.FaceRestoration): - def name(self): - return "CodeFormer" - - def __init__(self, dirname): - self.net = None - self.face_helper = None - self.cmd_dir = dirname - - def create_models(self): - - if self.net is not None and self.face_helper is not None: - self.net.to(devices.device_codeformer) - return self.net, self.face_helper - model_paths = modelloader.load_models(model_path, model_url, self.cmd_dir, download_name='codeformer-v0.1.0.pth', ext_filter=['.pth']) - if len(model_paths) != 0: - ckpt_path = model_paths[0] - else: - print("Unable to load codeformer model.") - return None, None - net = net_class(dim_embd=512, codebook_size=1024, n_head=8, n_layers=9, connect_list=['32', '64', '128', '256']).to(devices.device_codeformer) - checkpoint = torch.load(ckpt_path)['params_ema'] - net.load_state_dict(checkpoint) - net.eval() + np_image = np_image[:, :, ::-1] - if hasattr(retinaface, 'device'): - retinaface.device = devices.device_codeformer - face_helper = FaceRestoreHelper(1, face_size=512, crop_ratio=(1, 1), det_model='retinaface_resnet50', save_ext='png', use_parse=True, device=devices.device_codeformer) + original_resolution = np_image.shape[0:2] - self.net = net - self.face_helper = face_helper + self.create_models() + if self.net is None or self.face_helper is None: + return np_image - return net, face_helper + self.send_model_to(devices.device_codeformer) - def send_model_to(self, device): - self.net.to(device) - self.face_helper.face_det.to(device) - self.face_helper.face_parse.to(device) + self.face_helper.clean_all() + self.face_helper.read_image(np_image) + self.face_helper.get_face_landmarks_5(only_center_face=False, resize=640, eye_dist_threshold=5) + self.face_helper.align_warp_face() - def restore(self, np_image, w=None): - np_image = np_image[:, :, ::-1] + for cropped_face in self.face_helper.cropped_faces: + cropped_face_t = img2tensor(cropped_face / 255., bgr2rgb=True, float32=True) + normalize(cropped_face_t, (0.5, 0.5, 0.5), (0.5, 0.5, 0.5), inplace=True) + cropped_face_t = cropped_face_t.unsqueeze(0).to(devices.device_codeformer) - original_resolution = np_image.shape[0:2] + try: + with torch.no_grad(): + res = self.net(cropped_face_t, w=w if w is not None else shared.opts.code_former_weight, adain=True) + if isinstance(res, tuple): + output = res[0] + else: + output = res + if not isinstance(res, torch.Tensor): + raise TypeError(f"Expected torch.Tensor, got {type(res)}") + restored_face = tensor2img(output, rgb2bgr=True, min_max=(-1, 1)) + del output + devices.torch_gc() + except Exception: + errors.report('Failed inference for CodeFormer', exc_info=True) + restored_face = tensor2img(cropped_face_t, rgb2bgr=True, min_max=(-1, 1)) - self.create_models() - if self.net is None or self.face_helper is None: - return np_image + restored_face = restored_face.astype('uint8') + self.face_helper.add_restored_face(restored_face) - self.send_model_to(devices.device_codeformer) + self.face_helper.get_inverse_affine(None) - self.face_helper.clean_all() - self.face_helper.read_image(np_image) - self.face_helper.get_face_landmarks_5(only_center_face=False, resize=640, eye_dist_threshold=5) - self.face_helper.align_warp_face() + restored_img = self.face_helper.paste_faces_to_input_image() + restored_img = restored_img[:, :, ::-1] - for cropped_face in self.face_helper.cropped_faces: - cropped_face_t = img2tensor(cropped_face / 255., bgr2rgb=True, float32=True) - normalize(cropped_face_t, (0.5, 0.5, 0.5), (0.5, 0.5, 0.5), inplace=True) - cropped_face_t = cropped_face_t.unsqueeze(0).to(devices.device_codeformer) + if original_resolution != restored_img.shape[0:2]: + restored_img = cv2.resize( + restored_img, + (0, 0), + fx=original_resolution[1]/restored_img.shape[1], + fy=original_resolution[0]/restored_img.shape[0], + interpolation=cv2.INTER_LINEAR, + ) - try: - with torch.no_grad(): - output = self.net(cropped_face_t, w=w if w is not None else shared.opts.code_former_weight, adain=True)[0] - restored_face = tensor2img(output, rgb2bgr=True, min_max=(-1, 1)) - del output - devices.torch_gc() - except Exception: - errors.report('Failed inference for CodeFormer', exc_info=True) - restored_face = tensor2img(cropped_face_t, rgb2bgr=True, min_max=(-1, 1)) + self.face_helper.clean_all() - restored_face = restored_face.astype('uint8') - self.face_helper.add_restored_face(restored_face) + if shared.opts.face_restoration_unload: + self.send_model_to(devices.cpu) - self.face_helper.get_inverse_affine(None) + return restored_img - restored_img = self.face_helper.paste_faces_to_input_image() - restored_img = restored_img[:, :, ::-1] - - if original_resolution != restored_img.shape[0:2]: - restored_img = cv2.resize(restored_img, (0, 0), fx=original_resolution[1]/restored_img.shape[1], fy=original_resolution[0]/restored_img.shape[0], interpolation=cv2.INTER_LINEAR) - - self.face_helper.clean_all() - - if shared.opts.face_restoration_unload: - self.send_model_to(devices.cpu) - - return restored_img +def setup_model(dirname): + os.makedirs(model_path, exist_ok=True) + try: global codeformer codeformer = FaceRestorerCodeFormer(dirname) shared.face_restorers.append(codeformer) - except Exception: errors.report("Error setting up CodeFormer", exc_info=True) - - # sys.path = stored_sys_path diff --git a/modules/esrgan_model.py b/modules/esrgan_model.py index c0d22a99..a7c7c9e3 100644 --- a/modules/esrgan_model.py +++ b/modules/esrgan_model.py @@ -1,122 +1,9 @@ -import sys - -import torch - -import modules.esrgan_model_arch as arch -from modules import modelloader, devices +from modules import modelloader, devices, errors from modules.shared import opts from modules.upscaler import Upscaler, UpscalerData from modules.upscaler_utils import upscale_with_model -def mod2normal(state_dict): - # this code is copied from https://github.com/victorca25/iNNfer - if 'conv_first.weight' in state_dict: - crt_net = {} - items = list(state_dict) - - crt_net['model.0.weight'] = state_dict['conv_first.weight'] - crt_net['model.0.bias'] = state_dict['conv_first.bias'] - - for k in items.copy(): - if 'RDB' in k: - ori_k = k.replace('RRDB_trunk.', 'model.1.sub.') - if '.weight' in k: - ori_k = ori_k.replace('.weight', '.0.weight') - elif '.bias' in k: - ori_k = ori_k.replace('.bias', '.0.bias') - crt_net[ori_k] = state_dict[k] - items.remove(k) - - crt_net['model.1.sub.23.weight'] = state_dict['trunk_conv.weight'] - crt_net['model.1.sub.23.bias'] = state_dict['trunk_conv.bias'] - crt_net['model.3.weight'] = state_dict['upconv1.weight'] - crt_net['model.3.bias'] = state_dict['upconv1.bias'] - crt_net['model.6.weight'] = state_dict['upconv2.weight'] - crt_net['model.6.bias'] = state_dict['upconv2.bias'] - crt_net['model.8.weight'] = state_dict['HRconv.weight'] - crt_net['model.8.bias'] = state_dict['HRconv.bias'] - crt_net['model.10.weight'] = state_dict['conv_last.weight'] - crt_net['model.10.bias'] = state_dict['conv_last.bias'] - state_dict = crt_net - return state_dict - - -def resrgan2normal(state_dict, nb=23): - # this code is copied from https://github.com/victorca25/iNNfer - if "conv_first.weight" in state_dict and "body.0.rdb1.conv1.weight" in state_dict: - re8x = 0 - crt_net = {} - items = list(state_dict) - - crt_net['model.0.weight'] = state_dict['conv_first.weight'] - crt_net['model.0.bias'] = state_dict['conv_first.bias'] - - for k in items.copy(): - if "rdb" in k: - ori_k = k.replace('body.', 'model.1.sub.') - ori_k = ori_k.replace('.rdb', '.RDB') - if '.weight' in k: - ori_k = ori_k.replace('.weight', '.0.weight') - elif '.bias' in k: - ori_k = ori_k.replace('.bias', '.0.bias') - crt_net[ori_k] = state_dict[k] - items.remove(k) - - crt_net[f'model.1.sub.{nb}.weight'] = state_dict['conv_body.weight'] - crt_net[f'model.1.sub.{nb}.bias'] = state_dict['conv_body.bias'] - crt_net['model.3.weight'] = state_dict['conv_up1.weight'] - crt_net['model.3.bias'] = state_dict['conv_up1.bias'] - crt_net['model.6.weight'] = state_dict['conv_up2.weight'] - crt_net['model.6.bias'] = state_dict['conv_up2.bias'] - - if 'conv_up3.weight' in state_dict: - # modification supporting: https://github.com/ai-forever/Real-ESRGAN/blob/main/RealESRGAN/rrdbnet_arch.py - re8x = 3 - crt_net['model.9.weight'] = state_dict['conv_up3.weight'] - crt_net['model.9.bias'] = state_dict['conv_up3.bias'] - - crt_net[f'model.{8+re8x}.weight'] = state_dict['conv_hr.weight'] - crt_net[f'model.{8+re8x}.bias'] = state_dict['conv_hr.bias'] - crt_net[f'model.{10+re8x}.weight'] = state_dict['conv_last.weight'] - crt_net[f'model.{10+re8x}.bias'] = state_dict['conv_last.bias'] - - state_dict = crt_net - return state_dict - - -def infer_params(state_dict): - # this code is copied from https://github.com/victorca25/iNNfer - scale2x = 0 - scalemin = 6 - n_uplayer = 0 - plus = False - - for block in list(state_dict): - parts = block.split(".") - n_parts = len(parts) - if n_parts == 5 and parts[2] == "sub": - nb = int(parts[3]) - elif n_parts == 3: - part_num = int(parts[1]) - if (part_num > scalemin - and parts[0] == "model" - and parts[2] == "weight"): - scale2x += 1 - if part_num > n_uplayer: - n_uplayer = part_num - out_nc = state_dict[block].shape[0] - if not plus and "conv1x1" in block: - plus = True - - nf = state_dict["model.0.weight"].shape[0] - in_nc = state_dict["model.0.weight"].shape[1] - out_nc = out_nc - scale = 2 ** scale2x - - return in_nc, out_nc, nf, nb, plus, scale - - class UpscalerESRGAN(Upscaler): def __init__(self, dirname): self.name = "ESRGAN" @@ -142,12 +29,11 @@ class UpscalerESRGAN(Upscaler): def do_upscale(self, img, selected_model): try: model = self.load_model(selected_model) - except Exception as e: - print(f"Unable to load ESRGAN model {selected_model}: {e}", file=sys.stderr) + except Exception: + errors.report(f"Unable to load ESRGAN model {selected_model}", exc_info=True) return img model.to(devices.device_esrgan) - img = esrgan_upscale(model, img) - return img + return esrgan_upscale(model, img) def load_model(self, path: str): if path.startswith("http"): @@ -160,33 +46,10 @@ class UpscalerESRGAN(Upscaler): else: filename = path - state_dict = torch.load(filename, map_location='cpu' if devices.device_esrgan.type == 'mps' else None) - - if "params_ema" in state_dict: - state_dict = state_dict["params_ema"] - elif "params" in state_dict: - state_dict = state_dict["params"] - num_conv = 16 if "realesr-animevideov3" in filename else 32 - model = arch.SRVGGNetCompact(num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=num_conv, upscale=4, act_type='prelu') - model.load_state_dict(state_dict) - model.eval() - return model - - if "body.0.rdb1.conv1.weight" in state_dict and "conv_first.weight" in state_dict: - nb = 6 if "RealESRGAN_x4plus_anime_6B" in filename else 23 - state_dict = resrgan2normal(state_dict, nb) - elif "conv_first.weight" in state_dict: - state_dict = mod2normal(state_dict) - elif "model.0.weight" not in state_dict: - raise Exception("The file is not a recognized ESRGAN model.") - - in_nc, out_nc, nf, nb, plus, mscale = infer_params(state_dict) - - model = arch.RRDBNet(in_nc=in_nc, out_nc=out_nc, nf=nf, nb=nb, upscale=mscale, plus=plus) - model.load_state_dict(state_dict) - model.eval() - - return model + return modelloader.load_spandrel_model( + filename, + device=('cpu' if devices.device_esrgan.type == 'mps' else None), + ) def esrgan_upscale(model, img): diff --git a/modules/esrgan_model_arch.py b/modules/esrgan_model_arch.py deleted file mode 100644 index 2b9888ba..00000000 --- a/modules/esrgan_model_arch.py +++ /dev/null @@ -1,465 +0,0 @@ -# this file is adapted from https://github.com/victorca25/iNNfer - -from collections import OrderedDict -import math -import torch -import torch.nn as nn -import torch.nn.functional as F - - -#################### -# RRDBNet Generator -#################### - -class RRDBNet(nn.Module): - def __init__(self, in_nc, out_nc, nf, nb, nr=3, gc=32, upscale=4, norm_type=None, - act_type='leakyrelu', mode='CNA', upsample_mode='upconv', convtype='Conv2D', - finalact=None, gaussian_noise=False, plus=False): - super(RRDBNet, self).__init__() - n_upscale = int(math.log(upscale, 2)) - if upscale == 3: - n_upscale = 1 - - self.resrgan_scale = 0 - if in_nc % 16 == 0: - self.resrgan_scale = 1 - elif in_nc != 4 and in_nc % 4 == 0: - self.resrgan_scale = 2 - - fea_conv = conv_block(in_nc, nf, kernel_size=3, norm_type=None, act_type=None, convtype=convtype) - rb_blocks = [RRDB(nf, nr, kernel_size=3, gc=32, stride=1, bias=1, pad_type='zero', - norm_type=norm_type, act_type=act_type, mode='CNA', convtype=convtype, - gaussian_noise=gaussian_noise, plus=plus) for _ in range(nb)] - LR_conv = conv_block(nf, nf, kernel_size=3, norm_type=norm_type, act_type=None, mode=mode, convtype=convtype) - - if upsample_mode == 'upconv': - upsample_block = upconv_block - elif upsample_mode == 'pixelshuffle': - upsample_block = pixelshuffle_block - else: - raise NotImplementedError(f'upsample mode [{upsample_mode}] is not found') - if upscale == 3: - upsampler = upsample_block(nf, nf, 3, act_type=act_type, convtype=convtype) - else: - upsampler = [upsample_block(nf, nf, act_type=act_type, convtype=convtype) for _ in range(n_upscale)] - HR_conv0 = conv_block(nf, nf, kernel_size=3, norm_type=None, act_type=act_type, convtype=convtype) - HR_conv1 = conv_block(nf, out_nc, kernel_size=3, norm_type=None, act_type=None, convtype=convtype) - - outact = act(finalact) if finalact else None - - self.model = sequential(fea_conv, ShortcutBlock(sequential(*rb_blocks, LR_conv)), - *upsampler, HR_conv0, HR_conv1, outact) - - def forward(self, x, outm=None): - if self.resrgan_scale == 1: - feat = pixel_unshuffle(x, scale=4) - elif self.resrgan_scale == 2: - feat = pixel_unshuffle(x, scale=2) - else: - feat = x - - return self.model(feat) - - -class RRDB(nn.Module): - """ - Residual in Residual Dense Block - (ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks) - """ - - def __init__(self, nf, nr=3, kernel_size=3, gc=32, stride=1, bias=1, pad_type='zero', - norm_type=None, act_type='leakyrelu', mode='CNA', convtype='Conv2D', - spectral_norm=False, gaussian_noise=False, plus=False): - super(RRDB, self).__init__() - # This is for backwards compatibility with existing models - if nr == 3: - self.RDB1 = ResidualDenseBlock_5C(nf, kernel_size, gc, stride, bias, pad_type, - norm_type, act_type, mode, convtype, spectral_norm=spectral_norm, - gaussian_noise=gaussian_noise, plus=plus) - self.RDB2 = ResidualDenseBlock_5C(nf, kernel_size, gc, stride, bias, pad_type, - norm_type, act_type, mode, convtype, spectral_norm=spectral_norm, - gaussian_noise=gaussian_noise, plus=plus) - self.RDB3 = ResidualDenseBlock_5C(nf, kernel_size, gc, stride, bias, pad_type, - norm_type, act_type, mode, convtype, spectral_norm=spectral_norm, - gaussian_noise=gaussian_noise, plus=plus) - else: - RDB_list = [ResidualDenseBlock_5C(nf, kernel_size, gc, stride, bias, pad_type, - norm_type, act_type, mode, convtype, spectral_norm=spectral_norm, - gaussian_noise=gaussian_noise, plus=plus) for _ in range(nr)] - self.RDBs = nn.Sequential(*RDB_list) - - def forward(self, x): - if hasattr(self, 'RDB1'): - out = self.RDB1(x) - out = self.RDB2(out) - out = self.RDB3(out) - else: - out = self.RDBs(x) - return out * 0.2 + x - - -class ResidualDenseBlock_5C(nn.Module): - """ - Residual Dense Block - The core module of paper: (Residual Dense Network for Image Super-Resolution, CVPR 18) - Modified options that can be used: - - "Partial Convolution based Padding" arXiv:1811.11718 - - "Spectral normalization" arXiv:1802.05957 - - "ICASSP 2020 - ESRGAN+ : Further Improving ESRGAN" N. C. - {Rakotonirina} and A. {Rasoanaivo} - """ - - def __init__(self, nf=64, kernel_size=3, gc=32, stride=1, bias=1, pad_type='zero', - norm_type=None, act_type='leakyrelu', mode='CNA', convtype='Conv2D', - spectral_norm=False, gaussian_noise=False, plus=False): - super(ResidualDenseBlock_5C, self).__init__() - - self.noise = GaussianNoise() if gaussian_noise else None - self.conv1x1 = conv1x1(nf, gc) if plus else None - - self.conv1 = conv_block(nf, gc, kernel_size, stride, bias=bias, pad_type=pad_type, - norm_type=norm_type, act_type=act_type, mode=mode, convtype=convtype, - spectral_norm=spectral_norm) - self.conv2 = conv_block(nf+gc, gc, kernel_size, stride, bias=bias, pad_type=pad_type, - norm_type=norm_type, act_type=act_type, mode=mode, convtype=convtype, - spectral_norm=spectral_norm) - self.conv3 = conv_block(nf+2*gc, gc, kernel_size, stride, bias=bias, pad_type=pad_type, - norm_type=norm_type, act_type=act_type, mode=mode, convtype=convtype, - spectral_norm=spectral_norm) - self.conv4 = conv_block(nf+3*gc, gc, kernel_size, stride, bias=bias, pad_type=pad_type, - norm_type=norm_type, act_type=act_type, mode=mode, convtype=convtype, - spectral_norm=spectral_norm) - if mode == 'CNA': - last_act = None - else: - last_act = act_type - self.conv5 = conv_block(nf+4*gc, nf, 3, stride, bias=bias, pad_type=pad_type, - norm_type=norm_type, act_type=last_act, mode=mode, convtype=convtype, - spectral_norm=spectral_norm) - - def forward(self, x): - x1 = self.conv1(x) - x2 = self.conv2(torch.cat((x, x1), 1)) - if self.conv1x1: - x2 = x2 + self.conv1x1(x) - x3 = self.conv3(torch.cat((x, x1, x2), 1)) - x4 = self.conv4(torch.cat((x, x1, x2, x3), 1)) - if self.conv1x1: - x4 = x4 + x2 - x5 = self.conv5(torch.cat((x, x1, x2, x3, x4), 1)) - if self.noise: - return self.noise(x5.mul(0.2) + x) - else: - return x5 * 0.2 + x - - -#################### -# ESRGANplus -#################### - -class GaussianNoise(nn.Module): - def __init__(self, sigma=0.1, is_relative_detach=False): - super().__init__() - self.sigma = sigma - self.is_relative_detach = is_relative_detach - self.noise = torch.tensor(0, dtype=torch.float) - - def forward(self, x): - if self.training and self.sigma != 0: - self.noise = self.noise.to(x.device) - scale = self.sigma * x.detach() if self.is_relative_detach else self.sigma * x - sampled_noise = self.noise.repeat(*x.size()).normal_() * scale - x = x + sampled_noise - return x - -def conv1x1(in_planes, out_planes, stride=1): - return nn.Conv2d(in_planes, out_planes, kernel_size=1, stride=stride, bias=False) - - -#################### -# SRVGGNetCompact -#################### - -class SRVGGNetCompact(nn.Module): - """A compact VGG-style network structure for super-resolution. - This class is copied from https://github.com/xinntao/Real-ESRGAN - """ - - def __init__(self, num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=16, upscale=4, act_type='prelu'): - super(SRVGGNetCompact, self).__init__() - self.num_in_ch = num_in_ch - self.num_out_ch = num_out_ch - self.num_feat = num_feat - self.num_conv = num_conv - self.upscale = upscale - self.act_type = act_type - - self.body = nn.ModuleList() - # the first conv - self.body.append(nn.Conv2d(num_in_ch, num_feat, 3, 1, 1)) - # the first activation - if act_type == 'relu': - activation = nn.ReLU(inplace=True) - elif act_type == 'prelu': - activation = nn.PReLU(num_parameters=num_feat) - elif act_type == 'leakyrelu': - activation = nn.LeakyReLU(negative_slope=0.1, inplace=True) - self.body.append(activation) - - # the body structure - for _ in range(num_conv): - self.body.append(nn.Conv2d(num_feat, num_feat, 3, 1, 1)) - # activation - if act_type == 'relu': - activation = nn.ReLU(inplace=True) - elif act_type == 'prelu': - activation = nn.PReLU(num_parameters=num_feat) - elif act_type == 'leakyrelu': - activation = nn.LeakyReLU(negative_slope=0.1, inplace=True) - self.body.append(activation) - - # the last conv - self.body.append(nn.Conv2d(num_feat, num_out_ch * upscale * upscale, 3, 1, 1)) - # upsample - self.upsampler = nn.PixelShuffle(upscale) - - def forward(self, x): - out = x - for i in range(0, len(self.body)): - out = self.body[i](out) - - out = self.upsampler(out) - # add the nearest upsampled image, so that the network learns the residual - base = F.interpolate(x, scale_factor=self.upscale, mode='nearest') - out += base - return out - - -#################### -# Upsampler -#################### - -class Upsample(nn.Module): - r"""Upsamples a given multi-channel 1D (temporal), 2D (spatial) or 3D (volumetric) data. - The input data is assumed to be of the form - `minibatch x channels x [optional depth] x [optional height] x width`. - """ - - def __init__(self, size=None, scale_factor=None, mode="nearest", align_corners=None): - super(Upsample, self).__init__() - if isinstance(scale_factor, tuple): - self.scale_factor = tuple(float(factor) for factor in scale_factor) - else: - self.scale_factor = float(scale_factor) if scale_factor else None - self.mode = mode - self.size = size - self.align_corners = align_corners - - def forward(self, x): - return nn.functional.interpolate(x, size=self.size, scale_factor=self.scale_factor, mode=self.mode, align_corners=self.align_corners) - - def extra_repr(self): - if self.scale_factor is not None: - info = f'scale_factor={self.scale_factor}' - else: - info = f'size={self.size}' - info += f', mode={self.mode}' - return info - - -def pixel_unshuffle(x, scale): - """ Pixel unshuffle. - Args: - x (Tensor): Input feature with shape (b, c, hh, hw). - scale (int): Downsample ratio. - Returns: - Tensor: the pixel unshuffled feature. - """ - b, c, hh, hw = x.size() - out_channel = c * (scale**2) - assert hh % scale == 0 and hw % scale == 0 - h = hh // scale - w = hw // scale - x_view = x.view(b, c, h, scale, w, scale) - return x_view.permute(0, 1, 3, 5, 2, 4).reshape(b, out_channel, h, w) - - -def pixelshuffle_block(in_nc, out_nc, upscale_factor=2, kernel_size=3, stride=1, bias=True, - pad_type='zero', norm_type=None, act_type='relu', convtype='Conv2D'): - """ - Pixel shuffle layer - (Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional - Neural Network, CVPR17) - """ - conv = conv_block(in_nc, out_nc * (upscale_factor ** 2), kernel_size, stride, bias=bias, - pad_type=pad_type, norm_type=None, act_type=None, convtype=convtype) - pixel_shuffle = nn.PixelShuffle(upscale_factor) - - n = norm(norm_type, out_nc) if norm_type else None - a = act(act_type) if act_type else None - return sequential(conv, pixel_shuffle, n, a) - - -def upconv_block(in_nc, out_nc, upscale_factor=2, kernel_size=3, stride=1, bias=True, - pad_type='zero', norm_type=None, act_type='relu', mode='nearest', convtype='Conv2D'): - """ Upconv layer """ - upscale_factor = (1, upscale_factor, upscale_factor) if convtype == 'Conv3D' else upscale_factor - upsample = Upsample(scale_factor=upscale_factor, mode=mode) - conv = conv_block(in_nc, out_nc, kernel_size, stride, bias=bias, - pad_type=pad_type, norm_type=norm_type, act_type=act_type, convtype=convtype) - return sequential(upsample, conv) - - - - - - - - -#################### -# Basic blocks -#################### - - -def make_layer(basic_block, num_basic_block, **kwarg): - """Make layers by stacking the same blocks. - Args: - basic_block (nn.module): nn.module class for basic block. (block) - num_basic_block (int): number of blocks. (n_layers) - Returns: - nn.Sequential: Stacked blocks in nn.Sequential. - """ - layers = [] - for _ in range(num_basic_block): - layers.append(basic_block(**kwarg)) - return nn.Sequential(*layers) - - -def act(act_type, inplace=True, neg_slope=0.2, n_prelu=1, beta=1.0): - """ activation helper """ - act_type = act_type.lower() - if act_type == 'relu': - layer = nn.ReLU(inplace) - elif act_type in ('leakyrelu', 'lrelu'): - layer = nn.LeakyReLU(neg_slope, inplace) - elif act_type == 'prelu': - layer = nn.PReLU(num_parameters=n_prelu, init=neg_slope) - elif act_type == 'tanh': # [-1, 1] range output - layer = nn.Tanh() - elif act_type == 'sigmoid': # [0, 1] range output - layer = nn.Sigmoid() - else: - raise NotImplementedError(f'activation layer [{act_type}] is not found') - return layer - - -class Identity(nn.Module): - def __init__(self, *kwargs): - super(Identity, self).__init__() - - def forward(self, x, *kwargs): - return x - - -def norm(norm_type, nc): - """ Return a normalization layer """ - norm_type = norm_type.lower() - if norm_type == 'batch': - layer = nn.BatchNorm2d(nc, affine=True) - elif norm_type == 'instance': - layer = nn.InstanceNorm2d(nc, affine=False) - elif norm_type == 'none': - def norm_layer(x): return Identity() - else: - raise NotImplementedError(f'normalization layer [{norm_type}] is not found') - return layer - - -def pad(pad_type, padding): - """ padding layer helper """ - pad_type = pad_type.lower() - if padding == 0: - return None - if pad_type == 'reflect': - layer = nn.ReflectionPad2d(padding) - elif pad_type == 'replicate': - layer = nn.ReplicationPad2d(padding) - elif pad_type == 'zero': - layer = nn.ZeroPad2d(padding) - else: - raise NotImplementedError(f'padding layer [{pad_type}] is not implemented') - return layer - - -def get_valid_padding(kernel_size, dilation): - kernel_size = kernel_size + (kernel_size - 1) * (dilation - 1) - padding = (kernel_size - 1) // 2 - return padding - - -class ShortcutBlock(nn.Module): - """ Elementwise sum the output of a submodule to its input """ - def __init__(self, submodule): - super(ShortcutBlock, self).__init__() - self.sub = submodule - - def forward(self, x): - output = x + self.sub(x) - return output - - def __repr__(self): - return 'Identity + \n|' + self.sub.__repr__().replace('\n', '\n|') - - -def sequential(*args): - """ Flatten Sequential. It unwraps nn.Sequential. """ - if len(args) == 1: - if isinstance(args[0], OrderedDict): - raise NotImplementedError('sequential does not support OrderedDict input.') - return args[0] # No sequential is needed. - modules = [] - for module in args: - if isinstance(module, nn.Sequential): - for submodule in module.children(): - modules.append(submodule) - elif isinstance(module, nn.Module): - modules.append(module) - return nn.Sequential(*modules) - - -def conv_block(in_nc, out_nc, kernel_size, stride=1, dilation=1, groups=1, bias=True, - pad_type='zero', norm_type=None, act_type='relu', mode='CNA', convtype='Conv2D', - spectral_norm=False): - """ Conv layer with padding, normalization, activation """ - assert mode in ['CNA', 'NAC', 'CNAC'], f'Wrong conv mode [{mode}]' - padding = get_valid_padding(kernel_size, dilation) - p = pad(pad_type, padding) if pad_type and pad_type != 'zero' else None - padding = padding if pad_type == 'zero' else 0 - - if convtype=='PartialConv2D': - from torchvision.ops import PartialConv2d # this is definitely not going to work, but PartialConv2d doesn't work anyway and this shuts up static analyzer - c = PartialConv2d(in_nc, out_nc, kernel_size=kernel_size, stride=stride, padding=padding, - dilation=dilation, bias=bias, groups=groups) - elif convtype=='DeformConv2D': - from torchvision.ops import DeformConv2d # not tested - c = DeformConv2d(in_nc, out_nc, kernel_size=kernel_size, stride=stride, padding=padding, - dilation=dilation, bias=bias, groups=groups) - elif convtype=='Conv3D': - c = nn.Conv3d(in_nc, out_nc, kernel_size=kernel_size, stride=stride, padding=padding, - dilation=dilation, bias=bias, groups=groups) - else: - c = nn.Conv2d(in_nc, out_nc, kernel_size=kernel_size, stride=stride, padding=padding, - dilation=dilation, bias=bias, groups=groups) - - if spectral_norm: - c = nn.utils.spectral_norm(c) - - a = act(act_type) if act_type else None - if 'CNA' in mode: - n = norm(norm_type, out_nc) if norm_type else None - return sequential(p, c, n, a) - elif mode == 'NAC': - if norm_type is None and act_type is not None: - a = act(act_type, inplace=False) - n = norm(norm_type, in_nc) if norm_type else None - return sequential(n, a, p, c) diff --git a/modules/gfpgan_model.py b/modules/gfpgan_model.py index 01d668ec..6b6f17c4 100644 --- a/modules/gfpgan_model.py +++ b/modules/gfpgan_model.py @@ -1,8 +1,5 @@ import os -import facexlib -import gfpgan - import modules.face_restoration from modules import paths, shared, devices, modelloader, errors @@ -41,6 +38,8 @@ def gfpgann(): print("Unable to load gfpgan model!") return None + import facexlib.detection.retinaface + if hasattr(facexlib.detection.retinaface, 'device'): facexlib.detection.retinaface.device = devices.device_gfpgan model_file_path = model_file @@ -81,8 +80,10 @@ gfpgan_constructor = None def setup_model(dirname): try: os.makedirs(model_path, exist_ok=True) - from gfpgan import GFPGANer - from facexlib import detection, parsing # noqa: F401 + import gfpgan + import facexlib.detection + import facexlib.parsing + global user_path global have_gfpgan global gfpgan_constructor @@ -111,7 +112,7 @@ def setup_model(dirname): facexlib.parsing.load_file_from_url = facex_load_file_from_url2 user_path = dirname have_gfpgan = True - gfpgan_constructor = GFPGANer + gfpgan_constructor = gfpgan.GFPGANer class FaceRestorerGFPGAN(modules.face_restoration.FaceRestoration): def name(self): diff --git a/modules/launch_utils.py b/modules/launch_utils.py index dabef0f5..c2cbd8ce 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -345,13 +345,11 @@ def prepare_environment(): stable_diffusion_repo = os.environ.get('STABLE_DIFFUSION_REPO', "https://github.com/Stability-AI/stablediffusion.git") stable_diffusion_xl_repo = os.environ.get('STABLE_DIFFUSION_XL_REPO', "https://github.com/Stability-AI/generative-models.git") k_diffusion_repo = os.environ.get('K_DIFFUSION_REPO', 'https://github.com/crowsonkb/k-diffusion.git') - codeformer_repo = os.environ.get('CODEFORMER_REPO', 'https://github.com/sczhou/CodeFormer.git') blip_repo = os.environ.get('BLIP_REPO', 'https://github.com/salesforce/BLIP.git') stable_diffusion_commit_hash = os.environ.get('STABLE_DIFFUSION_COMMIT_HASH', "cf1d67a6fd5ea1aa600c4df58e5b47da45f6bdbf") stable_diffusion_xl_commit_hash = os.environ.get('STABLE_DIFFUSION_XL_COMMIT_HASH', "45c443b316737a4ab6e40413d7794a7f5657c19f") k_diffusion_commit_hash = os.environ.get('K_DIFFUSION_COMMIT_HASH', "ab527a9a6d347f364e3d185ba6d714e22d80cb3c") - codeformer_commit_hash = os.environ.get('CODEFORMER_COMMIT_HASH', "c5b4593074ba6214284d6acd5f1719b6c5d739af") blip_commit_hash = os.environ.get('BLIP_COMMIT_HASH', "48211a1594f1321b00f14c9f7a5b4813144b2fb9") try: @@ -408,15 +406,10 @@ def prepare_environment(): git_clone(stable_diffusion_repo, repo_dir('stable-diffusion-stability-ai'), "Stable Diffusion", stable_diffusion_commit_hash) git_clone(stable_diffusion_xl_repo, repo_dir('generative-models'), "Stable Diffusion XL", stable_diffusion_xl_commit_hash) git_clone(k_diffusion_repo, repo_dir('k-diffusion'), "K-diffusion", k_diffusion_commit_hash) - git_clone(codeformer_repo, repo_dir('CodeFormer'), "CodeFormer", codeformer_commit_hash) git_clone(blip_repo, repo_dir('BLIP'), "BLIP", blip_commit_hash) startup_timer.record("clone repositores") - if not is_installed("lpips"): - run_pip(f"install -r \"{os.path.join(repo_dir('CodeFormer'), 'requirements.txt')}\"", "requirements for CodeFormer") - startup_timer.record("install CodeFormer requirements") - if not os.path.isfile(requirements_file): requirements_file = os.path.join(script_path, requirements_file) diff --git a/modules/modelloader.py b/modules/modelloader.py index 098bcb79..30116932 100644 --- a/modules/modelloader.py +++ b/modules/modelloader.py @@ -1,5 +1,6 @@ from __future__ import annotations +import logging import os import shutil import importlib @@ -10,6 +11,9 @@ from modules.upscaler import Upscaler, UpscalerLanczos, UpscalerNearest, Upscale from modules.paths import script_path, models_path +logger = logging.getLogger(__name__) + + def load_file_from_url( url: str, *, @@ -177,3 +181,15 @@ def load_upscalers(): # Special case for UpscalerNone keeps it at the beginning of the list. key=lambda x: x.name.lower() if not isinstance(x.scaler, (UpscalerNone, UpscalerLanczos, UpscalerNearest)) else "" ) + + +def load_spandrel_model(path, *, device, half: bool = False, dtype=None): + import spandrel + model = spandrel.ModelLoader(device=device).load_from_file(path) + if half: + model = model.model.half() + if dtype: + model = model.model.to(dtype=dtype) + model.eval() + logger.debug("Loaded %s from %s (device=%s, half=%s, dtype=%s)", model, path, device, half, dtype) + return model diff --git a/modules/paths.py b/modules/paths.py index 187b9496..03064651 100644 --- a/modules/paths.py +++ b/modules/paths.py @@ -38,7 +38,6 @@ mute_sdxl_imports() path_dirs = [ (sd_path, 'ldm', 'Stable Diffusion', []), (os.path.join(sd_path, '../generative-models'), 'sgm', 'Stable Diffusion XL', ["sgm"]), - (os.path.join(sd_path, '../CodeFormer'), 'inference_codeformer.py', 'CodeFormer', []), (os.path.join(sd_path, '../BLIP'), 'models/blip.py', 'BLIP', []), (os.path.join(sd_path, '../k-diffusion'), 'k_diffusion/sampling.py', 'k_diffusion', ["atstart"]), ] diff --git a/modules/realesrgan_model.py b/modules/realesrgan_model.py index 02841c30..332d8f4b 100644 --- a/modules/realesrgan_model.py +++ b/modules/realesrgan_model.py @@ -1,9 +1,6 @@ import os -import numpy as np -from PIL import Image -from realesrgan import RealESRGANer - +from modules.upscaler_utils import upscale_with_model from modules.upscaler import Upscaler, UpscalerData from modules.shared import cmd_opts, opts from modules import modelloader, errors @@ -14,29 +11,20 @@ class UpscalerRealESRGAN(Upscaler): self.name = "RealESRGAN" self.user_path = path super().__init__() - try: - from basicsr.archs.rrdbnet_arch import RRDBNet # noqa: F401 - from realesrgan import RealESRGANer # noqa: F401 - from realesrgan.archs.srvgg_arch import SRVGGNetCompact # noqa: F401 - self.enable = True - self.scalers = [] - scalers = self.load_models(path) + self.enable = True + self.scalers = [] + scalers = get_realesrgan_models(self) - local_model_paths = self.find_models(ext_filter=[".pth"]) - for scaler in scalers: - if scaler.local_data_path.startswith("http"): - filename = modelloader.friendly_name(scaler.local_data_path) - local_model_candidates = [local_model for local_model in local_model_paths if local_model.endswith(f"{filename}.pth")] - if local_model_candidates: - scaler.local_data_path = local_model_candidates[0] + local_model_paths = self.find_models(ext_filter=[".pth"]) + for scaler in scalers: + if scaler.local_data_path.startswith("http"): + filename = modelloader.friendly_name(scaler.local_data_path) + local_model_candidates = [local_model for local_model in local_model_paths if local_model.endswith(f"{filename}.pth")] + if local_model_candidates: + scaler.local_data_path = local_model_candidates[0] - if scaler.name in opts.realesrgan_enabled_models: - self.scalers.append(scaler) - - except Exception: - errors.report("Error importing Real-ESRGAN", exc_info=True) - self.enable = False - self.scalers = [] + if scaler.name in opts.realesrgan_enabled_models: + self.scalers.append(scaler) def do_upscale(self, img, path): if not self.enable: @@ -48,20 +36,18 @@ class UpscalerRealESRGAN(Upscaler): errors.report(f"Unable to load RealESRGAN model {path}", exc_info=True) return img - upsampler = RealESRGANer( - scale=info.scale, - model_path=info.local_data_path, - model=info.model(), - half=not cmd_opts.no_half and not cmd_opts.upcast_sampling, - tile=opts.ESRGAN_tile, - tile_pad=opts.ESRGAN_tile_overlap, + mod = modelloader.load_spandrel_model( + info.local_data_path, device=self.device, + half=(not cmd_opts.no_half and not cmd_opts.upcast_sampling), + ) + return upscale_with_model( + mod, + img, + tile_size=opts.ESRGAN_tile, + tile_overlap=opts.ESRGAN_tile_overlap, + # TODO: `outscale`? ) - - upsampled = upsampler.enhance(np.array(img), outscale=info.scale)[0] - - image = Image.fromarray(upsampled) - return image def load_model(self, path): for scaler in self.scalers: @@ -76,58 +62,43 @@ class UpscalerRealESRGAN(Upscaler): return scaler raise ValueError(f"Unable to find model info: {path}") - def load_models(self, _): - return get_realesrgan_models(self) - -def get_realesrgan_models(scaler): - try: - from basicsr.archs.rrdbnet_arch import RRDBNet - from realesrgan.archs.srvgg_arch import SRVGGNetCompact - models = [ - UpscalerData( - name="R-ESRGAN General 4xV3", - path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-x4v3.pth", - scale=4, - upscaler=scaler, - model=lambda: SRVGGNetCompact(num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=32, upscale=4, act_type='prelu') - ), - UpscalerData( - name="R-ESRGAN General WDN 4xV3", - path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-wdn-x4v3.pth", - scale=4, - upscaler=scaler, - model=lambda: SRVGGNetCompact(num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=32, upscale=4, act_type='prelu') - ), - UpscalerData( - name="R-ESRGAN AnimeVideo", - path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-animevideov3.pth", - scale=4, - upscaler=scaler, - model=lambda: SRVGGNetCompact(num_in_ch=3, num_out_ch=3, num_feat=64, num_conv=16, upscale=4, act_type='prelu') - ), - UpscalerData( - name="R-ESRGAN 4x+", - path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth", - scale=4, - upscaler=scaler, - model=lambda: RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=23, num_grow_ch=32, scale=4) - ), - UpscalerData( - name="R-ESRGAN 4x+ Anime6B", - path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.2.4/RealESRGAN_x4plus_anime_6B.pth", - scale=4, - upscaler=scaler, - model=lambda: RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=6, num_grow_ch=32, scale=4) - ), - UpscalerData( - name="R-ESRGAN 2x+", - path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth", - scale=2, - upscaler=scaler, - model=lambda: RRDBNet(num_in_ch=3, num_out_ch=3, num_feat=64, num_block=23, num_grow_ch=32, scale=2) - ), - ] - return models - except Exception: - errors.report("Error making Real-ESRGAN models list", exc_info=True) +def get_realesrgan_models(scaler: UpscalerRealESRGAN): + return [ + UpscalerData( + name="R-ESRGAN General 4xV3", + path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-x4v3.pth", + scale=4, + upscaler=scaler, + ), + UpscalerData( + name="R-ESRGAN General WDN 4xV3", + path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-general-wdn-x4v3.pth", + scale=4, + upscaler=scaler, + ), + UpscalerData( + name="R-ESRGAN AnimeVideo", + path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.5.0/realesr-animevideov3.pth", + scale=4, + upscaler=scaler, + ), + UpscalerData( + name="R-ESRGAN 4x+", + path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth", + scale=4, + upscaler=scaler, + ), + UpscalerData( + name="R-ESRGAN 4x+ Anime6B", + path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.2.4/RealESRGAN_x4plus_anime_6B.pth", + scale=4, + upscaler=scaler, + ), + UpscalerData( + name="R-ESRGAN 2x+", + path="https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.1/RealESRGAN_x2plus.pth", + scale=2, + upscaler=scaler, + ), + ] diff --git a/modules/sysinfo.py b/modules/sysinfo.py index b669edd0..5abf616b 100644 --- a/modules/sysinfo.py +++ b/modules/sysinfo.py @@ -26,11 +26,9 @@ environment_whitelist = { "OPENCLIP_PACKAGE", "STABLE_DIFFUSION_REPO", "K_DIFFUSION_REPO", - "CODEFORMER_REPO", "BLIP_REPO", "STABLE_DIFFUSION_COMMIT_HASH", "K_DIFFUSION_COMMIT_HASH", - "CODEFORMER_COMMIT_HASH", "BLIP_COMMIT_HASH", "COMMANDLINE_ARGS", "IGNORE_CMD_ARGS_ERRORS", diff --git a/modules/upscaler.py b/modules/upscaler.py index b256e085..3aee69db 100644 --- a/modules/upscaler.py +++ b/modules/upscaler.py @@ -98,6 +98,9 @@ class UpscalerData: self.scale = scale self.model = model + def __repr__(self): + return f"" + class UpscalerNone(Upscaler): name = "None" -- cgit v1.2.1 From b621a63cf68c788487684250856707cb352b82d0 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Mon, 25 Dec 2023 23:01:02 +0200 Subject: Unify CodeFormer and GFPGAN restoration backends, use Spandrel for GFPGAN --- modules/codeformer_model.py | 158 +++++++++--------------------------- modules/face_restoration_utils.py | 163 +++++++++++++++++++++++++++++++++++++ modules/gfpgan_model.py | 166 +++++++++++++------------------------- 3 files changed, 257 insertions(+), 230 deletions(-) create mode 100644 modules/face_restoration_utils.py (limited to 'modules') diff --git a/modules/codeformer_model.py b/modules/codeformer_model.py index 517eadfd..ceda4bab 100644 --- a/modules/codeformer_model.py +++ b/modules/codeformer_model.py @@ -1,140 +1,62 @@ -import os +from __future__ import annotations + +import logging -import cv2 import torch -import modules.face_restoration -import modules.shared -from modules import shared, devices, modelloader, errors -from modules.paths import models_path +from modules import ( + devices, + errors, + face_restoration, + face_restoration_utils, + modelloader, + shared, +) + +logger = logging.getLogger(__name__) -model_dir = "Codeformer" -model_path = os.path.join(models_path, model_dir) model_url = 'https://github.com/sczhou/CodeFormer/releases/download/v0.1.0/codeformer.pth' +model_download_name = 'codeformer-v0.1.0.pth' -codeformer = None +# used by e.g. postprocessing_codeformer.py +codeformer: face_restoration.FaceRestoration | None = None -class FaceRestorerCodeFormer(modules.face_restoration.FaceRestoration): +class FaceRestorerCodeFormer(face_restoration_utils.CommonFaceRestoration): def name(self): return "CodeFormer" - def __init__(self, dirname): - self.net = None - self.face_helper = None - self.cmd_dir = dirname - - def create_models(self): - from facexlib.detection import retinaface - from facexlib.utils.face_restoration_helper import FaceRestoreHelper - - if self.net is not None and self.face_helper is not None: - self.net.to(devices.device_codeformer) - return self.net, self.face_helper - model_paths = modelloader.load_models( - model_path, - model_url, - self.cmd_dir, - download_name='codeformer-v0.1.0.pth', + def load_net(self) -> torch.Module: + for model_path in modelloader.load_models( + model_path=self.model_path, + model_url=model_url, + command_path=self.model_path, + download_name=model_download_name, ext_filter=['.pth'], - ) - - if len(model_paths) != 0: - ckpt_path = model_paths[0] - else: - print("Unable to load codeformer model.") - return None, None - net = modelloader.load_spandrel_model(ckpt_path, device=devices.device_codeformer) - - if hasattr(retinaface, 'device'): - retinaface.device = devices.device_codeformer - - face_helper = FaceRestoreHelper( - upscale_factor=1, - face_size=512, - crop_ratio=(1, 1), - det_model='retinaface_resnet50', - save_ext='png', - use_parse=True, - device=devices.device_codeformer, - ) - - self.net = net - self.face_helper = face_helper - - def send_model_to(self, device): - self.net.to(device) - self.face_helper.face_det.to(device) - self.face_helper.face_parse.to(device) - - def restore(self, np_image, w=None): - from torchvision.transforms.functional import normalize - from basicsr.utils import img2tensor, tensor2img - np_image = np_image[:, :, ::-1] - - original_resolution = np_image.shape[0:2] - - self.create_models() - if self.net is None or self.face_helper is None: - return np_image - - self.send_model_to(devices.device_codeformer) - - self.face_helper.clean_all() - self.face_helper.read_image(np_image) - self.face_helper.get_face_landmarks_5(only_center_face=False, resize=640, eye_dist_threshold=5) - self.face_helper.align_warp_face() - - for cropped_face in self.face_helper.cropped_faces: - cropped_face_t = img2tensor(cropped_face / 255., bgr2rgb=True, float32=True) - normalize(cropped_face_t, (0.5, 0.5, 0.5), (0.5, 0.5, 0.5), inplace=True) - cropped_face_t = cropped_face_t.unsqueeze(0).to(devices.device_codeformer) - - try: - with torch.no_grad(): - res = self.net(cropped_face_t, w=w if w is not None else shared.opts.code_former_weight, adain=True) - if isinstance(res, tuple): - output = res[0] - else: - output = res - if not isinstance(res, torch.Tensor): - raise TypeError(f"Expected torch.Tensor, got {type(res)}") - restored_face = tensor2img(output, rgb2bgr=True, min_max=(-1, 1)) - del output - devices.torch_gc() - except Exception: - errors.report('Failed inference for CodeFormer', exc_info=True) - restored_face = tensor2img(cropped_face_t, rgb2bgr=True, min_max=(-1, 1)) - - restored_face = restored_face.astype('uint8') - self.face_helper.add_restored_face(restored_face) - - self.face_helper.get_inverse_affine(None) - - restored_img = self.face_helper.paste_faces_to_input_image() - restored_img = restored_img[:, :, ::-1] + ): + return modelloader.load_spandrel_model( + model_path, + device=devices.device_codeformer, + ).model + raise ValueError("No codeformer model found") - if original_resolution != restored_img.shape[0:2]: - restored_img = cv2.resize( - restored_img, - (0, 0), - fx=original_resolution[1]/restored_img.shape[1], - fy=original_resolution[0]/restored_img.shape[0], - interpolation=cv2.INTER_LINEAR, - ) + def get_device(self): + return devices.device_codeformer - self.face_helper.clean_all() + def restore(self, np_image, w: float | None = None): + if w is None: + w = getattr(shared.opts, "code_former_weight", 0.5) - if shared.opts.face_restoration_unload: - self.send_model_to(devices.cpu) + def restore_face(cropped_face_t): + assert self.net is not None + return self.net(cropped_face_t, w=w, adain=True)[0] - return restored_img + return self.restore_with_helper(np_image, restore_face) -def setup_model(dirname): - os.makedirs(model_path, exist_ok=True) +def setup_model(dirname: str) -> None: + global codeformer try: - global codeformer codeformer = FaceRestorerCodeFormer(dirname) shared.face_restorers.append(codeformer) except Exception: diff --git a/modules/face_restoration_utils.py b/modules/face_restoration_utils.py new file mode 100644 index 00000000..c65c85ef --- /dev/null +++ b/modules/face_restoration_utils.py @@ -0,0 +1,163 @@ +from __future__ import annotations + +import logging +import os +from functools import cached_property +from typing import TYPE_CHECKING, Callable + +import cv2 +import numpy as np +import torch + +from modules import devices, errors, face_restoration, shared + +if TYPE_CHECKING: + from facexlib.utils.face_restoration_helper import FaceRestoreHelper + +logger = logging.getLogger(__name__) + + +def create_face_helper(device) -> FaceRestoreHelper: + from facexlib.detection import retinaface + from facexlib.utils.face_restoration_helper import FaceRestoreHelper + if hasattr(retinaface, 'device'): + retinaface.device = device + return FaceRestoreHelper( + upscale_factor=1, + face_size=512, + crop_ratio=(1, 1), + det_model='retinaface_resnet50', + save_ext='png', + use_parse=True, + device=device, + ) + + +def restore_with_face_helper( + np_image: np.ndarray, + face_helper: FaceRestoreHelper, + restore_face: Callable[[np.ndarray], np.ndarray], +) -> np.ndarray: + """ + Find faces in the image using face_helper, restore them using restore_face, and paste them back into the image. + + `restore_face` should take a cropped face image and return a restored face image. + """ + from basicsr.utils import img2tensor, tensor2img + from torchvision.transforms.functional import normalize + np_image = np_image[:, :, ::-1] + original_resolution = np_image.shape[0:2] + + try: + logger.debug("Detecting faces...") + face_helper.clean_all() + face_helper.read_image(np_image) + face_helper.get_face_landmarks_5(only_center_face=False, resize=640, eye_dist_threshold=5) + face_helper.align_warp_face() + logger.debug("Found %d faces, restoring", len(face_helper.cropped_faces)) + for cropped_face in face_helper.cropped_faces: + cropped_face_t = img2tensor(cropped_face / 255., bgr2rgb=True, float32=True) + normalize(cropped_face_t, (0.5, 0.5, 0.5), (0.5, 0.5, 0.5), inplace=True) + cropped_face_t = cropped_face_t.unsqueeze(0).to(devices.device_codeformer) + + try: + with torch.no_grad(): + restored_face = tensor2img( + restore_face(cropped_face_t), + rgb2bgr=True, + min_max=(-1, 1), + ) + devices.torch_gc() + except Exception: + errors.report('Failed face-restoration inference', exc_info=True) + restored_face = tensor2img(cropped_face_t, rgb2bgr=True, min_max=(-1, 1)) + + restored_face = restored_face.astype('uint8') + face_helper.add_restored_face(restored_face) + + logger.debug("Merging restored faces into image") + face_helper.get_inverse_affine(None) + img = face_helper.paste_faces_to_input_image() + img = img[:, :, ::-1] + if original_resolution != img.shape[0:2]: + img = cv2.resize( + img, + (0, 0), + fx=original_resolution[1] / img.shape[1], + fy=original_resolution[0] / img.shape[0], + interpolation=cv2.INTER_LINEAR, + ) + logger.debug("Face restoration complete") + finally: + face_helper.clean_all() + return img + + +class CommonFaceRestoration(face_restoration.FaceRestoration): + net: torch.Module | None + model_url: str + model_download_name: str + + def __init__(self, model_path: str): + super().