diff options
40 files changed, 1525 insertions, 174 deletions
diff --git a/configs/sd_xl_inpaint.yaml b/configs/sd_xl_inpaint.yaml new file mode 100644 index 00000000..3bad3721 --- /dev/null +++ b/configs/sd_xl_inpaint.yaml @@ -0,0 +1,98 @@ +model: + target: sgm.models.diffusion.DiffusionEngine + params: + scale_factor: 0.13025 + disable_first_stage_autocast: True + + denoiser_config: + target: sgm.modules.diffusionmodules.denoiser.DiscreteDenoiser + params: + num_idx: 1000 + + weighting_config: + target: sgm.modules.diffusionmodules.denoiser_weighting.EpsWeighting + scaling_config: + target: sgm.modules.diffusionmodules.denoiser_scaling.EpsScaling + discretization_config: + target: sgm.modules.diffusionmodules.discretizer.LegacyDDPMDiscretization + + network_config: + target: sgm.modules.diffusionmodules.openaimodel.UNetModel + params: + adm_in_channels: 2816 + num_classes: sequential + use_checkpoint: True + in_channels: 9 + out_channels: 4 + model_channels: 320 + attention_resolutions: [4, 2] + num_res_blocks: 2 + channel_mult: [1, 2, 4] + num_head_channels: 64 + use_spatial_transformer: True + use_linear_in_transformer: True + transformer_depth: [1, 2, 10] # note: the first is unused (due to attn_res starting at 2) 32, 16, 8 --> 64, 32, 16 + context_dim: 2048 + spatial_transformer_attn_type: softmax-xformers + legacy: False + + conditioner_config: + target: sgm.modules.GeneralConditioner + params: + emb_models: + # crossattn cond + - is_trainable: False + input_key: txt + target: sgm.modules.encoders.modules.FrozenCLIPEmbedder + params: + layer: hidden + layer_idx: 11 + # crossattn and vector cond + - is_trainable: False + input_key: txt + target: sgm.modules.encoders.modules.FrozenOpenCLIPEmbedder2 + params: + arch: ViT-bigG-14 + version: laion2b_s39b_b160k + freeze: True + layer: penultimate + always_return_pooled: True + legacy: False + # vector cond + - is_trainable: False + input_key: original_size_as_tuple + target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND + params: + outdim: 256 # multiplied by two + # vector cond + - is_trainable: False + input_key: crop_coords_top_left + target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND + params: + outdim: 256 # multiplied by two + # vector cond + - is_trainable: False + input_key: target_size_as_tuple + target: sgm.modules.encoders.modules.ConcatTimestepEmbedderND + params: + outdim: 256 # multiplied by two + + first_stage_config: + target: sgm.models.autoencoder.AutoencoderKLInferenceWrapper + params: + embed_dim: 4 + monitor: val/rec_loss + ddconfig: + attn_type: vanilla-xformers + double_z: true + z_channels: 4 + resolution: 256 + in_channels: 3 + out_ch: 3 + ch: 128 + ch_mult: [1, 2, 4, 4] + num_res_blocks: 2 + attn_resolutions: [] + dropout: 0.0 + lossconfig: + target: torch.nn.Identity diff --git a/extensions-builtin/Lora/network.py b/extensions-builtin/Lora/network.py index 6021fd8d..a62e5eff 100644 --- a/extensions-builtin/Lora/network.py +++ b/extensions-builtin/Lora/network.py @@ -137,7 +137,7 @@ class NetworkModule: def finalize_updown(self, updown, orig_weight, output_shape, ex_bias=None):
if self.bias is not None:
updown = updown.reshape(self.bias.shape)
- updown += self.bias.to(orig_weight.device, dtype=orig_weight.dtype)
+ updown += self.bias.to(orig_weight.device, dtype=updown.dtype)