__init__() + self.net = None + self.model_path = model_path + os.makedirs(model_path, exist_ok=True) + + @cached_property + def face_helper(self) -> FaceRestoreHelper: + return create_face_helper(self.get_device()) + + def send_model_to(self, device): + if self.net: + logger.debug("Sending %s to %s", self.net, device) + self.net.to(device) + if self.face_helper: + logger.debug("Sending face helper to %s", device) + self.face_helper.face_det.to(device) + self.face_helper.face_parse.to(device) + + def get_device(self): + raise NotImplementedError("get_device must be implemented by subclasses") + + def load_net(self) -> torch.Module: + raise NotImplementedError("load_net must be implemented by subclasses") + + def restore_with_helper( + self, + np_image: np.ndarray, + restore_face: Callable[[np.ndarray], np.ndarray], + ) -> np.ndarray: + try: + if self.net is None: + self.net = self.load_net() + except Exception: + logger.warning("Unable to load face-restoration model", exc_info=True) + return np_image + + try: + self.send_model_to(self.get_device()) + return restore_with_face_helper(np_image, self.face_helper, restore_face) + finally: + if shared.opts.face_restoration_unload: + self.send_model_to(devices.cpu) + + +def patch_facexlib(dirname: str) -> None: + import facexlib.detection + import facexlib.parsing + + det_facex_load_file_from_url = facexlib.detection.load_file_from_url + par_facex_load_file_from_url = facexlib.parsing.load_file_from_url + + def update_kwargs(kwargs): + return dict(kwargs, save_dir=dirname, model_dir=None) + + def facex_load_file_from_url(**kwargs): + return det_facex_load_file_from_url(**update_kwargs(kwargs)) + + def facex_load_file_from_url2(**kwargs): + return par_facex_load_file_from_url(**update_kwargs(kwargs)) + + facexlib.detection.load_file_from_url = facex_load_file_from_url + facexlib.parsing.load_file_from_url = facex_load_file_from_url2 diff --git a/modules/gfpgan_model.py b/modules/gfpgan_model.py index 6b6f17c4..a356b56f 100644 --- a/modules/gfpgan_model.py +++ b/modules/gfpgan_model.py @@ -1,126 +1,68 @@ +from __future__ import annotations + +import logging import os -import modules.face_restoration -from modules import paths, shared, devices, modelloader, errors +from modules import ( + devices, + errors, + face_restoration, + face_restoration_utils, + modelloader, + shared, +) -model_dir = "GFPGAN" -user_path = None -model_path = os.path.join(paths.models_path, model_dir) -model_file_path = None +logger = logging.getLogger(__name__) model_url = "https://github.com/TencentARC/GFPGAN/releases/download/v1.3.0/GFPGANv1.4.pth" -have_gfpgan = False -loaded_gfpgan_model = None - - -def gfpgann(): - global loaded_gfpgan_model - global model_path - global model_file_path - if loaded_gfpgan_model is not None: - loaded_gfpgan_model.gfpgan.to(devices.device_gfpgan) - return loaded_gfpgan_model - - if gfpgan_constructor is None: - return None - - models = modelloader.load_models(model_path, model_url, user_path, ext_filter=['.pth']) - - if len(models) == 1 and models[0].startswith("http"): - model_file = models[0] - elif len(models) != 0: - gfp_models = [] - for item in models: - if 'GFPGAN' in os.path.basename(item): - gfp_models.append(item) - latest_file = max(gfp_models, key=os.path.getctime) - model_file = latest_file - else: - print("Unable to load gfpgan model!") - return None - - import facexlib.detection.retinaface - - if hasattr(facexlib.detection.retinaface, 'device'): - facexlib.detection.retinaface.device = devices.device_gfpgan - model_file_path = model_file - model = gfpgan_constructor(model_path=model_file, upscale=1, arch='clean', channel_multiplier=2, bg_upsampler=None, device=devices.device_gfpgan) - loaded_gfpgan_model = model - - return model - - -def send_model_to(model, device): - model.gfpgan.to(device) - model.face_helper.face_det.to(device) - model.face_helper.face_parse.to(device) +model_download_name = "GFPGANv1.4.pth" +gfpgan_face_restorer: face_restoration.FaceRestoration | None = None + + +class FaceRestorerGFPGAN(face_restoration_utils.CommonFaceRestoration): + def name(self): + return "GFPGAN" + + def get_device(self): + return devices.device_gfpgan + + def load_net(self) -> None: + for model_path in modelloader.load_models( + model_path=self.model_path, + model_url=model_url, + command_path=self.model_path, + download_name=model_download_name, + ext_filter=['.pth'], + ): + if 'GFPGAN' in os.path.basename(model_path): + net = modelloader.load_spandrel_model( + model_path, + device=self.get_device(), + ).model + net.different_w = True # see https://github.com/chaiNNer-org/spandrel/pull/81 + return net + raise ValueError("No GFPGAN model found") + + def restore(self, np_image): + def restore_face(cropped_face_t): + assert self.net is not None + return self.net(cropped_face_t, return_rgb=False)[0] + + return self.restore_with_helper(np_image, restore_face) def gfpgan_fix_faces(np_image): - model = gfpgann() - if model is None: - return np_image - - send_model_to(model, devices.device_gfpgan) - - np_image_bgr = np_image[:, :, ::-1] - cropped_faces, restored_faces, gfpgan_output_bgr = model.enhance(np_image_bgr, has_aligned=False, only_center_face=False, paste_back=True) - np_image = gfpgan_output_bgr[:, :, ::-1] - - model.face_helper.clean_all() - - if shared.opts.face_restoration_unload: - send_model_to(model, devices.cpu) - + if gfpgan_face_restorer: + return gfpgan_face_restorer.restore(np_image) + logger.warning("GFPGAN face restorer not set up") return np_image -gfpgan_constructor = None +def setup_model(dirname: str) -> None: + global gfpgan_face_restorer - -def setup_model(dirname): try: - os.makedirs(model_path, exist_ok=True) - import gfpgan - import facexlib.detection - import facexlib.parsing - - global user_path - global have_gfpgan - global gfpgan_constructor - global model_file_path - - facexlib_path = model_path - - if dirname is not None: - facexlib_path = dirname - - load_file_from_url_orig = gfpgan.utils.load_file_from_url - facex_load_file_from_url_orig = facexlib.detection.load_file_from_url - facex_load_file_from_url_orig2 = facexlib.parsing.load_file_from_url - - def my_load_file_from_url(**kwargs): - return load_file_from_url_orig(**dict(kwargs, model_dir=model_file_path)) - - def facex_load_file_from_url(**kwargs): - return facex_load_file_from_url_orig(**dict(kwargs, save_dir=facexlib_path, model_dir=None)) - - def facex_load_file_from_url2(**kwargs): - return facex_load_file_from_url_orig2(**dict(kwargs, save_dir=facexlib_path, model_dir=None)) - - gfpgan.utils.load_file_from_url = my_load_file_from_url - facexlib.detection.load_file_from_url = facex_load_file_from_url - facexlib.parsing.load_file_from_url = facex_load_file_from_url2 - user_path = dirname - have_gfpgan = True - gfpgan_constructor = gfpgan.GFPGANer - - class FaceRestorerGFPGAN(modules.face_restoration.FaceRestoration): - def name(self): - return "GFPGAN" - - def restore(self, np_image): - return gfpgan_fix_faces(np_image) - - shared.face_restorers.append(FaceRestorerGFPGAN()) + face_restoration_utils.patch_facexlib(dirname) + gfpgan_face_restorer = FaceRestorerGFPGAN(model_path=dirname) + shared.face_restorers.append(gfpgan_face_restorer) except Exception: errors.report("Error setting up GFPGAN", exc_info=True) -- cgit v1.2.1 From c756133541da478a35a74cda416d114a8973cf8e Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Wed, 27 Dec 2023 10:55:01 +0200 Subject: Add experimental HAT model --- modules/hat_model.py | 42 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 42 insertions(+) create mode 100644 modules/hat_model.py (limited to 'modules') diff --git a/modules/hat_model.py b/modules/hat_model.py new file mode 100644 index 00000000..553e1941 --- /dev/null +++ b/modules/hat_model.py @@ -0,0 +1,42 @@ +import os +import sys + +from modules import modelloader, devices +from modules.shared import opts +from modules.upscaler import Upscaler, UpscalerData +from modules.upscaler_utils import upscale_with_model + + +class UpscalerHAT(Upscaler): + def __init__(self, dirname): + self.name = "HAT" + self.scalers = [] + self.user_path = dirname + super().__init__() + for file in self.find_models(ext_filter=[".pt", ".pth"]): + name = modelloader.friendly_name(file) + scale = 4 # TODO: scale might not be 4, but we can't know without loading the model + scaler_data = UpscalerData(name, file, upscaler=self, scale=scale) + self.scalers.append(scaler_data) + + def do_upscale(self, img, selected_model): + try: + model = self.load_model(selected_model) + except Exception as e: + print(f"Unable to load HAT model {selected_model}: {e}", file=sys.stderr) + return img + model.to(devices.device_esrgan) # TODO: should probably be device_hat + return upscale_with_model( + model, + img, + tile_size=opts.ESRGAN_tile, # TODO: should probably be HAT_tile + tile_overlap=opts.ESRGAN_tile_overlap, # TODO: should probably be HAT_tile_overlap + ) + + def load_model(self, path: str): + if not os.path.isfile(path): + raise FileNotFoundError(f"Model file {path} not found") + return modelloader.load_spandrel_model( + path, + device=devices.device_esrgan, # TODO: should probably be device_hat + ) -- cgit v1.2.1 From 4ad0c0c0a805da4bac03cff86ea17c25a1291546 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sat, 30 Dec 2023 16:37:03 +0200 Subject: Verify architecture for loaded Spandrel models --- modules/codeformer_model.py | 1 + modules/esrgan_model.py | 1 + modules/gfpgan_model.py | 1 + modules/hat_model.py | 1 + modules/modelloader.py | 13 ++++++++++++- modules/realesrgan_model.py | 7 ++++--- 6 files changed, 20 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/codeformer_model.py b/modules/codeformer_model.py index ceda4bab..44b84618 100644 --- a/modules/codeformer_model.py +++ b/modules/codeformer_model.py @@ -37,6 +37,7 @@ class FaceRestorerCodeFormer(face_restoration_utils.CommonFaceRestoration): return modelloader.load_spandrel_model( model_path, device=devices.device_codeformer, + expected_architecture='CodeFormer', ).model raise ValueError("No codeformer model found") diff --git a/modules/esrgan_model.py b/modules/esrgan_model.py index a7c7c9e3..70041ab0 100644 --- a/modules/esrgan_model.py +++ b/modules/esrgan_model.py @@ -49,6 +49,7 @@ class UpscalerESRGAN(Upscaler): return modelloader.load_spandrel_model( filename, device=('cpu' if devices.device_esrgan.type == 'mps' else None), + expected_architecture='ESRGAN', ) diff --git a/modules/gfpgan_model.py b/modules/gfpgan_model.py index a356b56f..48f8ad5e 100644 --- a/modules/gfpgan_model.py +++ b/modules/gfpgan_model.py @@ -37,6 +37,7 @@ class FaceRestorerGFPGAN(face_restoration_utils.CommonFaceRestoration): net = modelloader.load_spandrel_model( model_path, device=self.get_device(), + expected_architecture='GFPGAN', ).model net.different_w = True # see https://github.com/chaiNNer-org/spandrel/pull/81 return net diff --git a/modules/hat_model.py b/modules/hat_model.py index 553e1941..7f2abb41 100644 --- a/modules/hat_model.py +++ b/modules/hat_model.py @@ -39,4 +39,5 @@ class UpscalerHAT(Upscaler): return modelloader.load_spandrel_model( path, device=devices.device_esrgan, # TODO: should probably be device_hat + expected_architecture='HAT', ) diff --git a/modules/modelloader.py b/modules/modelloader.py index 30116932..f4182559 100644 --- a/modules/modelloader.py +++ b/modules/modelloader.py @@ -6,6 +6,8 @@ import shutil import importlib from urllib.parse import urlparse +import torch + from modules import shared from modules.upscaler import Upscaler, UpscalerLanczos, UpscalerNearest, UpscalerNone from modules.paths import script_path, models_path @@ -183,9 +185,18 @@ def load_upscalers(): ) -def load_spandrel_model(path, *, device, half: bool = False, dtype=None): +def load_spandrel_model( + path: str, + *, + device: str | torch.device | None, + half: bool = False, + dtype: str | None = None, + expected_architecture: str | None = None, +): import spandrel model = spandrel.ModelLoader(device=device).load_from_file(path) + if expected_architecture and model.architecture != expected_architecture: + raise TypeError(f"Model {path} is not a {expected_architecture} model") if half: model = model.model.half() if dtype: diff --git a/modules/realesrgan_model.py b/modules/realesrgan_model.py index 332d8f4b..2a2be5ad 100644 --- a/modules/realesrgan_model.py +++ b/modules/realesrgan_model.py @@ -1,9 +1,9 @@ import os -from modules.upscaler_utils import upscale_with_model -from modules.upscaler import Upscaler, UpscalerData -from modules.shared import cmd_opts, opts from modules import modelloader, errors +from modules.shared import cmd_opts, opts +from modules.upscaler import Upscaler, UpscalerData +from modules.upscaler_utils import upscale_with_model class UpscalerRealESRGAN(Upscaler): @@ -40,6 +40,7 @@ class UpscalerRealESRGAN(Upscaler): info.local_data_path, device=self.device, half=(not cmd_opts.no_half and not cmd_opts.upcast_sampling), + expected_architecture="RealESRGAN", ) return upscale_with_model( mod, -- cgit v1.2.1 From 05230c02606080527b65ace9eacb6fb835239877 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 30 Dec 2023 18:02:51 +0300 Subject: fix img2img api that i broke when implementing infotext support --- modules/api/api.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 2918f785..2e18c6b9 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -507,6 +507,7 @@ class Api: args.pop('script_name', None) args.pop('script_args', None) # will refeed them to the pipeline directly after initializing them args.pop('alwayson_scripts', None) + args.pop('infotext', None) script_args = self.init_script_args(img2imgreq, self.default_script_arg_img2img, selectable_scripts, selectable_script_idx, script_runner, input_script_args=infotext_script_args) -- cgit v1.2.1 From f476649c02cf3547d891fa08c50a92f92c4d73bd Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sat, 30 Dec 2023 17:41:19 +0200 Subject: Correct arg type for restore_face --- modules/face_restoration_utils.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/face_restoration_utils.py b/modules/face_restoration_utils.py index c65c85ef..85cb3057 100644 --- a/modules/face_restoration_utils.py +++ b/modules/face_restoration_utils.py @@ -36,7 +36,7 @@ def create_face_helper(device) -> FaceRestoreHelper: def restore_with_face_helper( np_image: np.ndarray, face_helper: FaceRestoreHelper, - restore_face: Callable[[np.ndarray], np.ndarray], + restore_face: Callable[[torch.Tensor], torch.Tensor], ) -> np.ndarray: """ Find faces in the image using face_helper, restore them using restore_face, and paste them back into the image. @@ -126,7 +126,7 @@ class CommonFaceRestoration(face_restoration.FaceRestoration): def restore_with_helper( self, np_image: np.ndarray, - restore_face: Callable[[np.ndarray], np.ndarray], + restore_face: Callable[[torch.Tensor], torch.Tensor], ) -> np.ndarray: try: if self.net is None: -- cgit v1.2.1 From 91560e98c47f8271d444556ef4ae6505dece9aba Mon Sep 17 00:00:00 2001 From: lanyeeee <1210347077@qq.com> Date: Sat, 30 Dec 2023 23:42:10 +0800 Subject: fix format issue --- modules/api/api.py | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 2f718ec2..d202cb8d 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -418,7 +418,6 @@ class Api: task_id = txt2imgreq.force_task_id or create_task_id("txt2img") script_runner = scripts.scripts_txt2img - with self.txt2img_script_arg_init_lock: if not script_runner.scripts: script_runner.initialize_scripts(False) @@ -489,14 +488,13 @@ class Api: mask = decode_base64_to_image(mask) script_runner = scripts.scripts_img2img - with self.img2img_script_arg_init_lock: if not script_runner.scripts: script_runner.initialize_scripts(True) ui.create_ui() - infotext_script_args = {} - self.apply_infotext(img2imgreq, "img2img", script_runner=script_runner, mentioned_script_args=infotext_script_args) + infotext_script_args = {} + self.apply_infotext(img2imgreq, "img2img", script_runner=script_runner, mentioned_script_args=infotext_script_args) if not self.default_script_arg_img2img: self.default_script_arg_img2img = self.init_default_script_args(script_runner) -- cgit v1.2.1 From c9174253fb603e6b2552e4c2721fd767b6ede87d Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sat, 30 Dec 2023 17:45:26 +0200 Subject: Drop dependency on basicsr --- modules/face_restoration_utils.py | 35 ++++++++++++++++++++++++++--------- 1 file changed, 26 insertions(+), 9 deletions(-) (limited to 'modules') diff --git a/modules/face_restoration_utils.py b/modules/face_restoration_utils.py index 85cb3057..1cbac236 100644 --- a/modules/face_restoration_utils.py +++ b/modules/face_restoration_utils.py @@ -17,6 +17,28 @@ if TYPE_CHECKING: logger = logging.getLogger(__name__) +def bgr_image_to_rgb_tensor(img: np.ndarray) -> torch.Tensor: + """Convert a BGR NumPy image in [0..1] range to a PyTorch RGB float32 tensor.""" + assert img.shape[2] == 3, "image must be RGB" + if img.dtype == "float64": + img = img.astype("float32") + img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) + return torch.from_numpy(img.transpose(2, 0, 1)).float() + + +def rgb_tensor_to_bgr_image(tensor: torch.Tensor, *, min_max=(0.0, 1.0)) -> np.ndarray: + """ + Convert a PyTorch RGB tensor in range `min_max` to a BGR NumPy image in [0..1] range. + """ + tensor = tensor.squeeze(0).float().detach().cpu().clamp_(*min_max) + tensor = (tensor - min_max[0]) / (min_max[1] - min_max[0]) + assert tensor.dim() == 3, "tensor must be RGB" + img_np = tensor.numpy().transpose(1, 2, 0) + if img_np.shape[2] == 1: # gray image, no RGB/BGR required + return np.squeeze(img_np, axis=2) + return cv2.cvtColor(img_np, cv2.COLOR_BGR2RGB) + + def create_face_helper(device) -> FaceRestoreHelper: from facexlib.detection import retinaface from facexlib.utils.face_restoration_helper import FaceRestoreHelper @@ -43,7 +65,6 @@ def restore_with_face_helper( `restore_face` should take a cropped face image and return a restored face image. """ - from basicsr.utils import img2tensor, tensor2img from torchvision.transforms.functional import normalize np_image = np_image[:, :, ::-1] original_resolution = np_image.shape[0:2] @@ -56,23 +77,19 @@ def restore_with_face_helper( face_helper.align_warp_face() logger.debug("Found %d faces, restoring", len(face_helper.cropped_faces)) for cropped_face in face_helper.cropped_faces: - cropped_face_t = img2tensor(cropped_face / 255., bgr2rgb=True, float32=True) + cropped_face_t = bgr_image_to_rgb_tensor(cropped_face / 255.0) normalize(cropped_face_t, (0.5, 0.5, 0.5), (0.5, 0.5, 0.5), inplace=True) cropped_face_t = cropped_face_t.unsqueeze(0).to(devices.device_codeformer) try: with torch.no_grad(): - restored_face = tensor2img( - restore_face(cropped_face_t), - rgb2bgr=True, - min_max=(-1, 1), - ) + cropped_face_t = restore_face(cropped_face_t) devices.torch_gc() except Exception: errors.report('Failed face-restoration inference', exc_info=True) - restored_face = tensor2img(cropped_face_t, rgb2bgr=True, min_max=(-1, 1)) - restored_face = restored_face.astype('uint8') + restored_face = rgb_tensor_to_bgr_image(cropped_face_t, min_max=(-1, 1)) + restored_face = (restored_face * 255.0).astype('uint8') face_helper.add_restored_face(restored_face) logger.debug("Merging restored faces into image") -- cgit v1.2.1 From f651405427dfc6d4ef96ecba7f9c2ceb580263fd Mon Sep 17 00:00:00 2001 From: lanyeeee <1210347077@qq.com> Date: Sun, 31 Dec 2023 01:09:13 +0800 Subject: remove locks, move init code to __init__ --- modules/api/api.py | 37 +++++++++++++++++++------------------ 1 file changed, 19 insertions(+), 18 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index d202cb8d..fc3921c2 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -251,8 +251,21 @@ class Api: self.default_script_arg_txt2img = [] self.default_script_arg_img2img = [] - self.txt2img_script_arg_init_lock = Lock() - self.img2img_script_arg_init_lock = Lock() + txt2img_script_runner = scripts.scripts_txt2img + img2img_script_runner = scripts.scripts_img2img + + if not txt2img_script_runner.scripts or not img2img_script_runner.scripts: + ui.create_ui() + + if not txt2img_script_runner.scripts: + txt2img_script_runner.initialize_scripts(False) + if not self.default_script_arg_txt2img: + self.default_script_arg_txt2img = self.init_default_script_args(txt2img_script_runner) + + if not img2img_script_runner.scripts: + img2img_script_runner.initialize_scripts(True) + if not self.default_script_arg_img2img: + self.default_script_arg_img2img = self.init_default_script_args(img2img_script_runner) @@ -418,16 +431,10 @@ class Api: task_id = txt2imgreq.force_task_id or create_task_id("txt2img") script_runner = scripts.scripts_txt2img - with self.txt2img_script_arg_init_lock: - if not script_runner.scripts: - script_runner.initialize_scripts(False) - ui.create_ui() - infotext_script_args = {} - self.apply_infotext(txt2imgreq, "txt2img", script_runner=script_runner, mentioned_script_args=infotext_script_args) + infotext_script_args = {} + self.apply_infotext(txt2imgreq, "txt2img", script_runner=script_runner, mentioned_script_args=infotext_script_args) - if not self.default_script_arg_txt2img: - self.default_script_arg_txt2img = self.init_default_script_args(script_runner) selectable_scripts, selectable_script_idx = self.get_selectable_script(txt2imgreq.script_name, script_runner) populate = txt2imgreq.copy(update={ # Override __init__ params @@ -488,16 +495,10 @@ class Api: mask = decode_base64_to_image(mask) script_runner = scripts.scripts_img2img - with self.img2img_script_arg_init_lock: - if not script_runner.scripts: - script_runner.initialize_scripts(True) - ui.create_ui() - infotext_script_args = {} - self.apply_infotext(img2imgreq, "img2img", script_runner=script_runner, mentioned_script_args=infotext_script_args) + infotext_script_args = {} + self.apply_infotext(img2imgreq, "img2img", script_runner=script_runner, mentioned_script_args=infotext_script_args) - if not self.default_script_arg_img2img: - self.default_script_arg_img2img = self.init_default_script_args(script_runner) selectable_scripts, selectable_script_idx = self.get_selectable_script(img2imgreq.script_name, script_runner) populate = img2imgreq.copy(update={ # Override __init__ params -- cgit v1.2.1 From 1465dab71564bb30091479ceabae6c69e3426bc6 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sat, 30 Dec 2023 19:44:05 +0200 Subject: Make Tensorboard a late import (it was implicitly installed by basicsr) --- modules/textual_inversion/textual_inversion.py | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/textual_inversion/textual_inversion.py b/modules/textual_inversion/textual_inversion.py index 04dda585..c6bcab15 100644 --- a/modules/textual_inversion/textual_inversion.py +++ b/modules/textual_inversion/textual_inversion.py @@ -11,7 +11,6 @@ import safetensors.torch import numpy as np from PIL import Image, PngImagePlugin -from torch.utils.tensorboard import SummaryWriter from modules import shared, devices, sd_hijack, sd_models, images, sd_samplers, sd_hijack_checkpoint, errors, hashes import modules.textual_inversion.dataset @@ -344,6 +343,7 @@ def write_loss(log_directory, filename, step, epoch_len, values): }) def tensorboard_setup(log_directory): + from torch.utils.tensorboard import SummaryWriter os.makedirs(os.path.join(log_directory, "tensorboard"), exist_ok=True) return SummaryWriter( log_dir=os.path.join(log_directory, "tensorboard"), @@ -448,8 +448,12 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st shared.state.textinfo = f"Preparing dataset from {html.escape(data_root)}..." old_parallel_processing_allowed = shared.parallel_processing_allowed + tensorboard_writer = None if shared.opts.training_enable_tensorboard: - tensorboard_writer = tensorboard_setup(log_directory) + try: + tensorboard_writer = tensorboard_setup(log_directory) + except ImportError: + errors.report("Error initializing tensorboard", exc_info=True) pin_memory = shared.opts.pin_memory @@ -622,7 +626,7 @@ def train_embedding(id_task, embedding_name, learn_rate, batch_size, gradient_st last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False) last_saved_image += f", prompt: {preview_text}" - if shared.opts.training_enable_tensorboard and shared.opts.training_tensorboard_save_images: + if tensorboard_writer and shared.opts.training_tensorboard_save_images: tensorboard_add_image(tensorboard_writer, f"Validation at epoch {epoch_num}", image, embedding.step) if save_image_with_stored_embedding and os.path.exists(last_saved_file) and embedding_yet_to_be_embedded: -- cgit v1.2.1 From 5fbb13e0da8eb2e26bd2c45ec8ffbb2de669ef47 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sat, 30 Dec 2023 20:46:44 +0200 Subject: Remove `cleanup_models` code --- modules/initialize.py | 3 --- modules/modelloader.py | 50 -------------------------------------------------- 2 files changed, 53 deletions(-) (limited to 'modules') diff --git a/modules/initialize.py b/modules/initialize.py index ac95fc6f..4a3cd98c 100644 --- a/modules/initialize.py +++ b/modules/initialize.py @@ -54,9 +54,6 @@ def initialize(): initialize_util.configure_sigint_handler() initialize_util.configure_opts_onchange() - from modules import modelloader - modelloader.cleanup_models() - from modules import sd_models sd_models.setup_model() startup_timer.record("setup SD model") diff --git a/modules/modelloader.py b/modules/modelloader.py index f4182559..5f7aec3e 100644 --- a/modules/modelloader.py +++ b/modules/modelloader.py @@ -2,7 +2,6 @@ from __future__ import annotations import logging import os -import shutil import importlib from urllib.parse import urlparse @@ -10,7 +9,6 @@ import torch from modules import shared from modules.upscaler import Upscaler, UpscalerLanczos, UpscalerNearest, UpscalerNone -from modules.paths import script_path, models_path logger = logging.getLogger(__name__) @@ -96,54 +94,6 @@ def friendly_name(file: str): return model_name -def cleanup_models(): - # This code could probably be more efficient if we used a tuple list or something to store the src/destinations - # and then enumerate that, but this works for now. In the future, it'd be nice to just have every "model" scaler - # somehow auto-register and just do these things... - root_path = script_path - src_path = models_path - dest_path = os.path.join(models_path, "Stable-diffusion") - move_files(src_path, dest_path, ".ckpt") - move_files(src_path, dest_path, ".safetensors") - src_path = os.path.join(root_path, "ESRGAN") - dest_path = os.path.join(models_path, "ESRGAN") - move_files(src_path, dest_path) - src_path = os.path.join(models_path, "BSRGAN") - dest_path = os.path.join(models_path, "ESRGAN") - move_files(src_path, dest_path, ".pth") - src_path = os.path.join(root_path, "gfpgan") - dest_path = os.path.join(models_path, "GFPGAN") - move_files(src_path, dest_path) - src_path = os.path.join(root_path, "SwinIR") - dest_path = os.path.join(models_path, "SwinIR") - move_files(src_path, dest_path) - src_path = os.path.join(root_path, "repositories/latent-diffusion/experiments/pretrained_models/") - dest_path = os.path.join(models_path, "LDSR") - move_files(src_path, dest_path) - - -def move_files(src_path: str, dest_path: str, ext_filter: str = None): - try: - os.makedirs(dest_path, exist_ok=True) - if os.path.exists(src_path): - for file in os.listdir(src_path): - fullpath = os.path.join(src_path, file) - if os.path.isfile(fullpath): - if ext_filter is not None: - if ext_filter not in file: - continue - print(f"Moving {file} from {src_path} to {dest_path}.") - try: - shutil.move(fullpath, dest_path) - except Exception: - pass - if len(os.listdir(src_path)) == 0: - print(f"Removing empty folder: {src_path}") - shutil.rmtree(src_path, True) - except Exception: - pass - - def load_upscalers(): # We can only do this 'magic' method to dynamically load upscalers if they are referenced, # so we'll try to import any _model.py files before looking in __subclasses__ -- cgit v1.2.1 From af050dcaa75ef40b6b1c3da3361f32fe52786aeb Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sat, 30 Dec 2023 21:05:59 +0200 Subject: Soften Spandrel model-architecture check to just a warning --- modules/modelloader.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/modelloader.py b/modules/modelloader.py index f4182559..6b7d697f 100644 --- a/modules/modelloader.py +++ b/modules/modelloader.py @@ -196,7 +196,9 @@ def load_spandrel_model( import spandrel model = spandrel.ModelLoader(device=device).load_from_file(path) if expected_architecture and model.architecture != expected_architecture: - raise TypeError(f"Model {path} is not a {expected_architecture} model") + logger.warning( + f"Model {path!r} is not a {expected_architecture!r} model (got {model.architecture!r})", + ) if half: model = model.model.half() if dtype: -- cgit v1.2.1 From 393a5b82ba6df06d85f2bf7bbbe0456d3d06115f Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sat, 30 Dec 2023 21:12:32 +0200 Subject: Correct RealESRGAN expected architecture type to ESRGAN --- modules/realesrgan_model.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/realesrgan_model.py b/modules/realesrgan_model.py index 2a2be5ad..65f2e880 100644 --- a/modules/realesrgan_model.py +++ b/modules/realesrgan_model.py @@ -40,7 +40,7 @@ class UpscalerRealESRGAN(Upscaler): info.local_data_path, device=self.device, half=(not cmd_opts.no_half and not cmd_opts.upcast_sampling), - expected_architecture="RealESRGAN", + expected_architecture="ESRGAN", # "RealESRGAN" isn't a specific thing for Spandrel ) return upscale_with_model( mod, -- cgit v1.2.1 From 8100e901ab0c5b04d289eebb722c8a653b8beef1 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sat, 30 Dec 2023 22:41:53 +0300 Subject: fix error with RealESRGAN model failing to upscale fp32 image --- modules/upscaler_utils.py | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index 8bdda51c..39f78a0b 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -16,9 +16,13 @@ def upscale_without_tiling(model, img: Image.Image): img = img[:, :, ::-1] img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255 img = torch.from_numpy(img).float() - img = img.unsqueeze(0).to(devices.device_esrgan) + + model_weight = next(iter(model.parameters())) + img = img.unsqueeze(0).to(device=model_weight.device, dtype=model_weight.dtype) + with torch.no_grad(): output = model(img) + output = output.squeeze().float().cpu().clamp_(0, 1).numpy() output = 255. * np.moveaxis(output, 0, 2) output = output.astype(np.uint8) -- cgit v1.2.1 From bc5ae74c7d8949bab37e260b16e76889b9968099 Mon Sep 17 00:00:00 2001 From: Learwin <6223515+Learwin@users.noreply.github.com> Date: Sat, 30 Dec 2023 21:52:27 +0100 Subject: Added negative prompts to extra networks lora --- modules/ui_extra_networks.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index fe5d3ba3..b8c02241 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -223,7 +223,10 @@ class ExtraNetworksPage: onclick = item.get("onclick", None) if onclick is None: - onclick = '"' + html.escape(f"""return cardClicked({quote_js(tabname)}, {item["prompt"]}, {"true" if self.allow_negative_prompt else "false"})""") + '"' + if "negative_prompt" in item: + onclick = '"' + html.escape(f"""return cardClicked({quote_js(tabname)}, {item["prompt"]}, {item["negative_prompt"]}, {"true" if self.allow_negative_prompt else "false"})""") + '"' + else: + onclick = '"' + html.escape(f"""return cardClicked({quote_js(tabname)}, {item["prompt"]}, {'""'}, {"true" if self.allow_negative_prompt else "false"})""") + '"' height = f"height: {shared.opts.extra_networks_card_height}px;" if shared.opts.extra_networks_card_height else '' width = f"width: {shared.opts.extra_networks_card_width}px;" if shared.opts.extra_networks_card_width else '' -- cgit v1.2.1 From a2f23f9d22dde87bf2529dcb2854a6a5d3d44278 Mon Sep 17 00:00:00 2001 From: Learwin <6223515+Learwin@users.noreply.github.com> Date: Sat, 30 Dec 2023 22:16:51 +0100 Subject: Code Style fixes --- modules/upscaler_utils.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index 39f78a0b..1d610dbf 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -6,7 +6,7 @@ import torch import tqdm from PIL import Image -from modules import devices, images +from modules import images logger = logging.getLogger(__name__) -- cgit v1.2.1 From 3be90740316f8fbb950b31d440458a5e8ed4beb3 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 31 Dec 2023 00:43:41 +0300 Subject: fix for the previous fix. --- modules/upscaler_utils.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index 39f78a0b..dde5d7ad 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -17,7 +17,7 @@ def upscale_without_tiling(model, img: Image.Image): img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255 img = torch.from_numpy(img).float() - model_weight = next(iter(model.parameters())) + model_weight = next(iter(model.model.parameters())) img = img.unsqueeze(0).to(device=model_weight.device, dtype=model_weight.dtype) with torch.no_grad(): -- cgit v1.2.1 From c0ca6348e8489651df861a101142805c213c66a0 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sun, 31 Dec 2023 00:04:47 +0200 Subject: load_spandrel_model: always return a model descriptor --- modules/modelloader.py | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) (limited to 'modules') diff --git a/modules/modelloader.py b/modules/modelloader.py index 0b89d682..8bcee08c 100644 --- a/modules/modelloader.py +++ b/modules/modelloader.py @@ -1,8 +1,9 @@ from __future__ import annotations +import importlib import logging import os -import importlib +from typing import TYPE_CHECKING from urllib.parse import urlparse import torch @@ -10,6 +11,8 @@ import torch from modules import shared from modules.upscaler import Upscaler, UpscalerLanczos, UpscalerNearest, UpscalerNone +if TYPE_CHECKING: + import spandrel logger = logging.getLogger(__name__) @@ -142,17 +145,17 @@ def load_spandrel_model( half: bool = False, dtype: str | None = None, expected_architecture: str | None = None, -): +) -> spandrel.ModelDescriptor: import spandrel - model = spandrel.ModelLoader(device=device).load_from_file(path) - if expected_architecture and model.architecture != expected_architecture: + model_descriptor = spandrel.ModelLoader(device=device).load_from_file(path) + if expected_architecture and model_descriptor.architecture != expected_architecture: logger.warning( - f"Model {path!r} is not a {expected_architecture!r} model (got {model.architecture!r})", + f"Model {path!r} is not a {expected_architecture!r} model (got {model_descriptor.architecture!r})", ) if half: - model = model.model.half() + model_descriptor.model.half() if dtype: - model = model.model.to(dtype=dtype) - model.eval() - logger.debug("Loaded %s from %s (device=%s, half=%s, dtype=%s)", model, path, device, half, dtype) - return model + model_descriptor.model.to(dtype=dtype) + model_descriptor.model.eval() + logger.debug("Loaded %s from %s (device=%s, half=%s, dtype=%s)", model_descriptor, path, device, half, dtype) + return model_descriptor -- cgit v1.2.1 From 777af661a21821994993df3ef566b01df2bb61a0 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sun, 31 Dec 2023 00:09:51 +0200 Subject: Be more clear about Spandrel model nomenclature --- modules/gfpgan_model.py | 10 ++++++---- modules/modelloader.py | 2 +- modules/realesrgan_model.py | 4 ++-- modules/upscaler_utils.py | 2 +- 4 files changed, 10 insertions(+), 8 deletions(-) (limited to 'modules') diff --git a/modules/gfpgan_model.py b/modules/gfpgan_model.py index 48f8ad5e..445b0409 100644 --- a/modules/gfpgan_model.py +++ b/modules/gfpgan_model.py @@ -3,6 +3,8 @@ from __future__ import annotations import logging import os +import torch + from modules import ( devices, errors, @@ -25,7 +27,7 @@ class FaceRestorerGFPGAN(face_restoration_utils.CommonFaceRestoration): def get_device(self): return devices.device_gfpgan - def load_net(self) -> None: + def load_net(self) -> torch.Module: for model_path in modelloader.load_models( model_path=self.model_path, model_url=model_url, @@ -34,13 +36,13 @@ class FaceRestorerGFPGAN(face_restoration_utils.CommonFaceRestoration): ext_filter=['.pth'], ): if 'GFPGAN' in os.path.basename(model_path): - net = modelloader.load_spandrel_model( + model = modelloader.load_spandrel_model( model_path, device=self.get_device(), expected_architecture='GFPGAN', ).model - net.different_w = True # see https://github.com/chaiNNer-org/spandrel/pull/81 - return net + model.different_w = True # see https://github.com/chaiNNer-org/spandrel/pull/81 + return model raise ValueError("No GFPGAN model found") def restore(self, np_image): diff --git a/modules/modelloader.py b/modules/modelloader.py index 8bcee08c..a7194137 100644 --- a/modules/modelloader.py +++ b/modules/modelloader.py @@ -143,7 +143,7 @@ def load_spandrel_model( *, device: str | torch.device | None, half: bool = False, - dtype: str | None = None, + dtype: str | torch.dtype | None = None, expected_architecture: str | None = None, ) -> spandrel.ModelDescriptor: import spandrel diff --git a/modules/realesrgan_model.py b/modules/realesrgan_model.py index 65f2e880..4d35b695 100644 --- a/modules/realesrgan_model.py +++ b/modules/realesrgan_model.py @@ -36,14 +36,14 @@ class UpscalerRealESRGAN(Upscaler): errors.report(f"Unable to load RealESRGAN model {path}", exc_info=True) return img - mod = modelloader.load_spandrel_model( + model_descriptor = modelloader.load_spandrel_model( info.local_data_path, device=self.device, half=(not cmd_opts.no_half and not cmd_opts.upcast_sampling), expected_architecture="ESRGAN", # "RealESRGAN" isn't a specific thing for Spandrel ) return upscale_with_model( - mod, + model_descriptor, img, tile_size=opts.ESRGAN_tile, tile_overlap=opts.ESRGAN_tile_overlap, diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index dde5d7ad..174c9bc3 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -6,7 +6,7 @@ import torch import tqdm from PIL import Image -from modules import devices, images +from modules import images logger = logging.getLogger(__name__) -- cgit v1.2.1 From 6f86b62a1be7993073ba3a789d522e0b8870605a Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sat, 30 Dec 2023 22:53:49 +0200 Subject: Deduplicate tiled inference code from SwinIR/ScuNET --- modules/upscaler_utils.py | 72 ++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 71 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index 174c9bc3..8e413854 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -6,7 +6,7 @@ import torch import tqdm from PIL import Image -from modules import images +from modules import images, shared logger = logging.getLogger(__name__) @@ -68,3 +68,73 @@ def upscale_with_model( overlap=grid.overlap * scale_factor, ) return images.combine_grid(newgrid) + + +def tiled_upscale_2( + img, + model, + *, + tile_size: int, + tile_overlap: int, + scale: int, + device, + desc="Tiled upscale", +): + # Alternative implementation of `upscale_with_model` originally used by + # SwinIR and ScuNET. It differs from `upscale_with_model` in that tiling and + # weighting is done in PyTorch space, as opposed to `images.Grid` doing it in + # Pillow space without weighting. + b, c, h, w = img.size() + tile_size = min(tile_size, h, w) + + if tile_size <= 0: + logger.debug("Upscaling %s without tiling", img.shape) + return model(img) + + stride = tile_size - tile_overlap + h_idx_list = list(range(0, h - tile_size, stride)) + [h - tile_size] + w_idx_list = list(range(0, w - tile_size, stride)) + [w - tile_size] + result = torch.zeros( + b, + c, + h * scale, + w * scale, + device=device, + ).type_as(img) + weights = torch.zeros_like(result) + logger.debug("Upscaling %s to %s with tiles", img.shape, result.shape) + with tqdm.tqdm(total=len(h_idx_list) * len(w_idx_list), desc=desc) as pbar: + for h_idx in h_idx_list: + if shared.state.interrupted or shared.state.skipped: + break + + for w_idx in w_idx_list: + if shared.state.interrupted or shared.state.skipped: + break + + in_patch = img[ + ..., + h_idx : h_idx + tile_size, + w_idx : w_idx + tile_size, + ] + out_patch = model(in_patch) + + result[ + ..., + h_idx * scale : (h_idx + tile_size) * scale, + w_idx * scale : (w_idx + tile_size) * scale, + ].add_(out_patch) + + out_patch_mask = torch.ones_like(out_patch) + + weights[ + ..., + h_idx * scale : (h_idx + tile_size) * scale, + w_idx * scale : (w_idx + tile_size) * scale, + ].add_(out_patch_mask) + + pbar.update(1) + + output = result.div_(weights) + + return output -- cgit v1.2.1 From 5768afc776a66bb94e77a9c1daebeea58fa731d5 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sun, 31 Dec 2023 00:20:30 +0200 Subject: Add utility to inspect a model's parameters (to get dtype/device) --- modules/devices.py | 3 ++- modules/interrogate.py | 3 ++- modules/sd_models_xl.py | 3 ++- modules/torch_utils.py | 17 +++++++++++++++++ modules/upscaler_utils.py | 5 +++-- modules/xlmr.py | 5 ++++- modules/xlmr_m18.py | 5 ++++- 7 files changed, 34 insertions(+), 7 deletions(-) create mode 100644 modules/torch_utils.py (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index c956207f..bd6bd579 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -4,6 +4,7 @@ from functools import lru_cache import torch from modules import errors, shared +from modules.torch_utils import get_param if sys.platform == "darwin": from modules import mac_specific @@ -131,7 +132,7 @@ patch_module_list = [ def manual_cast_forward(self, *args, **kwargs): - org_dtype = next(self.parameters()).dtype + org_dtype = get_param(self).dtype self.to(dtype) args = [arg.to(dtype) if isinstance(arg, torch.Tensor) else arg for arg in args] kwargs = {k: v.to(dtype) if isinstance(v, torch.Tensor) else v for k, v in kwargs.items()} diff --git a/modules/interrogate.py b/modules/interrogate.py index 3045560d..5be5a10f 100644 --- a/modules/interrogate.py +++ b/modules/interrogate.py @@ -11,6 +11,7 @@ from torchvision import transforms from torchvision.transforms.functional import InterpolationMode from modules import devices, paths, shared, lowvram, modelloader, errors +from modules.torch_utils import get_param blip_image_eval_size = 384 clip_model_name = 'ViT-L/14' @@ -131,7 +132,7 @@ class InterrogateModels: self.clip_model = self.clip_model.to(devices.device_interrogate) - self.dtype = next(self.clip_model.parameters()).dtype + self.dtype = get_param(self.clip_model).dtype def send_clip_to_ram(self): if not shared.opts.interrogate_keep_models_in_memory: diff --git a/modules/sd_models_xl.py b/modules/sd_models_xl.py index 1de31b0d..c3602a7e 100644 --- a/modules/sd_models_xl.py +++ b/modules/sd_models_xl.py @@ -6,6 +6,7 @@ import sgm.models.diffusion import sgm.modules.diffusionmodules.denoiser_scaling import sgm.modules.diffusionmodules.discretizer from modules import devices, shared, prompt_parser +from modules.torch_utils import get_param def get_learned_conditioning(self: sgm.models.diffusion.DiffusionEngine, batch: prompt_parser.SdConditioning | list[str]): @@ -90,7 +91,7 @@ sgm.modules.GeneralConditioner.get_target_prompt_token_count = get_target_prompt def extend_sdxl(model): """this adds a bunch of parameters to make SDXL model look a bit more like SD1.5 to the rest of the codebase.""" - dtype = next(model.model.diffusion_model.parameters()).dtype + dtype = get_param(model.model.diffusion_model).dtype model.model.diffusion_model.dtype = dtype model.model.conditioning_key = 'crossattn' model.cond_stage_key = 'txt' diff --git a/modules/torch_utils.py b/modules/torch_utils.py new file mode 100644 index 00000000..e5b52393 --- /dev/null +++ b/modules/torch_utils.py @@ -0,0 +1,17 @@ +from __future__ import annotations + +import torch.nn + + +def get_param(model) -> torch.nn.Parameter: + """ + Find the first parameter in a model or module. + """ + if hasattr(model, "model") and hasattr(model.model, "parameters"): + # Unpeel a model descriptor to get at the actual Torch module. + model = model.model + + for param in model.parameters(): + return param + + raise ValueError(f"No parameters found in model {model!r}") diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index 8e413854..c60e3beb 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -7,6 +7,7 @@ import tqdm from PIL import Image from modules import images, shared +from modules.torch_utils import get_param logger = logging.getLogger(__name__) @@ -17,8 +18,8 @@ def upscale_without_tiling(model, img: Image.Image): img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255 img = torch.from_numpy(img).float() - model_weight = next(iter(model.model.parameters())) - img = img.unsqueeze(0).to(device=model_weight.device, dtype=model_weight.dtype) + param = get_param(model) + img = img.unsqueeze(0).to(device=param.device, dtype=param.dtype) with torch.no_grad(): output = model(img) diff --git a/modules/xlmr.py b/modules/xlmr.py index a407a3ca..6e000a56 100644 --- a/modules/xlmr.py +++ b/modules/xlmr.py @@ -5,6 +5,9 @@ from transformers.models.xlm_roberta.configuration_xlm_roberta import XLMRoberta from transformers import XLMRobertaModel,XLMRobertaTokenizer from typing import Optional +from modules.torch_utils import get_param + + class BertSeriesConfig(BertConfig): def __init__(self, vocab_size=30522, hidden_size=768, num_hidden_layers=12, num_attention_heads=12, intermediate_size=3072, hidden_act="gelu", hidden_dropout_prob=0.1, attention_probs_dropout_prob=0.1, max_position_embeddings=512, type_vocab_size=2, initializer_range=0.02, layer_norm_eps=1e-12, pad_token_id=0, position_embedding_type="absolute", use_cache=True, classifier_dropout=None,project_dim=512, pooler_fn="average",learn_encoder=False,model_type='bert',**kwargs): @@ -62,7 +65,7 @@ class BertSeriesModelWithTransformation(BertPreTrainedModel): self.post_init() def encode(self,c): - device = next(self.parameters()).device + device = get_param(self).device text = self.tokenizer(c, truncation=True, max_length=77, diff --git a/modules/xlmr_m18.py b/modules/xlmr_m18.py index a727e865..e3e81961 100644 --- a/modules/xlmr_m18.py +++ b/modules/xlmr_m18.py @@ -5,6 +5,9 @@ from transformers.models.xlm_roberta.configuration_xlm_roberta import XLMRoberta from transformers import XLMRobertaModel,XLMRobertaTokenizer from typing import Optional +from modules.torch_utils import get_param + + class BertSeriesConfig(BertConfig): def __init__(self, vocab_size=30522, hidden_size=768, num_hidden_layers=12, num_attention_heads=12, intermediate_size=3072, hidden_act="gelu", hidden_dropout_prob=0.1, attention_probs_dropout_prob=0.1, max_position_embeddings=512, type_vocab_size=2, initializer_range=0.02, layer_norm_eps=1e-12, pad_token_id=0, position_embedding_type="absolute", use_cache=True, classifier_dropout=None,project_dim=512, pooler_fn="average",learn_encoder=False,model_type='bert',**kwargs): @@ -68,7 +71,7 @@ class BertSeriesModelWithTransformation(BertPreTrainedModel): self.post_init() def encode(self,c): - device = next(self.parameters()).device + device = get_param(self).device text = self.tokenizer(c, truncation=True, max_length=77, -- cgit v1.2.1 From b6f74e936e4de3b8d190bffaf3bed67d6d4bd211 Mon Sep 17 00:00:00 2001 From: Learwin <6223515+Learwin@users.noreply.github.com> Date: Sun, 31 Dec 2023 13:36:36 +0100 Subject: Revert change from linting for unrelated file --- modules/upscaler_utils.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index 1d610dbf..39f78a0b 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -6,7 +6,7 @@ import torch import tqdm from PIL import Image -from modules import images +from modules import devices, images logger = logging.getLogger(__name__) -- cgit v1.2.1 From a70dfb64a86b9b6d869deffdb0ffebe980365473 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 31 Dec 2023 22:38:30 +0300 Subject: change import statements for #14478 --- modules/devices.py | 4 ++-- modules/interrogate.py | 5 ++--- modules/sd_models_xl.py | 4 ++-- modules/upscaler_utils.py | 5 ++--- modules/xlmr.py | 4 ++-- modules/xlmr_m18.py | 5 ++--- 6 files changed, 12 insertions(+), 15 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index bd6bd579..ff279ac5 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -4,7 +4,7 @@ from functools import lru_cache import torch from modules import errors, shared -from modules.torch_utils import get_param +from modules import torch_utils if sys.platform == "darwin": from modules import mac_specific @@ -132,7 +132,7 @@ patch_module_list = [ def manual_cast_forward(self, *args, **kwargs): - org_dtype = get_param(self).dtype + org_dtype = torch_utils.get_param(self).dtype self.to(dtype) args = [arg.to(dtype) if isinstance(arg, torch.Tensor) else arg for arg in args] kwargs = {k: v.to(dtype) if isinstance(v, torch.Tensor) else v for k, v in kwargs.items()} diff --git a/modules/interrogate.py b/modules/interrogate.py index 5be5a10f..35a627ca 100644 --- a/modules/interrogate.py +++ b/modules/interrogate.py @@ -10,8 +10,7 @@ import torch.hub from torchvision import transforms from torchvision.transforms.functional import InterpolationMode -from modules import devices, paths, shared, lowvram, modelloader, errors -from modules.torch_utils import get_param +from modules import devices, paths, shared, lowvram, modelloader, errors, torch_utils blip_image_eval_size = 384 clip_model_name = 'ViT-L/14' @@ -132,7 +131,7 @@ class InterrogateModels: self.clip_model = self.clip_model.to(devices.device_interrogate) - self.dtype = get_param(self.clip_model).dtype + self.dtype = torch_utils.get_param(self.clip_model).dtype def send_clip_to_ram(self): if not shared.opts.interrogate_keep_models_in_memory: diff --git a/modules/sd_models_xl.py b/modules/sd_models_xl.py index c3602a7e..0de17af3 100644 --- a/modules/sd_models_xl.py +++ b/modules/sd_models_xl.py @@ -6,7 +6,7 @@ import sgm.models.diffusion import sgm.modules.diffusionmodules.denoiser_scaling import sgm.modules.diffusionmodules.discretizer from modules import devices, shared, prompt_parser -from modules.torch_utils import get_param +from modules import torch_utils def get_learned_conditioning(self: sgm.models.diffusion.DiffusionEngine, batch: prompt_parser.SdConditioning | list[str]): @@ -91,7 +91,7 @@ sgm.modules.GeneralConditioner.get_target_prompt_token_count = get_target_prompt def extend_sdxl(model): """this adds a bunch of parameters to make SDXL model look a bit more like SD1.5 to the rest of the codebase.""" - dtype = get_param(model.model.diffusion_model).dtype + dtype = torch_utils.get_param(model.model.diffusion_model).dtype model.model.diffusion_model.dtype = dtype model.model.conditioning_key = 'crossattn' model.cond_stage_key = 'txt' diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index c60e3beb..f5cb92d5 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -6,8 +6,7 @@ import torch import tqdm from PIL import Image -from modules import images, shared -from modules.torch_utils import get_param +from modules import images, shared, torch_utils logger = logging.getLogger(__name__) @@ -18,7 +17,7 @@ def upscale_without_tiling(model, img: Image.Image): img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255 img = torch.from_numpy(img).float() - param = get_param(model) + param = torch_utils.get_param(model) img = img.unsqueeze(0).to(device=param.device, dtype=param.dtype) with torch.no_grad(): diff --git a/modules/xlmr.py b/modules/xlmr.py index 6e000a56..319771b7 100644 --- a/modules/xlmr.py +++ b/modules/xlmr.py @@ -5,7 +5,7 @@ from transformers.models.xlm_roberta.configuration_xlm_roberta import XLMRoberta from transformers import XLMRobertaModel,XLMRobertaTokenizer from typing import Optional -from modules.torch_utils import get_param +from modules import torch_utils class BertSeriesConfig(BertConfig): @@ -65,7 +65,7 @@ class BertSeriesModelWithTransformation(BertPreTrainedModel): self.post_init() def encode(self,c): - device = get_param(self).device + device = torch_utils.get_param(self).device text = self.tokenizer(c, truncation=True, max_length=77, diff --git a/modules/xlmr_m18.py b/modules/xlmr_m18.py index e3e81961..f6055504 100644 --- a/modules/xlmr_m18.py +++ b/modules/xlmr_m18.py @@ -4,8 +4,7 @@ import torch from transformers.models.xlm_roberta.configuration_xlm_roberta import XLMRobertaConfig from transformers import XLMRobertaModel,XLMRobertaTokenizer from typing import Optional - -from modules.torch_utils import get_param +from modules import torch_utils class BertSeriesConfig(BertConfig): @@ -71,7 +70,7 @@ class BertSeriesModelWithTransformation(BertPreTrainedModel): self.post_init() def encode(self,c): - device = get_param(self).device + device = torch_utils.get_param(self).device text = self.tokenizer(c, truncation=True, max_length=77, -- cgit v1.2.1 From 00901bfbe0095303554f4440b4c12fac262e2e89 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Mon, 1 Jan 2024 15:47:57 +0900 Subject: handle selectable script_index is None --- modules/scripts.py | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/scripts.py b/modules/scripts.py index 3a766911..017aed5a 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -696,6 +696,8 @@ class ScriptRunner: self.setup_ui_for_section(None, self.selectable_scripts) def select_script(script_index): + if script_index is None: + script_index = 0 selected_script = self.selectable_scripts[script_index - 1] if script_index>0 else None return [gr.update(visible=selected_script == s) for s in self.selectable_scripts] @@ -739,7 +741,7 @@ class ScriptRunner: def run(self, p, *args): script_index = args[0] - if script_index == 0: + if script_index == 0 or script_index is None: return None script = self.selectable_scripts[script_index-1] -- cgit v1.2.1 From 5692bf1517c3409ad46262c56e65f256389825b1 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 11:11:14 +0300 Subject: add missing field for DDIM sampler that was breaking img2img --- modules/sd_samplers_timesteps.py | 1 + 1 file changed, 1 insertion(+) (limited to 'modules') diff --git a/modules/sd_samplers_timesteps.py b/modules/sd_samplers_timesteps.py index b17a8f93..f8afa8bd 100644 --- a/modules/sd_samplers_timesteps.py +++ b/modules/sd_samplers_timesteps.py @@ -80,6 +80,7 @@ class CompVisSampler(sd_samplers_common.Sampler): self.eta_default = 0.0 self.model_wrap_cfg = CFGDenoiserTimesteps(self) + self.model_wrap = self.model_wrap_cfg.inner_model def get_timesteps(self, p, steps): discard_next_to_last_sigma = self.config is not None and self.config.options.get('discard_next_to_last_sigma', False) -- cgit v1.2.1 From 003b91f08361c99ecdd97257624d81a2046d3823 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 13:45:01 +0300 Subject: rename generation_parameters_copypaste module to infotext --- modules/generation_parameters_copypaste.py | 497 ----------------------------- modules/infotext.py | 497 +++++++++++++++++++++++++++++ 2 files changed, 497 insertions(+), 497 deletions(-) delete mode 100644 modules/generation_parameters_copypaste.py create mode 100644 modules/infotext.py (limited to 'modules') diff --git a/modules/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py deleted file mode 100644 index 86a36c32..00000000 --- a/modules/generation_parameters_copypaste.py +++ /dev/null @@ -1,497 +0,0 @@ -from __future__ import annotations -import base64 -import io -import json -import os -import re - -import gradio as gr -from modules.paths import data_path -from modules import shared, ui_tempdir, script_callbacks, processing -from PIL import Image - -re_param_code = r'\s*(\w[\w \-/]+):\s*("(?:\\.|[^\\"])+"|[^,]*)(?:,|$)' -re_param = re.compile(re_param_code) -re_imagesize = re.compile(r"^(\d+)x(\d+)$") -re_hypernet_hash = re.compile("\(([0-9a-f]+)\)$") -type_of_gr_update = type(gr.update()) - - -class ParamBinding: - def __init__(self, paste_button, tabname, source_text_component=None, source_image_component=None, source_tabname=None, override_settings_component=None, paste_field_names=None): - self.paste_button = paste_button - self.tabname = tabname - self.source_text_component = source_text_component - self.source_image_component = source_image_component - self.source_tabname = source_tabname - self.override_settings_component = override_settings_component - self.paste_field_names = paste_field_names or [] - - -class PasteField(tuple): - def __new__(cls, component, target, *, api=None): - return super().__new__(cls, (component, target)) - - def __init__(self, component, target, *, api=None): - super().__init__() - - self.api = api - self.component = component - self.label = target if isinstance(target, str) else None - self.function = target if callable(target) else None - - -paste_fields: dict[str, dict] = {} -registered_param_bindings: list[ParamBinding] = [] - - -def reset(): - paste_fields.clear() - registered_param_bindings.clear() - - -def quote(text): - if ',' not in str(text) and '\n' not in str(text) and ':' not in str(text): - return text - - return json.dumps(text, ensure_ascii=False) - - -def unquote(text): - if len(text) == 0 or text[0] != '"' or text[-1] != '"': - return text - - try: - return json.loads(text) - except Exception: - return text - - -def image_from_url_text(filedata): - if filedata is None: - return None - - if type(filedata) == list and filedata and type(filedata[0]) == dict and filedata[0].get("is_file", False): - filedata = filedata[0] - - if type(filedata) == dict and filedata.get("is_file", False): - filename = filedata["name"] - is_in_right_dir = ui_tempdir.check_tmp_file(shared.demo, filename) - assert is_in_right_dir, 'trying to open image file outside of allowed directories' - - filename = filename.rsplit('?', 1)[0] - return Image.open(filename) - - if type(filedata) == list: - if len(filedata) == 0: - return None - - filedata = filedata[0] - - if filedata.startswith("data:image/png;base64,"): - filedata = filedata[len("data:image/png;base64,"):] - - filedata = base64.decodebytes(filedata.encode('utf-8')) - image = Image.open(io.BytesIO(filedata)) - return image - - -def add_paste_fields(tabname, init_img, fields, override_settings_component=None): - - if fields: - for i in range(len(fields)): - if not isinstance(fields[i], PasteField): - fields[i] = PasteField(*fields[i]) - - paste_fields[tabname] = {"init_img": init_img, "fields": fields, "override_settings_component": override_settings_component} - - # backwards compatibility for existing extensions - import modules.ui - if tabname == 'txt2img': - modules.ui.txt2img_paste_fields = fields - elif tabname == 'img2img': - modules.ui.img2img_paste_fields = fields - - -def create_buttons(tabs_list): - buttons = {} - for tab in tabs_list: - buttons[tab] = gr.Button(f"Send to {tab}", elem_id=f"{tab}_tab") - return buttons - - -def bind_buttons(buttons, send_image, send_generate_info): - """old function for backwards compatibility; do not use this, use register_paste_params_button""" - for tabname, button in buttons.items(): - source_text_component = send_generate_info if isinstance(send_generate_info, gr.components.Component) else None - source_tabname = send_generate_info if isinstance(send_generate_info, str) else None - - register_paste_params_button(ParamBinding(paste_button=button, tabname=tabname, source_text_component=source_text_component, source_image_component=send_image, source_tabname=source_tabname)) - - -def register_paste_params_button(binding: ParamBinding): - registered_param_bindings.append(binding) - - -def connect_paste_params_buttons(): - for binding in registered_param_bindings: - destination_image_component = paste_fields[binding.tabname]["init_img"] - fields = paste_fields[binding.tabname]["fields"] - override_settings_component = binding.override_settings_component or paste_fields[binding.tabname]["override_settings_component"] - - destination_width_component = next(iter([field for field, name in fields if name == "Size-1"] if fields else []), None) - destination_height_component = next(iter([field for field, name in fields if name == "Size-2"] if fields else []), None) - - if binding.source_image_component and destination_image_component: - if isinstance(binding.source_image_component, gr.Gallery): - func = send_image_and_dimensions if destination_width_component else image_from_url_text - jsfunc = "extract_image_from_gallery" - else: - func = send_image_and_dimensions if destination_width_component else lambda x: x - jsfunc = None - - binding.paste_button.click( - fn=func, - _js=jsfunc, - inputs=[binding.source_image_component], - outputs=[destination_image_component, destination_width_component, destination_height_component] if destination_width_component else [destination_image_component], - show_progress=False, - ) - - if binding.source_text_component is not None and fields is not None: - connect_paste(binding.paste_button, fields, binding.source_text_component, override_settings_component, binding.tabname) - - if binding.source_tabname is not None and fields is not None: - paste_field_names = ['Prompt', 'Negative prompt', 'Steps', 'Face restoration'] + (["Seed"] if shared.opts.send_seed else []) + binding.paste_field_names - binding.paste_button.click( - fn=lambda *x: x, - inputs=[field for field, name in paste_fields[binding.source_tabname]["fields"] if name in paste_field_names], - outputs=[field for field, name in fields if name in paste_field_names], - show_progress=False, - ) - - binding.paste_button.click( - fn=None, - _js=f"switch_to_{binding.tabname}", - inputs=None, - outputs=None, - show_progress=False, - ) - - -def send_image_and_dimensions(x): - if isinstance(x, Image.Image): - img = x - else: - img = image_from_url_text(x) - - if shared.opts.send_size and isinstance(img, Image.Image): - w = img.width - h = img.height - else: - w = gr.update() - h = gr.update() - - return img, w, h - - -def restore_old_hires_fix_params(res): - """for infotexts that specify old First pass size parameter, convert it into - width, height, and hr scale""" - - firstpass_width = res.get('First pass size-1', None) - firstpass_height = res.get('First pass size-2', None) - - if shared.opts.use_old_hires_fix_width_height: - hires_width = int(res.get("Hires resize-1", 0)) - hires_height = int(res.get("Hires resize-2", 0)) - - if hires_width and hires_height: - res['Size-1'] = hires_width - res['Size-2'] = hires_height - return - - if firstpass_width is None or firstpass_height is None: - return - - firstpass_width, firstpass_height = int(firstpass_width), int(firstpass_height) - width = int(res.get("Size-1", 512)) - height = int(res.get("Size-2", 512)) - - if firstpass_width == 0 or firstpass_height == 0: - firstpass_width, firstpass_height = processing.old_hires_fix_first_pass_dimensions(width, height) - - res['Size-1'] = firstpass_width - res['Size-2'] = firstpass_height - res['Hires resize-1'] = width - res['Hires resize-2'] = height - - -def parse_generation_parameters(x: str): - """parses generation parameters string, the one you see in text field under the picture in UI: -``` -girl with an artist's beret, determined, blue eyes, desert scene, computer monitors, heavy makeup, by Alphonse Mucha and Charlie Bowater, ((eyeshadow)), (coquettish), detailed, intricate -Negative prompt: ugly, fat, obese, chubby, (((deformed))), [blurry], bad anatomy, disfigured, poorly drawn face, mutation, mutated, (extra_limb), (ugly), (poorly drawn hands), messy drawing -Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model hash: 45dee52b -``` - - returns a dict with field values - """ - - res = {} - - prompt = "" - negative_prompt = "" - - done_with_prompt = False - - *lines, lastline = x.strip().split("\n") - if len(re_param.findall(lastline)) < 3: - lines.append(lastline) - lastline = '' - - for line in lines: - line = line.strip() - if line.startswith("Negative prompt:"): - done_with_prompt = True - line = line[16:].strip() - if done_with_prompt: - negative_prompt += ("" if negative_prompt == "" else "\n") + line - else: - prompt += ("" if prompt == "" else "\n") + line - - if shared.opts.infotext_styles != "Ignore": - found_styles, prompt, negative_prompt = shared.prompt_styles.extract_styles_from_prompt(prompt, negative_prompt) - - if shared.opts.infotext_styles == "Apply": - res["Styles array"] = found_styles - elif shared.opts.infotext_styles == "Apply if any" and found_styles: - res["Styles array"] = found_styles - - res["Prompt"] = prompt - res["Negative prompt"] = negative_prompt - - for k, v in re_param.findall(lastline): - try: - if v[0] == '"' and v[-1] == '"': - v = unquote(v) - - m = re_imagesize.match(v) - if m is not None: - res[f"{k}-1"] = m.group(1) - res[f"{k}-2"] = m.group(2) - else: - res[k] = v - except Exception: - print(f"Error parsing \"{k}: {v}\"") - - # Missing CLIP skip means it was set to 1 (the default) - if "Clip skip" not in res: - res["Clip skip"] = "1" - - hypernet = res.get("Hypernet", None) - if hypernet is not None: - res["Prompt"] += f"""""" - - if "Hires resize-1" not in res: - res["Hires resize-1"] = 0 - res["Hires resize-2"] = 0 - - if "Hires sampler" not in res: - res["Hires sampler"] = "Use same sampler" - - if "Hires checkpoint" not in res: - res["Hires checkpoint"] = "Use same checkpoint" - - if "Hires prompt" not in res: - res["Hires prompt"] = "" - - if "Hires negative prompt" not in res: - res["Hires negative prompt"] = "" - - restore_old_hires_fix_params(res) - - # Missing RNG means the default was set, which is GPU RNG - if "RNG" not in res: - res["RNG"] = "GPU" - - if "Schedule type" not in res: - res["Schedule type"] = "Automatic" - - if "Schedule max sigma" not in res: - res["Schedule max sigma"] = 0 - - if "Schedule min sigma" not in res: - res["Schedule min sigma"] = 0 - - if "Schedule rho" not in res: - res["Schedule rho"] = 0 - - if "VAE Encoder" not in res: - res["VAE Encoder"] = "Full" - - if "VAE Decoder" not in res: - res["VAE Decoder"] = "Full" - - if "FP8 weight" not in res: - res["FP8 weight"] = "Disable" - - if "Cache FP16 weight for LoRA" not in res and res["FP8 weight"] != "Disable": - res["Cache FP16 weight for LoRA"] = False - - skip = set(shared.opts.infotext_skip_pasting) - res = {k: v for k, v in res.items() if k not in skip} - - return res - - -infotext_to_setting_name_mapping = [ - -] -"""Mapping of infotext labels to setting names. Only left for backwards compatibility - use OptionInfo(..., infotext='...') instead. -Example content: - -infotext_to_setting_name_mapping = [ - ('Conditional mask weight', 'inpainting_mask_weight'), - ('Model hash', 'sd_model_checkpoint'), - ('ENSD', 'eta_noise_seed_delta'), - ('Schedule type', 'k_sched_type'), -] -""" - - -def create_override_settings_dict(text_pairs): - """creates processing's override_settings parameters from gradio's multiselect - - Example input: - ['Clip skip: 2', 'Model hash: e6e99610c4', 'ENSD: 31337'] - - Example output: - {'CLIP_stop_at_last_layers': 2, 'sd_model_checkpoint': 'e6e99610c4', 'eta_noise_seed_delta': 31337} - """ - - res = {} - - params = {} - for pair in text_pairs: - k, v = pair.split(":", maxsplit=1) - - params[k] = v.strip() - - mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] - for param_name, setting_name in mapping + infotext_to_setting_name_mapping: - value = params.get(param_name, None) - - if value is None: - continue - - res[setting_name] = shared.opts.cast_value(setting_name, value) - - return res - - -def get_override_settings(params, *, skip_fields=None): - """Returns a list of settings overrides from the infotext parameters dictionary. - - This function checks the `params` dictionary for any keys that correspond to settings in `shared.opts` and returns - a list of tuples containing the parameter name, setting name, and new value cast to correct type. - - It checks for conditions before adding an override: - - ignores settings that match the current value - - ignores parameter keys present in skip_fields argument. - - Example input: - {"Clip skip": "2"} - - Example output: - [("Clip skip", "CLIP_stop_at_last_layers", 2)] - """ - - res = [] - - mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] - for param_name, setting_name in mapping + infotext_to_setting_name_mapping: - if param_name in (skip_fields or {}): - continue - - v = params.get(param_name, None) - if v is None: - continue - - if setting_name == "sd_model_checkpoint" and shared.opts.disable_weights_auto_swap: - continue - - v = shared.opts.cast_value(setting_name, v) - current_value = getattr(shared.opts, setting_name, None) - - if v == current_value: - continue - - res.append((param_name, setting_name, v)) - - return res - - -def connect_paste(button, paste_fields, input_comp, override_settings_component, tabname): - def paste_func(prompt): - if not prompt and not shared.cmd_opts.hide_ui_dir_config: - filename = os.path.join(data_path, "params.txt") - if os.path.exists(filename): - with open(filename, "r", encoding="utf8") as file: - prompt = file.read() - - params = parse_generation_parameters(prompt) - script_callbacks.infotext_pasted_callback(prompt, params) - res = [] - - for output, key in paste_fields: - if callable(key): - v = key(params) - else: - v = params.get(key, None) - - if v is None: - res.append(gr.update()) - elif isinstance(v, type_of_gr_update): - res.append(v) - else: - try: - valtype = type(output.value) - - if valtype == bool and v == "False": - val = False - else: - val = valtype(v) - - res.append(gr.update(value=val)) - except Exception: - res.append(gr.update()) - - return res - - if override_settings_component is not None: - already_handled_fields = {key: 1 for _, key in paste_fields} - - def paste_settings(params): - vals = get_override_settings(params, skip_fields=already_handled_fields) - - vals_pairs = [f"{infotext_text}: {value}" for infotext_text, setting_name, value in vals] - - return gr.Dropdown.update(value=vals_pairs, choices=vals_pairs, visible=bool(vals_pairs)) - - paste_fields = paste_fields + [(override_settings_component, paste_settings)] - - button.click( - fn=paste_func, - inputs=[input_comp], - outputs=[x[0] for x in paste_fields], - show_progress=False, - ) - button.click( - fn=None, - _js=f"recalculate_prompts_{tabname}", - inputs=[], - outputs=[], - show_progress=False, - ) - diff --git a/modules/infotext.py b/modules/infotext.py new file mode 100644 index 00000000..86a36c32 --- /dev/null +++ b/modules/infotext.py @@ -0,0 +1,497 @@ +from __future__ import annotations +import base64 +import io +import json +import os +import re + +import gradio as gr +from modules.paths import data_path +from modules import shared, ui_tempdir, script_callbacks, processing +from PIL import Image + +re_param_code = r'\s*(\w[\w \-/]+):\s*("(?:\\.|[^\\"])+"|[^,]*)(?:,|$)' +re_param = re.compile(re_param_code) +re_imagesize = re.compile(r"^(\d+)x(\d+)$") +re_hypernet_hash = re.compile("\(([0-9a-f]+)\)$") +type_of_gr_update = type(gr.update()) + + +class ParamBinding: + def __init__(self, paste_button, tabname, source_text_component=None, source_image_component=None, source_tabname=None, override_settings_component=None, paste_field_names=None): + self.paste_button = paste_button + self.tabname = tabname + self.source_text_component = source_text_component + self.source_image_component = source_image_component + self.source_tabname = source_tabname + self.override_settings_component = override_settings_component + self.paste_field_names = paste_field_names or [] + + +class PasteField(tuple): + def __new__(cls, component, target, *, api=None): + return super().__new__(cls, (component, target)) + + def __init__(self, component, target, *, api=None): + super().__init__() + + self.api = api + self.component = component + self.label = target if isinstance(target, str) else None + self.function = target if callable(target) else None + + +paste_fields: dict[str, dict] = {} +registered_param_bindings: list[ParamBinding] = [] + + +def reset(): + paste_fields.clear() + registered_param_bindings.clear() + + +def quote(text): + if ',' not in str(text) and '\n' not in str(text) and ':' not in str(text): + return text + + return json.dumps(text, ensure_ascii=False) + + +def unquote(text): + if len(text) == 0 or text[0] != '"' or text[-1] != '"': + return text + + try: + return json.loads(text) + except Exception: + return text + + +def image_from_url_text(filedata): + if filedata is None: + return None + + if type(filedata) == list and filedata and type(filedata[0]) == dict and filedata[0].get("is_file", False): + filedata = filedata[0] + + if type(filedata) == dict and filedata.get("is_file", False): + filename = filedata["name"] + is_in_right_dir = ui_tempdir.check_tmp_file(shared.demo, filename) + assert is_in_right_dir, 'trying to open image file outside of allowed directories' + + filename = filename.rsplit('?', 1)[0] + return Image.open(filename) + + if type(filedata) == list: + if len(filedata) == 0: + return None + + filedata = filedata[0] + + if filedata.startswith("data:image/png;base64,"): + filedata = filedata[len("data:image/png;base64,"):] + + filedata = base64.decodebytes(filedata.encode('utf-8')) + image = Image.open(io.BytesIO(filedata)) + return image + + +def add_paste_fields(tabname, init_img, fields, override_settings_component=None): + + if fields: + for i in range(len(fields)): + if not isinstance(fields[i], PasteField): + fields[i] = PasteField(*fields[i]) + + paste_fields[tabname] = {"init_img": init_img, "fields": fields, "override_settings_component": override_settings_component} + + # backwards compatibility for existing extensions + import modules.ui + if tabname == 'txt2img': + modules.ui.txt2img_paste_fields = fields + elif tabname == 'img2img': + modules.ui.img2img_paste_fields = fields + + +def create_buttons(tabs_list): + buttons = {} + for tab in tabs_list: + buttons[tab] = gr.Button(f"Send to {tab}", elem_id=f"{tab}_tab") + return buttons + + +def bind_buttons(buttons, send_image, send_generate_info): + """old function for backwards compatibility; do not use this, use register_paste_params_button""" + for tabname, button in buttons.items(): + source_text_component = send_generate_info if isinstance(send_generate_info, gr.components.Component) else None + source_tabname = send_generate_info if isinstance(send_generate_info, str) else None + + register_paste_params_button(ParamBinding(paste_button=button, tabname=tabname, source_text_component=source_text_component, source_image_component=send_image, source_tabname=source_tabname)) + + +def register_paste_params_button(binding: ParamBinding): + registered_param_bindings.append(binding) + + +def connect_paste_params_buttons(): + for binding in registered_param_bindings: + destination_image_component = paste_fields[binding.tabname]["init_img"] + fields = paste_fields[binding.tabname]["fields"] + override_settings_component = binding.override_settings_component or paste_fields[binding.tabname]["override_settings_component"] + + destination_width_component = next(iter([field for field, name in fields if name == "Size-1"] if fields else []), None) + destination_height_component = next(iter([field for field, name in fields if name == "Size-2"] if fields else []), None) + + if binding.source_image_component and destination_image_component: + if isinstance(binding.source_image_component, gr.Gallery): + func = send_image_and_dimensions if destination_width_component else image_from_url_text + jsfunc = "extract_image_from_gallery" + else: + func = send_image_and_dimensions if destination_width_component else lambda x: x + jsfunc = None + + binding.paste_button.click( + fn=func, + _js=jsfunc, + inputs=[binding.source_image_component], + outputs=[destination_image_component, destination_width_component, destination_height_component] if destination_width_component else [destination_image_component], + show_progress=False, + ) + + if binding.source_text_component is not None and fields is not None: + connect_paste(binding.paste_button, fields, binding.source_text_component, override_settings_component, binding.tabname) + + if binding.source_tabname is not None and fields is not None: + paste_field_names = ['Prompt', 'Negative prompt', 'Steps', 'Face restoration'] + (["Seed"] if shared.opts.send_seed else []) + binding.paste_field_names + binding.paste_button.click( + fn=lambda *x: x, + inputs=[field for field, name in paste_fields[binding.source_tabname]["fields"] if name in paste_field_names], + outputs=[field for field, name in fields if name in paste_field_names], + show_progress=False, + ) + + binding.paste_button.click( + fn=None, + _js=f"switch_to_{binding.tabname}", + inputs=None, + outputs=None, + show_progress=False, + ) + + +def send_image_and_dimensions(x): + if isinstance(x, Image.Image): + img = x + else: + img = image_from_url_text(x) + + if shared.opts.send_size and isinstance(img, Image.Image): + w = img.width + h = img.height + else: + w = gr.update() + h = gr.update() + + return img, w, h + + +def restore_old_hires_fix_params(res): + """for infotexts that specify old First pass size parameter, convert it into + width, height, and hr scale""" + + firstpass_width = res.get('First pass size-1', None) + firstpass_height = res.get('First pass size-2', None) + + if shared.opts.use_old_hires_fix_width_height: + hires_width = int(res.get("Hires resize-1", 0)) + hires_height = int(res.get("Hires resize-2", 0)) + + if hires_width and hires_height: + res['Size-1'] = hires_width + res['Size-2'] = hires_height + return + + if firstpass_width is None or firstpass_height is None: + return + + firstpass_width, firstpass_height = int(firstpass_width), int(firstpass_height) + width = int(res.get("Size-1", 512)) + height = int(res.get("Size-2", 512)) + + if firstpass_width == 0 or firstpass_height == 0: + firstpass_width, firstpass_height = processing.old_hires_fix_first_pass_dimensions(width, height) + + res['Size-1'] = firstpass_width + res['Size-2'] = firstpass_height + res['Hires resize-1'] = width + res['Hires resize-2'] = height + + +def parse_generation_parameters(x: str): + """parses generation parameters string, the one you see in text field under the picture in UI: +``` +girl with an artist's beret, determined, blue eyes, desert scene, computer monitors, heavy makeup, by Alphonse Mucha and Charlie Bowater, ((eyeshadow)), (coquettish), detailed, intricate +Negative prompt: ugly, fat, obese, chubby, (((deformed))), [blurry], bad anatomy, disfigured, poorly drawn face, mutation, mutated, (extra_limb), (ugly), (poorly drawn hands), messy drawing +Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model hash: 45dee52b +``` + + returns a dict with field values + """ + + res = {} + + prompt = "" + negative_prompt = "" + + done_with_prompt = False + + *lines, lastline = x.strip().split("\n") + if len(re_param.findall(lastline)) < 3: + lines.append(lastline) + lastline = '' + + for line in lines: + line = line.strip() + if line.startswith("Negative prompt:"): + done_with_prompt = True + line = line[16:].strip() + if done_with_prompt: + negative_prompt += ("" if negative_prompt == "" else "\n") + line + else: + prompt += ("" if prompt == "" else "\n") + line + + if shared.opts.infotext_styles != "Ignore": + found_styles, prompt, negative_prompt = shared.prompt_styles.extract_styles_from_prompt(prompt, negative_prompt) + + if shared.opts.infotext_styles == "Apply": + res["Styles array"] = found_styles + elif shared.opts.infotext_styles == "Apply if any" and found_styles: + res["Styles array"] = found_styles + + res["Prompt"] = prompt + res["Negative prompt"] = negative_prompt + + for k, v in re_param.findall(lastline): + try: + if v[0] == '"' and v[-1] == '"': + v = unquote(v) + + m = re_imagesize.match(v) + if m is not None: + res[f"{k}-1"] = m.group(1) + res[f"{k}-2"] = m.group(2) + else: + res[k] = v + except Exception: + print(f"Error parsing \"{k}: {v}\"") + + # Missing CLIP skip means it was set to 1 (the default) + if "Clip skip" not in res: + res["Clip skip"] = "1" + + hypernet = res.get("Hypernet", None) + if hypernet is not None: + res["Prompt"] += f"""""" + + if "Hires resize-1" not in res: + res["Hires resize-1"] = 0 + res["Hires resize-2"] = 0 + + if "Hires sampler" not in res: + res["Hires sampler"] = "Use same sampler" + + if "Hires checkpoint" not in res: + res["Hires checkpoint"] = "Use same checkpoint" + + if "Hires prompt" not in res: + res["Hires prompt"] = "" + + if "Hires negative prompt" not in res: + res["Hires negative prompt"] = "" + + restore_old_hires_fix_params(res) + + # Missing RNG means the default was set, which is GPU RNG + if "RNG" not in res: + res["RNG"] = "GPU" + + if "Schedule type" not in res: + res["Schedule type"] = "Automatic" + + if "Schedule max sigma" not in res: + res["Schedule max sigma"] = 0 + + if "Schedule min sigma" not in res: + res["Schedule min sigma"] = 0 + + if "Schedule rho" not in res: + res["Schedule rho"] = 0 + + if "VAE Encoder" not in res: + res["VAE Encoder"] = "Full" + + if "VAE Decoder" not in res: + res["VAE Decoder"] = "Full" + + if "FP8 weight" not in res: + res["FP8 weight"] = "Disable" + + if "Cache FP16 weight for LoRA" not in res and res["FP8 weight"] != "Disable": + res["Cache FP16 weight for LoRA"] = False + + skip = set(shared.opts.infotext_skip_pasting) + res = {k: v for k, v in res.items() if k not in skip} + + return res + + +infotext_to_setting_name_mapping = [ + +] +"""Mapping of infotext labels to setting names. Only left for backwards compatibility - use OptionInfo(..., infotext='...') instead. +Example content: + +infotext_to_setting_name_mapping = [ + ('Conditional mask weight', 'inpainting_mask_weight'), + ('Model hash', 'sd_model_checkpoint'), + ('ENSD', 'eta_noise_seed_delta'), + ('Schedule type', 'k_sched_type'), +] +""" + + +def create_override_settings_dict(text_pairs): + """creates processing's override_settings parameters from gradio's multiselect + + Example input: + ['Clip skip: 2', 'Model hash: e6e99610c4', 'ENSD: 31337'] + + Example output: + {'CLIP_stop_at_last_layers': 2, 'sd_model_checkpoint': 'e6e99610c4', 'eta_noise_seed_delta': 31337} + """ + + res = {} + + params = {} + for pair in text_pairs: + k, v = pair.split(":", maxsplit=1) + + params[k] = v.strip() + + mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] + for param_name, setting_name in mapping + infotext_to_setting_name_mapping: + value = params.get(param_name, None) + + if value is None: + continue + + res[setting_name] = shared.opts.cast_value(setting_name, value) + + return res + + +def get_override_settings(params, *, skip_fields=None): + """Returns a list of settings overrides from the infotext parameters dictionary. + + This function checks the `params` dictionary for any keys that correspond to settings in `shared.opts` and returns + a list of tuples containing the parameter name, setting name, and new value cast to correct type. + + It checks for conditions before adding an override: + - ignores settings that match the current value + - ignores parameter keys present in skip_fields argument. + + Example input: + {"Clip skip": "2"} + + Example output: + [("Clip skip", "CLIP_stop_at_last_layers", 2)] + """ + + res = [] + + mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] + for param_name, setting_name in mapping + infotext_to_setting_name_mapping: + if param_name in (skip_fields or {}): + continue + + v = params.get(param_name, None) + if v is None: + continue + + if setting_name == "sd_model_checkpoint" and shared.opts.disable_weights_auto_swap: + continue + + v = shared.opts.cast_value(setting_name, v) + current_value = getattr(shared.opts, setting_name, None) + + if v == current_value: + continue + + res.append((param_name, setting_name, v)) + + return res + + +def connect_paste(button, paste_fields, input_comp, override_settings_component, tabname): + def paste_func(prompt): + if not prompt and not shared.cmd_opts.hide_ui_dir_config: + filename = os.path.join(data_path, "params.txt") + if os.path.exists(filename): + with open(filename, "r", encoding="utf8") as file: + prompt = file.read() + + params = parse_generation_parameters(prompt) + script_callbacks.infotext_pasted_callback(prompt, params) + res = [] + + for output, key in paste_fields: + if callable(key): + v = key(params) + else: + v = params.get(key, None) + + if v is None: + res.append(gr.update()) + elif isinstance(v, type_of_gr_update): + res.append(v) + else: + try: + valtype = type(output.value) + + if valtype == bool and v == "False": + val = False + else: + val = valtype(v) + + res.append(gr.update(value=val)) + except Exception: + res.append(gr.update()) + + return res + + if override_settings_component is not None: + already_handled_fields = {key: 1 for _, key in paste_fields} + + def paste_settings(params): + vals = get_override_settings(params, skip_fields=already_handled_fields) + + vals_pairs = [f"{infotext_text}: {value}" for infotext_text, setting_name, value in vals] + + return gr.Dropdown.update(value=vals_pairs, choices=vals_pairs, visible=bool(vals_pairs)) + + paste_fields = paste_fields + [(override_settings_component, paste_settings)] + + button.click( + fn=paste_func, + inputs=[input_comp], + outputs=[x[0] for x in paste_fields], + show_progress=False, + ) + button.click( + fn=None, + _js=f"recalculate_prompts_{tabname}", + inputs=[], + outputs=[], + show_progress=False, + ) + -- cgit v1.2.1 From c5496c76461c90bd186ae8804aa65a33cd136d48 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 13:52:37 +0300 Subject: infotext.py: add support for old modules.generation_parameters_copypaste name --- modules/infotext.py | 3 +++ 1 file changed, 3 insertions(+) (limited to 'modules') diff --git a/modules/infotext.py b/modules/infotext.py index 86a36c32..bcbeb0fd 100644 --- a/modules/infotext.py +++ b/modules/infotext.py @@ -4,12 +4,15 @@ import io import json import os import re +import sys import gradio as gr from modules.paths import data_path from modules import shared, ui_tempdir, script_callbacks, processing from PIL import Image +sys.modules['modules.generation_parameters_copypaste'] = sys.modules[__name__] # alias for old name + re_param_code = r'\s*(\w[\w \-/]+):\s*("(?:\\.