updown = updown.reshape(output_shape)
if len(output_shape) == 4:
diff --git a/extensions-builtin/Lora/network_full.py b/extensions-builtin/Lora/network_full.py index bf6930e9..f221c95f 100644 --- a/extensions-builtin/Lora/network_full.py +++ b/extensions-builtin/Lora/network_full.py @@ -18,9 +18,9 @@ class NetworkModuleFull(network.NetworkModule): def calc_updown(self, orig_weight):
output_shape = self.weight.shape
- updown = self.weight.to(orig_weight.device, dtype=orig_weight.dtype)
+ updown = self.weight.to(orig_weight.device)
if self.ex_bias is not None:
- ex_bias = self.ex_bias.to(orig_weight.device, dtype=orig_weight.dtype)
+ ex_bias = self.ex_bias.to(orig_weight.device)
else:
ex_bias = None
diff --git a/extensions-builtin/Lora/network_glora.py b/extensions-builtin/Lora/network_glora.py index 492d4870..efe5c681 100644 --- a/extensions-builtin/Lora/network_glora.py +++ b/extensions-builtin/Lora/network_glora.py @@ -22,12 +22,12 @@ class NetworkModuleGLora(network.NetworkModule): self.w2b = weights.w["b2.weight"] def calc_updown(self, orig_weight): - w1a = self.w1a.to(orig_weight.device, dtype=orig_weight.dtype) - w1b = self.w1b.to(orig_weight.device, dtype=orig_weight.dtype) - w2a = self.w2a.to(orig_weight.device, dtype=orig_weight.dtype) - w2b = self.w2b.to(orig_weight.device, dtype=orig_weight.dtype) + w1a = self.w1a.to(orig_weight.device) + w1b = self.w1b.to(orig_weight.device) + w2a = self.w2a.to(orig_weight.device) + w2b = self.w2b.to(orig_weight.device) output_shape = [w1a.size(0), w1b.size(1)] - updown = ((w2b @ w1b) + ((orig_weight @ w2a) @ w1a)) + updown = ((w2b @ w1b) + ((orig_weight.to(dtype = w1a.dtype) @ w2a) @ w1a)) return self.finalize_updown(updown, orig_weight, output_shape) diff --git a/extensions-builtin/Lora/network_hada.py b/extensions-builtin/Lora/network_hada.py index 5fcb0695..d95a0fd1 100644 --- a/extensions-builtin/Lora/network_hada.py +++ b/extensions-builtin/Lora/network_hada.py @@ -27,16 +27,16 @@ class NetworkModuleHada(network.NetworkModule): self.t2 = weights.w.get("hada_t2")
def calc_updown(self, orig_weight):
- w1a = self.w1a.to(orig_weight.device, dtype=orig_weight.dtype)
- w1b = self.w1b.to(orig_weight.device, dtype=orig_weight.dtype)
- w2a = self.w2a.to(orig_weight.device, dtype=orig_weight.dtype)
- w2b = self.w2b.to(orig_weight.device, dtype=orig_weight.dtype)
+ w1a = self.w1a.to(orig_weight.device)
+ w1b = self.w1b.to(orig_weight.device)
+ w2a = self.w2a.to(orig_weight.device)
+ w2b = self.w2b.to(orig_weight.device)
output_shape = [w1a.size(0), w1b.size(1)]
if self.t1 is not None:
output_shape = [w1a.size(1), w1b.size(1)]
- t1 = self.t1.to(orig_weight.device, dtype=orig_weight.dtype)
+ t1 = self.t1.to(orig_weight.device)
updown1 = lyco_helpers.make_weight_cp(t1, w1a, w1b)
output_shape += t1.shape[2:]
else:
@@ -45,7 +45,7 @@ class NetworkModuleHada(network.NetworkModule): updown1 = lyco_helpers.rebuild_conventional(w1a, w1b, output_shape)
if self.t2 is not None:
- t2 = self.t2.to(orig_weight.device, dtype=orig_weight.dtype)
+ t2 = self.t2.to(orig_weight.device)
updown2 = lyco_helpers.make_weight_cp(t2, w2a, w2b)
else:
updown2 = lyco_helpers.rebuild_conventional(w2a, w2b, output_shape)