|[^\\"])+"|[^,]*)(?:,|$)' re_param = re.compile(re_param_code) re_imagesize = re.compile(r"^(\d+)x(\d+)$") -- cgit v1.2.1 From d859cec696a953dbfd6f69f7735e68661748d579 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 13:53:12 +0300 Subject: infotext.py: rename usages in the codebase --- modules/api/api.py | 10 +++++----- modules/img2img.py | 2 +- modules/postprocessing.py | 4 ++-- modules/processing.py | 4 ++-- modules/processing_scripts/refiner.py | 2 +- modules/processing_scripts/seed.py | 2 +- modules/shared_items.py | 4 ++-- modules/txt2img.py | 2 +- modules/ui.py | 4 ++-- modules/ui_common.py | 4 ++-- modules/ui_extra_networks.py | 2 +- modules/ui_extra_networks_user_metadata.py | 4 ++-- modules/ui_postprocessing.py | 2 +- 13 files changed, 23 insertions(+), 23 deletions(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 843c59b0..0e2807de 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -17,7 +17,7 @@ from fastapi.encoders import jsonable_encoder from secrets import compare_digest import modules.shared as shared -from modules import sd_samplers, deepbooru, sd_hijack, images, scripts, ui, postprocessing, errors, restart, shared_items, script_callbacks, generation_parameters_copypaste, sd_models +from modules import sd_samplers, deepbooru, sd_hijack, images, scripts, ui, postprocessing, errors, restart, shared_items, script_callbacks, infotext, sd_models from modules.api import models from modules.shared import opts from modules.processing import StableDiffusionProcessingTxt2Img, StableDiffusionProcessingImg2Img, process_images @@ -369,9 +369,9 @@ class Api: if not request.infotext: return {} - possible_fields = generation_parameters_copypaste.paste_fields[tabname]["fields"] + possible_fields = infotext.paste_fields[tabname]["fields"] set_fields = request.model_dump(exclude_unset=True) if hasattr(request, "request") else request.dict(exclude_unset=True) # pydantic v1/v2 have differenrt names for this - params = generation_parameters_copypaste.parse_generation_parameters(request.infotext) + params = infotext.parse_generation_parameters(request.infotext) def get_field_value(field, params): value = field.function(params) if field.function else params.get(field.label) @@ -408,7 +408,7 @@ class Api: if request.override_settings is None: request.override_settings = {} - overriden_settings = generation_parameters_copypaste.get_override_settings(params) + overriden_settings = infotext.get_override_settings(params) for _, setting_name, value in overriden_settings: if setting_name not in request.override_settings: request.override_settings[setting_name] = value @@ -584,7 +584,7 @@ class Api: if geninfo is None: geninfo = "" - params = generation_parameters_copypaste.parse_generation_parameters(geninfo) + params = infotext.parse_generation_parameters(geninfo) script_callbacks.infotext_pasted_callback(geninfo, params) return models.PNGInfoResponse(info=geninfo, items=items, parameters=params) diff --git a/modules/img2img.py b/modules/img2img.py index c583290a..75b3d346 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -7,7 +7,7 @@ from PIL import Image, ImageOps, ImageFilter, ImageEnhance, UnidentifiedImageErr import gradio as gr from modules import images as imgutil -from modules.generation_parameters_copypaste import create_override_settings_dict, parse_generation_parameters +from modules.infotext import create_override_settings_dict, parse_generation_parameters from modules.processing import Processed, StableDiffusionProcessingImg2Img, process_images from modules.shared import opts, state from modules.sd_models import get_closet_checkpoint_match diff --git a/modules/postprocessing.py b/modules/postprocessing.py index 0c59fad4..f776f7b6 100644 --- a/modules/postprocessing.py +++ b/modules/postprocessing.py @@ -2,7 +2,7 @@ import os from PIL import Image -from modules import shared, images, devices, scripts, scripts_postprocessing, ui_common, generation_parameters_copypaste +from modules import shared, images, devices, scripts, scripts_postprocessing, ui_common, infotext from modules.shared import opts @@ -86,7 +86,7 @@ def run_postprocessing(extras_mode, image, image_folder, input_dir, output_dir, basename = '' forced_filename = None - infotext = ", ".join([k if k == v else f'{k}: {generation_parameters_copypaste.quote(v)}' for k, v in pp.info.items() if v is not None]) + infotext = ", ".join([k if k == v else f'{k}: {infotext.quote(v)}' for k, v in pp.info.items() if v is not None]) if opts.enable_pnginfo: pp.image.info = existing_pnginfo diff --git a/modules/processing.py b/modules/processing.py index 7789f9a4..b30df60d 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -16,7 +16,7 @@ from skimage import exposure from typing import Any import modules.sd_hijack -from modules import devices, prompt_parser, masking, sd_samplers, lowvram, generation_parameters_copypaste, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors, rng +from modules import devices, prompt_parser, masking, sd_samplers, lowvram, infotext, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors, rng from modules.rng import slerp # noqa: F401 from modules.sd_hijack import model_hijack from modules.sd_samplers_common import images_tensor_to_samples, decode_first_stage, approximation_indexes @@ -733,7 +733,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "User": p.user if opts.add_user_name_to_info else None, } - generation_params_text = ", ".join([k if k == v else f'{k}: {generation_parameters_copypaste.quote(v)}' for k, v in generation_params.items() if v is not None]) + generation_params_text = ", ".join([k if k == v else f'{k}: {infotext.quote(v)}' for k, v in generation_params.items() if v is not None]) prompt_text = p.main_prompt if use_main_prompt else all_prompts[index] negative_prompt_text = f"\nNegative prompt: {p.main_negative_prompt if use_main_prompt else all_negative_prompts[index]}" if all_negative_prompts[index] else "" diff --git a/modules/processing_scripts/refiner.py b/modules/processing_scripts/refiner.py index cefad32b..e9941413 100644 --- a/modules/processing_scripts/refiner.py +++ b/modules/processing_scripts/refiner.py @@ -1,7 +1,7 @@ import gradio as gr from modules import scripts, sd_models -from modules.generation_parameters_copypaste import PasteField +from modules.infotext import PasteField from modules.ui_common import create_refresh_button from modules.ui_components import InputAccordion diff --git a/modules/processing_scripts/seed.py b/modules/processing_scripts/seed.py index a3e16a12..60293278 100644 --- a/modules/processing_scripts/seed.py +++ b/modules/processing_scripts/seed.py @@ -3,7 +3,7 @@ import json import gradio as gr from modules import scripts, ui, errors -from modules.generation_parameters_copypaste import PasteField +from modules.infotext import PasteField from modules.shared import cmd_opts from modules.ui_components import ToolButton diff --git a/modules/shared_items.py b/modules/shared_items.py index 991971ad..e1392472 100644 --- a/modules/shared_items.py +++ b/modules/shared_items.py @@ -67,14 +67,14 @@ def reload_hypernetworks(): def get_infotext_names(): - from modules import generation_parameters_copypaste, shared + from modules import infotext, shared res = {} for info in shared.opts.data_labels.values(): if info.infotext: res[info.infotext] = 1 - for tab_data in generation_parameters_copypaste.paste_fields.values(): + for tab_data in infotext.paste_fields.values(): for _, name in tab_data.get("fields") or []: if isinstance(name, str): res[name] = 1 diff --git a/modules/txt2img.py b/modules/txt2img.py index e4e18ceb..3a481915 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -2,7 +2,7 @@ from contextlib import closing import modules.scripts from modules import processing -from modules.generation_parameters_copypaste import create_override_settings_dict +from modules.infotext import create_override_settings_dict from modules.shared import opts import modules.shared as shared from modules.ui import plaintext_to_html diff --git a/modules/ui.py b/modules/ui.py index 9db2407e..6451e14c 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -21,14 +21,14 @@ from modules.ui_gradio_extensions import reload_javascript from modules.shared import opts, cmd_opts -import modules.generation_parameters_copypaste as parameters_copypaste +import modules.infotext as parameters_copypaste import modules.hypernetworks.ui as hypernetworks_ui import modules.textual_inversion.ui as textual_inversion_ui import modules.textual_inversion.textual_inversion as textual_inversion import modules.shared as shared from modules import prompt_parser from modules.sd_hijack import model_hijack -from modules.generation_parameters_copypaste import image_from_url_text, PasteField +from modules.infotext import image_from_url_text, PasteField create_setting_component = ui_settings.create_setting_component diff --git a/modules/ui_common.py b/modules/ui_common.py index 032ec4af..fd32676f 100644 --- a/modules/ui_common.py +++ b/modules/ui_common.py @@ -8,10 +8,10 @@ import gradio as gr import subprocess as sp from modules import call_queue, shared -from modules.generation_parameters_copypaste import image_from_url_text +from modules.infotext import image_from_url_text import modules.images from modules.ui_components import ToolButton -import modules.generation_parameters_copypaste as parameters_copypaste +import modules.infotext as parameters_copypaste folder_symbol = '\U0001f4c2' # 📂 refresh_symbol = '\U0001f504' # 🔄 diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index b8c02241..790af135 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -10,7 +10,7 @@ import json import html from fastapi.exceptions import HTTPException -from modules.generation_parameters_copypaste import image_from_url_text +from modules.infotext import image_from_url_text from modules.ui_components import ToolButton extra_pages = [] diff --git a/modules/ui_extra_networks_user_metadata.py b/modules/ui_extra_networks_user_metadata.py index 36a807fc..87aeb6f3 100644 --- a/modules/ui_extra_networks_user_metadata.py +++ b/modules/ui_extra_networks_user_metadata.py @@ -5,7 +5,7 @@ import os.path import gradio as gr -from modules import generation_parameters_copypaste, images, sysinfo, errors, ui_extra_networks +from modules import infotext, images, sysinfo, errors, ui_extra_networks class UserMetadataEditor: @@ -181,7 +181,7 @@ class UserMetadataEditor: index = len(gallery) - 1 if index >= len(gallery) else index img_info = gallery[index if index >= 0 else 0] - image = generation_parameters_copypaste.image_from_url_text(img_info) + image = infotext.image_from_url_text(img_info) geninfo, items = images.read_info_from_image(image) images.save_image_with_geninfo(image, geninfo, item["local_preview"]) diff --git a/modules/ui_postprocessing.py b/modules/ui_postprocessing.py index 13d888e4..b74a1532 100644 --- a/modules/ui_postprocessing.py +++ b/modules/ui_postprocessing.py @@ -1,6 +1,6 @@ import gradio as gr from modules import scripts, shared, ui_common, postprocessing, call_queue, ui_toprow -import modules.generation_parameters_copypaste as parameters_copypaste +import modules.infotext as parameters_copypaste def create_ui(): -- cgit v1.2.1 From d613cd17c72c753bd1e314dff74dc22d9a949374 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 14:38:29 +0300 Subject: add automatic backwards version compatibility --- modules/infotext.py | 4 +++- modules/infotext_versions.py | 35 +++++++++++++++++++++++++++++++++++ modules/shared_options.py | 1 + 3 files changed, 39 insertions(+), 1 deletion(-) create mode 100644 modules/infotext_versions.py (limited to 'modules') diff --git a/modules/infotext.py b/modules/infotext.py index bcbeb0fd..7f30446b 100644 --- a/modules/infotext.py +++ b/modules/infotext.py @@ -8,7 +8,7 @@ import sys import gradio as gr from modules.paths import data_path -from modules import shared, ui_tempdir, script_callbacks, processing +from modules import shared, ui_tempdir, script_callbacks, processing, infotext_versions, errors from PIL import Image sys.modules['modules.generation_parameters_copypaste'] = sys.modules[__name__] # alias for old name @@ -342,6 +342,8 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model if "Cache FP16 weight for LoRA" not in res and res["FP8 weight"] != "Disable": res["Cache FP16 weight for LoRA"] = False + infotext_versions.backcompat(res) + skip = set(shared.opts.infotext_skip_pasting) res = {k: v for k, v in res.items() if k not in skip} diff --git a/modules/infotext_versions.py b/modules/infotext_versions.py new file mode 100644 index 00000000..01e885a2 --- /dev/null +++ b/modules/infotext_versions.py @@ -0,0 +1,35 @@ +from modules import shared +from packaging import version +import re + + +v160 = version.parse("1.6.0") + + +def parse_version(text): + if text is None: + return None + + m = re.match(r'([^-]+-[^-]+)-.*', text) + if m: + text = m.group(1) + + try: + return version.parse(text) + except Exception as e: + return None + + +def backcompat(d): + """Checks infotext Version field, and enables backwards compatibility options according to it.""" + + if not shared.opts.auto_backcompat: + return + + ver = parse_version(d.get("Version")) + if ver is None: + return + + if ver < v160: + d["Old prompt editing timelines"] = True + diff --git a/modules/shared_options.py b/modules/shared_options.py index 752a4f12..281591da 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -212,6 +212,7 @@ options_templates.update(options_section(('optimizations', "Optimizations", "sd" })) options_templates.update(options_section(('compatibility', "Compatibility", "sd"), { + "auto_backcompat": OptionInfo(True, "Automatic backward compatibility").info("automatically enable options for backwards compatibility when importing generation parameters from infotext that has program version."), "use_old_emphasis_implementation": OptionInfo(False, "Use old emphasis implementation. Can be useful to reproduce old seeds."), "use_old_karras_scheduler_sigmas": OptionInfo(False, "Use old karras scheduler sigmas (0.1 to 10)."), "no_dpmpp_sde_batch_determinism": OptionInfo(False, "Do not make DPM++ SDE deterministic across different batch sizes."), -- cgit v1.2.1 From 45b7bba3d06f2d4bc2fffc210cbfcb357b86add6 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 14:51:56 +0300 Subject: add automatic version support for zero terminal SNR noise schedule option from #14145 --- modules/infotext_versions.py | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'modules') diff --git a/modules/infotext_versions.py b/modules/infotext_versions.py index 01e885a2..9a204d84 100644 --- a/modules/infotext_versions.py +++ b/modules/infotext_versions.py @@ -4,6 +4,7 @@ import re v160 = version.parse("1.6.0") +v170_tsnr = version.parse("v1.7.0-225") def parse_version(text): @@ -33,3 +34,6 @@ def backcompat(d): if ver < v160: d["Old prompt editing timelines"] = True + if ver < v170_tsnr: + d["Downcast alphas_cumprod"] = True + -- cgit v1.2.1 From d8126be578c7d4579c0f2ee4adbe35500bc71ce6 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 15:00:39 +0300 Subject: linter --- modules/infotext.py | 2 +- modules/infotext_versions.py | 2 +- modules/postprocessing.py | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/infotext.py b/modules/infotext.py index 7f30446b..26e9b949 100644 --- a/modules/infotext.py +++ b/modules/infotext.py @@ -8,7 +8,7 @@ import sys import gradio as gr from modules.paths import data_path -from modules import shared, ui_tempdir, script_callbacks, processing, infotext_versions, errors +from modules import shared, ui_tempdir, script_callbacks, processing, infotext_versions from PIL import Image sys.modules['modules.generation_parameters_copypaste'] = sys.modules[__name__] # alias for old name diff --git a/modules/infotext_versions.py b/modules/infotext_versions.py index 9a204d84..a5afeebf 100644 --- a/modules/infotext_versions.py +++ b/modules/infotext_versions.py @@ -17,7 +17,7 @@ def parse_version(text): try: return version.parse(text) - except Exception as e: + except Exception: return None diff --git a/modules/postprocessing.py b/modules/postprocessing.py index f776f7b6..facea899 100644 --- a/modules/postprocessing.py +++ b/modules/postprocessing.py @@ -2,7 +2,7 @@ import os from PIL import Image -from modules import shared, images, devices, scripts, scripts_postprocessing, ui_common, infotext +from modules import shared, images, devices, scripts, scripts_postprocessing, ui_common from modules.shared import opts -- cgit v1.2.1 From ac0ecf3b4b9d147743c04f0ff4ddc4cf4595e11d Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 16:28:58 +0300 Subject: option to convert VAE to bfloat16 (implementation of #9295) --- modules/processing.py | 23 ++++++++++++++++++----- modules/shared_options.py | 1 + 2 files changed, 19 insertions(+), 5 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 846e4796..f0656882 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -628,20 +628,33 @@ def decode_latent_batch(model, batch, target_device=None, check_for_nans=False): sample = decode_first_stage(model, batch[i:i + 1])[0] if check_for_nans: + try: devices.test_for_nans(sample, "vae") except devices.NansException as e: - if devices.dtype_vae == torch.float32 or not shared.opts.auto_vae_precision: + if shared.opts.auto_vae_precision_bfloat16: + autofix_dtype = torch.bfloat16 + autofix_dtype_text = "bfloat16" + autofix_dtype_setting = "Automatically convert VAE to bfloat16" + autofix_dtype_comment = "" + elif shared.opts.auto_vae_precision: + autofix_dtype = torch.float32 + autofix_dtype_text = "32-bit float" + autofix_dtype_setting = "Automatically revert VAE to 32-bit floats" + autofix_dtype_comment = "\nTo always start with 32-bit VAE, use --no-half-vae commandline flag." + else: + raise e + + if devices.dtype_vae == autofix_dtype: raise e errors.print_error_explanation( "A tensor with all NaNs was produced in VAE.\n" - "Web UI will now convert VAE into 32-bit float and retry.\n" - "To disable this behavior, disable the 'Automatically revert VAE to 32-bit floats' setting.\n" - "To always start with 32-bit VAE, use --no-half-vae commandline flag." + f"Web UI will now convert VAE into {autofix_dtype_text} and retry.\n" + f"To disable this behavior, disable the '{autofix_dtype_setting}' setting.{autofix_dtype_comment}" ) - devices.dtype_vae = torch.float32 + devices.dtype_vae = autofix_dtype model.first_stage_model.to(devices.dtype_vae) batch = batch.to(devices.dtype_vae) diff --git a/modules/shared_options.py b/modules/shared_options.py index ce06f022..e813546f 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -177,6 +177,7 @@ For img2img, VAE is used to process user's input image before the sampling, and "sd_vae_checkpoint_cache": OptionInfo(0, "VAE Checkpoints to cache in RAM", gr.Slider, {"minimum": 0, "maximum": 10, "step": 1}), "sd_vae": OptionInfo("Automatic", "SD VAE", gr.Dropdown, lambda: {"choices": shared_items.sd_vae_items()}, refresh=shared_items.refresh_vae_list, infotext='VAE').info("choose VAE model: Automatic = use one with same filename as checkpoint; None = use VAE from checkpoint"), "sd_vae_overrides_per_model_preferences": OptionInfo(True, "Selected VAE overrides per-model preferences").info("you can set per-model VAE either by editing user metadata for checkpoints, or by making the VAE have same name as checkpoint"), + "auto_vae_precision_bfloat16": OptionInfo(False, "Automatically convert VAE to bfloat16").info("triggers when a tensor with NaNs is produced in VAE; disabling the option in this case will result in a black square image; if enabled, overrides the option below"), "auto_vae_precision": OptionInfo(True, "Automatically revert VAE to 32-bit floats").info("triggers when a tensor with NaNs is produced in VAE; disabling the option in this case will result in a black square image"), "sd_vae_encode_method": OptionInfo("Full", "VAE type for encode", gr.Radio, {"choices": ["Full", "TAESD"]}, infotext='VAE Encoder').info("method to encode image to latent (use in img2img, hires-fix or inpaint mask)"), "sd_vae_decode_method": OptionInfo("Full", "VAE type for decode", gr.Radio, {"choices": ["Full", "TAESD"]}, infotext='VAE Decoder').info("method to decode latent to image"), -- cgit v1.2.1 From 0aa7c53c0b9469849377aff83f43c9f75c19b3fa Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 16:50:59 +0300 Subject: fix borked merge, rename fields to better match what they do, change setting default to true for #13653 --- modules/call_queue.py | 2 +- modules/img2img.py | 2 +- modules/processing.py | 2 +- modules/shared_options.py | 2 +- modules/shared_state.py | 12 ++++++------ modules/ui_toprow.py | 8 +++++++- 6 files changed, 17 insertions(+), 11 deletions(-) (limited to 'modules') diff --git a/modules/call_queue.py b/modules/call_queue.py index 01c6d17f..bcd7c546 100644 --- a/modules/call_queue.py +++ b/modules/call_queue.py @@ -78,7 +78,7 @@ def wrap_gradio_call(func, extra_outputs=None, add_stats=False): shared.state.skipped = False shared.state.interrupted = False - shared.state.interrupted_next = False + shared.state.stopping_generation = False shared.state.job_count = 0 if not add_stats: diff --git a/modules/img2img.py b/modules/img2img.py index 829faa81..e7e8e251 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -51,7 +51,7 @@ def process_batch(p, input_dir, output_dir, inpaint_mask_dir, args, to_scale=Fal if state.skipped: state.skipped = False - if state.interrupted or state.interrupted_next: + if state.interrupted or state.stopping_generation: break try: diff --git a/modules/processing.py b/modules/processing.py index 00de2ed2..f55b85ed 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -865,7 +865,7 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if state.skipped: state.skipped = False - if state.interrupted or state.interrupted_next: + if state.interrupted or state.stopping_generation: break sd_models.reload_model_weights() # model can be changed for example by refiner diff --git a/modules/shared_options.py b/modules/shared_options.py index 7852e0ea..7581e276 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -120,7 +120,6 @@ options_templates.update(options_section(('system', "System", "system"), { "disable_mmap_load_safetensors": OptionInfo(False, "Disable memmapping for loading .safetensors files.").info("fixes very slow loading speed in some cases"), "hide_ldm_prints": OptionInfo(True, "Prevent Stability-AI's ldm/sgm modules from printing noise to console."), "dump_stacks_on_signal": OptionInfo(False, "Print stack traces before exiting the program with ctrl+c."), - "interrupt_after_current": OptionInfo(False, "Interrupt generation after current image is finished on batch processing"), })) options_templates.update(options_section(('API', "API", "system"), { @@ -286,6 +285,7 @@ options_templates.update(options_section(('ui_alternatives', "UI alternatives", "hires_fix_show_prompts": OptionInfo(False, "Hires fix: show hires prompt and negative prompt").needs_reload_ui(), "txt2img_settings_accordion": OptionInfo(False, "Settings in txt2img hidden under Accordion").needs_reload_ui(), "img2img_settings_accordion": OptionInfo(False, "Settings in img2img hidden under Accordion").needs_reload_ui(), + "interrupt_after_current": OptionInfo(True, "Don't Interrupt in the middle").info("when using Interrupt button, if generating more than one image, stop after the generation of an image has finished, instead of immediately"), })) options_templates.update(options_section(('ui', "User interface", "ui"), { diff --git a/modules/shared_state.py b/modules/shared_state.py index 532fdcd8..33996691 100644 --- a/modules/shared_state.py +++ b/modules/shared_state.py @@ -12,7 +12,7 @@ log = logging.getLogger(__name__) class State: skipped = False interrupted = False - interrupted_next = False + stopping_generation = False job = "" job_no = 0 job_count = 0 @@ -80,9 +80,9 @@ class State: self.interrupted = True log.info("Received interrupt request") - def interrupt_next(self): - self.interrupted_next = True - log.info("Received interrupt request, interrupt after current job") + def stop_generating(self): + self.stopping_generation = True + log.info("Received stop generating request") def nextjob(self): if shared.opts.live_previews_enable and shared.opts.show_progress_every_n_steps == -1: @@ -96,7 +96,7 @@ class State: obj = { "skipped": self.skipped, "interrupted": self.interrupted, - "interrupted_next": self.interrupted_next, + "stopping_generation": self.stopping_generation, "job": self.job, "job_count": self.job_count, "job_timestamp": self.job_timestamp, @@ -120,7 +120,7 @@ class State: self.id_live_preview = 0 self.skipped = False self.interrupted = False - self.interrupted_next = False + self.stopping_generation = False self.textinfo = None self.job = job devices.torch_gc() diff --git a/modules/ui_toprow.py b/modules/ui_toprow.py index 9caf8faa..1abc9117 100644 --- a/modules/ui_toprow.py +++ b/modules/ui_toprow.py @@ -106,8 +106,14 @@ class Toprow: outputs=[], ) + def interrupt_function(): + if shared.state.job_count > 1 and shared.opts.interrupt_after_current: + shared.state.stop_generating() + else: + shared.state.interrupt() + self.interrupt.click( - fn=lambda: shared.state.interrupt(), + fn=interrupt_function, inputs=[], outputs=[], ) -- cgit v1.2.1 From 1ffdedc11d49862cf0d030fb0bcc25eb0449939b Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 17:03:08 +0300 Subject: restore lines lost from #13789 merge --- modules/cmd_args.py | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/cmd_args.py b/modules/cmd_args.py index 3775e670..e58059a1 100644 --- a/modules/cmd_args.py +++ b/modules/cmd_args.py @@ -93,7 +93,7 @@ parser.add_argument("--autolaunch", action='store_true', help="open the webui UR parser.add_argument("--theme", type=str, help="launches the UI with light or dark theme", default=None) parser.add_argument("--use-textbox-seed", action='store_true', help="use textbox for seeds in UI (no up/down, but possible to input long seeds)", default=False) parser.add_argument("--disable-console-progressbars", action='store_true', help="do not output progressbars to console", default=False) -parser.add_argument("--enable-console-prompts", action='store_true', help="print prompts to console when generating with txt2img and img2img", default=False) +parser.add_argument("--enable-console-prompts", action='store_true', help="does not do anything", default=False) # Legacy compatibility, use as default value shared.opts.enable_console_prompts parser.add_argument('--vae-path', type=str, help='Checkpoint to use as VAE; setting this argument disables all settings related to VAE', default=None) parser.add_argument("--disable-safe-unpickle", action='store_true', help="disable checking pytorch models for malicious code", default=False) parser.add_argument("--api", action='store_true', help="use api=True to launch the API together with the webui (use --nowebui instead for only the API)") @@ -115,8 +115,9 @@ parser.add_argument("--skip-version-check", action='store_true', help="Do not ch parser.add_argument("--no-hashing", action='store_true', help="disable sha256 hashing of checkpoints to help loading performance", default=False) parser.add_argument("--no-download-sd-model", action='store_true', help="don't download SD1.5 model even if no model is found in --ckpt-dir", default=False) parser.add_argument('--subpath', type=str, help='customize the subpath for gradio, use with reverse proxy') -parser.add_argument('--add-stop-route', action='store_true', help='add /_stop route to stop server') +parser.add_argument('--add-stop-route', action='store_true', help='does not do anything') parser.add_argument('--api-server-stop', action='store_true', help='enable server stop/restart/kill via api') parser.add_argument('--timeout-keep-alive', type=int, default=30, help='set timeout_keep_alive for uvicorn') parser.add_argument("--disable-all-extensions", action='store_true', help="prevent all extensions from running regardless of any other settings", default=False) -parser.add_argument("--disable-extra-extensions", action='store_true', help=" prevent all extensions except built-in from running regardless of any other settings", default=False) +parser.add_argument("--disable-extra-extensions", action='store_true', help="prevent all extensions except built-in from running regardless of any other settings", default=False) +parser.add_argument("--skip-load-model-at-start", action='store_true', help="if load a model at web start, only take effect when --nowebui", ) -- cgit v1.2.1 From 5d7d1823afab0a051a3fbbdb3213bae8051350b7 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 17:25:30 +0300 Subject: rename infotext.py again, this time to infotext_utils.py; I didn't realize infotext would be used for variable names in multiple places, which makes it awkward to import the module; also fix the bug I caused by this rename that breaks tests --- modules/api/api.py | 10 +- modules/img2img.py | 2 +- modules/infotext.py | 502 ----------------------------- modules/infotext_utils.py | 502 +++++++++++++++++++++++++++++ modules/postprocessing.py | 4 +- modules/processing.py | 4 +- modules/processing_scripts/refiner.py | 2 +- modules/processing_scripts/seed.py | 2 +- modules/shared_items.py | 4 +- modules/txt2img.py | 2 +- modules/ui.py | 4 +- modules/ui_common.py | 4 +- modules/ui_extra_networks.py | 2 +- modules/ui_extra_networks_user_metadata.py | 4 +- modules/ui_postprocessing.py | 2 +- 15 files changed, 525 insertions(+), 525 deletions(-) delete mode 100644 modules/infotext.py create mode 100644 modules/infotext_utils.py (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 0e2807de..9d1292e9 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -17,7 +17,7 @@ from fastapi.encoders import jsonable_encoder from secrets import compare_digest import modules.shared as shared -from modules import sd_samplers, deepbooru, sd_hijack, images, scripts, ui, postprocessing, errors, restart, shared_items, script_callbacks, infotext, sd_models +from modules import sd_samplers, deepbooru, sd_hijack, images, scripts, ui, postprocessing, errors, restart, shared_items, script_callbacks, infotext_utils, sd_models from modules.api import models from modules.shared import opts from modules.processing import StableDiffusionProcessingTxt2Img, StableDiffusionProcessingImg2Img, process_images @@ -369,9 +369,9 @@ class Api: if not request.infotext: return {} - possible_fields = infotext.paste_fields[tabname]["fields"] + possible_fields = infotext_utils.paste_fields[tabname]["fields"] set_fields = request.model_dump(exclude_unset=True) if hasattr(request, "request") else request.dict(exclude_unset=True) # pydantic v1/v2 have differenrt names for this - params = infotext.parse_generation_parameters(request.infotext) + params = infotext_utils.parse_generation_parameters(request.infotext) def get_field_value(field, params): value = field.function(params) if field.function else params.get(field.label) @@ -408,7 +408,7 @@ class Api: if request.override_settings is None: request.override_settings = {} - overriden_settings = infotext.get_override_settings(params) + overriden_settings = infotext_utils.get_override_settings(params) for _, setting_name, value in overriden_settings: if setting_name not in request.override_settings: request.override_settings[setting_name] = value @@ -584,7 +584,7 @@ class Api: if geninfo is None: geninfo = "" - params = infotext.parse_generation_parameters(geninfo) + params = infotext_utils.parse_generation_parameters(geninfo) script_callbacks.infotext_pasted_callback(geninfo, params) return models.PNGInfoResponse(info=geninfo, items=items, parameters=params) diff --git a/modules/img2img.py b/modules/img2img.py index e7e8e251..04de8e62 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -7,7 +7,7 @@ from PIL import Image, ImageOps, ImageFilter, ImageEnhance, UnidentifiedImageErr import gradio as gr from modules import images as imgutil -from modules.infotext import create_override_settings_dict, parse_generation_parameters +from modules.infotext_utils import create_override_settings_dict, parse_generation_parameters from modules.processing import Processed, StableDiffusionProcessingImg2Img, process_images from modules.shared import opts, state from modules.sd_models import get_closet_checkpoint_match diff --git a/modules/infotext.py b/modules/infotext.py deleted file mode 100644 index 26e9b949..00000000 --- a/modules/infotext.py +++ /dev/null @@ -1,502 +0,0 @@ -from __future__ import annotations -import base64 -import io -import json -import os -import re -import sys - -import gradio as gr -from modules.paths import data_path -from modules import shared, ui_tempdir, script_callbacks, processing, infotext_versions -from PIL import Image - -sys.modules['modules.generation_parameters_copypaste'] = sys.modules[__name__] # alias for old name - -re_param_code = r'\s*(\w[\w \-/]+):\s*("(?:\\.|[^\\"])+"|[^,]*)(?:,|$)' -re_param = re.compile(re_param_code) -re_imagesize = re.compile(r"^(\d+)x(\d+)$") -re_hypernet_hash = re.compile("\(([0-9a-f]+)\)$") -type_of_gr_update = type(gr.update()) - - -class ParamBinding: - def __init__(self, paste_button, tabname, source_text_component=None, source_image_component=None, source_tabname=None, override_settings_component=None, paste_field_names=None): - self.paste_button = paste_button - self.tabname = tabname - self.source_text_component = source_text_component - self.source_image_component = source_image_component - self.source_tabname = source_tabname - self.override_settings_component = override_settings_component - self.paste_field_names = paste_field_names or [] - - -class PasteField(tuple): - def __new__(cls, component, target, *, api=None): - return super().__new__(cls, (component, target)) - - def __init__(self, component, target, *, api=None): - super().__init__() - - self.api = api - self.component = component - self.label = target if isinstance(target, str) else None - self.function = target if callable(target) else None - - -paste_fields: dict[str, dict] = {} -registered_param_bindings: list[ParamBinding] = [] - - -def reset(): - paste_fields.clear() - registered_param_bindings.clear() - - -def quote(text): - if ',' not in str(text) and '\n' not in str(text) and ':' not in str(text): - return text - - return json.dumps(text, ensure_ascii=False) - - -def unquote(text): - if len(text) == 0 or text[0] != '"' or text[-1] != '"': - return text - - try: - return json.loads(text) - except Exception: - return text - - -def image_from_url_text(filedata): - if filedata is None: - return None - - if type(filedata) == list and filedata and type(filedata[0]) == dict and filedata[0].get("is_file", False): - filedata = filedata[0] - - if type(filedata) == dict and filedata.get("is_file", False): - filename = filedata["name"] - is_in_right_dir = ui_tempdir.check_tmp_file(shared.demo, filename) - assert is_in_right_dir, 'trying to open image file outside of allowed directories' - - filename = filename.rsplit('?', 1)[0] - return Image.open(filename) - - if type(filedata) == list: - if len(filedata) == 0: - return None - - filedata = filedata[0] - - if filedata.startswith("data:image/png;base64,"): - filedata = filedata[len("data:image/png;base64,"):] - - filedata = base64.decodebytes(filedata.encode('utf-8')) - image = Image.open(io.BytesIO(filedata)) - return image - - -def add_paste_fields(tabname, init_img, fields, override_settings_component=None): - - if fields: - for i in range(len(fields)): - if not isinstance(fields[i], PasteField): - fields[i] = PasteField(*fields[i]) - - paste_fields[tabname] = {"init_img": init_img, "fields": fields, "override_settings_component": override_settings_component} - - # backwards compatibility for existing extensions - import modules.ui - if tabname == 'txt2img': - modules.ui.txt2img_paste_fields = fields - elif tabname == 'img2img': - modules.ui.img2img_paste_fields = fields - - -def create_buttons(tabs_list): - buttons = {} - for tab in tabs_list: - buttons[tab] = gr.Button(f"Send to {tab}", elem_id=f"{tab}_tab") - return buttons - - -def bind_buttons(buttons, send_image, send_generate_info): - """old function for backwards compatibility; do not use this, use register_paste_params_button""" - for tabname, button in buttons.items(): - source_text_component = send_generate_info if isinstance(send_generate_info, gr.components.Component) else None - source_tabname = send_generate_info if isinstance(send_generate_info, str) else None - - register_paste_params_button(ParamBinding(paste_button=button, tabname=tabname, source_text_component=source_text_component, source_image_component=send_image, source_tabname=source_tabname)) - - -def register_paste_params_button(binding: ParamBinding): - registered_param_bindings.append(binding) - - -def connect_paste_params_buttons(): - for binding in registered_param_bindings: - destination_image_component = paste_fields[binding.tabname]["init_img"] - fields = paste_fields[binding.tabname]["fields"] - override_settings_component = binding.override_settings_component or paste_fields[binding.tabname]["override_settings_component"] - - destination_width_component = next(iter([field for field, name in fields if name == "Size-1"] if fields else []), None) - destination_height_component = next(iter([field for field, name in fields if name == "Size-2"] if fields else []), None) - - if binding.source_image_component and destination_image_component: - if isinstance(binding.source_image_component, gr.Gallery): - func = send_image_and_dimensions if destination_width_component else image_from_url_text - jsfunc = "extract_image_from_gallery" - else: - func = send_image_and_dimensions if destination_width_component else lambda x: x - jsfunc = None - - binding.paste_button.click( - fn=func, - _js=jsfunc, - inputs=[binding.source_image_component], - outputs=[destination_image_component, destination_width_component, destination_height_component] if destination_width_component else [destination_image_component], - show_progress=False, - ) - - if binding.source_text_component is not None and fields is not None: - connect_paste(binding.paste_button, fields, binding.source_text_component, override_settings_component, binding.tabname) - - if binding.source_tabname is not None and fields is not None: - paste_field_names = ['Prompt', 'Negative prompt', 'Steps', 'Face restoration'] + (["Seed"] if shared.opts.send_seed else []) + binding.paste_field_names - binding.paste_button.click( - fn=lambda *x: x, - inputs=[field for field, name in paste_fields[binding.source_tabname]["fields"] if name in paste_field_names], - outputs=[field for field, name in fields if name in paste_field_names], - show_progress=False, - ) - - binding.paste_button.click( - fn=None, - _js=f"switch_to_{binding.tabname}", - inputs=None, - outputs=None, - show_progress=False, - ) - - -def send_image_and_dimensions(x): - if isinstance(x, Image.Image): - img = x - else: - img = image_from_url_text(x) - - if shared.opts.send_size and isinstance(img, Image.Image): - w = img.width - h = img.height - else: - w = gr.update() - h = gr.update() - - return img, w, h - - -def restore_old_hires_fix_params(res): - """for infotexts that specify old First pass size parameter, convert it into - width, height, and hr scale""" - - firstpass_width = res.get('First pass size-1', None) - firstpass_height = res.get('First pass size-2', None) - - if shared.opts.use_old_hires_fix_width_height: - hires_width = int(res.get("Hires resize-1", 0)) - hires_height = int(res.get("Hires resize-2", 0)) - - if hires_width and hires_height: - res['Size-1'] = hires_width - res['Size-2'] = hires_height - return - - if firstpass_width is None or firstpass_height is None: - return - - firstpass_width, firstpass_height = int(firstpass_width), int(firstpass_height) - width = int(res.get("Size-1", 512)) - height = int(res.get("Size-2", 512)) - - if firstpass_width == 0 or firstpass_height == 0: - firstpass_width, firstpass_height = processing.old_hires_fix_first_pass_dimensions(width, height) - - res['Size-1'] = firstpass_width - res['Size-2'] = firstpass_height - res['Hires resize-1'] = width - res['Hires resize-2'] = height - - -def parse_generation_parameters(x: str): - """parses generation parameters string, the one you see in text field under the picture in UI: -``` -girl with an artist's beret, determined, blue eyes, desert scene, computer monitors, heavy makeup, by Alphonse Mucha and Charlie Bowater, ((eyeshadow)), (coquettish), detailed, intricate -Negative prompt: ugly, fat, obese, chubby, (((deformed))), [blurry], bad anatomy, disfigured, poorly drawn face, mutation, mutated, (extra_limb), (ugly), (poorly drawn hands), messy drawing -Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model hash: 45dee52b -``` - - returns a dict with field values - """ - - res = {} - - prompt = "" - negative_prompt = "" - - done_with_prompt = False - - *lines, lastline = x.strip().split("\n") - if len(re_param.findall(lastline)) < 3: - lines.append(lastline) - lastline = '' - - for line in lines: - line = line.strip() - if line.startswith("Negative prompt:"): - done_with_prompt = True - line = line[16:].strip() - if done_with_prompt: - negative_prompt += ("" if negative_prompt == "" else "\n") + line - else: - prompt += ("" if prompt == "" else "\n") + line - - if shared.opts.infotext_styles != "Ignore": - found_styles, prompt, negative_prompt = shared.prompt_styles.extract_styles_from_prompt(prompt, negative_prompt) - - if shared.opts.infotext_styles == "Apply": - res["Styles array"] = found_styles - elif shared.opts.infotext_styles == "Apply if any" and found_styles: - res["Styles array"] = found_styles - - res["Prompt"] = prompt - res["Negative prompt"] = negative_prompt - - for k, v in re_param.findall(lastline): - try: - if v[0] == '"' and v[-1] == '"': - v = unquote(v) - - m = re_imagesize.match(v) - if m is not None: - res[f"{k}-1"] = m.group(1) - res[f"{k}-2"] = m.group(2) - else: - res[k] = v - except Exception: - print(f"Error parsing \"{k}: {v}\"") - - # Missing CLIP skip means it was set to 1 (the default) - if "Clip skip" not in res: - res["Clip skip"] = "1" - - hypernet = res.get("Hypernet", None) - if hypernet is not None: - res["Prompt"] += f"""""" - - if "Hires resize-1" not in res: - res["Hires resize-1"] = 0 - res["Hires resize-2"] = 0 - - if "Hires sampler" not in res: - res["Hires sampler"] = "Use same sampler" - - if "Hires checkpoint" not in res: - res["Hires checkpoint"] = "Use same checkpoint" - - if "Hires prompt" not in res: - res["Hires prompt"] = "" - - if "Hires negative prompt" not in res: - res["Hires negative prompt"] = "" - - restore_old_hires_fix_params(res) - - # Missing RNG means the default was set, which is GPU RNG - if "RNG" not in res: - res["RNG"] = "GPU" - - if "Schedule type" not in res: - res["Schedule type"] = "Automatic" - - if "Schedule max sigma" not in res: - res["Schedule max sigma"] = 0 - - if "Schedule min sigma" not in res: - res["Schedule min sigma"] = 0 - - if "Schedule rho" not in res: - res["Schedule rho"] = 0 - - if "VAE Encoder" not in res: - res["VAE Encoder"] = "Full" - - if "VAE Decoder" not in res: - res["VAE Decoder"] = "Full" - - if "FP8 weight" not in res: - res["FP8 weight"] = "Disable" - - if "Cache FP16 weight for LoRA" not in res and res["FP8 weight"] != "Disable": - res["Cache FP16 weight for LoRA"] = False - - infotext_versions.backcompat(res) - - skip = set(shared.opts.infotext_skip_pasting) - res = {k: v for k, v in res.items() if k not in skip} - - return res - - -infotext_to_setting_name_mapping = [ - -] -"""Mapping of infotext labels to setting names. Only left for backwards compatibility - use OptionInfo(..., infotext='...') instead. -Example content: - -infotext_to_setting_name_mapping = [ - ('Conditional mask weight', 'inpainting_mask_weight'), - ('Model hash', 'sd_model_checkpoint'), - ('ENSD', 'eta_noise_seed_delta'), - ('Schedule type', 'k_sched_type'), -] -""" - - -def create_override_settings_dict(text_pairs): - """creates processing's override_settings parameters from gradio's multiselect - - Example input: - ['Clip skip: 2', 'Model hash: e6e99610c4', 'ENSD: 31337'] - - Example output: - {'CLIP_stop_at_last_layers': 2, 'sd_model_checkpoint': 'e6e99610c4', 'eta_noise_seed_delta': 31337} - """ - - res = {} - - params = {} - for pair in text_pairs: - k, v = pair.split(":", maxsplit=1) - - params[k] = v.strip() - - mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] - for param_name, setting_name in mapping + infotext_to_setting_name_mapping: - value = params.get(param_name, None) - - if value is None: - continue - - res[setting_name] = shared.opts.cast_value(setting_name, value) - - return res - - -def get_override_settings(params, *, skip_fields=None): - """Returns a list of settings overrides from the infotext parameters dictionary. - - This function checks the `params` dictionary for any keys that correspond to settings in `shared.opts` and returns - a list of tuples containing the parameter name, setting name, and new value cast to correct type. - - It checks for conditions before adding an override: - - ignores settings that match the current value - - ignores parameter keys present in skip_fields argument. - - Example input: - {"Clip skip": "2"} - - Example output: - [("Clip skip", "CLIP_stop_at_last_layers", 2)] - """ - - res = [] - - mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] - for param_name, setting_name in mapping + infotext_to_setting_name_mapping: - if param_name in (skip_fields or {}): - continue - - v = params.get(param_name, None) - if v is None: - continue - - if setting_name == "sd_model_checkpoint" and shared.opts.disable_weights_auto_swap: - continue - - v = shared.opts.cast_value(setting_name, v) - current_value = getattr(shared.opts, setting_name, None) - - if v == current_value: - continue - - res.append((param_name, setting_name, v)) - - return res - - -def connect_paste(button, paste_fields, input_comp, override_settings_component, tabname): - def paste_func(prompt): - if not prompt and not shared.cmd_opts.hide_ui_dir_config: - filename = os.path.join(data_path, "params.txt") - if os.path.exists(filename): - with open(filename, "r", encoding="utf8") as file: - prompt = file.read() - - params = parse_generation_parameters(prompt) - script_callbacks.infotext_pasted_callback(prompt, params) - res = [] - - for output, key in paste_fields: - if callable(key): - v = key(params) - else: - v = params.get(key, None) - - if v is None: - res.append(gr.update()) - elif isinstance(v, type_of_gr_update): - res.append(v) - else: - try: - valtype = type(output.value) - - if valtype == bool and v == "False": - val = False - else: - val = valtype(v) - - res.append(gr.update(value=val)) - except Exception: - res.append(gr.update()) - - return res - - if override_settings_component is not None: - already_handled_fields = {key: 1 for _, key in paste_fields} - - def paste_settings(params): - vals = get_override_settings(params, skip_fields=already_handled_fields) - - vals_pairs = [f"{infotext_text}: {value}" for infotext_text, setting_name, value in vals] - - return gr.Dropdown.update(value=vals_pairs, choices=vals_pairs, visible=bool(vals_pairs)) - - paste_fields = paste_fields + [(override_settings_component, paste_settings)] - - button.click( - fn=paste_func, - inputs=[input_comp], - outputs=[x[0] for x in paste_fields], - show_progress=False, - ) - button.click( - fn=None, - _js=f"recalculate_prompts_{tabname}", - inputs=[], - outputs=[], - show_progress=False, - ) - diff --git a/modules/infotext_utils.py b/modules/infotext_utils.py new file mode 100644 index 00000000..26e9b949 --- /dev/null +++ b/modules/infotext_utils.py @@ -0,0 +1,502 @@ +from __future__ import annotations +import base64 +import io +import json +import os +import re +import sys + +import gradio as gr +from modules.paths import data_path +from modules import shared, ui_tempdir, script_callbacks, processing, infotext_versions +from PIL import Image + +sys.modules['modules.generation_parameters_copypaste'] = sys.modules[__name__] # alias for old name + +re_param_code = r'\s*(\w[\w \-/]+):\s*("(?:\\.|[^\\"])+"|[^,]*)(?:,|$)' +re_param = re.compile(re_param_code) +re_imagesize = re.compile(r"^(\d+)x(\d+)$") +re_hypernet_hash = re.compile("\(([0-9a-f]+)\)$") +type_of_gr_update = type(gr.update()) + + +class ParamBinding: + def __init__(self, paste_button, tabname, source_text_component=None, source_image_component=None, source_tabname=None, override_settings_component=None, paste_field_names=None): + self.paste_button = paste_button + self.tabname = tabname + self.source_text_component = source_text_component + self.source_image_component = source_image_component + self.source_tabname = source_tabname + self.override_settings_component = override_settings_component + self.paste_field_names = paste_field_names or [] + + +class PasteField(tuple): + def __new__(cls, component, target, *, api=None): + return super().__new__(cls, (component, target)) + + def __init__(self, component, target, *, api=None): + super().__init__() + + self.api = api + self.component = component + self.label = target if isinstance(target, str) else None + self.function = target if callable(target) else None + + +paste_fields: dict[str, dict] = {} +registered_param_bindings: list[ParamBinding] = [] + + +def reset(): + paste_fields.clear() + registered_param_bindings.clear() + + +def quote(text): + if ',' not in str(text) and '\n' not in str(text) and ':' not in str(text): + return text + + return json.dumps(text, ensure_ascii=False) + + +def unquote(text): + if len(text) == 0 or text[0] != '"' or text[-1] != '"': + return text + + try: + return json.loads(text) + except Exception: + return text + + +def image_from_url_text(filedata): + if filedata is None: + return None + + if type(filedata) == list and filedata and type(filedata[0]) == dict and filedata[0].get("is_file", False): + filedata = filedata[0] + + if type(filedata) == dict and filedata.get("is_file", False): + filename = filedata["name"] + is_in_right_dir = ui_tempdir.check_tmp_file(shared.demo, filename) + assert is_in_right_dir, 'trying to open image file outside of allowed directories' + + filename = filename.rsplit('?', 1)[0] + return Image.open(filename) + + if type(filedata) == list: + if len(filedata) == 0: + return None + + filedata = filedata[0] + + if filedata.startswith("data:image/png;base64,"): + filedata = filedata[len("data:image/png;base64,"):] + + filedata = base64.decodebytes(filedata.encode('utf-8')) + image = Image.open(io.BytesIO(filedata)) + return image + + +def add_paste_fields(tabname, init_img, fields, override_settings_component=None): + + if fields: + for i in range(len(fields)): + if not isinstance(fields[i], PasteField): + fields[i] = PasteField(*fields[i]) + + paste_fields[tabname] = {"init_img": init_img, "fields": fields, "override_settings_component": override_settings_component} + + # backwards compatibility for existing extensions + import modules.ui + if tabname == 'txt2img': + modules.ui.txt2img_paste_fields = fields + elif tabname == 'img2img': + modules.ui.img2img_paste_fields = fields + + +def create_buttons(tabs_list): + buttons = {} + for tab in tabs_list: + buttons[tab] = gr.Button(f"Send to {tab}", elem_id=f"{tab}_tab") + return buttons + + +def bind_buttons(buttons, send_image, send_generate_info): + """old function for backwards compatibility; do not use this, use register_paste_params_button""" + for tabname, button in buttons.items(): + source_text_component = send_generate_info if isinstance(send_generate_info, gr.components.Component) else None + source_tabname = send_generate_info if isinstance(send_generate_info, str) else None + + register_paste_params_button(ParamBinding(paste_button=button, tabname=tabname, source_text_component=source_text_component, source_image_component=send_image, source_tabname=source_tabname)) + + +def register_paste_params_button(binding: ParamBinding): + registered_param_bindings.append(binding) + + +def connect_paste_params_buttons(): + for binding in registered_param_bindings: + destination_image_component = paste_fields[binding.tabname]["init_img"] + fields = paste_fields[binding.tabname]["fields"] + override_settings_component = binding.override_settings_component or paste_fields[binding.tabname]["override_settings_component"] + + destination_width_component = next(iter([field for field, name in fields if name == "Size-1"] if fields else []), None) + destination_height_component = next(iter([field for field, name in fields if name == "Size-2"] if fields else []), None) + + if binding.source_image_component and destination_image_component: + if isinstance(binding.source_image_component, gr.Gallery): + func = send_image_and_dimensions if destination_width_component else image_from_url_text + jsfunc = "extract_image_from_gallery" + else: + func = send_image_and_dimensions if destination_width_component else lambda x: x + jsfunc = None + + binding.paste_button.click( + fn=func, + _js=jsfunc, + inputs=[binding.source_image_component], + outputs=[destination_image_component, destination_width_component, destination_height_component] if destination_width_component else [destination_image_component], + show_progress=False, + ) + + if binding.source_text_component is not None and fields is not None: + connect_paste(binding.paste_button, fields, binding.source_text_component, override_settings_component, binding.tabname) + + if binding.source_tabname is not None and fields is not None: + paste_field_names = ['Prompt', 'Negative prompt', 'Steps', 'Face restoration'] + (["Seed"] if shared.opts.send_seed else []) + binding.paste_field_names + binding.paste_button.click( + fn=lambda *x: x, + inputs=[field for field, name in paste_fields[binding.source_tabname]["fields"] if name in paste_field_names], + outputs=[field for field, name in fields if name in paste_field_names], + show_progress=False, + ) + + binding.paste_button.click( + fn=None, + _js=f"switch_to_{binding.tabname}", + inputs=None, + outputs=None, + show_progress=False, + ) + + +def send_image_and_dimensions(x): + if isinstance(x, Image.Image): + img = x + else: + img = image_from_url_text(x) + + if shared.opts.send_size and isinstance(img, Image.Image): + w = img.width + h = img.height + else: + w = gr.update() + h = gr.update() + + return img, w, h + + +def restore_old_hires_fix_params(res): + """for infotexts that specify old First pass size parameter, convert it into + width, height, and hr scale""" + + firstpass_width = res.get('First pass size-1', None) + firstpass_height = res.get('First pass size-2', None) + + if shared.opts.use_old_hires_fix_width_height: + hires_width = int(res.get("Hires resize-1", 0)) + hires_height = int(res.get("Hires resize-2", 0)) + + if hires_width and hires_height: + res['Size-1'] = hires_width + res['Size-2'] = hires_height + return + + if firstpass_width is None or firstpass_height is None: + return + + firstpass_width, firstpass_height = int(firstpass_width), int(firstpass_height) + width = int(res.get("Size-1", 512)) + height = int(res.get("Size-2", 512)) + + if firstpass_width == 0 or firstpass_height == 0: + firstpass_width, firstpass_height = processing.old_hires_fix_first_pass_dimensions(width, height) + + res['Size-1'] = firstpass_width + res['Size-2'] = firstpass_height + res['Hires resize-1'] = width + res['Hires resize-2'] = height + + +def parse_generation_parameters(x: str): + """parses generation parameters string, the one you see in text field under the picture in UI: +``` +girl with an artist's beret, determined, blue eyes, desert scene, computer monitors, heavy makeup, by Alphonse Mucha and Charlie Bowater, ((eyeshadow)), (coquettish), detailed, intricate +Negative prompt: ugly, fat, obese, chubby, (((deformed))), [blurry], bad anatomy, disfigured, poorly drawn face, mutation, mutated, (extra_limb), (ugly), (poorly drawn hands), messy drawing +Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model hash: 45dee52b +``` + + returns a dict with field values + """ + + res = {} + + prompt = "" + negative_prompt = "" + + done_with_prompt = False + + *lines, lastline = x.strip().split("\n") + if len(re_param.findall(lastline)) < 3: + lines.append(lastline) + lastline = '' + + for line in lines: + line = line.strip() + if line.startswith("Negative prompt:"): + done_with_prompt = True + line = line[16:].strip() + if done_with_prompt: + negative_prompt += ("" if negative_prompt == "" else "\n") + line + else: + prompt += ("" if prompt == "" else "\n") + line + + if shared.opts.infotext_styles != "Ignore": + found_styles, prompt, negative_prompt = shared.prompt_styles.extract_styles_from_prompt(prompt, negative_prompt) + + if shared.opts.infotext_styles == "Apply": + res["Styles array"] = found_styles + elif shared.opts.infotext_styles == "Apply if any" and found_styles: + res["Styles array"] = found_styles + + res["Prompt"] = prompt + res["Negative prompt"] = negative_prompt + + for k, v in re_param.findall(lastline): + try: + if v[0] == '"' and v[-1] == '"': + v = unquote(v) + + m = re_imagesize.match(v) + if m is not None: + res[f"{k}-1"] = m.group(1) + res[f"{k}-2"] = m.group(2) + else: + res[k] = v + except Exception: + print(f"Error parsing \"{k}: {v}\"") + + # Missing CLIP skip means it was set to 1 (the default) + if "Clip skip" not in res: + res["Clip skip"] = "1" + + hypernet = res.get("Hypernet", None) + if hypernet is not None: + res["Prompt"] += f"""""" + + if "Hires resize-1" not in res: + res["Hires resize-1"] = 0 + res["Hires resize-2"] = 0 + + if "Hires sampler" not in res: + res["Hires sampler"] = "Use same sampler" + + if "Hires checkpoint" not in res: + res["Hires checkpoint"] = "Use same checkpoint" + + if "Hires prompt" not in res: + res["Hires prompt"] = "" + + if "Hires negative prompt" not in res: + res["Hires negative prompt"] = "" + + restore_old_hires_fix_params(res) + + # Missing RNG means the default was set, which is GPU RNG + if "RNG" not in res: + res["RNG"] = "GPU" + + if "Schedule type" not in res: + res["Schedule type"] = "Automatic" + + if "Schedule max sigma" not in res: + res["Schedule max sigma"] = 0 + + if "Schedule min sigma" not in res: + res["Schedule min sigma"] = 0 + + if "Schedule rho" not in res: + res["Schedule rho"] = 0 + + if "VAE Encoder" not in res: + res["VAE Encoder"] = "Full" + + if "VAE Decoder" not in res: + res["VAE Decoder"] = "Full" + + if "FP8 weight" not in res: + res["FP8 weight"] = "Disable" + + if "Cache FP16 weight for LoRA" not in res and res["FP8 weight"] != "Disable": + res["Cache FP16 weight for LoRA"] = False + + infotext_versions.backcompat(res) + + skip = set(shared.opts.infotext_skip_pasting) + res = {k: v for k, v in res.items() if k not in skip} + + return res + + +infotext_to_setting_name_mapping = [ + +] +"""Mapping of infotext labels to setting names. Only left for backwards compatibility - use OptionInfo(..., infotext='...') instead. +Example content: + +infotext_to_setting_name_mapping = [ + ('Conditional mask weight', 'inpainting_mask_weight'), + ('Model hash', 'sd_model_checkpoint'), + ('ENSD', 'eta_noise_seed_delta'), + ('Schedule type', 'k_sched_type'), +] +""" + + +def create_override_settings_dict(text_pairs): + """creates processing's override_settings parameters from gradio's multiselect + + Example input: + ['Clip skip: 2', 'Model hash: e6e99610c4', 'ENSD: 31337'] + + Example output: + {'CLIP_stop_at_last_layers': 2, 'sd_model_checkpoint': 'e6e99610c4', 'eta_noise_seed_delta': 31337} + """ + + res = {} + + params = {} + for pair in text_pairs: + k, v = pair.split(":", maxsplit=1) + + params[k] = v.strip() + + mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] + for param_name, setting_name in mapping + infotext_to_setting_name_mapping: + value = params.get(param_name, None) + + if value is None: + continue + + res[setting_name] = shared.opts.cast_value(setting_name, value) + + return res + + +def get_override_settings(params, *, skip_fields=None): + """Returns a list of settings overrides from the infotext parameters dictionary. + + This function checks the `params` dictionary for any keys that correspond to settings in `shared.opts` and returns + a list of tuples containing the parameter name, setting name, and new value cast to correct type. + + It checks for conditions before adding an override: + - ignores settings that match the current value + - ignores parameter keys present in skip_fields argument. + + Example input: + {"Clip skip": "2"} + + Example output: + [("Clip skip", "CLIP_stop_at_last_layers", 2)] + """ + + res = [] + + mapping = [(info.infotext, k) for k, info in shared.opts.data_labels.items() if info.infotext] + for param_name, setting_name in mapping + infotext_to_setting_name_mapping: + if param_name in (skip_fields or {}): + continue + + v = params.get(param_name, None) + if v is None: + continue + + if setting_name == "sd_model_checkpoint" and shared.opts.disable_weights_auto_swap: + continue + + v = shared.opts.cast_value(setting_name, v) + current_value = getattr(shared.opts, setting_name, None) + + if v == current_value: + continue + + res.append((param_name, setting_name, v)) + + return res + + +def connect_paste(button, paste_fields, input_comp, override_settings_component, tabname): + def paste_func(prompt): + if not prompt and not shared.cmd_opts.hide_ui_dir_config: + filename = os.path.join(data_path, "params.txt") + if os.path.exists(filename): + with open(filename, "r", encoding="utf8") as file: + prompt = file.read() + + params = parse_generation_parameters(prompt) + script_callbacks.infotext_pasted_callback(prompt, params) + res = [] + + for output, key in paste_fields: + if callable(key): + v = key(params) + else: + v = params.get(key, None) + + if v is None: + res.append(gr.update()) + elif isinstance(v, type_of_gr_update): + res.append(v) + else: + try: + valtype = type(output.value) + + if valtype == bool and v == "False": + val = False + else: + val = valtype(v) + + res.append(gr.update(value=val)) + except Exception: + res.append(gr.update()) + + return res + + if override_settings_component is not None: + already_handled_fields = {key: 1 for _, key in paste_fields} + + def paste_settings(params): + vals = get_override_settings(params, skip_fields=already_handled_fields) + + vals_pairs = [f"{infotext_text}: {value}" for infotext_text, setting_name, value in vals] + + return gr.Dropdown.update(value=vals_pairs, choices=vals_pairs, visible=bool(vals_pairs)) + + paste_fields = paste_fields + [(override_settings_component, paste_settings)] + + button.click( + fn=paste_func, + inputs=[input_comp], + outputs=[x[0] for x in paste_fields], + show_progress=False, + ) + button.click( + fn=None, + _js=f"recalculate_prompts_{tabname}", + inputs=[], + outputs=[], + show_progress=False, + ) + diff --git a/modules/postprocessing.py b/modules/postprocessing.py index facea899..7850328f 100644 --- a/modules/postprocessing.py +++ b/modules/postprocessing.py @@ -2,7 +2,7 @@ import os from PIL import Image -from modules import shared, images, devices, scripts, scripts_postprocessing, ui_common +from modules import shared, images, devices, scripts, scripts_postprocessing, ui_common, infotext_utils from modules.shared import opts @@ -86,7 +86,7 @@ def run_postprocessing(extras_mode, image, image_folder, input_dir, output_dir, basename = '' forced_filename = None - infotext = ", ".join([k if k == v else f'{k}: {infotext.quote(v)}' for k, v in pp.info.items() if v is not None]) + infotext = ", ".join([k if k == v else f'{k}: {infotext_utils.quote(v)}' for k, v in pp.info.items() if v is not None]) if opts.enable_pnginfo: pp.image.info = existing_pnginfo diff --git a/modules/processing.py b/modules/processing.py index f55b85ed..213a2879 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -16,7 +16,7 @@ from skimage import exposure from typing import Any import modules.sd_hijack -from modules import devices, prompt_parser, masking, sd_samplers, lowvram, infotext, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors, rng +from modules import devices, prompt_parser, masking, sd_samplers, lowvram, infotext_utils, extra_networks, sd_vae_approx, scripts, sd_samplers_common, sd_unet, errors, rng from modules.rng import slerp # noqa: F401 from modules.sd_hijack import model_hijack from modules.sd_samplers_common import images_tensor_to_samples, decode_first_stage, approximation_indexes @@ -746,7 +746,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "User": p.user if opts.add_user_name_to_info else None, } - generation_params_text = ", ".join([k if k == v else f'{k}: {infotext.quote(v)}' for k, v in generation_params.items() if v is not None]) + generation_params_text = ", ".join([k if k == v else f'{k}: {infotext_utils.quote(v)}' for k, v in generation_params.items() if v is not None]) prompt_text = p.main_prompt if use_main_prompt else all_prompts[index] negative_prompt_text = f"\nNegative prompt: {p.main_negative_prompt if use_main_prompt else all_negative_prompts[index]}" if all_negative_prompts[index] else "" diff --git a/modules/processing_scripts/refiner.py b/modules/processing_scripts/refiner.py index e9941413..ba33d8a4 100644 --- a/modules/processing_scripts/refiner.py +++ b/modules/processing_scripts/refiner.py @@ -1,7 +1,7 @@ import gradio as gr from modules import scripts, sd_models -from modules.infotext import PasteField +from modules.infotext_utils import PasteField from modules.ui_common import create_refresh_button from modules.ui_components import InputAccordion diff --git a/modules/processing_scripts/seed.py b/modules/processing_scripts/seed.py index 60293278..2d3cbb97 100644 --- a/modules/processing_scripts/seed.py +++ b/modules/processing_scripts/seed.py @@ -3,7 +3,7 @@ import json import gradio as gr from modules import scripts, ui, errors -from modules.infotext import PasteField +from modules.infotext_utils import PasteField from modules.shared import cmd_opts from modules.ui_components import ToolButton diff --git a/modules/shared_items.py b/modules/shared_items.py index e1392472..13fb2814 100644 --- a/modules/shared_items.py +++ b/modules/shared_items.py @@ -67,14 +67,14 @@ def reload_hypernetworks(): def get_infotext_names(): - from modules import infotext, shared + from modules import infotext_utils, shared res = {} for info in shared.opts.data_labels.values(): if info.infotext: res[info.infotext] = 1 - for tab_data in infotext.paste_fields.values(): + for tab_data in infotext_utils.paste_fields.values(): for _, name in tab_data.get("fields") or []: if isinstance(name, str): res[name] = 1 diff --git a/modules/txt2img.py b/modules/txt2img.py index 3a481915..49660e89 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -2,7 +2,7 @@ from contextlib import closing import modules.scripts from modules import processing -from modules.infotext import create_override_settings_dict +from modules.infotext_utils import create_override_settings_dict from modules.shared import opts import modules.shared as shared from modules.ui import plaintext_to_html diff --git a/modules/ui.py b/modules/ui.py index 378529c7..52b15646 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -21,14 +21,14 @@ from modules.ui_gradio_extensions import reload_javascript from modules.shared import opts, cmd_opts -import modules.infotext as parameters_copypaste +import modules.infotext_utils as parameters_copypaste import modules.hypernetworks.ui as hypernetworks_ui import modules.textual_inversion.ui as textual_inversion_ui import modules.textual_inversion.textual_inversion as textual_inversion import modules.shared as shared from modules import prompt_parser from modules.sd_hijack import model_hijack -from modules.infotext import image_from_url_text, PasteField +from modules.infotext_utils import image_from_url_text, PasteField create_setting_component = ui_settings.create_setting_component diff --git a/modules/ui_common.py b/modules/ui_common.py index fd32676f..f48ad426 100644 --- a/modules/ui_common.py +++ b/modules/ui_common.py @@ -8,10 +8,10 @@ import gradio as gr import subprocess as sp from modules import call_queue, shared -from modules.infotext import image_from_url_text +from modules.infotext_utils import image_from_url_text import modules.images from modules.ui_components import ToolButton -import modules.infotext as parameters_copypaste +import modules.infotext_utils as parameters_copypaste folder_symbol = '\U0001f4c2' # 📂 refresh_symbol = '\U0001f504' # 🔄 diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index 790af135..beea1316 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -10,7 +10,7 @@ import json import html from fastapi.exceptions import HTTPException -from modules.infotext import image_from_url_text +from modules.infotext_utils import image_from_url_text from modules.ui_components import ToolButton extra_pages = [] diff --git a/modules/ui_extra_networks_user_metadata.py b/modules/ui_extra_networks_user_metadata.py index 87aeb6f3..989a649b 100644 --- a/modules/ui_extra_networks_user_metadata.py +++ b/modules/ui_extra_networks_user_metadata.py @@ -5,7 +5,7 @@ import os.path import gradio as gr -from modules import infotext, images, sysinfo, errors, ui_extra_networks +from modules import infotext_utils, images, sysinfo, errors, ui_extra_networks class UserMetadataEditor: @@ -181,7 +181,7 @@ class UserMetadataEditor: index = len(gallery) - 1 if index >= len(gallery) else index img_info = gallery[index if index >= 0 else 0] - image = infotext.image_from_url_text(img_info) + image = infotext_utils.image_from_url_text(img_info) geninfo, items = images.read_info_from_image(image) images.save_image_with_geninfo(image, geninfo, item["local_preview"]) diff --git a/modules/ui_postprocessing.py b/modules/ui_postprocessing.py index b74a1532..1edb68c5 100644 --- a/modules/ui_postprocessing.py +++ b/modules/ui_postprocessing.py @@ -1,6 +1,6 @@ import gradio as gr from modules import scripts, shared, ui_common, postprocessing, call_queue, ui_toprow -import modules.infotext as parameters_copypaste +import modules.infotext_utils as parameters_copypaste def create_ui(): -- cgit v1.2.1 From 501993ebf210bf3b55173ec1910f0c84c7e75424 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Mon, 1 Jan 2024 19:31:06 +0300 Subject: added a button to run hires fix on selected image in the gallery --- modules/processing.py | 46 ++++++++++++++---- modules/txt2img.py | 19 +++++++- modules/ui.py | 108 +++++++++++++++++++++++-------------------- modules/ui_common.py | 57 ++++++++++++++--------- modules/ui_postprocessing.py | 8 ++-- 5 files changed, 152 insertions(+), 86 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 213a2879..045c7d79 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -179,6 +179,7 @@ class StableDiffusionProcessing: token_merging_ratio = 0 token_merging_ratio_hr = 0 disable_extra_networks: bool = False + firstpass_image: Image = None scripts_value: scripts.ScriptRunner = field(default=None, init=False) script_args_value: list = field(default=None, init=False) @@ -1238,18 +1239,45 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def sample(self, conditioning, unconditional_conditioning, seeds, subseeds, subseed_strength, prompts): self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model) - x = self.rng.next() - samples = self.sampler.sample(self, x, conditioning, unconditional_conditioning, image_conditioning=self.txt2img_image_conditioning(x)) - del x + if self.firstpass_image is not None and self.enable_hr: + # here we don't need to generate image, we just take self.firstpass_image and prepare it for hires fix - if not self.enable_hr: - return samples - devices.torch_gc() + if self.latent_scale_mode is None: + image = np.array(self.firstpass_image).astype(np.float32) / 255.0 * 2.0 - 1.0 + image = np.moveaxis(image, 2, 0) + + samples = None + decoded_samples = torch.asarray(np.expand_dims(image, 0)) + + else: + image = np.array(self.firstpass_image).astype(np.float32) / 255.0 + image = np.moveaxis(image, 2, 0) + image = torch.from_numpy(np.expand_dims(image, axis=0)) + image = image.to(shared.device, dtype=devices.dtype_vae) + + if opts.sd_vae_encode_method != 'Full': + self.extra_generation_params['VAE Encoder'] = opts.sd_vae_encode_method + + samples = images_tensor_to_samples(image, approximation_indexes.get(opts.sd_vae_encode_method), self.sd_model) + decoded_samples = None + devices.torch_gc() - if self.latent_scale_mode is None: - decoded_samples = torch.stack(decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True)).to(dtype=torch.float32) else: - decoded_samples = None + # here we generate an image normally + + x = self.rng.next() + samples = self.sampler.sample(self, x, conditioning, unconditional_conditioning, image_conditioning=self.txt2img_image_conditioning(x)) + del x + + if not self.enable_hr: + return samples + + devices.torch_gc() + + if self.latent_scale_mode is None: + decoded_samples = torch.stack(decode_latent_batch(self.sd_model, samples, target_device=devices.cpu, check_for_nans=True)).to(dtype=torch.float32) + else: + decoded_samples = None with sd_models.SkipWritingToConfig(): sd_models.reload_model_weights(info=self.hr_checkpoint_info) diff --git a/modules/txt2img.py b/modules/txt2img.py index 49660e89..4a6fe72a 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -1,7 +1,7 @@ from contextlib import closing import modules.scripts -from modules import processing +from modules import processing, infotext_utils from modules.infotext_utils import create_override_settings_dict from modules.shared import opts import modules.shared as shared @@ -9,9 +9,23 @@ from modules.ui import plaintext_to_html import gradio as gr -def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_name: str, n_iter: int, batch_size: int, cfg_scale: float, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, hr_checkpoint_name: str, hr_sampler_name: str, hr_prompt: str, hr_negative_prompt, override_settings_texts, request: gr.Request, *args): +def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, *args): + assert len(gallery) > 0, 'No image to upscale' + + image_info = gallery[gallery_index] if 0 <= gallery_index < len(gallery) else gallery[0] + image = infotext_utils.image_from_url_text(image_info) + + return txt2img(id_task, request, *args, firstpass_image=image) + + +def txt2img(id_task: str, request: gr.Request, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_name: str, n_iter: int, batch_size: int, cfg_scale: float, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, hr_checkpoint_name: str, hr_sampler_name: str, hr_prompt: str, hr_negative_prompt, override_settings_texts, *args, firstpass_image=None): override_settings = create_override_settings_dict(override_settings_texts) + if firstpass_image is not None: + enable_hr = True + batch_size = 1 + n_iter = 1 + p = processing.StableDiffusionProcessingTxt2Img( sd_model=shared.sd_model, outpath_samples=opts.outdir_samples or opts.outdir_txt2img_samples, @@ -38,6 +52,7 @@ def txt2img(id_task: str, prompt: str, negative_prompt: str, prompt_styles, step hr_prompt=hr_prompt, hr_negative_prompt=hr_negative_prompt, override_settings=override_settings, + firstpass_image=firstpass_image, ) p.scripts = modules.scripts.scripts_txt2img diff --git a/modules/ui.py b/modules/ui.py index 52b15646..3d548430 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -375,50 +375,60 @@ def create_ui(): show_progress=False, ) - txt2img_gallery, generation_info, html_info, html_log = create_output_panel("txt2img", opts.outdir_txt2img_samples, toprow) + output_panel = create_output_panel("txt2img", opts.outdir_txt2img_samples, toprow) + + txt2img_inputs = [ + dummy_component, + toprow.prompt, + toprow.negative_prompt, + toprow.ui_styles.dropdown, + steps, + sampler_name, + batch_count, + batch_size, + cfg_scale, + height, + width, + enable_hr, + denoising_strength, + hr_scale, + hr_upscaler, + hr_second_pass_steps, + hr_resize_x, + hr_resize_y, + hr_checkpoint_name, + hr_sampler_name, + hr_prompt, + hr_negative_prompt, + override_settings, + ] + custom_inputs + + txt2img_outputs = [ + output_panel.gallery, + output_panel.infotext, + output_panel.html_info, + output_panel.html_log, + ] txt2img_args = dict( fn=wrap_gradio_gpu_call(modules.txt2img.txt2img, extra_outputs=[None, '', '']), _js="submit", - inputs=[ - dummy_component, - toprow.prompt, - toprow.negative_prompt, - toprow.ui_styles.dropdown, - steps, - sampler_name, - batch_count, - batch_size, - cfg_scale, - height, - width, - enable_hr, - denoising_strength, - hr_scale, - hr_upscaler, - hr_second_pass_steps, - hr_resize_x, - hr_resize_y, - hr_checkpoint_name, - hr_sampler_name, - hr_prompt, - hr_negative_prompt, - override_settings, - - ] + custom_inputs, - - outputs=[ - txt2img_gallery, - generation_info, - html_info, - html_log, - ], + inputs=txt2img_inputs, + outputs=txt2img_outputs, show_progress=False, ) toprow.prompt.submit(**txt2img_args) toprow.submit.click(**txt2img_args) + output_panel.button_upscale.click( + fn=wrap_gradio_gpu_call(modules.txt2img.txt2img_upscale, extra_outputs=[None, '', '']), + _js="submit_txt2img_upscale", + inputs=txt2img_inputs[0:1] + [output_panel.gallery, dummy_component] + txt2img_inputs[1:], + outputs=txt2img_outputs, + show_progress=False, + ) + res_switch_btn.click(fn=None, _js="function(){switchWidthHeight('txt2img')}", inputs=None, outputs=None, show_progress=False) toprow.restore_progress_button.click( @@ -426,10 +436,10 @@ def create_ui(): _js="restoreProgressTxt2img", inputs=[dummy_component], outputs=[ - txt2img_gallery, - generation_info, - html_info, - html_log, + output_panel.gallery, + output_panel.infotext, + output_panel.html_info, + output_panel.html_log, ], show_progress=False, ) @@ -479,7 +489,7 @@ def create_ui(): toprow.negative_token_button.click(fn=wrap_queued_call(update_negative_prompt_token_counter), inputs=[toprow.negative_prompt, steps], outputs=[toprow.negative_token_counter]) extra_networks_ui = ui_extra_networks.create_ui(txt2img_interface, [txt2img_generation_tab], 'txt2img') - ui_extra_networks.setup_ui(extra_networks_ui, txt2img_gallery) + ui_extra_networks.setup_ui(extra_networks_ui, output_panel.gallery) extra_tabs.__exit__() @@ -710,7 +720,7 @@ def create_ui(): outputs=[inpaint_controls, mask_alpha], ) - img2img_gallery, generation_info, html_info, html_log = create_output_panel("img2img", opts.outdir_img2img_samples, toprow) + output_panel = create_output_panel("img2img", opts.outdir_img2img_samples, toprow) img2img_args = dict( fn=wrap_gradio_gpu_call(modules.img2img.img2img, extra_outputs=[None, '', '']), @@ -755,10 +765,10 @@ def create_ui(): img2img_batch_png_info_dir, ] + custom_inputs, outputs=[ - img2img_gallery, - generation_info, - html_info, - html_log, + output_panel.gallery, + output_panel.infotext, + output_panel.html_info, + output_panel.html_log, ], show_progress=False, ) @@ -796,10 +806,10 @@ def create_ui(): _js="restoreProgressImg2img", inputs=[dummy_component], outputs=[ - img2img_gallery, - generation_info, - html_info, - html_log, + output_panel.gallery, + output_panel.infotext, + output_panel.html_info, + output_panel.html_log, ], show_progress=False, ) @@ -839,7 +849,7 @@ def create_ui(): )) extra_networks_ui_img2img = ui_extra_networks.create_ui(img2img_interface, [img2img_generation_tab], 'img2img') - ui_extra_networks.setup_ui(extra_networks_ui_img2img, img2img_gallery) + ui_extra_networks.setup_ui(extra_networks_ui_img2img, output_panel.gallery) extra_tabs.__exit__() diff --git a/modules/ui_common.py b/modules/ui_common.py index f48ad426..ff84197c 100644 --- a/modules/ui_common.py +++ b/modules/ui_common.py @@ -1,3 +1,4 @@ +import dataclasses import json import html import os @@ -104,7 +105,17 @@ def save_files(js_data, images, do_make_zip, index): return gr.File.update(value=fullfns, visible=True), plaintext_to_html(f"Saved: {filenames[0]}") +@dataclasses.dataclass +class OutputPanel: + gallery = None + infotext = None + html_info = None + html_log = None + button_upscale = None + + def create_output_panel(tabname, outdir, toprow=None): + res = OutputPanel() def open_folder(f): if not os.path.exists(f): @@ -136,9 +147,8 @@ Requested path was: {f} with gr.Column(variant='panel', elem_id=f"{tabname}_results_panel"): with gr.Group(elem_id=f"{tabname}_gallery_container"): - result_gallery = gr.Gallery(label='Output', show_label=False, elem_id=f"{tabname}_gallery", columns=4, preview=True, height=shared.opts.gallery_height or None) + res.gallery = gr.Gallery(label='Output', show_label=False, elem_id=f"{tabname}_gallery", columns=4, preview=True, height=shared.opts.gallery_height or None) - generation_info = None with gr.Row(elem_id=f"image_buttons_{tabname}", elem_classes="image-buttons"): open_folder_button = ToolButton(folder_symbol, elem_id=f'{tabname}_open_folder', visible=not shared.cmd_opts.hide_ui_dir_config, tooltip="Open images output directory.") @@ -152,6 +162,9 @@ Requested path was: {f} 'extras': ToolButton('📐', elem_id=f'{tabname}_send_to_extras', tooltip="Send image and generation parameters to extras tab.") } + if tabname == 'txt2img': + res.button_upscale = ToolButton('✨', elem_id=f'{tabname}_upscale', tooltip="Create an upscaled version of the current image using hires fix settings.") + open_folder_button.click( fn=lambda: open_folder(shared.opts.outdir_samples or outdir), inputs=[], @@ -162,17 +175,17 @@ Requested path was: {f} download_files = gr.File(None, file_count="multiple", interactive=False, show_label=False, visible=False, elem_id=f'download_files_{tabname}') with gr.Group(): - html_info = gr.HTML(elem_id=f'html_info_{tabname}', elem_classes="infotext") - html_log = gr.HTML(elem_id=f'html_log_{tabname}', elem_classes="html-log") + res.html_info = gr.HTML(elem_id=f'html_info_{tabname}', elem_classes="infotext") + res.html_log = gr.HTML(elem_id=f'html_log_{tabname}', elem_classes="html-log") - generation_info = gr.Textbox(visible=False, elem_id=f'generation_info_{tabname}') + res.infotext = gr.Textbox(visible=False, elem_id=f'generation_info_{tabname}') if tabname == 'txt2img' or tabname == 'img2img': generation_info_button = gr.Button(visible=False, elem_id=f"{tabname}_generation_info_button") generation_info_button.click( fn=update_generation_info, _js="function(x, y, z){ return [x, y, selected_gallery_index()] }", - inputs=[generation_info, html_info, html_info], - outputs=[html_info, html_info], + inputs=[res.infotext, res.html_info, res.html_info], + outputs=[res.html_info, res.html_info], show_progress=False, ) @@ -180,14 +193,14 @@ Requested path was: {f} fn=call_queue.wrap_gradio_call(save_files), _js="(x, y, z, w) => [x, y, false, selected_gallery_index()]", inputs=[ - generation_info, - result_gallery, - html_info, - html_info, + res.infotext, + res.gallery, + res.html_info, + res.html_info, ], outputs=[ download_files, - html_log, + res.html_log, ], show_progress=False, ) @@ -196,21 +209,21 @@ Requested path was: {f} fn=call_queue.wrap_gradio_call(save_files), _js="(x, y, z, w) => [x, y, true, selected_gallery_index()]", inputs=[ - generation_info, - result_gallery, - html_info, - html_info, + res.infotext, + res.gallery, + res.html_info, + res.html_info, ], outputs=[ download_files, - html_log, + res.html_log, ] ) else: - html_info_x = gr.HTML(elem_id=f'html_info_x_{tabname}') - html_info = gr.HTML(elem_id=f'html_info_{tabname}', elem_classes="infotext") - html_log = gr.HTML(elem_id=f'html_log_{tabname}') + res.infotext = gr.HTML(elem_id=f'html_info_x_{tabname}') + res.html_info = gr.HTML(elem_id=f'html_info_{tabname}', elem_classes="infotext") + res.html_log = gr.HTML(elem_id=f'html_log_{tabname}') paste_field_names = [] if tabname == "txt2img": @@ -220,11 +233,11 @@ Requested path was: {f} for paste_tabname, paste_button in buttons.items(): parameters_copypaste.register_paste_params_button(parameters_copypaste.ParamBinding( - paste_button=paste_button, tabname=paste_tabname, source_tabname="txt2img" if tabname == "txt2img" else None, source_image_component=result_gallery, + paste_button=paste_button, tabname=paste_tabname, source_tabname="txt2img" if tabname == "txt2img" else None, source_image_component=res.gallery, paste_field_names=paste_field_names )) - return result_gallery, generation_info if tabname != "extras" else html_info_x, html_info, html_log + return res def create_refresh_button(refresh_component, refresh_method, refreshed_args, elem_id): diff --git a/modules/ui_postprocessing.py b/modules/ui_postprocessing.py index 1edb68c5..8f09e658 100644 --- a/modules/ui_postprocessing.py +++ b/modules/ui_postprocessing.py @@ -28,7 +28,7 @@ def create_ui(): toprow.create_inline_toprow_image() submit = toprow.submit - result_images, html_info_x, html_info, html_log = ui_common.create_output_panel("extras", shared.opts.outdir_extras_samples) + output_panel = ui_common.create_output_panel("extras", shared.opts.outdir_extras_samples) tab_single.select(fn=lambda: 0, inputs=[], outputs=[tab_index]) tab_batch.select(fn=lambda: 1, inputs=[], outputs=[tab_index]) @@ -48,9 +48,9 @@ def create_ui(): *script_inputs ], outputs=[ - result_images, - html_info_x, - html_log, + output_panel.gallery, + output_panel.infotext, + output_panel.html_log, ], show_progress=False, ) -- cgit v1.2.1 From c2ea571005dab29b285e31a0ad4a97258360bf2d Mon Sep 17 00:00:00 2001 From: Jibaku789 <151478027+Jibaku789@users.noreply.github.com> Date: Mon, 1 Jan 2024 14:57:41 -0600 Subject: Add inpaint options to paste fields --- modules/ui.py | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index 3d548430..4cdb0e9c 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -840,6 +840,10 @@ def create_ui(): (toprow.ui_styles.dropdown, lambda d: d["Styles array"] if isinstance(d.get("Styles array"), list) else gr.update()), (denoising_strength, "Denoising strength"), (mask_blur, "Mask blur"), + (inpainting_mask_invert, 'Mask mode'), + (inpainting_fill, 'Masked content'), + (inpaint_full_res, 'Inpaint area'), + (inpaint_full_res_padding, 'Only masked padding, pixels'), *scripts.scripts_img2img.infotext_fields ] parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields, override_settings) -- cgit v1.2.1 From a5b6a5a3adcc845237d872750ded34240cc6a810 Mon Sep 17 00:00:00 2001 From: Jibaku789 <151478027+Jibaku789@users.noreply.github.com> Date: Mon, 1 Jan 2024 14:58:55 -0600 Subject: Add inpaint options to img2img.py --- modules/img2img.py | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'modules') diff --git a/modules/img2img.py b/modules/img2img.py index 04de8e62..9e09c0a0 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -225,6 +225,18 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s if mask: p.extra_generation_params["Mask blur"] = mask_blur + if inpainting_mask_invert is not None: + p.extra_generation_params["Mask mode"] = inpainting_mask_invert + + if inpainting_fill is not None: + p.extra_generation_params["Masked content"] = inpainting_fill + + if inpaint_full_res is not None: + p.extra_generation_params["Inpaint area"] = inpaint_full_res + + if inpaint_full_res_padding is not None: + p.extra_generation_params["Only masked padding, pixels"] = inpaint_full_res_padding + with closing(p): if is_batch: assert not shared.cmd_opts.hide_ui_dir_config, "Launched with --hide-ui-dir-config, batch img2img disabled" -- cgit v1.2.1 From 1341b2208185cd89b0019bda2df63b406ec0cb5e Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 2 Jan 2024 06:47:26 +0300 Subject: add an option to hide upscaling progressbar --- modules/shared_options.py | 1 + modules/upscaler_utils.py | 4 ++-- 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index cca3f7be..63488f4e 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -115,6 +115,7 @@ options_templates.update(options_section(('system', "System", "system"), { "memmon_poll_rate": OptionInfo(8, "VRAM usage polls per second during generation.", gr.Slider, {"minimum": 0, "maximum": 40, "step": 1}).info("0 = disable"), "samples_log_stdout": OptionInfo(False, "Always print all generation info to standard output"), "multiple_tqdm": OptionInfo(True, "Add a second progress bar to the console that shows progress for an entire job."), + "enable_upscale_progressbar": OptionInfo(True, "Show a progress bar in the console for tiled upscaling."), "print_hypernet_extra": OptionInfo(False, "Print extra hypernetwork information to console."), "list_hidden_files": OptionInfo(True, "Load models/files in hidden directories").info("directory is hidden if its name starts with \".\""), "disable_mmap_load_safetensors": OptionInfo(False, "Disable memmapping for loading .safetensors files.").info("fixes very slow loading speed in some cases"), diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index f5cb92d5..9379f512 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -47,7 +47,7 @@ def upscale_with_model( grid = images.split_grid(img, tile_size, tile_size, tile_overlap) newtiles = [] - with tqdm.tqdm(total=grid.tile_count, desc=desc) as p: + with tqdm.tqdm(total=grid.tile_count, desc=desc, disable=not shared.opts.enable_upscale_progressbar) as p: for y, h, row in grid.tiles: newrow = [] for x, w, tile in row: @@ -103,7 +103,7 @@ def tiled_upscale_2( ).type_as(img) weights = torch.zeros_like(result) logger.debug("Upscaling %s to %s with tiles", img.shape, result.shape) - with tqdm.tqdm(total=len(h_idx_list) * len(w_idx_list), desc=desc) as pbar: + with tqdm.tqdm(total=len(h_idx_list) * len(w_idx_list), desc=desc, disable=not shared.opts.enable_upscale_progressbar) as pbar: for h_idx in h_idx_list: if shared.state.interrupted or shared.state.skipped: break -- cgit v1.2.1 From 80873b1538e6ca0c7ebe558f8ce4213b06fd8307 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 2 Jan 2024 07:05:05 +0300 Subject: fix #14497 --- modules/img2img.py | 15 --------------- modules/infotext_utils.py | 12 ++++++++++++ modules/processing.py | 13 +++++++++++++ 3 files changed, 25 insertions(+), 15 deletions(-) (limited to 'modules') diff --git a/modules/img2img.py b/modules/img2img.py index 9e09c0a0..f81405df 100644 --- a/modules/img2img.py +++ b/modules/img2img.py @@ -222,21 +222,6 @@ def img2img(id_task: str, mode: int, prompt: str, negative_prompt: str, prompt_s if shared.opts.enable_console_prompts: print(f"\nimg2img: {prompt}", file=shared.progress_print_out) - if mask: - p.extra_generation_params["Mask blur"] = mask_blur - - if inpainting_mask_invert is not None: - p.extra_generation_params["Mask mode"] = inpainting_mask_invert - - if inpainting_fill is not None: - p.extra_generation_params["Masked content"] = inpainting_fill - - if inpaint_full_res is not None: - p.extra_generation_params["Inpaint area"] = inpaint_full_res - - if inpaint_full_res_padding is not None: - p.extra_generation_params["Only masked padding, pixels"] = inpaint_full_res_padding - with closing(p): if is_batch: assert not shared.cmd_opts.hide_ui_dir_config, "Launched with --hide-ui-dir-config, batch img2img disabled" diff --git a/modules/infotext_utils.py b/modules/infotext_utils.py index 26e9b949..e582ee47 100644 --- a/modules/infotext_utils.py +++ b/modules/infotext_utils.py @@ -312,6 +312,18 @@ Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 965400086, Size: 512x512, Model if "Hires negative prompt" not in res: res["Hires negative prompt"] = "" + if "Mask mode" not in res: + res["Mask mode"] = "Inpaint masked" + + if "Masked content" not in res: + res["Masked content"] = 'original' + + if "Inpaint area" not in res: + res["Inpaint area"] = "Whole picture" + + if "Masked area padding" not in res: + res["Masked area padding"] = 32 + restore_old_hires_fix_params(res) # Missing RNG means the default was set, which is GPU RNG diff --git a/modules/processing.py b/modules/processing.py index 045c7d79..84e7b1b4 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -1530,6 +1530,7 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): if self.inpainting_mask_invert: image_mask = ImageOps.invert(image_mask) + self.extra_generation_params["Mask mode"] = "Inpaint not masked" if self.mask_blur_x > 0: np_mask = np.array(image_mask) @@ -1543,6 +1544,9 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): np_mask = cv2.GaussianBlur(np_mask, (1, kernel_size), self.mask_blur_y) image_mask = Image.fromarray(np_mask) + if self.mask_blur_x > 0 or self.mask_blur_y > 0: + self.extra_generation_params["Mask blur"] = self.mask_blur + if self.inpaint_full_res: self.mask_for_overlay = image_mask mask = image_mask.convert('L') @@ -1553,6 +1557,9 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): mask = mask.crop(crop_region) image_mask = images.resize_image(2, mask, self.width, self.height) self.paste_to = (x1, y1, x2-x1, y2-y1) + + self.extra_generation_params["Inpaint area"] = "Only masked" + self.extra_generation_params["Masked area padding"] = self.inpaint_full_res_padding else: image_mask = images.resize_image(self.resize_mode, image_mask, self.width, self.height) np_mask = np.array(image_mask) @@ -1594,6 +1601,9 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): if self.inpainting_fill != 1: image = masking.fill(image, latent_mask) + if self.inpainting_fill == 0: + self.extra_generation_params["Masked content"] = 'fill' + if add_color_corrections: self.color_corrections.append(setup_color_correction(image)) @@ -1643,8 +1653,11 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): # this needs to be fixed to be done in sample() using actual seeds for batches if self.inpainting_fill == 2: self.init_latent = self.init_latent * self.mask + create_random_tensors(self.init_latent.shape[1:], all_seeds[0:self.init_latent.shape[0]]) * self.nmask + self.extra_generation_params["Masked content"] = 'latent noise' + elif self.inpainting_fill == 3: self.init_latent = self.init_latent * self.mask + self.extra_generation_params["Masked content"] = 'latent nothing' self.image_conditioning = self.img2img_image_conditioning(image * 2 - 1, self.init_latent, image_mask, self.mask_round) -- cgit v1.2.1 From 980970d39091e572500434c69660bc6eed22498d Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Tue, 2 Jan 2024 07:08:32 +0300 Subject: final touches --- modules/ui.