diff --git a/extensions-builtin/Lora/network_ia3.py b/extensions-builtin/Lora/network_ia3.py index 7edc4249..96faeaf3 100644 --- a/extensions-builtin/Lora/network_ia3.py +++ b/extensions-builtin/Lora/network_ia3.py @@ -17,7 +17,7 @@ class NetworkModuleIa3(network.NetworkModule): self.on_input = weights.w["on_input"].item()
def calc_updown(self, orig_weight):
- w = self.w.to(orig_weight.device, dtype=orig_weight.dtype)
+ w = self.w.to(orig_weight.device)
output_shape = [w.size(0), orig_weight.size(1)]
if self.on_input:
diff --git a/extensions-builtin/Lora/network_lokr.py b/extensions-builtin/Lora/network_lokr.py index 340acdab..fcdaeafd 100644 --- a/extensions-builtin/Lora/network_lokr.py +++ b/extensions-builtin/Lora/network_lokr.py @@ -37,22 +37,22 @@ class NetworkModuleLokr(network.NetworkModule): def calc_updown(self, orig_weight):
if self.w1 is not None:
- w1 = self.w1.to(orig_weight.device, dtype=orig_weight.dtype)
+ w1 = self.w1.to(orig_weight.device)
else:
- w1a = self.w1a.to(orig_weight.device, dtype=orig_weight.dtype)
- w1b = self.w1b.to(orig_weight.device, dtype=orig_weight.dtype)
+ w1a = self.w1a.to(orig_weight.device)
+ w1b = self.w1b.to(orig_weight.device)
w1 = w1a @ w1b
if self.w2 is not None:
- w2 = self.w2.to(orig_weight.device, dtype=orig_weight.dtype)
+ w2 = self.w2.to(orig_weight.device)
elif self.t2 is None:
- w2a = self.w2a.to(orig_weight.device, dtype=orig_weight.dtype)
- w2b = self.w2b.to(orig_weight.device, dtype=orig_weight.dtype)
+ w2a = self.w2a.to(orig_weight.device)
+ w2b = self.w2b.to(orig_weight.device)
w2 = w2a @ w2b
else:
- t2 = self.t2.to(orig_weight.device, dtype=orig_weight.dtype)
- w2a = self.w2a.to(orig_weight.device, dtype=orig_weight.dtype)
- w2b = self.w2b.to(orig_weight.device, dtype=orig_weight.dtype)
+ t2 = self.t2.to(orig_weight.device)
+ w2a = self.w2a.to(orig_weight.device)
+ w2b = self.w2b.to(orig_weight.device)
w2 = lyco_helpers.make_weight_cp(t2, w2a, w2b)
output_shape = [w1.size(0) * w2.size(0), w1.size(1) * w2.size(1)]
diff --git a/extensions-builtin/Lora/network_lora.py b/extensions-builtin/Lora/network_lora.py index 26c0a72c..4cc40295 100644 --- a/extensions-builtin/Lora/network_lora.py +++ b/extensions-builtin/Lora/network_lora.py @@ -61,13 +61,13 @@ class NetworkModuleLora(network.NetworkModule): return module
def calc_updown(self, orig_weight):
- up = self.up_model.weight.to(orig_weight.device, dtype=orig_weight.dtype)
- down = self.down_model.weight.to(orig_weight.device, dtype=orig_weight.dtype)
+ up = self.up_model.weight.to(orig_weight.device)
+ down = self.down_model.weight.to(orig_weight.device)
output_shape = [up.size(0), down.size(1)]
if self.mid_model is not None:
# cp-decomposition
- mid = self.mid_model.weight.to(orig_weight.device, dtype=orig_weight.dtype)
+ mid = self.mid_model.weight.to(orig_weight.device)
updown = lyco_helpers.rebuild_cp_decomposition(up, down, mid)
output_shape += mid.shape[2:]
else:
diff --git a/extensions-builtin/Lora/network_norm.py b/extensions-builtin/Lora/network_norm.py index ce450158..d25afcbb 100644 --- a/extensions-builtin/Lora/network_norm.py +++ b/extensions-builtin/Lora/network_norm.py @@ -18,10 +18,10 @@ class NetworkModuleNorm(network.NetworkModule): def calc_updown(self, orig_weight): output_shape = self.w_norm.shape - updown = self.w_norm.to(orig_weight.device, dtype=orig_weight.dtype) + updown = self.w_norm.to(orig_weight.device) if self.b_norm is not None: - ex_bias = self.b_norm.to(orig_weight.device, dtype=orig_weight.dtype) + ex_bias = self.b_norm.to(orig_weight.device) else: ex_bias = None diff --git a/extensions-builtin/Lora/networks.py b/extensions-builtin/Lora/networks.py index 629bf853..72ebd624 100644 --- a/extensions-builtin/Lora/networks.py +++ b/extensions-builtin/Lora/networks.py @@ -1,3 +1,4 @@ +import gradio as gr