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/ui.py b/modules/ui.py index 4cdb0e9c..7116d71c 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -843,7 +843,7 @@ def create_ui(): (inpainting_mask_invert, 'Mask mode'), (inpainting_fill, 'Masked content'), (inpaint_full_res, 'Inpaint area'), - (inpaint_full_res_padding, 'Only masked padding, pixels'), + (inpaint_full_res_padding, 'Masked area padding'), *scripts.scripts_img2img.infotext_fields ] parameters_copypaste.add_paste_fields("img2img", init_img, img2img_paste_fields, override_settings) -- cgit v1.2.1 From cf14a6a7aaf8ccb40552990785d5c9e400d93610 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sun, 31 Dec 2023 16:11:18 +0200 Subject: Refactor upscale_2 helper out of ScuNET/SwinIR; make sure devices are right --- modules/upscaler_utils.py | 89 ++++++++++++++++++++++++++++++++++++----------- 1 file changed, 69 insertions(+), 20 deletions(-) (limited to 'modules') diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index 9379f512..e4c63f09 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -11,23 +11,40 @@ from modules import images, shared, torch_utils logger = logging.getLogger(__name__) -def upscale_without_tiling(model, img: Image.Image): - img = np.array(img) - img = img[:, :, ::-1] - img = np.ascontiguousarray(np.transpose(img, (2, 0, 1))) / 255 - img = torch.from_numpy(img).float() - +def pil_image_to_torch_bgr(img: Image.Image) -> torch.Tensor: + img = np.array(img.convert("RGB")) + img = img[:, :, ::-1] # flip RGB to BGR + img = np.transpose(img, (2, 0, 1)) # HWC to CHW + img = np.ascontiguousarray(img) / 255 # Rescale to [0, 1] + return torch.from_numpy(img) + + +def torch_bgr_to_pil_image(tensor: torch.Tensor) -> Image.Image: + if tensor.ndim == 4: + # If we're given a tensor with a batch dimension, squeeze it out + # (but only if it's a batch of size 1). + if tensor.shape[0] != 1: + raise ValueError(f"{tensor.shape} does not describe a BCHW tensor") + tensor = tensor.squeeze(0) + assert tensor.ndim == 3, f"{tensor.shape} does not describe a CHW tensor" + # TODO: is `tensor.float().cpu()...numpy()` the most efficient idiom? + arr = tensor.float().cpu().clamp_(0, 1).numpy() # clamp + arr = 255.0 * np.moveaxis(arr, 0, 2) # CHW to HWC, rescale + arr = arr.astype(np.uint8) + arr = arr[:, :, ::-1] # flip BGR to RGB + return Image.fromarray(arr, "RGB") + + +def upscale_pil_patch(model, img: Image.Image) -> Image.Image: + """ + Upscale a given PIL image using the given model. + """ param = torch_utils.get_param(model) - img = img.unsqueeze(0).to(device=param.device, dtype=param.dtype) with torch.no_grad(): - output = model(img) - - output = output.squeeze().float().cpu().clamp_(0, 1).numpy() - output = 255. * np.moveaxis(output, 0, 2) - output = output.astype(np.uint8) - output = output[:, :, ::-1] - return Image.fromarray(output, 'RGB') + tensor = pil_image_to_torch_bgr(img).unsqueeze(0) # add batch dimension + tensor = tensor.to(device=param.device, dtype=param.dtype) + return torch_bgr_to_pil_image(model(tensor)) def upscale_with_model( @@ -40,7 +57,7 @@ def upscale_with_model( ) -> Image.Image: if tile_size <= 0: logger.debug("Upscaling %s without tiling", img) - output = upscale_without_tiling(model, img) + output = upscale_pil_patch(model, img) logger.debug("=> %s", output) return output @@ -52,7 +69,7 @@ def upscale_with_model( newrow = [] for x, w, tile in row: logger.debug("Tile (%d, %d) %s...", x, y, tile) - output = upscale_without_tiling(model, tile) + output = upscale_pil_patch(model, tile) scale_factor = output.width // tile.width logger.debug("=> %s (scale factor %s)", output, scale_factor) newrow.append([x * scale_factor, w * scale_factor, output]) @@ -71,19 +88,22 @@ def upscale_with_model( def tiled_upscale_2( - img, + img: torch.Tensor, model, *, tile_size: int, tile_overlap: int, scale: int, - device, desc="Tiled upscale", ): # Alternative implementation of `upscale_with_model` originally used by # SwinIR and ScuNET. It differs from `upscale_with_model` in that tiling and # weighting is done in PyTorch space, as opposed to `images.Grid` doing it in # Pillow space without weighting. + + # Grab the device the model is on, and use it. + device = torch_utils.get_param(model).device + b, c, h, w = img.size() tile_size = min(tile_size, h, w) @@ -100,7 +120,8 @@ def tiled_upscale_2( h * scale, w * scale, device=device, - ).type_as(img) + dtype=img.dtype, + ) weights = torch.zeros_like(result) logger.debug("Upscaling %s to %s with tiles", img.shape, result.shape) with tqdm.tqdm(total=len(h_idx_list) * len(w_idx_list), desc=desc, disable=not shared.opts.enable_upscale_progressbar) as pbar: @@ -112,11 +133,13 @@ def tiled_upscale_2( if shared.state.interrupted or shared.state.skipped: break + # Only move this patch to the device if it's not already there. in_patch = img[ ..., h_idx : h_idx + tile_size, w_idx : w_idx + tile_size, - ] + ].to(device=device) + out_patch = model(in_patch) result[ @@ -138,3 +161,29 @@ def tiled_upscale_2( output = result.div_(weights) return output + + +def upscale_2( + img: Image.Image, + model, + *, + tile_size: int, + tile_overlap: int, + scale: int, + desc: str, +): + """ + Convenience wrapper around `tiled_upscale_2` that handles PIL images. + """ + tensor = pil_image_to_torch_bgr(img).float().unsqueeze(0) # add batch dimension + + with torch.no_grad(): + output = tiled_upscale_2( + tensor, + model, + tile_size=tile_size, + tile_overlap=tile_overlap, + scale=scale, + desc=desc, + ) + return torch_bgr_to_pil_image(output) -- cgit v1.2.1 From 2cacbc124c49f45da5b66b79d9b0a3ab943472eb Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Sun, 31 Dec 2023 19:52:32 +0200 Subject: load_spandrel_model: make `half` `prefer_half` As discussed with the Spandrel folks, it's good to heed Spandrel's "supports half precision" flag to avoid e.g. black blotches and what-not. --- modules/modelloader.py | 20 ++++++++++++++------ modules/realesrgan_model.py | 2 +- 2 files changed, 15 insertions(+), 7 deletions(-) (limited to 'modules') diff --git a/modules/modelloader.py b/modules/modelloader.py index a7194137..e100bb24 100644 --- a/modules/modelloader.py +++ b/modules/modelloader.py @@ -139,23 +139,31 @@ def load_upscalers(): def load_spandrel_model( - path: str, + path: str | os.PathLike, *, device: str | torch.device | None, - half: bool = False, + prefer_half: bool = False, dtype: str | torch.dtype | None = None, expected_architecture: str | None = None, ) -> spandrel.ModelDescriptor: import spandrel - model_descriptor = spandrel.ModelLoader(device=device).load_from_file(path) + model_descriptor = spandrel.ModelLoader(device=device).load_from_file(str(path)) if expected_architecture and model_descriptor.architecture != expected_architecture: logger.warning( f"Model {path!r} is not a {expected_architecture!r} model (got {model_descriptor.architecture!r})", ) - if half: - model_descriptor.model.half() + half = False + if prefer_half: + if model_descriptor.supports_half: + model_descriptor.model.half() + half = True + else: + logger.info("Model %s does not support half precision, ignoring --half", path) if dtype: model_descriptor.model.to(dtype=dtype) model_descriptor.model.eval() - logger.debug("Loaded %s from %s (device=%s, half=%s, dtype=%s)", model_descriptor, path, device, half, dtype) + logger.debug( + "Loaded %s from %s (device=%s, half=%s, dtype=%s)", + model_descriptor, path, device, half, dtype, + ) return model_descriptor diff --git a/modules/realesrgan_model.py b/modules/realesrgan_model.py index 4d35b695..ff9d8ac0 100644 --- a/modules/realesrgan_model.py +++ b/modules/realesrgan_model.py @@ -39,7 +39,7 @@ class UpscalerRealESRGAN(Upscaler): model_descriptor = modelloader.load_spandrel_model( info.local_data_path, device=self.device, - half=(not cmd_opts.no_half and not cmd_opts.upcast_sampling), + prefer_half=(not cmd_opts.no_half and not cmd_opts.upcast_sampling), expected_architecture="ESRGAN", # "RealESRGAN" isn't a specific thing for Spandrel ) return upscale_with_model( -- cgit v1.2.1 From 7ad6899bf987a8ee615efbcfc99562457f89cd8b Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Tue, 2 Jan 2024 17:14:05 +0200 Subject: torch_bgr_to_pil_image: round, don't truncate This matches what `realesrgan` does. --- modules/upscaler_utils.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index e4c63f09..4f1417cf 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -30,7 +30,7 @@ def torch_bgr_to_pil_image(tensor: torch.Tensor) -> Image.Image: # TODO: is `tensor.float().cpu()...numpy()` the most efficient idiom? arr = tensor.float().cpu().clamp_(0, 1).numpy() # clamp arr = 255.0 * np.moveaxis(arr, 0, 2) # CHW to HWC, rescale - arr = arr.astype(np.uint8) + arr = arr.round().astype(np.uint8) arr = arr[:, :, ::-1] # flip BGR to RGB return Image.fromarray(arr, "RGB") -- cgit v1.2.1 From fccd0b00c2ca17360b7b956cd2e9bd1fb42c017d Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Wed, 3 Jan 2024 18:55:43 +0900 Subject: reduce unnecessary re-indexing extra networks dir --- modules/ui_extra_networks.py | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'modules') diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index beea1316..e1c679ec 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -417,21 +417,21 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): dropdown_sort.change(fn=lambda: None, _js="function(){ applyExtraNetworkSort('" + tabname + "'); }") + def create_html(): + ui.pages_contents = [pg.create_html(ui.tabname) for pg in ui.stored_extra_pages] + def pages_html(): if not ui.pages_contents: - return refresh() - + create_html() return ui.pages_contents def refresh(): for pg in ui.stored_extra_pages: pg.refresh() - - ui.pages_contents = [pg.create_html(ui.tabname) for pg in ui.stored_extra_pages] - + create_html() return ui.pages_contents - interface.load(fn=pages_html, inputs=[], outputs=[*ui.pages]) + interface.load(fn=pages_html, inputs=[], outputs=ui.pages) button_refresh.click(fn=refresh, inputs=[], outputs=ui.pages) return ui -- cgit v1.2.1 From bfc48fbc244130770991fab284f6fedcef2054e7 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Thu, 4 Jan 2024 03:46:05 +0900 Subject: paste infotext cast int as float --- modules/infotext_utils.py | 2 ++ 1 file changed, 2 insertions(+) (limited to 'modules') diff --git a/modules/infotext_utils.py b/modules/infotext_utils.py index e582ee47..a21329e6 100644 --- a/modules/infotext_utils.py +++ b/modules/infotext_utils.py @@ -477,6 +477,8 @@ def connect_paste(button, paste_fields, input_comp, override_settings_component, if valtype == bool and v == "False": val = False + elif valtype == int: + val = float(v) else: val = valtype(v) -- cgit v1.2.1 From 62470ee23443cb2ad3943a152ccae26a689c86e1 Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Wed, 3 Jan 2024 22:39:12 +0200 Subject: upscale_2: cast image to model's dtype --- modules/upscaler_utils.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/upscaler_utils.py b/modules/upscaler_utils.py index e4c63f09..5db74877 100644 --- a/modules/upscaler_utils.py +++ b/modules/upscaler_utils.py @@ -94,6 +94,7 @@ def tiled_upscale_2( tile_size: int, tile_overlap: int, scale: int, + device: torch.device, desc="Tiled upscale", ): # Alternative implementation of `upscale_with_model` originally used by @@ -101,9 +102,6 @@ def tiled_upscale_2( # weighting is done in PyTorch space, as opposed to `images.Grid` doing it in # Pillow space without weighting. - # Grab the device the model is on, and use it. - device = torch_utils.get_param(model).device - b, c, h, w = img.size() tile_size = min(tile_size, h, w) @@ -175,7 +173,8 @@ def upscale_2( """ Convenience wrapper around `tiled_upscale_2` that handles PIL images. """ - tensor = pil_image_to_torch_bgr(img).float().unsqueeze(0) # add batch dimension + param = torch_utils.get_param(model) + tensor = pil_image_to_torch_bgr(img).to(dtype=param.dtype).unsqueeze(0) # add batch dimension with torch.no_grad(): output = tiled_upscale_2( @@ -185,5 +184,6 @@ def upscale_2( tile_overlap=tile_overlap, scale=scale, desc=desc, + device=param.device, ) return torch_bgr_to_pil_image(output) -- cgit v1.2.1 From 50158a1fc9b4dd47a7bef70d34fbb0b30d5e8b47 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Thu, 4 Jan 2024 06:21:53 +0900 Subject: handle config.json failed to load --- modules/launch_utils.py | 4 +++- modules/options.py | 12 +++++++++--- 2 files changed, 12 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/launch_utils.py b/modules/launch_utils.py index c2cbd8ce..e2ad412a 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -249,7 +249,9 @@ def list_extensions(settings_file): with open(settings_file, "r", encoding="utf8") as file: settings = json.load(file) except Exception: - errors.report("Could not load settings", exc_info=True) + errors.report(f'\nCould not load settings\nThe config file "{settings_file}" is likely corrupted\nIt has been moved to the "tmp/config.json"\nReverting config to default\n\n''', exc_info=True) + os.replace(settings_file, os.path.join(script_path, "tmp", "config.json")) + settings = {} disabled_extensions = set(settings.get('disabled_extensions', [])) disable_all_extensions = settings.get('disable_all_extensions', 'none') diff --git a/modules/options.py b/modules/options.py index 09ff9403..503b40e9 100644 --- a/modules/options.py +++ b/modules/options.py @@ -1,3 +1,4 @@ +import os import json import sys from dataclasses import dataclass @@ -6,6 +7,7 @@ import gradio as gr from modules import errors from modules.shared_cmd_options import cmd_opts +from modules.paths_internal import script_path class OptionInfo: @@ -193,9 +195,13 @@ class Options: return type_x == type_y def load(self, filename): - with open(filename, "r", encoding="utf8") as file: - self.data = json.load(file) - + try: + with open(filename, "r", encoding="utf8") as file: + self.data = json.load(file) + except Exception: + errors.report(f'\nCould not load settings\nThe config file "{filename}" is likely corrupted\nIt has been moved to the "tmp/config.json"\nReverting config to default\n\n''', exc_info=True) + os.replace(filename, os.path.join(script_path, "tmp", "config.json")) + self.data = {} # 1.6.0 VAE defaults if self.data.get('sd_vae_as_default') is not None and self.data.get('sd_vae_overrides_per_model_preferences') is None: self.data['sd_vae_overrides_per_model_preferences'] = not self.data.get('sd_vae_as_default') -- cgit v1.2.1 From d9034b48a526f0a0c3e8f0dbf7c171bf4f0597fd Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Thu, 4 Jan 2024 00:16:58 +0200 Subject: Avoid unnecessary `isfile`/`exists` calls --- modules/cache.py | 17 ++++++++--------- modules/extensions.py | 11 ++++++----- modules/extra_networks.py | 7 ++++--- modules/infotext_utils.py | 4 +++- modules/launch_utils.py | 7 ++++--- modules/postprocessing.py | 7 ++++--- modules/shared_init.py | 4 +++- modules/ui_gradio_extensions.py | 8 +++----- modules/ui_loadsave.py | 5 +++-- modules/util.py | 6 +++--- 10 files changed, 41 insertions(+), 35 deletions(-) (limited to 'modules') diff --git a/modules/cache.py b/modules/cache.py index 2d37e7b9..a9822a0e 100644 --- a/modules/cache.py +++ b/modules/cache.py @@ -62,16 +62,15 @@ def cache(subsection): if cache_data is None: with cache_lock: if cache_data is None: - if not os.path.isfile(cache_filename): + try: + with open(cache_filename, "r", encoding="utf8") as file: + cache_data = json.load(file) + except FileNotFoundError: + cache_data = {} + except Exception: + os.replace(cache_filename, os.path.join(script_path, "tmp", "cache.json")) + print('[ERROR] issue occurred while trying to read cache.json, move current cache to tmp/cache.json and create new cache') cache_data = {} - else: - try: - with open(cache_filename, "r", encoding="utf8") as file: - cache_data = json.load(file) - except Exception: - os.replace(cache_filename, os.path.join(script_path, "tmp", "cache.json")) - print('[ERROR] issue occurred while trying to read cache.json, move current cache to tmp/cache.json and create new cache') - cache_data = {} s = cache_data.get(subsection, {}) cache_data[subsection] = s diff --git a/modules/extensions.py b/modules/extensions.py index 1899cd52..99e7ee60 100644 --- a/modules/extensions.py +++ b/modules/extensions.py @@ -32,11 +32,12 @@ class ExtensionMetadata: self.config = configparser.ConfigParser() filepath = os.path.join(path, self.filename) - if os.path.isfile(filepath): - try: - self.config.read(filepath) - except Exception: - errors.report(f"Error reading {self.filename} for extension {canonical_name}.", exc_info=True) + # `self.config.read()` will quietly swallow OSErrors (which FileNotFoundError is), + # so no need to check whether the file exists beforehand. + try: + self.config.read(filepath) + except Exception: + errors.report(f"Error reading {self.filename} for extension {canonical_name}.", exc_info=True) self.canonical_name = self.config.get("Extension", "Name", fallback=canonical_name) self.canonical_name = canonical_name.lower().strip() diff --git a/modules/extra_networks.py b/modules/extra_networks.py index b9533677..cd030fa3 100644 --- a/modules/extra_networks.py +++ b/modules/extra_networks.py @@ -215,9 +215,10 @@ def get_user_metadata(filename): metadata = {} try: - if os.path.isfile(metadata_filename): - with open(metadata_filename, "r", encoding="utf8") as file: - metadata = json.load(file) + with open(metadata_filename, "r", encoding="utf8") as file: + metadata = json.load(file) + except FileNotFoundError: + pass except Exception as e: errors.display(e, f"reading extra network user metadata from {metadata_filename}") diff --git a/modules/infotext_utils.py b/modules/infotext_utils.py index e582ee47..6978a0bf 100644 --- a/modules/infotext_utils.py +++ b/modules/infotext_utils.py @@ -453,9 +453,11 @@ def connect_paste(button, paste_fields, input_comp, override_settings_component, def paste_func(prompt): if not prompt and not shared.cmd_opts.hide_ui_dir_config: filename = os.path.join(data_path, "params.txt") - if os.path.exists(filename): + try: with open(filename, "r", encoding="utf8") as file: prompt = file.read() + except OSError: + pass params = parse_generation_parameters(prompt) script_callbacks.infotext_pasted_callback(prompt, params) diff --git a/modules/launch_utils.py b/modules/launch_utils.py index c2cbd8ce..febd8c24 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -245,9 +245,10 @@ def list_extensions(settings_file): settings = {} try: - if os.path.isfile(settings_file): - with open(settings_file, "r", encoding="utf8") as file: - settings = json.load(file) + with open(settings_file, "r", encoding="utf8") as file: + settings = json.load(file) + except FileNotFoundError: + pass except Exception: errors.report("Could not load settings", exc_info=True) diff --git a/modules/postprocessing.py b/modules/postprocessing.py index 7850328f..7449b0dc 100644 --- a/modules/postprocessing.py +++ b/modules/postprocessing.py @@ -97,11 +97,12 @@ def run_postprocessing(extras_mode, image, image_folder, input_dir, output_dir, if pp.caption: caption_filename = os.path.splitext(fullfn)[0] + ".txt" - if os.path.isfile(caption_filename): + existing_caption = "" + try: with open(caption_filename, encoding="utf8") as file: existing_caption = file.read().strip() - else: - existing_caption = "" + except FileNotFoundError: + pass action = shared.opts.postprocessing_existing_caption_action if action == 'Prepend' and existing_caption: diff --git a/modules/shared_init.py b/modules/shared_init.py index d3fb687e..586be342 100644 --- a/modules/shared_init.py +++ b/modules/shared_init.py @@ -18,8 +18,10 @@ def initialize(): shared.options_templates = shared_options.options_templates shared.opts = options.Options(shared_options.options_templates, shared_options.restricted_opts) shared.restricted_opts = shared_options.restricted_opts - if os.path.exists(shared.config_filename): + try: shared.opts.load(shared.config_filename) + except FileNotFoundError: + pass from modules import devices devices.device, devices.device_interrogate, devices.device_gfpgan, devices.device_esrgan, devices.device_codeformer = \ diff --git a/modules/ui_gradio_extensions.py b/modules/ui_gradio_extensions.py index a86c368e..f5278d22 100644 --- a/modules/ui_gradio_extensions.py +++ b/modules/ui_gradio_extensions.py @@ -35,13 +35,11 @@ def css_html(): return f'' for cssfile in scripts.list_files_with_name("style.css"): - if not os.path.isfile(cssfile): - continue - head += stylesheet(cssfile) - if os.path.exists(os.path.join(data_path, "user.css")): - head += stylesheet(os.path.join(data_path, "user.css")) + user_css = os.path.join(data_path, "user.css") + if os.path.exists(user_css): + head += stylesheet(user_css) return head diff --git a/modules/ui_loadsave.py b/modules/ui_loadsave.py index 693ff75c..2555cdb6 100644 --- a/modules/ui_loadsave.py +++ b/modules/ui_loadsave.py @@ -26,8 +26,9 @@ class UiLoadsave: self.ui_defaults_review = None try: - if os.path.exists(self.filename): - self.ui_settings = self.read_from_file() + self.ui_settings = self.read_from_file() + except FileNotFoundError: + pass except Exception as e: self.error_loading = True errors.display(e, "loading settings") diff --git a/modules/util.py b/modules/util.py index 4861bcb0..d503f267 100644 --- a/modules/util.py +++ b/modules/util.py @@ -21,11 +21,11 @@ def html_path(filename): def html(filename): path = html_path(filename) - if os.path.exists(path): + try: with open(path, encoding="utf8") as file: return file.read() - - return "" + except OSError: + return "" def walk_files(path, allowed_extensions=None): -- cgit v1.2.1 From 420f56c2e85ebbd3f530cf2c7b22022fda13ae13 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 4 Jan 2024 02:28:05 +0300 Subject: mass file lister as an attempt to tackle #14507 --- modules/extra_networks.py | 5 ++-- modules/ui_extra_networks.py | 18 ++++++++---- modules/util.py | 70 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 85 insertions(+), 8 deletions(-) (limited to 'modules') diff --git a/modules/extra_networks.py b/modules/extra_networks.py index b9533677..04249dff 100644 --- a/modules/extra_networks.py +++ b/modules/extra_networks.py @@ -206,7 +206,7 @@ def parse_prompts(prompts): return res, extra_data -def get_user_metadata(filename): +def get_user_metadata(filename, lister=None): if filename is None: return {} @@ -215,7 +215,8 @@ def get_user_metadata(filename): metadata = {} try: - if os.path.isfile(metadata_filename): + exists = lister.exists(metadata_filename) if lister else os.path.exists(metadata_filename) + if exists: with open(metadata_filename, "r", encoding="utf8") as file: metadata = json.load(file) except Exception as e: diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index e1c679ec..c06c8664 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -3,7 +3,7 @@ import os.path import urllib.parse from pathlib import Path -from modules import shared, ui_extra_networks_user_metadata, errors, extra_networks +from modules import shared, ui_extra_networks_user_metadata, errors, extra_networks, util from modules.images import read_info_from_image, save_image_with_geninfo import gradio as gr import json @@ -107,6 +107,7 @@ class ExtraNetworksPage: self.allow_negative_prompt = False self.metadata = {} self.items = {} + self.lister = util.MassFileLister() def refresh(self): pass @@ -123,7 +124,7 @@ class ExtraNetworksPage: def link_preview(self, filename): quoted_filename = urllib.parse.quote(filename.replace('\\', '/')) - mtime = os.path.getmtime(filename) + mtime, _ = self.lister.mctime(filename) return f"./sd_extra_networks/thumb?filename={quoted_filename}&mtime={mtime}" def search_terms_from_path(self, filename, possible_directories=None): @@ -137,6 +138,8 @@ class ExtraNetworksPage: return "" def create_html(self, tabname): + self.lister.reset() + items_html = '' self.metadata = {} @@ -282,10 +285,10 @@ class ExtraNetworksPage: List of default keys used for sorting in the UI. """ pth = Path(path) - stat = pth.stat() + mtime, ctime = self.lister.mctime(path) return { - "date_created": int(stat.st_ctime or 0), - "date_modified": int(stat.st_mtime or 0), + "date_created": int(mtime), + "date_modified": int(ctime), "name": pth.name.lower(), "path": str(pth.parent).lower(), } @@ -298,7 +301,7 @@ class ExtraNetworksPage: potential_files = sum([[path + "." + ext, path + ".preview." + ext] for ext in allowed_preview_extensions()], []) for file in potential_files: - if os.path.isfile(file): + if self.lister.exists(file): return self.link_preview(file) return None @@ -308,6 +311,9 @@ class ExtraNetworksPage: Find and read a description file for a given path (without extension). """ for file in [f"{path}.txt", f"{path}.description.txt"]: + if not self.lister.exists(file): + continue + try: with open(file, "r", encoding="utf-8", errors="replace") as f: return f.read() diff --git a/modules/util.py b/modules/util.py index 4861bcb0..c2a27590 100644 --- a/modules/util.py +++ b/modules/util.py @@ -66,3 +66,73 @@ def truncate_path(target_path, base_path=cwd): except ValueError: pass return abs_target + + +class MassFileListerCachedDir: + """A class that caches file metadata for a specific directory.""" + + def __init__(self, dirname): + self.files = None + self.files_cased = None + self.dirname = dirname + + stats = ((x.name, x.stat(follow_symlinks=False)) for x in os.scandir(self.dirname)) + files = [(n, s.st_mtime, s.st_ctime) for n, s in stats] + self.files = {x[0].lower(): x for x in files} + self.files_cased = {x[0]: x for x in files} + + +class MassFileLister: + """A class that provides a way to check for the existence and mtime/ctile of files without doing more than one stat call per file.""" + + def __init__(self): + self.cached_dirs = {} + + def find(self, path): + """ + Find the metadata for a file at the given path. + + Returns: + tuple or None: A tuple of (name, mtime, ctime) if the file exists, or None if it does not. + """ + + dirname, filename = os.path.split(path) + + cached_dir = self.cached_dirs.get(dirname) + if cached_dir is None: + cached_dir = MassFileListerCachedDir(dirname) + self.cached_dirs[dirname] = cached_dir + + stats = cached_dir.files_cased.get(filename) + if stats is not None: + return stats + + stats = cached_dir.files.get(filename.lower()) + if stats is None: + return None + + try: + os_stats = os.stat(path, follow_symlinks=False) + return filename, os_stats.st_mtime, os_stats.st_ctime + except Exception: + return None + + def exists(self, path): + """Check if a file exists at the given path.""" + + return self.find(path) is not None + + def mctime(self, path): + """ + Get the modification and creation times for a file at the given path. + + Returns: + tuple: A tuple of (mtime, ctime) if the file exists, or (0, 0) if it does not. + """ + + stats = self.find(path) + return (0, 0) if stats is None else stats[1:3] + + def reset(self): + """Clear the cache of all directories.""" + self.cached_dirs.clear() -- cgit v1.2.1 From 320a217b78047f30e1aa5e735742669a7f4c6bd8 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 4 Jan 2024 02:39:02 +0300 Subject: forgot something --- modules/ui_extra_networks.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index c06c8664..62db36f5 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -114,7 +114,7 @@ class ExtraNetworksPage: def read_user_metadata(self, item): filename = item.get("filename", None) - metadata = extra_networks.get_user_metadata(filename) + metadata = extra_networks.get_user_metadata(filename, lister=self.lister) desc = metadata.get("description", None) if desc is not None: -- cgit v1.2.1 From 15ec54dd969d6dc3fea7790ca5cce5badcfda426 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Thu, 4 Jan 2024 19:47:00 +0300 Subject: Have upscale button use the same seed as hires fix. --- modules/scripts.py | 21 +++++++++++++++++++++ modules/txt2img.py | 14 +++++++++++++- modules/ui.py | 10 +++++----- modules/ui_common.py | 26 +++++++++++++------------- modules/ui_postprocessing.py | 2 +- 5 files changed, 53 insertions(+), 20 deletions(-) (limited to 'modules') diff --git a/modules/scripts.py b/modules/scripts.py index 017aed5a..cf938ebb 100644 --- a/modules/scripts.py +++ b/modules/scripts.py @@ -91,6 +91,9 @@ class Script: setup_for_ui_only = False """If true, the script setup will only be run in Gradio UI, not in API""" + controls = None + """A list of controls retured by the ui().""" + def title(self): """this function should return the title of the script. This is what will be displayed in the dropdown menu.""" @@ -624,6 +627,7 @@ class ScriptRunner: import modules.api.models as api_models controls = wrap_call(script.ui, script.filename, "ui", script.is_img2img) + script.controls = controls if controls is None: return @@ -918,6 +922,23 @@ class ScriptRunner: except Exception: errors.report(f"Error running setup: {script.filename}", exc_info=True) + def set_named_arg(self, args, script_type, arg_elem_id, value): + script = next((x for x in self.scripts if type(x).__name__ == script_type), None) + if script is None: + return + + for i, control in enumerate(script.controls): + if arg_elem_id in control.elem_id: + index = script.args_from + i + + if isinstance(args, list): + args[index] = value + return args + elif isinstance(args, tuple): + return args[:index] + (value,) + args[index+1:] + else: + return None + scripts_txt2img: ScriptRunner = None scripts_img2img: ScriptRunner = None diff --git a/modules/txt2img.py b/modules/txt2img.py index 4a6fe72a..41bb9da3 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -1,3 +1,4 @@ +import json from contextlib import closing import modules.scripts @@ -9,12 +10,19 @@ from modules.ui import plaintext_to_html import gradio as gr -def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, *args): +def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, generation_info, *args): assert len(gallery) > 0, 'No image to upscale' + assert 0 <= gallery_index < len(gallery), f'Bad image index: {gallery_index}' + + geninfo = json.loads(generation_info) + all_seeds = geninfo["all_seeds"] image_info = gallery[gallery_index] if 0 <= gallery_index < len(gallery) else gallery[0] image = infotext_utils.image_from_url_text(image_info) + gallery_index_from_end = len(gallery) - gallery_index + image.seed = all_seeds[-gallery_index_from_end if gallery_index_from_end < len(all_seeds) + 1 else 0] + return txt2img(id_task, request, *args, firstpass_image=image) @@ -22,6 +30,10 @@ def txt2img(id_task: str, request: gr.Request, prompt: str, negative_prompt: str override_settings = create_override_settings_dict(override_settings_texts) if firstpass_image is not None: + seed = getattr(firstpass_image, 'seed', None) + if seed: + args = modules.scripts.scripts_txt2img.set_named_arg(args, 'ScriptSeed', 'seed', seed) + enable_hr = True batch_size = 1 n_iter = 1 diff --git a/modules/ui.py b/modules/ui.py index 7116d71c..2d2e333b 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -405,8 +405,8 @@ def create_ui(): txt2img_outputs = [ output_panel.gallery, + output_panel.generation_info, output_panel.infotext, - output_panel.html_info, output_panel.html_log, ] @@ -424,7 +424,7 @@ def create_ui(): output_panel.button_upscale.click( fn=wrap_gradio_gpu_call(modules.txt2img.txt2img_upscale, extra_outputs=[None, '', '']), _js="submit_txt2img_upscale", - inputs=txt2img_inputs[0:1] + [output_panel.gallery, dummy_component] + txt2img_inputs[1:], + inputs=txt2img_inputs[0:1] + [output_panel.gallery, dummy_component, output_panel.generation_info] + txt2img_inputs[1:], outputs=txt2img_outputs, show_progress=False, ) @@ -437,8 +437,8 @@ def create_ui(): inputs=[dummy_component], outputs=[ output_panel.gallery, + output_panel.generation_info, output_panel.infotext, - output_panel.html_info, output_panel.html_log, ], show_progress=False, @@ -766,8 +766,8 @@ def create_ui(): ] + custom_inputs, outputs=[ output_panel.gallery, + output_panel.generation_info, output_panel.infotext, - output_panel.html_info, output_panel.html_log, ], show_progress=False, @@ -807,8 +807,8 @@ def create_ui(): inputs=[dummy_component], outputs=[ output_panel.gallery, + output_panel.generation_info, output_panel.infotext, - output_panel.html_info, output_panel.html_log, ], show_progress=False, diff --git a/modules/ui_common.py b/modules/ui_common.py index ff84197c..f17259c2 100644 --- a/modules/ui_common.py +++ b/modules/ui_common.py @@ -108,8 +108,8 @@ def save_files(js_data, images, do_make_zip, index): @dataclasses.dataclass class OutputPanel: gallery = None + generation_info = None infotext = None - html_info = None html_log = None button_upscale = None @@ -175,17 +175,17 @@ Requested path was: {f} download_files = gr.File(None, file_count="multiple", interactive=False, show_label=False, visible=False, elem_id=f'download_files_{tabname}') with gr.Group(): - res.html_info = gr.HTML(elem_id=f'html_info_{tabname}', elem_classes="infotext") + res.infotext = gr.HTML(elem_id=f'html_info_{tabname}', elem_classes="infotext") res.html_log = gr.HTML(elem_id=f'html_log_{tabname}', elem_classes="html-log") - res.infotext = gr.Textbox(visible=False, elem_id=f'generation_info_{tabname}') + res.generation_info = gr.Textbox(visible=False, elem_id=f'generation_info_{tabname}') if tabname == 'txt2img' or tabname == 'img2img': generation_info_button = gr.Button(visible=False, elem_id=f"{tabname}_generation_info_button") generation_info_button.click( fn=update_generation_info, _js="function(x, y, z){ return [x, y, selected_gallery_index()] }", - inputs=[res.infotext, res.html_info, res.html_info], - outputs=[res.html_info, res.html_info], + inputs=[res.generation_info, res.infotext, res.infotext], + outputs=[res.infotext, res.infotext], show_progress=False, ) @@ -193,10 +193,10 @@ Requested path was: {f} fn=call_queue.wrap_gradio_call(save_files), _js="(x, y, z, w) => [x, y, false, selected_gallery_index()]", inputs=[ - res.infotext, + res.generation_info, res.gallery, - res.html_info, - res.html_info, + res.infotext, + res.infotext, ], outputs=[ download_files, @@ -209,10 +209,10 @@ Requested path was: {f} fn=call_queue.wrap_gradio_call(save_files), _js="(x, y, z, w) => [x, y, true, selected_gallery_index()]", inputs=[ - res.infotext, + res.generation_info, res.gallery, - res.html_info, - res.html_info, + res.infotext, + res.infotext, ], outputs=[ download_files, @@ -221,8 +221,8 @@ Requested path was: {f} ) else: - res.infotext = gr.HTML(elem_id=f'html_info_x_{tabname}') - res.html_info = gr.HTML(elem_id=f'html_info_{tabname}', elem_classes="infotext") + res.generation_info = gr.HTML(elem_id=f'html_info_x_{tabname}') + res.infotext = gr.HTML(elem_id=f'html_info_{tabname}', elem_classes="infotext") res.html_log = gr.HTML(elem_id=f'html_log_{tabname}') paste_field_names = [] diff --git a/modules/ui_postprocessing.py b/modules/ui_postprocessing.py index 8f09e658..7a132ac2 100644 --- a/modules/ui_postprocessing.py +++ b/modules/ui_postprocessing.py @@ -49,7 +49,7 @@ def create_ui(): ], outputs=[ output_panel.gallery, - output_panel.infotext, + output_panel.generation_info, output_panel.html_log, ], show_progress=False, -- cgit v1.2.1 From 9805f35c6f3ef0b0fc4e3648aa3d6eddf0a907af Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Thu, 4 Jan 2024 19:13:36 +0200 Subject: Ensure GRADIO_ANALYTICS_ENABLED is set early enough --- modules/initialize.py | 2 ++ modules/launch_utils.py | 3 +-- 2 files changed, 3 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/initialize.py b/modules/initialize.py index 4a3cd98c..7c1ac99e 100644 --- a/modules/initialize.py +++ b/modules/initialize.py @@ -1,5 +1,6 @@ import importlib import logging +import os import sys import warnings from threading import Thread @@ -18,6 +19,7 @@ def imports(): warnings.filterwarnings(action="ignore", category=DeprecationWarning, module="pytorch_lightning") warnings.filterwarnings(action="ignore", category=UserWarning, module="torchvision") + os.environ.setdefault('GRADIO_ANALYTICS_ENABLED', 'False') import gradio # noqa: F401 startup_timer.record("import gradio") diff --git a/modules/launch_utils.py b/modules/launch_utils.py index 7ebdf0b4..c2a7ae93 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -27,8 +27,7 @@ dir_repos = "repositories" # Whether to default to printing command output default_command_live = (os.environ.get('WEBUI_LAUNCH_LIVE_OUTPUT') == "1") -if 'GRADIO_ANALYTICS_ENABLED' not in os.environ: - os.environ['GRADIO_ANALYTICS_ENABLED'] = 'False' +os.environ.setdefault('GRADIO_ANALYTICS_ENABLED', 'False') def check_python_version(): -- cgit v1.2.1 From 6fa42e919f27d7316a2c7b61674fb3eb17f3a1bb Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Thu, 4 Jan 2024 16:13:08 +0200 Subject: Fix logging configuration again * Only use `tqdm.write()` if `tqdm` is active, defer to stderr * Correct log formatter for TqdmLoggingHandler * If `rich` is installed and `SD_WEBUI_RICH_LOG` is set, use `rich`'s formatter --- modules/logging_config.py | 62 +++++++++++++++++++++++++++++------------------ 1 file changed, 39 insertions(+), 23 deletions(-) (limited to 'modules') diff --git a/modules/logging_config.py b/modules/logging_config.py index 79269875..11eee9a6 100644 --- a/modules/logging_config.py +++ b/modules/logging_config.py @@ -1,41 +1,57 @@ -import os import logging +import os try: - from tqdm.auto import tqdm + from tqdm import tqdm + class TqdmLoggingHandler(logging.Handler): - def __init__(self, level=logging.INFO): - super().__init__(level) + def __init__(self, fallback_handler: logging.Handler): + super().__init__() + self.fallback_handler = fallback_handler def emit(self, record): try: - msg = self.format(record) - tqdm.write(msg) - self.flush() + # If there are active tqdm progress bars, + # attempt to not interfere with them. + if tqdm._instances: + tqdm.write(self.format(record)) + else: + self.fallback_handler.emit(record) except Exception: - self.handleError(record) + self.fallback_handler.emit(record) - TQDM_IMPORTED = True except ImportError: - # tqdm does not exist before first launch - # I will import once the UI finishes seting up the enviroment and reloads. - TQDM_IMPORTED = False + TqdmLoggingHandler = None + def setup_logging(loglevel): if loglevel is None: loglevel = os.environ.get("SD_WEBUI_LOG_LEVEL") - loghandlers = [] + if not loglevel: + return + + if logging.root.handlers: + # Already configured, do not interfere + return + + if os.environ.get("SD_WEBUI_RICH_LOG"): + from rich.logging import RichHandler + handler = RichHandler() + else: + handler = logging.StreamHandler() + + if TqdmLoggingHandler: + handler = TqdmLoggingHandler(handler) + + formatter = logging.Formatter( + '%(asctime)s %(levelname)s [%(name)s] %(message)s', + '%Y-%m-%d %H:%M:%S', + ) - if TQDM_IMPORTED: - loghandlers.append(TqdmLoggingHandler()) + handler.setFormatter(formatter) - if loglevel: - log_level = getattr(logging, loglevel.upper(), None) or logging.INFO - logging.basicConfig( - level=log_level, - format='%(asctime)s %(levelname)s [%(name)s] %(message)s', - datefmt='%Y-%m-%d %H:%M:%S', - handlers=loghandlers - ) + log_level = getattr(logging, loglevel.upper(), None) or logging.INFO + logging.root.setLevel(log_level) + logging.root.addHandler(handler) -- cgit v1.2.1 From 233c66b36eba07b905f9743c2ad807aec33d9ccb Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Fri, 5 Jan 2024 12:28:32 +0300 Subject: Make the upscale button update the gallery with the new image rather than replace it. --- modules/processing.py | 6 +++- modules/txt2img.py | 85 +++++++++++++++++++++++++++++++++++---------------- 2 files changed, 63 insertions(+), 28 deletions(-) (limited to 'modules') diff --git a/modules/processing.py b/modules/processing.py index 84e7b1b4..dcc807fe 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -732,7 +732,7 @@ def create_infotext(p, all_prompts, all_seeds, all_subseeds, comments=None, iter "Variation seed": (None if p.subseed_strength == 0 else (p.all_subseeds[0] if use_main_prompt else all_subseeds[index])), "Variation seed strength": (None if p.subseed_strength == 0 else p.subseed_strength), "Seed resize from": (None if p.seed_resize_from_w <= 0 or p.seed_resize_from_h <= 0 else f"{p.seed_resize_from_w}x{p.seed_resize_from_h}"), - "Denoising strength": getattr(p, 'denoising_strength', None), + "Denoising strength": p.extra_generation_params.get("Denoising strength"), "Conditional mask weight": getattr(p, "inpainting_mask_weight", shared.opts.inpainting_mask_weight) if p.is_using_inpainting_conditioning else None, "Clip skip": None if clip_skip <= 1 else clip_skip, "ENSD": opts.eta_noise_seed_delta if uses_ensd else None, @@ -1198,6 +1198,8 @@ class StableDiffusionProcessingTxt2Img(StableDiffusionProcessing): def init(self, all_prompts, all_seeds, all_subseeds): if self.enable_hr: + self.extra_generation_params["Denoising strength"] = self.denoising_strength + if self.hr_checkpoint_name and self.hr_checkpoint_name != 'Use same checkpoint': self.hr_checkpoint_info = sd_models.get_closet_checkpoint_match(self.hr_checkpoint_name) @@ -1516,6 +1518,8 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): self.mask_blur_y = value def init(self, all_prompts, all_seeds, all_subseeds): + self.extra_generation_params["Denoising strength"] = self.denoising_strength + self.image_cfg_scale: float = self.image_cfg_scale if shared.sd_model.cond_stage_key == "edit" else None self.sampler = sd_samplers.create_sampler(self.sampler_name, self.sd_model) diff --git a/modules/txt2img.py b/modules/txt2img.py index 41bb9da3..c4cc12d2 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -7,36 +7,15 @@ from modules.infotext_utils import create_override_settings_dict from modules.shared import opts import modules.shared as shared from modules.ui import plaintext_to_html +from PIL import Image import gradio as gr -def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, generation_info, *args): - assert len(gallery) > 0, 'No image to upscale' - assert 0 <= gallery_index < len(gallery), f'Bad image index: {gallery_index}' - - geninfo = json.loads(generation_info) - all_seeds = geninfo["all_seeds"] - - image_info = gallery[gallery_index] if 0 <= gallery_index < len(gallery) else gallery[0] - image = infotext_utils.image_from_url_text(image_info) - - gallery_index_from_end = len(gallery) - gallery_index - image.seed = all_seeds[-gallery_index_from_end if gallery_index_from_end < len(all_seeds) + 1 else 0] - - return txt2img(id_task, request, *args, firstpass_image=image) - - -def txt2img(id_task: str, request: gr.Request, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_name: str, n_iter: int, batch_size: int, cfg_scale: float, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, hr_checkpoint_name: str, hr_sampler_name: str, hr_prompt: str, hr_negative_prompt, override_settings_texts, *args, firstpass_image=None): +def txt2img_create_processing(id_task: str, request: gr.Request, prompt: str, negative_prompt: str, prompt_styles, steps: int, sampler_name: str, n_iter: int, batch_size: int, cfg_scale: float, height: int, width: int, enable_hr: bool, denoising_strength: float, hr_scale: float, hr_upscaler: str, hr_second_pass_steps: int, hr_resize_x: int, hr_resize_y: int, hr_checkpoint_name: str, hr_sampler_name: str, hr_prompt: str, hr_negative_prompt, override_settings_texts, *args, force_enable_hr=False): override_settings = create_override_settings_dict(override_settings_texts) - if firstpass_image is not None: - seed = getattr(firstpass_image, 'seed', None) - if seed: - args = modules.scripts.scripts_txt2img.set_named_arg(args, 'ScriptSeed', 'seed', seed) - + if force_enable_hr: enable_hr = True - batch_size = 1 - n_iter = 1 p = processing.StableDiffusionProcessingTxt2Img( sd_model=shared.sd_model, @@ -53,7 +32,7 @@ def txt2img(id_task: str, request: gr.Request, prompt: str, negative_prompt: str width=width, height=height, enable_hr=enable_hr, - denoising_strength=denoising_strength if enable_hr else None, + denoising_strength=denoising_strength, hr_scale=hr_scale, hr_upscaler=hr_upscaler, hr_second_pass_steps=hr_second_pass_steps, @@ -64,7 +43,6 @@ def txt2img(id_task: str, request: gr.Request, prompt: str, negative_prompt: str hr_prompt=hr_prompt, hr_negative_prompt=hr_negative_prompt, override_settings=override_settings, - firstpass_image=firstpass_image, ) p.scripts = modules.scripts.scripts_txt2img @@ -75,8 +53,61 @@ def txt2img(id_task: str, request: gr.Request, prompt: str, negative_prompt: str if shared.opts.enable_console_prompts: print(f"\ntxt2img: {prompt}", file=shared.progress_print_out) + return p + + +def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, generation_info, *args): + assert len(gallery) > 0, 'No image to upscale' + assert 0 <= gallery_index < len(gallery), f'Bad image index: {gallery_index}' + + p = txt2img_create_processing(id_task, request, *args) + p.enable_hr = True + p.batch_size = 1 + p.n_iter = 1 + + geninfo = json.loads(generation_info) + all_seeds = geninfo["all_seeds"] + + image_info = gallery[gallery_index] if 0 <= gallery_index < len(gallery) else gallery[0] + p.firstpass_image = infotext_utils.image_from_url_text(image_info) + + gallery_index_from_end = len(gallery) - gallery_index + seed = all_seeds[-gallery_index_from_end if gallery_index_from_end < len(all_seeds) + 1 else 0] + p.script_args = modules.scripts.scripts_txt2img.set_named_arg(p.script_args, 'ScriptSeed', 'seed', seed) + + with closing(p): + processed = modules.scripts.scripts_txt2img.run(p, *p.script_args) + + if processed is None: + processed = processing.process_images(p) + + shared.total_tqdm.clear() + + new_gallery = [] + for i, image in enumerate(gallery): + fake_image = Image.new(mode="RGB", size=(1, 1)) + + if i == gallery_index: + already_saved_as = getattr(processed.images[0], 'already_saved_as', None) + if already_saved_as is not None: + fake_image.already_saved_as = already_saved_as + else: + fake_image = processed.images[0] + else: + fake_image.already_saved_as = image["name"] + + new_gallery.append(fake_image) + + geninfo["infotexts"][gallery_index] = processed.info + + return new_gallery, json.dumps(geninfo), plaintext_to_html(processed.info), plaintext_to_html(processed.comments, classname="comments") + + +def txt2img(id_task: str, request: gr.Request, *args): + p = txt2img_create_processing(id_task, request, *args) + with closing(p): - processed = modules.scripts.scripts_txt2img.run(p, *args) + processed = modules.scripts.scripts_txt2img.run(p, *p.script_args) if processed is None: processed = processing.process_images(p) -- cgit v1.2.1 From 16b4d2cf3f51f1d88b97d1d459dec59d3a2d0642 Mon Sep 17 00:00:00 2001 From: Nuullll Date: Sat, 6 Jan 2024 16:32:18 +0800 Subject: [IPEX] Fix SDPA attn_mask dtype --- modules/xpu_specific.py | 2 ++ 1 file changed, 2 insertions(+) (limited to 'modules') diff --git a/modules/xpu_specific.py b/modules/xpu_specific.py index f7687a66..4e11125b 100644 --- a/modules/xpu_specific.py +++ b/modules/xpu_specific.py @@ -41,6 +41,8 @@ def torch_xpu_scaled_dot_product_attention( # cast to same dtype first key = key.to(query.dtype) value = value.to(query.dtype) + if attn_mask is not None and attn_mask.dtype != torch.bool: + attn_mask = attn_mask.to(query.dtype) N = query.shape[:-2] # Batch size L = query.size(-2) # Target sequence length -- cgit v1.2.1 From ec9acb31450536a9258192351d6a26421efd7eb4 Mon Sep 17 00:00:00 2001 From: Nuullll Date: Sat, 6 Jan 2024 17:17:53 +0800 Subject: Handle CondFunc exception when resolving attributes --- modules/sd_hijack_utils.py | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/sd_hijack_utils.py b/modules/sd_hijack_utils.py index f8684475..79bf6e46 100644 --- a/modules/sd_hijack_utils.py +++ b/modules/sd_hijack_utils.py @@ -11,10 +11,14 @@ class CondFunc: break except ImportError: pass - for attr_name in func_path[i:-1]: - resolved_obj = getattr(resolved_obj, attr_name) - orig_func = getattr(resolved_obj, func_path[-1]) - setattr(resolved_obj, func_path[-1], lambda *args, **kwargs: self(*args, **kwargs)) + try: + for attr_name in func_path[i:-1]: + resolved_obj = getattr(resolved_obj, attr_name) + orig_func = getattr(resolved_obj, func_path[-1]) + setattr(resolved_obj, func_path[-1], lambda *args, **kwargs: self(*args, **kwargs)) + except AttributeError: + print(f"Warning: Failed to resolve {orig_func} for CondFunc hijack") + pass self.__init__(orig_func, sub_func, cond_func) return lambda *args, **kwargs: self(*args, **kwargs) def __init__(self, orig_func, sub_func, cond_func): -- cgit v1.2.1 From 73786c047f14d6ae658b2c12f493f05486ba1789 Mon Sep 17 00:00:00 2001 From: Nuullll Date: Sat, 6 Jan 2024 19:09:56 +0800 Subject: [IPEX] Fix torch.Generator hijack --- modules/xpu_specific.py | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/xpu_specific.py b/modules/xpu_specific.py index 4e11125b..1137891a 100644 --- a/modules/xpu_specific.py +++ b/modules/xpu_specific.py @@ -94,11 +94,23 @@ def torch_xpu_scaled_dot_product_attention( return torch.reshape(result, (*N, L, Ev)) +def is_xpu_device(device: str | torch.device = None): + if device is None: + return False + if isinstance(device, str): + return device.startswith("xpu") + return device.type == "xpu" + + if has_xpu: - # W/A for https://github.com/intel/intel-extension-for-pytorch/issues/452: torch.Generator API doesn't support XPU device - CondFunc('torch.Generator', - lambda orig_func, device=None: torch.xpu.Generator(device), - lambda orig_func, device=None: device is not None and device.type == "xpu") + try: + # torch.Generator supports "xpu" device since 2.1 + torch.Generator("xpu") + except: + # W/A for https://github.com/intel/intel-extension-for-pytorch/issues/452: torch.Generator API doesn't support XPU device (for IPEX < 2.1) + CondFunc('torch.Generator', + lambda orig_func, device=None: torch.xpu.Generator(device), + lambda orig_func, device=None: is_xpu_device(device)) # W/A for some OPs that could not handle different input dtypes CondFunc('torch.nn.functional.layer_norm', -- cgit v1.2.1 From 818d6a11e709bf07d48606bdccab944c46a5f4b0 Mon Sep 17 00:00:00 2001 From: Nuullll Date: Sat, 6 Jan 2024 19:14:06 +0800 Subject: Fix format --- modules/xpu_specific.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/xpu_specific.py b/modules/xpu_specific.py index 1137891a..2971dbc3 100644 --- a/modules/xpu_specific.py +++ b/modules/xpu_specific.py @@ -106,8 +106,8 @@ if has_xpu: try: # torch.Generator supports "xpu" device since 2.1 torch.Generator("xpu") - except: - # W/A for https://github.com/intel/intel-extension-for-pytorch/issues/452: torch.Generator API doesn't support XPU device (for IPEX < 2.1) + except RuntimeError: + # W/A for https://github.com/intel/intel-extension-for-pytorch/issues/452: torch.Generator API doesn't support XPU device (for torch < 2.1) CondFunc('torch.Generator', lambda orig_func, device=None: torch.xpu.Generator(device), lambda orig_func, device=None: is_xpu_device(device)) -- cgit v1.2.1 From a183de04e3f965083e7f3462201327d30c36b958 Mon Sep 17 00:00:00 2001 From: Nuullll Date: Sat, 6 Jan 2024 20:03:33 +0800 Subject: Execute model_loaded_callback after moving to target device --- modules/sd_models.py | 6 +++--- modules/sd_vae.py | 3 ++- 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/sd_models.py b/modules/sd_models.py index 50bc209e..2c045771 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -842,13 +842,13 @@ def reload_model_weights(sd_model=None, info=None, forced_reload=False): sd_hijack.model_hijack.hijack(sd_model) timer.record("hijack") - script_callbacks.model_loaded_callback(sd_model) - timer.record("script callbacks") - if not sd_model.lowvram: sd_model.to(devices.device) timer.record("move model to device") + script_callbacks.model_loaded_callback(sd_model) + timer.record("script callbacks") + print(f"Weights loaded in {timer.summary()}.") model_data.set_sd_model(sd_model) diff --git a/modules/sd_vae.py b/modules/sd_vae.py index 31306d8b..43687e48 100644 --- a/modules/sd_vae.py +++ b/modules/sd_vae.py @@ -273,10 +273,11 @@ def reload_vae_weights(sd_model=None, vae_file=unspecified): load_vae(sd_model, vae_file, vae_source) sd_hijack.model_hijack.hijack(sd_model) - script_callbacks.model_loaded_callback(sd_model) if not sd_model.lowvram: sd_model.to(devices.device) + script_callbacks.model_loaded_callback(sd_model) + print("VAE weights loaded.") return sd_model -- cgit v1.2.1 From 2f98a35fc4508494355c01ec45f5bec725f570a6 Mon Sep 17 00:00:00 2001 From: AUTOMATIC1111 <16777216c@gmail.com> Date: Sun, 7 Jan 2024 09:21:21 +0300 Subject: add assets repo; serve fonts locally rather than from google's servers --- modules/launch_utils.py | 3 +++ modules/sysinfo.py | 2 ++ modules/ui.py | 4 +++- 3 files changed, 8 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/launch_utils.py b/modules/launch_utils.py index c2a7ae93..8e58d714 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -344,11 +344,13 @@ def prepare_environment(): clip_package = os.environ.get('CLIP_PACKAGE', "https://github.com/openai/CLIP/archive/d50d76daa670286dd6cacf3bcd80b5e4823fc8e1.zip") openclip_package = os.environ.get('OPENCLIP_PACKAGE', "https://github.com/mlfoundations/open_clip/archive/bb6e834e9c70d9c27d0dc3ecedeebeaeb1ffad6b.zip") + assets_repo = os.environ.get('ASSETS_REPO', "https://github.com/AUTOMATIC1111/stable-diffusion-webui-assets.git") stable_diffusion_repo = os.environ.get('STABLE_DIFFUSION_REPO', "https://github.com/Stability-AI/stablediffusion.git") stable_diffusion_xl_repo = os.environ.get('STABLE_DIFFUSION_XL_REPO', "https://github.com/Stability-AI/generative-models.git") k_diffusion_repo = os.environ.get('K_DIFFUSION_REPO', 'https://github.com/crowsonkb/k-diffusion.git') blip_repo = os.environ.get('BLIP_REPO', 'https://github.com/salesforce/BLIP.git') + assets_commit_hash = os.environ.get('ASSETS_COMMIT_HASH', "6f7db241d2f8ba7457bac5ca9753331f0c266917") stable_diffusion_commit_hash = os.environ.get('STABLE_DIFFUSION_COMMIT_HASH', "cf1d67a6fd5ea1aa600c4df58e5b47da45f6bdbf") stable_diffusion_xl_commit_hash = os.environ.get('STABLE_DIFFUSION_XL_COMMIT_HASH', "45c443b316737a4ab6e40413d7794a7f5657c19f") k_diffusion_commit_hash = os.environ.get('K_DIFFUSION_COMMIT_HASH', "ab527a9a6d347f364e3d185ba6d714e22d80cb3c") @@ -405,6 +407,7 @@ def prepare_environment(): os.makedirs(os.path.join(script_path, dir_repos), exist_ok=True) + git_clone(assets_repo, repo_dir('stable-diffusion-webui-assets'), "assets", assets_commit_hash) git_clone(stable_diffusion_repo, repo_dir('stable-diffusion-stability-ai'), "Stable Diffusion", stable_diffusion_commit_hash) git_clone(stable_diffusion_xl_repo, repo_dir('generative-models'), "Stable Diffusion XL", stable_diffusion_xl_commit_hash) git_clone(k_diffusion_repo, repo_dir('k-diffusion'), "K-diffusion", k_diffusion_commit_hash) diff --git a/modules/sysinfo.py b/modules/sysinfo.py index 5abf616b..f336251e 100644 --- a/modules/sysinfo.py +++ b/modules/sysinfo.py @@ -24,9 +24,11 @@ environment_whitelist = { "XFORMERS_PACKAGE", "CLIP_PACKAGE", "OPENCLIP_PACKAGE", + "ASSETS_REPO", "STABLE_DIFFUSION_REPO", "K_DIFFUSION_REPO", "BLIP_REPO", + "ASSETS_COMMIT_HASH", "STABLE_DIFFUSION_COMMIT_HASH", "K_DIFFUSION_COMMIT_HASH", "BLIP_COMMIT_HASH", diff --git a/modules/ui.py b/modules/ui.py index 2d2e333b..a716a040 100644 --- a/modules/ui.py +++ b/modules/ui.py @@ -13,7 +13,7 @@ from PIL import Image, PngImagePlugin # noqa: F401 from modules.call_queue import wrap_gradio_gpu_call, wrap_queued_call, wrap_gradio_call from modules import gradio_extensons # noqa: F401 -from modules import sd_hijack, sd_models, script_callbacks, ui_extensions, deepbooru, extra_networks, ui_common, ui_postprocessing, progress, ui_loadsave, shared_items, ui_settings, timer, sysinfo, ui_checkpoint_merger, scripts, sd_samplers, processing, ui_extra_networks, ui_toprow +from modules import sd_hijack, sd_models, script_callbacks, ui_extensions, deepbooru, extra_networks, ui_common, ui_postprocessing, progress, ui_loadsave, shared_items, ui_settings, timer, sysinfo, ui_checkpoint_merger, scripts, sd_samplers, processing, ui_extra_networks, ui_toprow, launch_utils from modules.ui_components import FormRow, FormGroup, ToolButton, FormHTML, InputAccordion, ResizeHandleRow from modules.paths import script_path from modules.ui_common import create_refresh_button @@ -1223,3 +1223,5 @@ def setup_ui_api(app): app.add_api_route("/internal/sysinfo", download_sysinfo, methods=["GET"]) app.add_api_route("/internal/sysinfo-download", lambda: download_sysinfo(attachment=True), methods=["GET"]) + import fastapi.staticfiles + app.mount("/webui-assets", fastapi.staticfiles.StaticFiles(directory=launch_utils.repo_dir('stable-diffusion-webui-assets')), name="webui-assets") -- cgit v1.2.1 From 425507bd10c55f1f804eb5015db74520668f46f9 Mon Sep 17 00:00:00 2001 From: continue-revolution Date: Sun, 7 Jan 2024 10:25:01 -0600 Subject: add p to cfgdenoiserparams --- modules/script_callbacks.py | 5 ++++- modules/sd_samplers_cfg_denoiser.py | 2 +- 2 files changed, 5 insertions(+), 2 deletions(-) (limited to 'modules') diff --git a/modules/script_callbacks.py b/modules/script_callbacks.py index 9ed7ad21..bb47c18d 100644 --- a/modules/script_callbacks.py +++ b/modules/script_callbacks.py @@ -41,7 +41,7 @@ class ExtraNoiseParams: class CFGDenoiserParams: - def __init__(self, x, image_cond, sigma, sampling_step, total_sampling_steps, text_cond, text_uncond): + def __init__(self, x, image_cond, sigma, sampling_step, total_sampling_steps, text_cond, text_uncond, p): self.x = x """Latent image representation in the process of being denoised""" @@ -63,6 +63,9 @@ class CFGDenoiserParams: self.text_uncond = text_uncond """ Encoder hidden states of text conditioning from negative prompt""" + self.p = p + """StableDiffusionProcessing object with processing parameters""" + class CFGDenoisedParams: def __init__(self, x, sampling_step, total_sampling_steps, inner_model): diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index eb9d5daf..f4ded6bd 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -146,7 +146,7 @@ class CFGDenoiser(torch.nn.Module): sigma_in = torch.cat([torch.stack([sigma[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [sigma] + [sigma]) image_cond_in = torch.cat([torch.stack([image_cond[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [image_uncond] + [torch.zeros_like(self.init_latent)]) - denoiser_params = CFGDenoiserParams(x_in, image_cond_in, sigma_in, state.sampling_step, state.sampling_steps, tensor, uncond) + denoiser_params = CFGDenoiserParams(x_in, image_cond_in, sigma_in, state.sampling_step, state.sampling_steps, tensor, uncond, self.p) cfg_denoiser_callback(denoiser_params) x_in = denoiser_params.x image_cond_in = denoiser_params.image_cond -- cgit v1.2.1 From f56cebf5ba24313447b2204c3f804379767201c9 Mon Sep 17 00:00:00 2001 From: continue-revolution Date: Sun, 7 Jan 2024 12:35:35 -0600 Subject: add self instead --- modules/script_callbacks.py | 6 +++--- modules/sd_samplers_cfg_denoiser.py | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/script_callbacks.py b/modules/script_callbacks.py index bb47c18d..053dfc96 100644 --- a/modules/script_callbacks.py +++ b/modules/script_callbacks.py @@ -41,7 +41,7 @@ class ExtraNoiseParams: class CFGDenoiserParams: - def __init__(self, x, image_cond, sigma, sampling_step, total_sampling_steps, text_cond, text_uncond, p): + def __init__(self, x, image_cond, sigma, sampling_step, total_sampling_steps, text_cond, text_uncond, denoiser): self.x = x """Latent image representation in the process of being denoised""" @@ -63,8 +63,8 @@ class CFGDenoiserParams: self.text_uncond = text_uncond """ Encoder hidden states of text conditioning from negative prompt""" - self.p = p - """StableDiffusionProcessing object with processing parameters""" + self.denoiser = denoiser + """Current CFGDenoiser object with processing parameters""" class CFGDenoisedParams: diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index f4ded6bd..6d76aa96 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -146,7 +146,7 @@ class CFGDenoiser(torch.nn.Module): sigma_in = torch.cat([torch.stack([sigma[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [sigma] + [sigma]) image_cond_in = torch.cat([torch.stack([image_cond[i] for _ in range(n)]) for i, n in enumerate(repeats)] + [image_uncond] + [torch.zeros_like(self.init_latent)]) - denoiser_params = CFGDenoiserParams(x_in, image_cond_in, sigma_in, state.sampling_step, state.sampling_steps, tensor, uncond, self.p) + denoiser_params = CFGDenoiserParams(x_in, image_cond_in, sigma_in, state.sampling_step, state.sampling_steps, tensor, uncond, self) cfg_denoiser_callback(denoiser_params) x_in = denoiser_params.x image_cond_in = denoiser_params.image_cond -- cgit v1.2.1 From 37906e429ae5a92f9ea96ab6dc2157b1d7c4d8b6 Mon Sep 17 00:00:00 2001 From: Chengsong Zhang Date: Sun, 7 Jan 2024 20:17:42 -0600 Subject: make denoiser None by default --- modules/script_callbacks.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/script_callbacks.py b/modules/script_callbacks.py index 053dfc96..a54cb3eb 100644 --- a/modules/script_callbacks.py +++ b/modules/script_callbacks.py @@ -41,7 +41,7 @@ class ExtraNoiseParams: class CFGDenoiserParams: - def __init__(self, x, image_cond, sigma, sampling_step, total_sampling_steps, text_cond, text_uncond, denoiser): + def __init__(self, x, image_cond, sigma, sampling_step, total_sampling_steps, text_cond, text_uncond, denoiser=None): self.x = x """Latent image representation in the process of being denoised""" -- cgit v1.2.1 From 8e292373ec5c54493ce48af7c76f5eaa79dc8abd Mon Sep 17 00:00:00 2001 From: continue-revolution Date: Mon, 8 Jan 2024 06:43:39 -0600 Subject: lcm sampler --- modules/sd_samplers.py | 3 +- modules/sd_samplers_lcm.py | 104 +++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 106 insertions(+), 1 deletion(-) create mode 100644 modules/sd_samplers_lcm.py (limited to 'modules') diff --git a/modules/sd_samplers.py b/modules/sd_samplers.py index 45faae62..a58528a0 100644 --- a/modules/sd_samplers.py +++ b/modules/sd_samplers.py @@ -1,4 +1,4 @@ -from modules import sd_samplers_kdiffusion, sd_samplers_timesteps, shared +from modules import sd_samplers_kdiffusion, sd_samplers_timesteps, sd_samplers_lcm, shared # imports for functions that previously were here and are used by other modules from modules.sd_samplers_common import samples_to_image_grid, sample_to_image # noqa: F401 @@ -6,6 +6,7 @@ from modules.sd_samplers_common import samples_to_image_grid, sample_to_image # all_samplers = [ *sd_samplers_kdiffusion.samplers_data_k_diffusion, *sd_samplers_timesteps.samplers_data_timesteps, + *sd_samplers_lcm.samplers_data_lcm, ] all_samplers_map = {x.name: x for x in all_samplers} diff --git a/modules/sd_samplers_lcm.py b/modules/sd_samplers_lcm.py new file mode 100644 index 00000000..59839b72 --- /dev/null +++ b/modules/sd_samplers_lcm.py @@ -0,0 +1,104 @@ +import torch + +from k_diffusion import utils, sampling +from k_diffusion.external import DiscreteEpsDDPMDenoiser +from k_diffusion.sampling import default_noise_sampler, trange + +from modules import shared, sd_samplers_cfg_denoiser, sd_samplers_kdiffusion, sd_samplers_common + + +class LCMCompVisDenoiser(DiscreteEpsDDPMDenoiser): + def __init__(self, model): + timesteps = 1000 + original_timesteps = 50 # LCM Original Timesteps (default=50, for current version of LCM) + self.skip_steps = timesteps // original_timesteps + + alphas_cumprod_valid = torch.zeros((original_timesteps), dtype=torch.float32, device=model.device) + for x in range(original_timesteps): + alphas_cumprod_valid[original_timesteps - 1 - x] = model.alphas_cumprod[timesteps - 1 - x * self.skip_steps] + + super().__init__(model, alphas_cumprod_valid, quantize=None) + + + def get_sigmas(self, n=None,): + if n is None: + return sampling.append_zero(self.sigmas.flip(0)) + + start = self.sigma_to_t(self.sigma_max) + end = self.sigma_to_t(self.sigma_min) + + t = torch.linspace(start, end, n, device=shared.sd_model.device) + + return sampling.append_zero(self.t_to_sigma(t)) + + + def sigma_to_t(self, sigma, quantize=None): + log_sigma = sigma.log() + dists = log_sigma - self.log_sigmas[:, None] + return dists.abs().argmin(dim=0).view(sigma.shape) * self.skip_steps + (self.skip_steps - 1) + + + def t_to_sigma(self, timestep): + t = torch.clamp(((timestep - (self.skip_steps - 1)) / self.skip_steps).float(), min=0, max=(len(self.sigmas) - 1)) + return super().t_to_sigma(t) + + + def get_eps(self, *args, **kwargs): + return self.inner_model.apply_model(*args, **kwargs) + + + def get_scaled_out(self, sigma, output, input): + sigma_data = 0.5 + scaled_timestep = utils.append_dims(self.sigma_to_t(sigma), output.ndim) * 10.0 + + c_skip = sigma_data**2 / (scaled_timestep**2 + sigma_data**2) + c_out = scaled_timestep / (scaled_timestep**2 + sigma_data**2) ** 0.5 + + return c_out * output + c_skip * input + + + def forward(self, input, sigma, **kwargs): + c_out, c_in = [utils.append_dims(x, input.ndim) for x in self.get_scalings(sigma)] + eps = self.get_eps(input * c_in, self.sigma_to_t(sigma), **kwargs) + return self.get_scaled_out(sigma, input + eps * c_out, input) + + +def sample_lcm(model, x, sigmas, extra_args=None, callback=None, disable=None, noise_sampler=None): + extra_args = {} if extra_args is None else extra_args + noise_sampler = default_noise_sampler(x) if noise_sampler is None else noise_sampler + s_in = x.new_ones([x.shape[0]]) + + for i in trange(len(sigmas) - 1, disable=disable): + denoised = model(x, sigmas[i] * s_in, **extra_args) + + if callback is not None: + callback({'x': x, 'i': i, 'sigma': sigmas[i], 'sigma_hat': sigmas[i], 'denoised': denoised}) + + x = denoised + if sigmas[i + 1] > 0: + x += sigmas[i + 1] * noise_sampler(sigmas[i], sigmas[i + 1]) + return x + + +class CFGDenoiserLCM(sd_samplers_cfg_denoiser.CFGDenoiser): + @property + def inner_model(self): + if self.model_wrap is None: + denoiser = LCMCompVisDenoiser + self.model_wrap = denoiser(shared.sd_model) + + return self.model_wrap + + +class LCMSampler(sd_samplers_kdiffusion.KDiffusionSampler): + def __init__(self, funcname, sd_model, options=None): + super().__init__(funcname, sd_model, options) + self.model_wrap_cfg = CFGDenoiserLCM(self) + self.model_wrap = self.model_wrap_cfg.inner_model + + +samplers_lcm = [('LCM', sample_lcm, ['k_lcm'], {})] +samplers_data_lcm = [ + sd_samplers_common.SamplerData(label, lambda model, funcname=funcname: LCMSampler(funcname, model), aliases, options) + for label, funcname, aliases, options in samplers_lcm +] -- cgit v1.2.1 From df8aa69a99e38ae59a4e599b9dff11eccf3490f4 Mon Sep 17 00:00:00 2001 From: Sj-Si Date: Mon, 8 Jan 2024 14:10:03 -0500 Subject: Add tree-view display for extra networks. --- modules/shared_options.py | 1 + modules/ui_extra_networks.py | 352 +++++++++++++++++++++++++++++++------------ 2 files changed, 254 insertions(+), 99 deletions(-) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index d2e86ff1..e698c264 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -238,6 +238,7 @@ options_templates.update(options_section(('extra_networks', "Extra Networks", "s "extra_networks_dir_button_function": OptionInfo(False, "Add a '/' to the beginning of directory buttons").info("Buttons will display the contents of the selected directory without acting as a search filter."), "extra_networks_hidden_models": OptionInfo("When searched", "Show cards for models in hidden directories", gr.Radio, {"choices": ["Always", "When searched", "Never"]}).info('"When searched" option will only show the item when the search string has 4 characters or more'), "extra_networks_default_multiplier": OptionInfo(1.0, "Default multiplier for extra networks", gr.Slider, {"minimum": 0.0, "maximum": 2.0, "step": 0.01}), + "extra_networks_tree_view": OptionInfo(False, "Show extra networks using a directory tree view.").needs_reload_ui(), "extra_networks_card_width": OptionInfo(0, "Card width for Extra Networks").info("in pixels"), "extra_networks_card_height": OptionInfo(0, "Card height for Extra Networks").info("in pixels"), "extra_networks_card_text_scale": OptionInfo(1.0, "Card text scale", gr.Slider, {"minimum": 0.0, "maximum": 2.0, "step": 0.01}).info("1 = original size"), diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index fe5d3ba3..8667617b 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -2,6 +2,8 @@ import functools import os.path import urllib.parse from pathlib import Path +from typing import Optional, Union +from dataclasses import dataclass from modules import shared, ui_extra_networks_user_metadata, errors, extra_networks from modules.images import read_info_from_image, save_image_with_geninfo @@ -15,10 +17,8 @@ from modules.ui_components import ToolButton extra_pages = [] allowed_dirs = set() - default_allowed_preview_extensions = ["png", "jpg", "jpeg", "webp", "gif"] - @functools.cache def allowed_preview_extensions_with_extra(extra_extensions=None): return set(default_allowed_preview_extensions) | set(extra_extensions or []) @@ -28,6 +28,58 @@ def allowed_preview_extensions(): return allowed_preview_extensions_with_extra((shared.opts.samples_format, )) +@dataclass +class ExtraNetworksItem: + """Wrapper for dictionaries representing ExtraNetworks items.""" + item: dict + + +def get_tree(paths: Union[str, list[str]], items: dict[str, ExtraNetworksItem]) -> dict: + """Recursively builds a directory tree. + + Args: + paths: Path or list of paths to directories. These paths are treated as roots from which + the tree will be built. + items: A dictionary associating filepaths to an ExtraNetworksItem instance. + + Returns: + The result directory tree. + """ + if isinstance(paths, (str,)): + paths = [paths] + + def _get_tree(_paths: list[str]): + _res = {} + for path in _paths: + if os.path.isdir(path): + dir_items = os.listdir(path) + # Ignore empty directories. + if not dir_items: + continue + dir_tree = _get_tree([os.path.join(path, x) for x in dir_items]) + # We only want to store non-empty folders in the tree. + if dir_tree: + _res[os.path.basename(path)] = dir_tree + else: + if path not in items: + continue + # Add the ExtraNetworksItem to the result. + _res[os.path.basename(path)] = items[path] + return _res + + res = {} + # Handle each root directory separately. + # Each root WILL have a key/value at the root of the result dict though + # the value can be an empty dict if the directory is empty. We want these + # placeholders for empty dirs so we can inform the user later. + for path in paths: + # Wrap the path in a list since that is what the `_get_tree` expects. + res[path] = _get_tree([path]) + if res[path]: + res[path] = res[path][os.path.basename(path)] + + return res + def register_page(page): """registers extra networks page for the UI; recommend doing it in on_before_ui() callback for extensions""" @@ -80,7 +132,7 @@ def get_single_card(page: str = "", tabname: str = "", name: str = ""): item = page.items.get(name) page.read_user_metadata(item) - item_html = page.create_html_for_item(item, tabname) + item_html = page.create_item_html(tabname, item) return JSONResponse({"html": item_html}) @@ -96,13 +148,15 @@ def quote_js(s): s = s.replace('"', '\\"') return f'"{s}"' - class ExtraNetworksPage: def __init__(self, title): self.title = title self.name = title.lower() self.id_page = self.name.replace(" ", "_") - self.card_page = shared.html("extra-networks-card.html") + if shared.opts.extra_networks_tree_view: + self.card_page = shared.html("extra-networks-card-minimal.html") + else: + self.card_page = shared.html("extra-networks-card.html") self.allow_prompt = True self.allow_negative_prompt = False self.metadata = {} @@ -136,12 +190,141 @@ class ExtraNetworksPage: return "" - def create_html(self, tabname): - items_html = '' + def create_item_html(self, tabname: str, item: dict) -> str: + """Generates HTML for a single ExtraNetworks Item + + Args: + tabname: The name of the active tab. + item: Dictionary containing item information. + + Returns: + HTML string generated for this item. + Can be empty if the item is not meant to be shown. + """ + metadata = item.get("metadata") + if metadata: + self.metadata[item["name"]] = metadata + + if "user_metadata" not in item: + self.read_user_metadata(item) + + preview = item.get("preview", None) + height = f"height: {shared.opts.extra_networks_card_height}px;" if shared.opts.extra_networks_card_height else '' + width = f"width: {shared.opts.extra_networks_card_width}px;" if shared.opts.extra_networks_card_width else '' + background_image = f'' if preview else '' + + onclick = item.get("onclick", None) + if onclick is None: + onclick = '"' + html.escape(f"""return cardClicked({quote_js(tabname)}, {item["prompt"]}, {"true" if self.allow_negative_prompt else "false"})""") + '"' + + copy_path_button = f"
" + + metadata_button = "" + metadata = item.get("metadata") + if metadata: + metadata_button = f"" + + edit_button = f"
" + + local_path = "" + filename = item.get("filename", "") + for reldir in self.allowed_directories_for_previews(): + absdir = os.path.abspath(reldir) + + if filename.startswith(absdir): + local_path = filename[len(absdir):] + + # if this is true, the item must not be shown in the default view, and must instead only be + # shown when searching for it + if shared.opts.extra_networks_hidden_models == "Always": + search_only = False + else: + search_only = "/." in local_path or "\\." in local_path + + if search_only and shared.opts.extra_networks_hidden_models == "Never": + return "" + + sort_keys = " ".join([f'data-sort-{k}="{html.escape(str(v))}"' for k, v in item.get("sort_keys", {}).items()]).strip() + + # Some items here might not be used depending on HTML template used. + args = { + "background_image": background_image, + "card_clicked": onclick, + "copy_path_button": copy_path_button, + "description": (item.get("description") or "" if shared.opts.extra_networks_card_show_desc else ""), + "edit_button": edit_button, + "local_preview": quote_js(item["local_preview"]), + "metadata_button": metadata_button, + "name": html.escape(item["name"]), + "prompt": item.get("prompt", None), + "save_card_preview": '"' + html.escape(f"""return saveCardPreview(event, {quote_js(tabname)}, {quote_js(item["local_preview"])})""") + '"', + "search_only": " search_only" if search_only else "", + "search_term": item.get("search_term", ""), + "sort_keys": sort_keys, + "style": f"'display: none; {height}{width}; font-size: {shared.opts.extra_networks_card_text_scale*100}%'", + "tabname": quote_js(tabname), + } + + return self.card_page.format(**args) + + def create_tree_view_html(self, tabname: str) -> str: + """Generates HTML for displaying folders in a tree view. + + Args: + tabname: The name of the active tab. + + Returns: + HTML string generated for this tree view. + """ + self_name_id = self.name.replace(" ", "_") + res = f"
" self.metadata = {} + self.items = {x["name"]: x for x in self.list_items()} + roots = self.allowed_directories_for_previews() + tree_items = {v["filename"]: ExtraNetworksItem(v) for v in self.items.values()} + tree = get_tree([os.path.abspath(x) for x in roots], items=tree_items) + + if not tree: + return res + "
" + + file_template = "
  • {}
  • " + dir_template = ( + "
    " + "{}" + "{}" + "
    " + ) + + def _build_tree(data: Optional[dict[str, ExtraNetworksItem]] = None) -> str: + """Recursively builds HTML for a tree.""" + _res = "
      " + if not data: + return "
      • DIRECTORY IS EMPTY
      " + + for k, v in sorted(data.items(), key=lambda x: shared.natural_sort_key(x[0])): + if isinstance(v, (ExtraNetworksItem,)): + _res += file_template.format(self.create_item_html(tabname, v.item)) + else: + _res += dir_template.format("", k, _build_tree(v)) + return _res + + res += "
        " + # Add each root directory to the tree. + for k, v in sorted(tree.items(), key=lambda x: shared.natural_sort_key(x[0])): + # If root is empty, append the "disabled" attribute to the template details tag. + res += dir_template.format("open" if v else "open disabled", k, _build_tree(v)) + res += "
      " + res += "" + + return res + + def create_card_view_html(self, tabname): + items_html = "" + self.metadata = {} subdirs = {} + for parentdir in [os.path.abspath(x) for x in self.allowed_directories_for_previews()]: for root, dirs, _ in sorted(os.walk(parentdir, followlinks=True), key=lambda x: shared.natural_sort_key(x[0])): for dirname in sorted(dirs, key=shared.natural_sort_key): @@ -171,40 +354,45 @@ class ExtraNetworksPage: if subdirs: subdirs = {"": 1, **subdirs} - subdirs_html = "".join([f""" - -""" for subdir in subdirs]) + subdirs_html_template = ( + "" + ) + subdirs_html = "".join( + [ + subdirs_html_template.format( + " search-all" if subdir == "" else "", + tabname, + html.escape(subdir if subdir != "" else "all"), + ) for subdir in subdirs + ] + ) self.items = {x["name"]: x for x in self.list_items()} for item in self.items.values(): - metadata = item.get("metadata") - if metadata: - self.metadata[item["name"]] = metadata - - if "user_metadata" not in item: - self.read_user_metadata(item) - - items_html += self.create_html_for_item(item, tabname) + items_html += self.create_item_html(tabname, item) - if items_html == '': + if items_html == "": dirs = "".join([f"
    • {x}
    • " for x in self.allowed_directories_for_previews()]) items_html = shared.html("extra-networks-no-cards.html").format(dirs=dirs) self_name_id = self.name.replace(" ", "_") - res = f""" -
      -{subdirs_html} -
      -
      -{items_html} -
      -""" + res = ( + f"
      {subdirs_html}
      " + f"
      {items_html}
      " + ) return res + def create_html(self, tabname): + if shared.opts.extra_networks_tree_view: + return self.create_tree_view_html(tabname) + else: + return self.create_card_view_html(tabname) + def create_item(self, name, index=None): raise NotImplementedError() @@ -214,66 +402,6 @@ class ExtraNetworksPage: def allowed_directories_for_previews(self): return [] - def create_html_for_item(self, item, tabname): - """ - Create HTML for card item in tab tabname; can return empty string if the item is not meant to be shown. - """ - - preview = item.get("preview", None) - - onclick = item.get("onclick", None) - if onclick is None: - onclick = '"' + html.escape(f"""return cardClicked({quote_js(tabname)}, {item["prompt"]}, {"true" if self.allow_negative_prompt else "false"})""") + '"' - - height = f"height: {shared.opts.extra_networks_card_height}px;" if shared.opts.extra_networks_card_height else '' - width = f"width: {shared.opts.extra_networks_card_width}px;" if shared.opts.extra_networks_card_width else '' - background_image = f'' if preview else '' - metadata_button = "" - metadata = item.get("metadata") - if metadata: - metadata_button = f"" - - edit_button = f"
      " - - local_path = "" - filename = item.get("filename", "") - for reldir in self.allowed_directories_for_previews(): - absdir = os.path.abspath(reldir) - - if filename.startswith(absdir): - local_path = filename[len(absdir):] - - # if this is true, the item must not be shown in the default view, and must instead only be - # shown when searching for it - if shared.opts.extra_networks_hidden_models == "Always": - search_only = False - else: - search_only = "/." in local_path or "\\." in local_path - - if search_only and shared.opts.extra_networks_hidden_models == "Never": - return "" - - sort_keys = " ".join([f'data-sort-{k}="{html.escape(str(v))}"' for k, v in item.get("sort_keys", {}).items()]).strip() - - args = { - "background_image": background_image, - "style": f"'display: none; {height}{width}; font-size: {shared.opts.extra_networks_card_text_scale*100}%'", - "prompt": item.get("prompt", None), - "tabname": quote_js(tabname), - "local_preview": quote_js(item["local_preview"]), - "name": html.escape(item["name"]), - "description": (item.get("description") or "" if shared.opts.extra_networks_card_show_desc else ""), - "card_clicked": onclick, - "save_card_preview": '"' + html.escape(f"""return saveCardPreview(event, {quote_js(tabname)}, {quote_js(item["local_preview"])})""") + '"', - "search_term": item.get("search_term", ""), - "metadata_button": metadata_button, - "edit_button": edit_button, - "search_only": " search_only" if search_only else "", - "sort_keys": sort_keys, - } - - return self.card_page.format(**args) - def get_sort_keys(self, path): """ List of default keys used for sorting in the UI. @@ -360,7 +488,6 @@ def pages_in_preferred_order(pages): return sorted(pages, key=lambda x: tab_scores[x.name]) - def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): from modules.ui import switch_values_symbol @@ -381,7 +508,6 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): elem_id = f"{tabname}_{page.id_page}_cards_html" page_elem = gr.HTML('Loading...', elem_id=elem_id) ui.pages.append(page_elem) - page_elem.change(fn=lambda: None, _js='function(){applyExtraNetworkFilter(' + quote_js(tabname) + '); return []}', inputs=[], outputs=[]) editor = page.create_user_metadata_editor(ui, tabname) @@ -390,30 +516,60 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): related_tabs.append(tab) + tab_controls = {} + edit_search = gr.Textbox('', show_label=False, elem_id=tabname+"_extra_search", elem_classes="search", placeholder="Search...", visible=False, interactive=True) dropdown_sort = gr.Dropdown(choices=['Path', 'Name', 'Date Created', 'Date Modified', ], value=shared.opts.extra_networks_card_order_field, elem_id=tabname+"_extra_sort", elem_classes="sort", multiselect=False, visible=False, show_label=False, interactive=True, label=tabname+"_extra_sort_order") button_sortorder = ToolButton(switch_values_symbol, elem_id=tabname+"_extra_sortorder", elem_classes=["sortorder"] + ([] if shared.opts.extra_networks_card_order == "Ascending" else ["sortReverse"]), visible=False, tooltip="Invert sort order") button_refresh = gr.Button('Refresh', elem_id=tabname+"_extra_refresh", visible=False) checkbox_show_dirs = gr.Checkbox(True, label='Show dirs', elem_id=tabname+"_extra_show_dirs", elem_classes="show-dirs", visible=False) + + tab_controls["edit_search"] = edit_search + tab_controls["dropdown_sort"] = dropdown_sort + tab_controls["button_sortorder"] = button_sortorder + tab_controls["button_refresh"] = button_refresh + tab_controls["checkbox_show_dirs"] = checkbox_show_dirs ui.button_save_preview = gr.Button('Save preview', elem_id=tabname+"_save_preview", visible=False) ui.preview_target_filename = gr.Textbox('Preview save filename', elem_id=tabname+"_preview_filename", visible=False) - tab_controls = [edit_search, dropdown_sort, button_sortorder, button_refresh, checkbox_show_dirs] - for tab in unrelated_tabs: - tab.select(fn=lambda: [gr.update(visible=False) for _ in tab_controls], _js='function(){ extraNetworksUrelatedTabSelected("' + tabname + '"); }', inputs=[], outputs=tab_controls, show_progress=False) + tab.select( + fn=lambda: [gr.update(visible=False) for _ in tab_controls], + _js="function(){ extraNetworksUrelatedTabSelected('" + tabname + "'); }", + inputs=[], + outputs=list(tab_controls.values()), + show_progress=False, + ) + + visible_controls = list(tab_controls.keys()) + if shared.opts.extra_networks_tree_view: + visible_controls = ["button_refresh"] for page, tab in zip(ui.stored_extra_pages, related_tabs): allow_prompt = "true" if page.allow_prompt else "false" allow_negative_prompt = "true" if page.allow_negative_prompt else "false" - jscode = 'extraNetworksTabSelected("' + tabname + '", "' + f"{tabname}_{page.id_page}_prompts" + '", ' + allow_prompt + ', ' + allow_negative_prompt + ');' - - tab.select(fn=lambda: [gr.update(visible=True) for _ in tab_controls], _js='function(){ ' + jscode + ' }', inputs=[], outputs=tab_controls, show_progress=False) + jscode = ( + "extraNetworksTabSelected(" + f"'{tabname}', " + f"'{tabname}_{page.id_page}_prompts', " + f"'{allow_prompt}', " + f"'{allow_negative_prompt}'" + ");" + ) + + tab.select( + fn=lambda: [gr.update(visible=k in visible_controls) for k in tab_controls], + _js="function(){ " + jscode + " }", + inputs=[], + outputs=list(tab_controls.values()), + show_progress=False, + ) dropdown_sort.change(fn=lambda: None, _js="function(){ applyExtraNetworkSort('" + tabname + "'); }") + def pages_html(): if not ui.pages_contents: return refresh() @@ -478,5 +634,3 @@ def setup_ui(ui, gallery): for editor in ui.user_metadata_editors: editor.setup_ui(gallery) - - -- cgit v1.2.1 From 34fc215249e2bc0acc66cda47319e40b6e46a05f Mon Sep 17 00:00:00 2001 From: Sj-Si Date: Mon, 8 Jan 2024 14:23:01 -0500 Subject: fix linting --- modules/ui_extra_networks.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index 8667617b..ab484a5d 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -216,7 +216,7 @@ class ExtraNetworksPage: onclick = item.get("onclick", None) if onclick is None: onclick = '"' + html.escape(f"""return cardClicked({quote_js(tabname)}, {item["prompt"]}, {"true" if self.allow_negative_prompt else "false"})""") + '"' - + copy_path_button = f"
      " metadata_button = "" @@ -523,7 +523,7 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): button_sortorder = ToolButton(switch_values_symbol, elem_id=tabname+"_extra_sortorder", elem_classes=["sortorder"] + ([] if shared.opts.extra_networks_card_order == "Ascending" else ["sortReverse"]), visible=False, tooltip="Invert sort order") button_refresh = gr.Button('Refresh', elem_id=tabname+"_extra_refresh", visible=False) checkbox_show_dirs = gr.Checkbox(True, label='Show dirs', elem_id=tabname+"_extra_show_dirs", elem_classes="show-dirs", visible=False) - + tab_controls["edit_search"] = edit_search tab_controls["dropdown_sort"] = dropdown_sort tab_controls["button_sortorder"] = button_sortorder @@ -560,7 +560,7 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): ) tab.select( - fn=lambda: [gr.update(visible=k in visible_controls) for k in tab_controls], + fn=lambda: [gr.update(visible=k in visible_controls) for k in tab_controls], _js="function(){ " + jscode + " }", inputs=[], outputs=list(tab_controls.values()), -- cgit v1.2.1 From 8d986727b39ee6616190559efa1c41c1942b99b0 Mon Sep 17 00:00:00 2001 From: unknown Date: Tue, 9 Jan 2024 03:01:20 -0600 Subject: include tls arguments in api uvicorn init --- modules/api/api.py | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/api/api.py b/modules/api/api.py index 9d1292e9..59e46335 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -879,7 +879,15 @@ class Api: def launch(self, server_name, port, root_path): self.app.include_router(self.router) - uvicorn.run(self.app, host=server_name, port=port, timeout_keep_alive=shared.cmd_opts.timeout_keep_alive, root_path=root_path) + uvicorn.run( + self.app, + host=server_name, + port=port, + timeout_keep_alive=shared.cmd_opts.timeout_keep_alive, + root_path=root_path, + ssl_keyfile=shared.cmd_opts.tls_keyfile, + ssl_certfile=shared.cmd_opts.tls_certfile + ) def kill_webui(self): restart.stop_program() -- cgit v1.2.1 From 209c26a1cb9e4be357ab3c5e7613caf3cbc26183 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 9 Jan 2024 22:11:44 +0800 Subject: improve efficiency and support more device --- modules/devices.py | 60 ++++++++++++++++++++++++++++++++++++-------------- modules/shared_init.py | 1 + 2 files changed, 44 insertions(+), 17 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index ff279ac5..6edfb127 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -110,6 +110,7 @@ device_codeformer: torch.device = None dtype: torch.dtype = torch.float16 dtype_vae: torch.dtype = torch.float16 dtype_unet: torch.dtype = torch.float16 +dtype_inference: torch.dtype = torch.float16 unet_needs_upcast = False @@ -131,21 +132,49 @@ patch_module_list = [ ] -def manual_cast_forward(self, *args, **kwargs): - org_dtype = torch_utils.get_param(self).dtype - self.to(dtype) - args = [arg.to(dtype) if isinstance(arg, torch.Tensor) else arg for arg in args] - kwargs = {k: v.to(dtype) if isinstance(v, torch.Tensor) else v for k, v in kwargs.items()} - result = self.org_forward(*args, **kwargs) - self.to(org_dtype) - return result +def manual_cast_forward(target_dtype): + def forward_wrapper(self, *args, **kwargs): + org_dtype = torch_utils.get_param(self).dtype + if not target_dtype == org_dtype == dtype_inference: + self.to(target_dtype) + args = [ + arg.to(target_dtype) + if isinstance(arg, torch.Tensor) + else arg + for arg in args + ] + kwargs = { + k: v.to(target_dtype) + if isinstance(v, torch.Tensor) + else v + for k, v in kwargs.items() + } + + result = self.org_forward(*args, **kwargs) + self.to(org_dtype) + + if target_dtype != dtype_inference: + if isinstance(result, tuple): + result = tuple( + i.to(dtype_inference) + if isinstance(i, torch.Tensor) + else i + for i in result + ) + elif isinstance(result, torch.Tensor): + result = result.to(dtype_inference) + return result + return forward_wrapper @contextlib.contextmanager -def manual_cast(): +def manual_cast(target_dtype): for module_type in patch_module_list: org_forward = module_type.forward - module_type.forward = manual_cast_forward + if module_type == torch.nn.MultiheadAttention and has_xpu(): + module_type.forward = manual_cast_forward(torch.float32) + else: + module_type.forward = manual_cast_forward(target_dtype) module_type.org_forward = org_forward try: yield None @@ -161,15 +190,12 @@ def autocast(disable=False): if fp8 and device==cpu: return torch.autocast("cpu", dtype=torch.bfloat16, enabled=True) - if fp8 and (dtype == torch.float32 or shared.cmd_opts.precision == "full" or cuda_no_autocast()): - return manual_cast() - - if has_mps() and shared.cmd_opts.precision != "full": - return manual_cast() - - if dtype == torch.float32 or shared.cmd_opts.precision == "full": + if dtype == torch.float32 and shared.cmd_opts.precision == "full": return contextlib.nullcontext() + if has_xpu() or has_mps() or cuda_no_autocast(): + return manual_cast(dtype_inference) + return torch.autocast("cuda") diff --git a/modules/shared_init.py b/modules/shared_init.py index 586be342..935e3a21 100644 --- a/modules/shared_init.py +++ b/modules/shared_init.py @@ -29,6 +29,7 @@ def initialize(): devices.dtype = torch.float32 if cmd_opts.no_half else torch.float16 devices.dtype_vae = torch.float32 if cmd_opts.no_half or cmd_opts.no_half_vae else torch.float16 + devices.dtype_inference = torch.float32 if cmd_opts.precision == 'full' else devices.dtype shared.device = devices.device shared.weight_load_location = None if cmd_opts.lowram else "cpu" -- cgit v1.2.1 From 42e6df723c68af775b73c9fa4f43f99345348689 Mon Sep 17 00:00:00 2001 From: KohakuBlueleaf Date: Tue, 9 Jan 2024 22:39:39 +0800 Subject: Fix bugs when arg dtype doesn't match --- modules/devices.py | 25 ++++++++++--------------- 1 file changed, 10 insertions(+), 15 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index 6edfb127..e0574052 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -134,24 +134,19 @@ patch_module_list = [ def manual_cast_forward(target_dtype): def forward_wrapper(self, *args, **kwargs): + if any( + isinstance(arg, torch.Tensor) and arg.dtype != target_dtype + for arg in args + ): + args = [arg.to(target_dtype) if isinstance(arg, torch.Tensor) else arg for arg in args] + kwargs = {k: v.to(target_dtype) if isinstance(v, torch.Tensor) else v for k, v in kwargs.items()} + org_dtype = torch_utils.get_param(self).dtype - if not target_dtype == org_dtype == dtype_inference: + if org_dtype != target_dtype: self.to(target_dtype) - args = [ - arg.to(target_dtype) - if isinstance(arg, torch.Tensor) - else arg - for arg in args - ] - kwargs = { - k: v.to(target_dtype) - if isinstance(v, torch.Tensor) - else v - for k, v in kwargs.items() - } - result = self.org_forward(*args, **kwargs) - self.to(org_dtype) + if org_dtype != target_dtype: + self.to(org_dtype) if target_dtype != dtype_inference: if isinstance(result, tuple): -- cgit v1.2.1 From