import logging
import os
import re
@@ -314,7 +315,12 @@ def load_networks(names, te_multipliers=None, unet_multipliers=None, dyn_dims=No emb_db.skipped_embeddings[name] = embedding
if failed_to_load_networks:
- sd_hijack.model_hijack.comments.append("Networks not found: " + ", ".join(failed_to_load_networks))
+ lora_not_found_message = f'Lora not found: {", ".join(failed_to_load_networks)}'
+ sd_hijack.model_hijack.comments.append(lora_not_found_message)
+ if shared.opts.lora_not_found_warning_console:
+ print(f'\n{lora_not_found_message}\n')
+ if shared.opts.lora_not_found_gradio_warning:
+ gr.Warning(lora_not_found_message)
purge_networks_from_memory()
@@ -389,18 +395,26 @@ def network_apply_weights(self: Union[torch.nn.Conv2d, torch.nn.Linear, torch.nn if module is not None and hasattr(self, 'weight'):
try:
with torch.no_grad():
- updown, ex_bias = module.calc_updown(self.weight)
+ if getattr(self, 'fp16_weight', None) is None:
+ weight = self.weight
+ bias = self.bias
+ else:
+ weight = self.fp16_weight.clone().to(self.weight.device)
+ bias = getattr(self, 'fp16_bias', None)
+ if bias is not None:
+ bias = bias.clone().to(self.bias.device)
+ updown, ex_bias = module.calc_updown(weight)
- if len(self.weight.shape) == 4 and self.weight.shape[1] == 9:
+ if len(weight.shape) == 4 and weight.shape[1] == 9:
# inpainting model. zero pad updown to make channel[1] 4 to 9
updown = torch.nn.functional.pad(updown, (0, 0, 0, 0, 0, 5))
- self.weight += updown
+ self.weight.copy_((weight.to(dtype=updown.dtype) + updown).to(dtype=self.weight.dtype))
if ex_bias is not None and hasattr(self, 'bias'):
if self.bias is None:
- self.bias = torch.nn.Parameter(ex_bias)
+ self.bias = torch.nn.Parameter(ex_bias).to(self.weight.dtype)
else:
- self.bias += ex_bias
+ self.bias.copy_((bias + ex_bias).to(dtype=self.bias.dtype))
except RuntimeError as e:
logging.debug(f"Network {net.name} layer {network_layer_name}: {e}")
extra_network_lora.errors[net.name] = extra_network_lora.errors.get(net.name, 0) + 1
diff --git a/extensions-builtin/Lora/scripts/lora_script.py b/extensions-builtin/Lora/scripts/lora_script.py index ef23968c..1518f7e5 100644 --- a/extensions-builtin/Lora/scripts/lora_script.py +++ b/extensions-builtin/Lora/scripts/lora_script.py @@ -39,6 +39,8 @@ shared.options_templates.update(shared.options_section(('extra_networks', "Extra "lora_show_all": shared.OptionInfo(False, "Always show all networks on the Lora page").info("otherwise, those detected as for incompatible version of Stable Diffusion will be hidden"),
"lora_hide_unknown_for_versions": shared.OptionInfo([], "Hide networks of unknown versions for model versions", gr.CheckboxGroup, {"choices": ["SD1", "SD2", "SDXL"]}),
"lora_in_memory_limit": shared.OptionInfo(0, "Number of Lora networks to keep cached in memory", gr.Number, {"precision": 0}),
+ "lora_not_found_warning_console": shared.OptionInfo(False, "Lora not found warning in console"),