c2c05fcca8f3547783c5440c04ec10cc63c65db5 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 9 Jan 2024 22:53:58 +0800 Subject: linting and debugs --- modules/devices.py | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index e0574052..ad36f656 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -140,20 +140,20 @@ def manual_cast_forward(target_dtype): ): args = [arg.to(target_dtype) if isinstance(arg, torch.Tensor) else arg for arg in args] kwargs = {k: v.to(target_dtype) if isinstance(v, torch.Tensor) else v for k, v in kwargs.items()} - + org_dtype = torch_utils.get_param(self).dtype if org_dtype != target_dtype: self.to(target_dtype) result = self.org_forward(*args, **kwargs) if org_dtype != target_dtype: self.to(org_dtype) - + if target_dtype != dtype_inference: if isinstance(result, tuple): result = tuple( - i.to(dtype_inference) - if isinstance(i, torch.Tensor) - else i + i.to(dtype_inference) + if isinstance(i, torch.Tensor) + else i for i in result ) elif isinstance(result, torch.Tensor): @@ -185,7 +185,7 @@ def autocast(disable=False): if fp8 and device==cpu: return torch.autocast("cpu", dtype=torch.bfloat16, enabled=True) - if dtype == torch.float32 and shared.cmd_opts.precision == "full": + if dtype == torch.float32: return contextlib.nullcontext() if has_xpu() or has_mps() or cuda_no_autocast(): -- cgit v1.2.1 From e00365962b17550a42235d1fbe2ad2c7cc4b8961 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 9 Jan 2024 23:13:34 +0800 Subject: Apply correct inference precision implementation --- modules/devices.py | 42 +++++++++++++++++++++++++++++++++--------- 1 file changed, 33 insertions(+), 9 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index ad36f656..9e1f207c 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -132,6 +132,21 @@ patch_module_list = [ ] +def cast_output(result): + if isinstance(result, tuple): + result = tuple(i.to(dtype_inference) if isinstance(i, torch.Tensor) else i for i in result) + elif isinstance(result, torch.Tensor): + result = result.to(dtype_inference) + return result + + +def autocast_with_cast_output(self, *args, **kwargs): + result = self.org_forward(*args, **kwargs) + if dtype_inference != dtype: + result = cast_output(result) + return result + + def manual_cast_forward(target_dtype): def forward_wrapper(self, *args, **kwargs): if any( @@ -149,15 +164,7 @@ def manual_cast_forward(target_dtype): self.to(org_dtype) if target_dtype != dtype_inference: - if isinstance(result, tuple): - result = tuple( - i.to(dtype_inference) - if isinstance(i, torch.Tensor) - else i - for i in result - ) - elif isinstance(result, torch.Tensor): - result = result.to(dtype_inference) + result = cast_output(result) return result return forward_wrapper @@ -178,6 +185,20 @@ def manual_cast(target_dtype): module_type.forward = module_type.org_forward +@contextlib.contextmanager +def precision_full_with_autocast(autocast_ctx): + for module_type in patch_module_list: + org_forward = module_type.forward + module_type.forward = autocast_with_cast_output + module_type.org_forward = org_forward + try: + with autocast_ctx: + yield None + finally: + for module_type in patch_module_list: + module_type.forward = module_type.org_forward + + def autocast(disable=False): if disable: return contextlib.nullcontext() @@ -191,6 +212,9 @@ def autocast(disable=False): if has_xpu() or has_mps() or cuda_no_autocast(): return manual_cast(dtype_inference) + if dtype_inference == torch.float32 and dtype != torch.float32: + return precision_full_with_autocast(torch.autocast("cuda")) + return torch.autocast("cuda") -- cgit v1.2.1 From 1fd69655fe340325863cbd7bf5297e034a6a3a0a Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 9 Jan 2024 23:15:05 +0800 Subject: Revert "Apply correct inference precision implementation" This reverts commit e00365962b17550a42235d1fbe2ad2c7cc4b8961. --- modules/devices.py | 42 +++++++++--------------------------------- 1 file changed, 9 insertions(+), 33 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index 9e1f207c..ad36f656 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -132,21 +132,6 @@ patch_module_list = [ ] -def cast_output(result): - if isinstance(result, tuple): - result = tuple(i.to(dtype_inference) if isinstance(i, torch.Tensor) else i for i in result) - elif isinstance(result, torch.Tensor): - result = result.to(dtype_inference) - return result - - -def autocast_with_cast_output(self, *args, **kwargs): - result = self.org_forward(*args, **kwargs) - if dtype_inference != dtype: - result = cast_output(result) - return result - - def manual_cast_forward(target_dtype): def forward_wrapper(self, *args, **kwargs): if any( @@ -164,7 +149,15 @@ def manual_cast_forward(target_dtype): self.to(org_dtype) if target_dtype != dtype_inference: - result = cast_output(result) + if isinstance(result, tuple): + result = tuple( + i.to(dtype_inference) + if isinstance(i, torch.Tensor) + else i + for i in result + ) + elif isinstance(result, torch.Tensor): + result = result.to(dtype_inference) return result return forward_wrapper @@ -185,20 +178,6 @@ def manual_cast(target_dtype): module_type.forward = module_type.org_forward -@contextlib.contextmanager -def precision_full_with_autocast(autocast_ctx): - for module_type in patch_module_list: - org_forward = module_type.forward - module_type.forward = autocast_with_cast_output - module_type.org_forward = org_forward - try: - with autocast_ctx: - yield None - finally: - for module_type in patch_module_list: - module_type.forward = module_type.org_forward - - def autocast(disable=False): if disable: return contextlib.nullcontext() @@ -212,9 +191,6 @@ def autocast(disable=False): if has_xpu() or has_mps() or cuda_no_autocast(): return manual_cast(dtype_inference) - if dtype_inference == torch.float32 and dtype != torch.float32: - return precision_full_with_autocast(torch.autocast("cuda")) - return torch.autocast("cuda") -- cgit v1.2.1 From 58d5b042cd02f287faabef399134b97d323691f2 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 9 Jan 2024 23:23:40 +0800 Subject: Apply the correct behavior of precision='full' --- modules/devices.py | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index ad36f656..29a270d1 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -185,11 +185,14 @@ def autocast(disable=False): if fp8 and device==cpu: return torch.autocast("cpu", dtype=torch.bfloat16, enabled=True) - if dtype == torch.float32: - return contextlib.nullcontext() - if has_xpu() or has_mps() or cuda_no_autocast(): - return manual_cast(dtype_inference) + return manual_cast(dtype) + + if fp8 and dtype_inference == torch.float32: + return manual_cast(dtype) + + if dtype == torch.float32 or dtype_inference == torch.float32: + return contextlib.nullcontext() return torch.autocast("cuda") -- cgit v1.2.1 From ca671e5d7b9d03227f01e6bcb350032b6d14e722 Mon Sep 17 00:00:00 2001 From: Kohaku-Blueleaf <59680068+KohakuBlueleaf@users.noreply.github.com> Date: Tue, 9 Jan 2024 23:30:55 +0800 Subject: rearrange if-statements for cpu --- modules/devices.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'modules') diff --git a/modules/devices.py b/modules/devices.py index 29a270d1..0321d12c 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -185,15 +185,15 @@ def autocast(disable=False): if fp8 and device==cpu: return torch.autocast("cpu", dtype=torch.bfloat16, enabled=True) - if has_xpu() or has_mps() or cuda_no_autocast(): - return manual_cast(dtype) - if fp8 and dtype_inference == torch.float32: return manual_cast(dtype) if dtype == torch.float32 or dtype_inference == torch.float32: return contextlib.nullcontext() + if has_xpu() or has_mps() or cuda_no_autocast(): + return manual_cast(dtype) + return torch.autocast("cuda") -- cgit v1.2.1 From 4d9f2c3ec8e791b9c354292c4333fc85b8f8d740 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sat, 6 Jan 2024 21:41:29 +0900 Subject: update p.seed and p.subseed --- modules/txt2img.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/txt2img.py b/modules/txt2img.py index c4cc12d2..d22a1f31 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -67,13 +67,16 @@ def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, g geninfo = json.loads(generation_info) all_seeds = geninfo["all_seeds"] + all_subseeds = geninfo["all_subseeds"] image_info = gallery[gallery_index] if 0 <= gallery_index < len(gallery) else gallery[0] p.firstpass_image = infotext_utils.image_from_url_text(image_info) gallery_index_from_end = len(gallery) - gallery_index seed = all_seeds[-gallery_index_from_end if gallery_index_from_end < len(all_seeds) + 1 else 0] - p.script_args = modules.scripts.scripts_txt2img.set_named_arg(p.script_args, 'ScriptSeed', 'seed', seed) + subseed = all_subseeds[-gallery_index_from_end if gallery_index_from_end < len(all_seeds) + 1 else 0] + p.seed = seed + p.subseed = subseed with closing(p): processed = modules.scripts.scripts_txt2img.run(p, *p.script_args) -- cgit v1.2.1 From 3db6938caa719aaa38b52edecf42740ef62b0c3c Mon Sep 17 00:00:00 2001 From: Sj-Si Date: Wed, 10 Jan 2024 18:11:48 -0500 Subject: begin redesign of tree module. --- modules/shared_options.py | 1 - modules/ui_extra_networks.py | 164 ++++++++++++++++++++++++------------------- 2 files changed, 93 insertions(+), 72 deletions(-) (limited to 'modules') diff --git a/modules/shared_options.py b/modules/shared_options.py index e698c264..d2e86ff1 100644 --- a/modules/shared_options.py +++ b/modules/shared_options.py @@ -238,7 +238,6 @@ options_templates.update(options_section(('extra_networks', "Extra Networks", "s "extra_networks_dir_button_function": OptionInfo(False, "Add a '/' to the beginning of directory buttons").info("Buttons will display the contents of the selected directory without acting as a search filter."), "extra_networks_hidden_models": OptionInfo("When searched", "Show cards for models in hidden directories", gr.Radio, {"choices": ["Always", "When searched", "Never"]}).info('"When searched" option will only show the item when the search string has 4 characters or more'), "extra_networks_default_multiplier": OptionInfo(1.0, "Default multiplier for extra networks", gr.Slider, {"minimum": 0.0, "maximum": 2.0, "step": 0.01}), - "extra_networks_tree_view": OptionInfo(False, "Show extra networks using a directory tree view.").needs_reload_ui(), "extra_networks_card_width": OptionInfo(0, "Card width for Extra Networks").info("in pixels"), "extra_networks_card_height": OptionInfo(0, "Card height for Extra Networks").info("in pixels"), "extra_networks_card_text_scale": OptionInfo(1.0, "Card text scale", gr.Slider, {"minimum": 0.0, "maximum": 2.0, "step": 0.01}).info("1 = original size"), diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index ab484a5d..6318594f 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -73,10 +73,11 @@ def get_tree(paths: Union[str, list[str]], items: dict[str, ExtraNetworksItem]) # the value can be an empty dict if the directory is empty. We want these # placeholders for empty dirs so we can inform the user later. for path in paths: + short_path = os.path.basename(path) # Wrap the path in a list since that is what the `_get_tree` expects. - res[path] = _get_tree([path]) - if res[path]: - res[path] = res[path][os.path.basename(path)] + res[short_path] = _get_tree([path]) + if res[short_path]: + res[short_path] = res[short_path][os.path.basename(path)] return res @@ -153,10 +154,9 @@ class ExtraNetworksPage: self.title = title self.name = title.lower() self.id_page = self.name.replace(" ", "_") - if shared.opts.extra_networks_tree_view: - self.card_page = shared.html("extra-networks-card-minimal.html") - else: - self.card_page = shared.html("extra-networks-card.html") + self.extra_networks_pane_template = shared.html("extra-networks-pane.html") + self.card_page_template = shared.html("extra-networks-card.html") + self.card_page_minimal_template = shared.html("extra-networks-card-minimal.html") self.allow_prompt = True self.allow_negative_prompt = False self.metadata = {} @@ -182,15 +182,14 @@ class ExtraNetworksPage: def search_terms_from_path(self, filename, possible_directories=None): abspath = os.path.abspath(filename) - for parentdir in (possible_directories if possible_directories is not None else self.allowed_directories_for_previews()): - parentdir = os.path.abspath(parentdir) + parentdir = os.path.dirname(os.path.abspath(parentdir)) if abspath.startswith(parentdir): - return abspath[len(parentdir):].replace('\\', '/') + return os.path.relpath(abspath, parentdir) return "" - def create_item_html(self, tabname: str, item: dict) -> str: + def create_item_html(self, tabname: str, item: dict, template: Optional[str] = None) -> str: """Generates HTML for a single ExtraNetworks Item Args: @@ -265,7 +264,10 @@ class ExtraNetworksPage: "tabname": quote_js(tabname), } - return self.card_page.format(**args) + if template: + return template.format(**args) + else: + return self.card_page.format(**args) def create_tree_view_html(self, tabname: str) -> str: """Generates HTML for displaying folders in a tree view. @@ -276,53 +278,67 @@ class ExtraNetworksPage: Returns: HTML string generated for this tree view. """ - self_name_id = self.name.replace(" ", "_") - res = f"
      " - - self.metadata = {} - self.items = {x["name"]: x for x in self.list_items()} + res = f"" + # Generate HTML for the tree. roots = self.allowed_directories_for_previews() tree_items = {v["filename"]: ExtraNetworksItem(v) for v in self.items.values()} tree = get_tree([os.path.abspath(x) for x in roots], items=tree_items) if not tree: - return res + "
      " + return res - file_template = "
    • {}
    • " + file_template = "
    • {card}
    • " dir_template = ( - "
      " - "{}" - "{}
    " + "
    " + "" + "{folder_name}" + "" + "
      {content}
    " "
    " ) def _build_tree(data: Optional[dict[str, ExtraNetworksItem]] = None) -> str: """Recursively builds HTML for a tree.""" - _res = "
      " + _res = "" if not data: - return "
      • DIRECTORY IS EMPTY
      " + return "
    • DIRECTORY IS EMPTY
    • " for k, v in sorted(data.items(), key=lambda x: shared.natural_sort_key(x[0])): if isinstance(v, (ExtraNetworksItem,)): - _res += file_template.format(self.create_item_html(tabname, v.item)) + item_html = self.create_item_html(tabname, v.item, self.card_page_minimal_template) + _res += file_template.format(**{"card": item_html}) else: - _res += dir_template.format("", k, _build_tree(v)) + tmp = dir_template.format( + **{ + "attributes": "", + "tabname": tabname, + "folder_name": k, + "content": _build_tree(v), + } + ) + _res += tmp + return _res - res += "
        " # Add each root directory to the tree. for k, v in sorted(tree.items(), key=lambda x: shared.natural_sort_key(x[0])): # If root is empty, append the "disabled" attribute to the template details tag. - res += dir_template.format("open" if v else "open disabled", k, _build_tree(v)) + res += "
          " + res += dir_template.format( + **{ + "attributes": "open" if v else "open", + "tabname": tabname, + "folder_name": k, + "content": _build_tree(v), + } + ) + res += "
        " res += "
      " - res += "" return res - def create_card_view_html(self, tabname): - items_html = "" - self.metadata = {} + def create_subdirs_html(self, tabname): subdirs = {} for parentdir in [os.path.abspath(x) for x in self.allowed_directories_for_previews()]: @@ -355,43 +371,53 @@ class ExtraNetworksPage: subdirs = {"": 1, **subdirs} subdirs_html_template = ( - "" ) - subdirs_html = "".join( + return "".join( [ subdirs_html_template.format( - " search-all" if subdir == "" else "", - tabname, - html.escape(subdir if subdir != "" else "all"), + **{ + "classes": "search-all" if not subdir else "", + "tabname": tabname, + "content": html.escape(subdir if subdir else "all"), + } ) for subdir in subdirs ] ) + def create_card_view_html(self, tabname): + res = "" self.items = {x["name"]: x for x in self.list_items()} for item in self.items.values(): - items_html += self.create_item_html(tabname, item) + res += self.create_item_html(tabname, item, self.card_page_template) - if items_html == "": + if res == "": dirs = "".join([f"
    • {x}
    • " for x in self.allowed_directories_for_previews()]) - items_html = shared.html("extra-networks-no-cards.html").format(dirs=dirs) - - self_name_id = self.name.replace(" ", "_") - - res = ( - f"
      {subdirs_html}
      " - f"
      {items_html}
      " - ) + res = shared.html("extra-networks-no-cards.html").format(dirs=dirs) return res def create_html(self, tabname): - if shared.opts.extra_networks_tree_view: - return self.create_tree_view_html(tabname) - else: - return self.create_card_view_html(tabname) + self.metadata = {} + self.items = {x["name"]: x for x in self.list_items()} + + tree_view_html = self.create_tree_view_html(tabname) + subdirs_html = self.create_subdirs_html(tabname) + card_view_html = self.create_card_view_html(tabname) + network_type_id = self.name.replace(" ", "_") + + return self.extra_networks_pane_template.format( + **{ + "tabname": tabname, + "network_type_id": network_type_id, + "tree_html": tree_view_html, + "subdirs_html": subdirs_html, + "items_html": card_view_html, + } + ) def create_item(self, name, index=None): raise NotImplementedError() @@ -516,19 +542,19 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): related_tabs.append(tab) - tab_controls = {} - edit_search = gr.Textbox('', show_label=False, elem_id=tabname+"_extra_search", elem_classes="search", placeholder="Search...", visible=False, interactive=True) dropdown_sort = gr.Dropdown(choices=['Path', 'Name', 'Date Created', 'Date Modified', ], value=shared.opts.extra_networks_card_order_field, elem_id=tabname+"_extra_sort", elem_classes="sort", multiselect=False, visible=False, show_label=False, interactive=True, label=tabname+"_extra_sort_order") button_sortorder = ToolButton(switch_values_symbol, elem_id=tabname+"_extra_sortorder", elem_classes=["sortorder"] + ([] if shared.opts.extra_networks_card_order == "Ascending" else ["sortReverse"]), visible=False, tooltip="Invert sort order") button_refresh = gr.Button('Refresh', elem_id=tabname+"_extra_refresh", visible=False) checkbox_show_dirs = gr.Checkbox(True, label='Show dirs', elem_id=tabname+"_extra_show_dirs", elem_classes="show-dirs", visible=False) - tab_controls["edit_search"] = edit_search - tab_controls["dropdown_sort"] = dropdown_sort - tab_controls["button_sortorder"] = button_sortorder - tab_controls["button_refresh"] = button_refresh - tab_controls["checkbox_show_dirs"] = checkbox_show_dirs + tab_controls = [ + edit_search, + dropdown_sort, + button_sortorder, + button_refresh, + checkbox_show_dirs, + ] ui.button_save_preview = gr.Button('Save preview', elem_id=tabname+"_save_preview", visible=False) ui.preview_target_filename = gr.Textbox('Preview save filename', elem_id=tabname+"_preview_filename", visible=False) @@ -538,32 +564,28 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): fn=lambda: [gr.update(visible=False) for _ in tab_controls], _js="function(){ extraNetworksUrelatedTabSelected('" + tabname + "'); }", inputs=[], - outputs=list(tab_controls.values()), + outputs=tab_controls, show_progress=False, ) - visible_controls = list(tab_controls.keys()) - if shared.opts.extra_networks_tree_view: - visible_controls = ["button_refresh"] - for page, tab in zip(ui.stored_extra_pages, related_tabs): allow_prompt = "true" if page.allow_prompt else "false" allow_negative_prompt = "true" if page.allow_negative_prompt else "false" jscode = ( "extraNetworksTabSelected(" - f"'{tabname}', " - f"'{tabname}_{page.id_page}_prompts', " - f"'{allow_prompt}', " - f"'{allow_negative_prompt}'" + f"'{tabname}', " + f"'{tabname}_{page.id_page}_prompts', " + f"'{allow_prompt}', " + f"'{allow_negative_prompt}'" ");" ) tab.select( - fn=lambda: [gr.update(visible=k in visible_controls) for k in tab_controls], + fn=lambda: [gr.update(visible=True) for _ in tab_controls], _js="function(){ " + jscode + " }", inputs=[], - outputs=list(tab_controls.values()), + outputs=tab_controls, show_progress=False, ) -- cgit v1.2.1 From 0011640ab187e5a59098bb80d165a23f9d08568c Mon Sep 17 00:00:00 2001 From: Aarni Koskela Date: Thu, 11 Jan 2024 08:29:42 +0200 Subject: Logging: set formatter correctly for fallback logger too --- modules/logging_config.py | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'modules') diff --git a/modules/logging_config.py b/modules/logging_config.py index 11eee9a6..8e31d8c9 100644 --- a/modules/logging_config.py +++ b/modules/logging_config.py @@ -36,20 +36,21 @@ def setup_logging(loglevel): # Already configured, do not interfere return + formatter = logging.Formatter( + '%(asctime)s %(levelname)s [%(name)s] %(message)s', + '%Y-%m-%d %H:%M:%S', + ) + if os.environ.get("SD_WEBUI_RICH_LOG"): from rich.logging import RichHandler handler = RichHandler() else: handler = logging.StreamHandler() + handler.setFormatter(formatter) if TqdmLoggingHandler: handler = TqdmLoggingHandler(handler) - formatter = logging.Formatter( - '%(asctime)s %(levelname)s [%(name)s] %(message)s', - '%Y-%m-%d %H:%M:%S', - ) - handler.setFormatter(formatter) log_level = getattr(logging, loglevel.upper(), None) or logging.INFO -- cgit v1.2.1 From 0726a6e12e85a37d1e514f5603acf9f058c11783 Mon Sep 17 00:00:00 2001 From: Sj-Si Date: Thu, 11 Jan 2024 15:06:57 -0500 Subject: Finish base layout. Fix bugs. Need to test for stability and clean up. --- modules/ui_extra_networks.py | 113 +++++++++---------------- modules/ui_extra_networks_checkpoints.py | 5 +- modules/ui_extra_networks_hypernets.py | 6 +- modules/ui_extra_networks_textual_inversion.py | 5 +- 4 files changed, 51 insertions(+), 78 deletions(-) (limited to 'modules') diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index 6318594f..2e226ba0 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -48,23 +48,24 @@ def get_tree(paths: Union[str, list[str]], items: dict[str, ExtraNetworksItem]) if isinstance(paths, (str,)): paths = [paths] - def _get_tree(_paths: list[str]): + def _get_tree(_paths: list[str], _root: str): _res = {} for path in _paths: + relpath = os.path.relpath(path, _root) if os.path.isdir(path): dir_items = os.listdir(path) # Ignore empty directories. if not dir_items: continue - dir_tree = _get_tree([os.path.join(path, x) for x in dir_items]) + dir_tree = _get_tree([os.path.join(path, x) for x in dir_items], _root) # We only want to store non-empty folders in the tree. if dir_tree: - _res[os.path.basename(path)] = dir_tree + _res[relpath] = dir_tree else: if path not in items: continue # Add the ExtraNetworksItem to the result. - _res[os.path.basename(path)] = items[path] + _res[relpath] = items[path] return _res res = {} @@ -73,11 +74,13 @@ def get_tree(paths: Union[str, list[str]], items: dict[str, ExtraNetworksItem]) # the value can be an empty dict if the directory is empty. We want these # placeholders for empty dirs so we can inform the user later. for path in paths: - short_path = os.path.basename(path) + root = os.path.dirname(path) + relpath = os.path.relpath(path, root) # Wrap the path in a list since that is what the `_get_tree` expects. - res[short_path] = _get_tree([path]) - if res[short_path]: - res[short_path] = res[short_path][os.path.basename(path)] + res[relpath] = _get_tree([path], root) + if res[relpath]: + # We need to pull the inner path out one for these root dirs. + res[relpath] = res[relpath][relpath] return res @@ -245,6 +248,17 @@ class ExtraNetworksPage: sort_keys = " ".join([f'data-sort-{k}="{html.escape(str(v))}"' for k, v in item.get("sort_keys", {}).items()]).strip() + search_terms_html = "" + search_term_template = "{search_term}" + for search_term in item.get("search_terms", []): + search_terms_html += search_term_template.format( + **{ + "style": "display: none;", + "class": "search_terms" + (" search_only" if search_only else ""), + "search_term": search_term, + } + ) + # Some items here might not be used depending on HTML template used. args = { "background_image": background_image, @@ -258,7 +272,7 @@ class ExtraNetworksPage: "prompt": item.get("prompt", None), "save_card_preview": '"' + html.escape(f"""return saveCardPreview(event, {quote_js(tabname)}, {quote_js(item["local_preview"])})""") + '"', "search_only": " search_only" if search_only else "", - "search_term": item.get("search_term", ""), + "search_terms": search_terms_html, "sort_keys": sort_keys, "style": f"'display: none; {height}{width}; font-size: {shared.opts.extra_networks_card_text_scale*100}%'", "tabname": quote_js(tabname), @@ -278,7 +292,7 @@ class ExtraNetworksPage: Returns: HTML string generated for this tree view. """ - res = f"" + res = "" # Generate HTML for the tree. roots = self.allowed_directories_for_previews() @@ -291,7 +305,8 @@ class ExtraNetworksPage: file_template = "
    • {card}
    • " dir_template = ( "
      " - "" + "" "{folder_name}" "" "
        {content}
      " @@ -309,16 +324,15 @@ class ExtraNetworksPage: item_html = self.create_item_html(tabname, v.item, self.card_page_minimal_template) _res += file_template.format(**{"card": item_html}) else: - tmp = dir_template.format( + _res += dir_template.format( **{ "attributes": "", "tabname": tabname, - "folder_name": k, + "folder_name": os.path.basename(k), + "data_path": k, "content": _build_tree(v), } ) - _res += tmp - return _res # Add each root directory to the tree. @@ -329,65 +343,15 @@ class ExtraNetworksPage: **{ "attributes": "open" if v else "open", "tabname": tabname, - "folder_name": k, + "folder_name": os.path.basename(k), + "data_path": k, "content": _build_tree(v), } ) res += "
    " res += "" - return res - def create_subdirs_html(self, tabname): - subdirs = {} - - for parentdir in [os.path.abspath(x) for x in self.allowed_directories_for_previews()]: - for root, dirs, _ in sorted(os.walk(parentdir, followlinks=True), key=lambda x: shared.natural_sort_key(x[0])): - for dirname in sorted(dirs, key=shared.natural_sort_key): - x = os.path.join(root, dirname) - - if not os.path.isdir(x): - continue - - subdir = os.path.abspath(x)[len(parentdir):].replace("\\", "/") - - if shared.opts.extra_networks_dir_button_function: - if not subdir.startswith("/"): - subdir = "/" + subdir - else: - while subdir.startswith("/"): - subdir = subdir[1:] - - is_empty = len(os.listdir(x)) == 0 - if not is_empty and not subdir.endswith("/"): - subdir = subdir + "/" - - if ("/." in subdir or subdir.startswith(".")) and not shared.opts.extra_networks_show_hidden_directories: - continue - - subdirs[subdir] = 1 - - if subdirs: - subdirs = {"": 1, **subdirs} - - subdirs_html_template = ( - "" - ) - return "".join( - [ - subdirs_html_template.format( - **{ - "classes": "search-all" if not subdir else "", - "tabname": tabname, - "content": html.escape(subdir if subdir else "all"), - } - ) for subdir in subdirs - ] - ) - def create_card_view_html(self, tabname): res = "" self.items = {x["name"]: x for x in self.list_items()} @@ -405,7 +369,6 @@ class ExtraNetworksPage: self.items = {x["name"]: x for x in self.list_items()} tree_view_html = self.create_tree_view_html(tabname) - subdirs_html = self.create_subdirs_html(tabname) card_view_html = self.create_card_view_html(tabname) network_type_id = self.name.replace(" ", "_") @@ -414,7 +377,6 @@ class ExtraNetworksPage: "tabname": tabname, "network_type_id": network_type_id, "tree_html": tree_view_html, - "subdirs_html": subdirs_html, "items_html": card_view_html, } ) @@ -534,7 +496,12 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): elem_id = f"{tabname}_{page.id_page}_cards_html" page_elem = gr.HTML('Loading...', elem_id=elem_id) ui.pages.append(page_elem) - page_elem.change(fn=lambda: None, _js='function(){applyExtraNetworkFilter(' + quote_js(tabname) + '); return []}', inputs=[], outputs=[]) + page_elem.change( + fn=lambda: None, + _js=f"function(){{applyExtraNetworkFilter({tabname}_extra_search); return []}}", + inputs=[], + outputs=[], + ) editor = page.create_user_metadata_editor(ui, tabname) editor.create_ui() @@ -542,18 +509,16 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): related_tabs.append(tab) - edit_search = gr.Textbox('', show_label=False, elem_id=tabname+"_extra_search", elem_classes="search", placeholder="Search...", visible=False, interactive=True) + edit_search = gr.Textbox('', show_label=False, elem_id=f"{tabname}_extra_search", elem_classes="search", placeholder="Search...", visible=False, interactive=True) dropdown_sort = gr.Dropdown(choices=['Path', 'Name', 'Date Created', 'Date Modified', ], value=shared.opts.extra_networks_card_order_field, elem_id=tabname+"_extra_sort", elem_classes="sort", multiselect=False, visible=False, show_label=False, interactive=True, label=tabname+"_extra_sort_order") button_sortorder = ToolButton(switch_values_symbol, elem_id=tabname+"_extra_sortorder", elem_classes=["sortorder"] + ([] if shared.opts.extra_networks_card_order == "Ascending" else ["sortReverse"]), visible=False, tooltip="Invert sort order") button_refresh = gr.Button('Refresh', elem_id=tabname+"_extra_refresh", visible=False) - checkbox_show_dirs = gr.Checkbox(True, label='Show dirs', elem_id=tabname+"_extra_show_dirs", elem_classes="show-dirs", visible=False) tab_controls = [ edit_search, dropdown_sort, button_sortorder, button_refresh, - checkbox_show_dirs, ] ui.button_save_preview = gr.Button('Save preview', elem_id=tabname+"_save_preview", visible=False) @@ -562,7 +527,7 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): for tab in unrelated_tabs: tab.select( fn=lambda: [gr.update(visible=False) for _ in tab_controls], - _js="function(){ extraNetworksUrelatedTabSelected('" + tabname + "'); }", + _js=f"function(){{ extraNetworksUnrelatedTabSelected('{tabname}'); }}", inputs=[], outputs=tab_controls, show_progress=False, diff --git a/modules/ui_extra_networks_checkpoints.py b/modules/ui_extra_networks_checkpoints.py index 1693e71f..e7976ba1 100644 --- a/modules/ui_extra_networks_checkpoints.py +++ b/modules/ui_extra_networks_checkpoints.py @@ -21,13 +21,16 @@ class ExtraNetworksPageCheckpoints(ui_extra_networks.ExtraNetworksPage): return path, ext = os.path.splitext(checkpoint.filename) + search_terms = [self.search_terms_from_path(checkpoint.filename)] + if checkpoint.sha256: + search_terms.append(checkpoint.sha256) return { "name": checkpoint.name_for_extra, "filename": checkpoint.filename, "shorthash": checkpoint.shorthash, "preview": self.find_preview(path), "description": self.find_description(path), - "search_term": self.search_terms_from_path(checkpoint.filename) + " " + (checkpoint.sha256 or ""), + "search_terms": search_terms, "onclick": '"' + html.escape(f"""return selectCheckpoint({quote_js(name)})""") + '"', "local_preview": f"{path}.{shared.opts.samples_format}", "metadata": checkpoint.metadata, diff --git a/modules/ui_extra_networks_hypernets.py b/modules/ui_extra_networks_hypernets.py index c96c4fa3..2fb4bd19 100644 --- a/modules/ui_extra_networks_hypernets.py +++ b/modules/ui_extra_networks_hypernets.py @@ -20,14 +20,16 @@ class ExtraNetworksPageHypernetworks(ui_extra_networks.ExtraNetworksPage): path, ext = os.path.splitext(full_path) sha256 = sha256_from_cache(full_path, f'hypernet/{name}') shorthash = sha256[0:10] if sha256 else None - + search_terms = [self.search_terms_from_path(path)] + if sha256: + search_terms.append(sha256) return { "name": name, "filename": full_path, "shorthash": shorthash, "preview": self.find_preview(path), "description": self.find_description(path), - "search_term": self.search_terms_from_path(path) + " " + (sha256 or ""), + "search_terms": search_terms, "prompt": quote_js(f""), "local_preview": f"{path}.preview.{shared.opts.samples_format}", "sort_keys": {'default': index, **self.get_sort_keys(path + ext)}, diff --git a/modules/ui_extra_networks_textual_inversion.py b/modules/ui_extra_networks_textual_inversion.py index 1b334fda..deb7cb87 100644 --- a/modules/ui_extra_networks_textual_inversion.py +++ b/modules/ui_extra_networks_textual_inversion.py @@ -18,13 +18,16 @@ class ExtraNetworksPageTextualInversion(ui_extra_networks.ExtraNetworksPage): return path, ext = os.path.splitext(embedding.filename) + search_terms = [self.search_terms_from_path(embedding.filename)] + if embedding.hash: + search_terms.append(embedding.hash) return { "name": name, "filename": embedding.filename, "shorthash": embedding.shorthash, "preview": self.find_preview(path), "description": self.find_description(path), - "search_term": self.search_terms_from_path(embedding.filename) + " " + (embedding.hash or ""), + "search_terms": search_terms, "prompt": quote_js(embedding.name), "local_preview": f"{path}.preview.{shared.opts.samples_format}", "sort_keys": {'default': index, **self.get_sort_keys(embedding.filename)}, -- cgit v1.2.1 From 47b52d9b28aaabb603358fae5d9b824c49aa627b Mon Sep 17 00:00:00 2001 From: WebDev <6970043+WebDev9000@users.noreply.github.com> Date: Sat, 13 Jan 2024 02:31:26 -0800 Subject: Add # to the invalid_filename_chars list --- modules/images.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/images.py b/modules/images.py index 87a7bf22..b6f2358c 100644 --- a/modules/images.py +++ b/modules/images.py @@ -321,7 +321,7 @@ def resize_image(resize_mode, im, width, height, upscaler_name=None): return res -invalid_filename_chars = '<>:"/\\|?*\n\r\t' +invalid_filename_chars = '#<>:"/\\|?*\n\r\t' invalid_filename_prefix = ' ' invalid_filename_postfix = ' .' re_nonletters = re.compile(r'[\s' + string.punctuation + ']+') -- cgit v1.2.1 From b6dc307c99a25bc3f3f3e96ce640a4710c14e133 Mon Sep 17 00:00:00 2001 From: Andray Date: Sat, 13 Jan 2024 14:45:15 +0400 Subject: fix_extension_check_for_requirements --- modules/extensions.py | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/extensions.py b/modules/extensions.py index 99e7ee60..04bda297 100644 --- a/modules/extensions.py +++ b/modules/extensions.py @@ -224,13 +224,16 @@ def list_extensions(): # check for requirements for extension in extensions: + if not extension.enabled: + continue + for req in extension.metadata.requires: required_extension = loaded_extensions.get(req) if required_extension is None: errors.report(f'Extension "{extension.name}" requires "{req}" which is not installed.', exc_info=False) continue - if not extension.enabled: + if not required_extension.enabled: errors.report(f'Extension "{extension.name}" requires "{required_extension.name}" which is disabled.', exc_info=False) continue -- cgit v1.2.1 From 02e6963325e5221e0efb96a63f3dc849550489b7 Mon Sep 17 00:00:00 2001 From: Sj-Si Date: Sat, 13 Jan 2024 13:16:39 -0500 Subject: continue cleanup and redesign. --- modules/ui_extra_networks.py | 177 +++++++++++++++++++++++++++++++++++-------- 1 file changed, 144 insertions(+), 33 deletions(-) (limited to 'modules') diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index 093ac7b4..9cf5b57f 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -19,6 +19,90 @@ extra_pages = [] allowed_dirs = set() default_allowed_preview_extensions = ["png", "jpg", "jpeg", "webp", "gif"] +tree_tpl = ( + "" + "
      " + "{content}" + "
    " +) + +tree_ul_tpl = ( + "
      " + "{content}" + "
    " +) + +tree_li_dir_tpl = ( + "
  • " + "{content}" + "
  • " +) +tree_li_file_tpl = ( + "
  • " + "{content}" + "
  • " +) + +tree_btn_dir_tpl = ( + "" +) + +tree_btn_file_action_buttons_tpl = ( + "
    " + "
    " + "
    " + "
    " + "
    " + "
    " +) + +tree_btn_file_tpl = ( + "" + "" +) + + @functools.cache def allowed_preview_extensions_with_extra(extra_extensions=None): return set(default_allowed_preview_extensions) | set(extra_extensions or []) @@ -160,6 +244,7 @@ class ExtraNetworksPage: self.extra_networks_pane_template = shared.html("extra-networks-pane.html") self.card_page_template = shared.html("extra-networks-card.html") self.card_page_minimal_template = shared.html("extra-networks-card-minimal.html") + self.tree_button_template = shared.html("extra-networks-tree-button.html") self.allow_prompt = True self.allow_negative_prompt = False self.metadata = {} @@ -279,7 +364,9 @@ class ExtraNetworksPage: "search_terms": search_terms_html, "sort_keys": sort_keys, "style": f"'display: none; {height}{width}; font-size: {shared.opts.extra_networks_card_text_scale*100}%'", - "tabname": quote_js(tabname), + "tabname": tabname, + "tab_id": self.id_page, + } if template: @@ -306,55 +393,81 @@ class ExtraNetworksPage: if not tree: return res - file_template = "
  • {card}
  • " - dir_template = ( - "
    " - "" - "{folder_name}" - "" - "
      {content}
    " - "
    " - ) - def _build_tree(data: Optional[dict[str, ExtraNetworksItem]] = None) -> str: """Recursively builds HTML for a tree.""" _res = "" if not data: - return "
  • DIRECTORY IS EMPTY
  • " + return ( + "
    " + "Directory is empty" + "
    " + ) for k, v in sorted(data.items(), key=lambda x: shared.natural_sort_key(x[0])): if isinstance(v, (ExtraNetworksItem,)): - item_html = self.create_item_html(tabname, v.item, self.card_page_minimal_template) - _res += file_template.format(**{"card": item_html}) + _action_buttons = tree_btn_file_action_buttons_tpl.format( + **{ + "path": quote_js(k), + "filename": quote_js(v.item["name"]), + "tabname": quote_js(tabname), + "tab_id": quote_js(self.id_page), + } + ) + _btn = tree_btn_file_tpl.format( + **{ + "label": v.item["name"], + "filter": v.item["search_terms"], + "tabname": tabname, + "tab_id": self.id_page, + "buttons": _action_buttons, + } + ) + _li = tree_li_file_tpl.format( + **{ + "hash": v.item["shorthash"], + "path": k, + "type": "file", + #"content": _btn, + "content": self.create_item_html(tabname, v.item, self.tree_button_template), + } + ) + _res += _li + #item_html = self.create_item_html(tabname, v.item, self.card_page_minimal_template) + #_res += file_template.format(**{"card": item_html}) else: - _res += dir_template.format( + _btn = tree_btn_dir_tpl.format( **{ - "attributes": "", + "label": os.path.basename(k), "tabname": tabname, - "folder_name": os.path.basename(k), - "data_path": k, - "content": _build_tree(v), + "tab_id": self.id_page, } ) + _ul = tree_ul_tpl.format(**{"content": _build_tree(v)}) + _li = tree_li_dir_tpl.format(**{"content": _btn + _ul, "path": k}) + _res += _li return _res # Add each root directory to the tree. for k, v in sorted(tree.items(), key=lambda x: shared.natural_sort_key(x[0])): # If root is empty, append the "disabled" attribute to the template details tag. - res += "
      " - res += dir_template.format( + btn = tree_btn_dir_tpl.format( **{ - "attributes": "open" if v else "open", + "label": os.path.basename(k), "tabname": tabname, - "folder_name": os.path.basename(k), - "data_path": k, - "content": _build_tree(v), + "tab_id": self.id_page, } ) - res += "
    " - res += "" - return res + ul = tree_ul_tpl.format(**{"content": _build_tree(v)}) + li = tree_li_dir_tpl.format(**{"content": btn + ul, "path": k}) + res += li + + return tree_tpl.format( + **{ + "content": res, + "tabname": tabname, + "tab_id": self.id_page, + } + ) def create_card_view_html(self, tabname): res = "" @@ -375,7 +488,7 @@ class ExtraNetworksPage: tree_view_html = self.create_tree_view_html(tabname) card_view_html = self.create_card_view_html(tabname) - network_type_id = self.name.replace(" ", "_") + network_type_id = self.id_page return self.extra_networks_pane_template.format( **{ @@ -506,7 +619,7 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): ui.pages.append(page_elem) page_elem.change( fn=lambda: None, - _js=f"function(){{applyExtraNetworkFilter({tabname}_extra_search); return []}}", + _js=f"function(){{applyExtraNetworkFilter({tabname}_{page.id_page}_extra_search); return []}}", inputs=[], outputs=[], ) @@ -517,13 +630,11 @@ def create_ui(interface: gr.Blocks, unrelated_tabs, tabname): related_tabs.append(tab) - edit_search = gr.Textbox('', show_label=False, elem_id=f"{tabname}_extra_search", elem_classes="search", placeholder="Search...", visible=False, interactive=True) dropdown_sort = gr.Dropdown(choices=['Path', 'Name', 'Date Created', 'Date Modified', ], value=shared.opts.extra_networks_card_order_field, elem_id=tabname+"_extra_sort", elem_classes="sort", multiselect=False, visible=False, show_label=False, interactive=True, label=tabname+"_extra_sort_order") button_sortorder = ToolButton(switch_values_symbol, elem_id=tabname+"_extra_sortorder", elem_classes=["sortorder"] + ([] if shared.opts.extra_networks_card_order == "Ascending" else ["sortReverse"]), visible=False, tooltip="Invert sort order") button_refresh = gr.Button('Refresh', elem_id=tabname+"_extra_refresh", visible=False) tab_controls = [ - edit_search, dropdown_sort, button_sortorder, button_refresh, -- cgit v1.2.1 From 92501d4f8006866cec3438c038e1588315d835bc Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sat, 13 Jan 2024 02:20:06 +0900 Subject: disable saving images before highres fix --- modules/txt2img.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/txt2img.py b/modules/txt2img.py index d22a1f31..11558785 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -77,7 +77,7 @@ def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, g subseed = all_subseeds[-gallery_index_from_end if gallery_index_from_end < len(all_seeds) + 1 else 0] p.seed = seed p.subseed = subseed - + p.override_settings['save_images_before_highres_fix'] = False with closing(p): processed = modules.scripts.scripts_txt2img.run(p, *p.script_args) -- cgit v1.2.1 From cfb90a938eff6d5d4cfa39f58ebc0ab32ffedfb3 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sat, 13 Jan 2024 02:22:32 +0900 Subject: allowe hr pass to return multiple images --- modules/txt2img.py | 13 ++++--------- 1 file changed, 4 insertions(+), 9 deletions(-) (limited to 'modules') diff --git a/modules/txt2img.py b/modules/txt2img.py index 11558785..daf8f51a 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -88,18 +88,13 @@ def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, g new_gallery = [] for i, image in enumerate(gallery): - fake_image = Image.new(mode="RGB", size=(1, 1)) - if i == gallery_index: - already_saved_as = getattr(processed.images[0], 'already_saved_as', None) - if already_saved_as is not None: - fake_image.already_saved_as = already_saved_as - else: - fake_image = processed.images[0] + geninfo["infotexts"][gallery_index: gallery_index+1] = processed.infotexts + new_gallery.extend(processed.images) else: + fake_image = Image.new(mode="RGB", size=(1, 1)) fake_image.already_saved_as = image["name"] - - new_gallery.append(fake_image) + new_gallery.append(fake_image) geninfo["infotexts"][gallery_index] = processed.info -- cgit v1.2.1 From ee9d4870811a34533ad5d20ed8aca2ff116fd3b9 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sat, 13 Jan 2024 02:23:16 +0900 Subject: fix gallery black image issue --- modules/txt2img.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'modules') diff --git a/modules/txt2img.py b/modules/txt2img.py index daf8f51a..7a1a1626 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -93,7 +93,7 @@ def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, g new_gallery.extend(processed.images) else: fake_image = Image.new(mode="RGB", size=(1, 1)) - fake_image.already_saved_as = image["name"] + fake_image.already_saved_as = image["name"].rsplit('?', 1)[0] new_gallery.append(fake_image) geninfo["infotexts"][gallery_index] = processed.info -- cgit v1.2.1 From 208ccfbe7c04e1dd5b262f705a904204e8297102 Mon Sep 17 00:00:00 2001 From: w-e-w <40751091+w-e-w@users.noreply.github.com> Date: Sun, 14 Jan 2024 02:11:39 +0900 Subject: seed info from infotexts --- modules/txt2img.py | 14 ++++++-------- 1 file changed, 6 insertions(+), 8 deletions(-) (limited to 'modules') diff --git a/modules/txt2img.py b/modules/txt2img.py index 7a1a1626..e617cb1c 100644 --- a/modules/txt2img.py +++ b/modules/txt2img.py @@ -3,7 +3,7 @@ from contextlib import closing import modules.scripts from modules import processing, infotext_utils -from modules.infotext_utils import create_override_settings_dict +from modules.infotext_utils import create_override_settings_dict, parse_generation_parameters from modules.shared import opts import modules.shared as shared from modules.ui import plaintext_to_html @@ -66,18 +66,16 @@ def txt2img_upscale(id_task: str, request: gr.Request, gallery, gallery_index, g p.n_iter = 1 geninfo = json.loads(generation_info) - all_seeds = geninfo["all_seeds"] - all_subseeds = geninfo["all_subseeds"] image_info = gallery[gallery_index] if 0 <= gallery_index < len(gallery) else gallery[0] p.firstpass_image = infotext_utils.image_from_url_text(image_info) - gallery_index_from_end = len(gallery) - gallery_index - seed = all_seeds[-gallery_index_from_end if gallery_index_from_end < len(all_seeds) + 1 else 0] - subseed = all_subseeds[-gallery_index_from_end if gallery_index_from_end < len(all_seeds) + 1 else 0] - p.seed = seed - p.subseed = subseed + parameters = parse_generation_parameters(geninfo.get('infotexts')[gallery_index]) + p.seed = parameters.get('Seed', -1) + p.subseed = parameters.get('Variation seed', -1) + p.override_settings['save_images_before_highres_fix'] = False + with closing(p): processed = modules.scripts.scripts_txt2img.run(p, *p.script_args) -- cgit v1.2.1 From d88424ef2a5d175fb3cb8b557da0abefb468bae0 Mon Sep 17 00:00:00 2001 From: Sj-Si Date: Mon, 15 Jan 2024 13:40:47 -0500 Subject: fix bugs. introduce new ones. --- modules/ui_extra_networks.py | 187 ++++++++++++++++++++++--------------------- 1 file changed, 95 insertions(+), 92 deletions(-) (limited to 'modules') diff --git a/modules/ui_extra_networks.py b/modules/ui_extra_networks.py index 9cf5b57f..a49c6c1c 100644 --- a/modules/ui_extra_networks.py +++ b/modules/ui_extra_networks.py @@ -20,28 +20,28 @@ allowed_dirs = set() default_allowed_preview_extensions = ["png", "jpg", "jpeg", "webp", "gif"] tree_tpl = ( - "