+ "lora_not_found_gradio_warning": shared.OptionInfo(False, "Lora not found warning popup in webui"),
}))
diff --git a/extensions-builtin/soft-inpainting/scripts/soft_inpainting.py b/extensions-builtin/soft-inpainting/scripts/soft_inpainting.py new file mode 100644 index 00000000..d9024344 --- /dev/null +++ b/extensions-builtin/soft-inpainting/scripts/soft_inpainting.py @@ -0,0 +1,747 @@ +import numpy as np +import gradio as gr +import math +from modules.ui_components import InputAccordion +import modules.scripts as scripts + + +class SoftInpaintingSettings: + def __init__(self, + mask_blend_power, + mask_blend_scale, + inpaint_detail_preservation, + composite_mask_influence, + composite_difference_threshold, + composite_difference_contrast): + self.mask_blend_power = mask_blend_power + self.mask_blend_scale = mask_blend_scale + self.inpaint_detail_preservation = inpaint_detail_preservation + self.composite_mask_influence = composite_mask_influence + self.composite_difference_threshold = composite_difference_threshold + self.composite_difference_contrast = composite_difference_contrast + + 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 + dest[gen_param_labels.composite_mask_influence] = self.composite_mask_influence + dest[gen_param_labels.composite_difference_threshold] = self.composite_difference_threshold + dest[gen_param_labels.composite_difference_contrast] = self.composite_difference_contrast + + +# ------------------- Methods ------------------- + +def processing_uses_inpainting(p): + # TODO: Figure out a better way to determine if inpainting is being used by p + if getattr(p, "image_mask", None) is not None: + return True + + if getattr(p, "mask", None) is not None: + return True + + if getattr(p, "nmask", None) is not None: + return True + + return False + + +def latent_blend(settings, 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_( + settings.inpaint_detail_preservation) * one_minus_t3 + b_magnitude = torch.norm(b, p=2, dim=1, keepdim=True).to(torch.float64).pow_( + settings.inpaint_detail_preservation) * t3 + desired_magnitude = a_magnitude + desired_magnitude.add_(b_magnitude).pow_(1 / settings.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(settings, 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 ** settings.mask_blend_power) * settings.mask_blend_scale) + + +def apply_adaptive_masks( + settings: SoftInpaintingSettings, + nmask, + latent_orig, + latent_processed, + overlay_images, + width, height, + paste_to): + import torch + 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 = nmask[0].float() + # 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) ** (settings.mask_blend_scale / 2)) + mask_scalar = (0.5 * (1 - settings.composite_mask_influence) + + mask_scalar * settings.composite_mask_influence) + mask_scalar = mask_scalar / (1.00001 - mask_scalar) + mask_scalar = mask_scalar.cpu().numpy() + + latent_distance = torch.norm(latent_processed - latent_orig, p=2, dim=1) + + kernel, kernel_center = get_gaussian_kernel(stddev_radius=1.5, max_radius=2) + + masks_for_overlay = [] + + for i, (distance_map, overlay_image) in enumerate(zip(latent_distance, overlay_images)): + converted_mask = distance_map.float().cpu().numpy() + converted_mask = weighted_histogram_filter(converted_mask, kernel, kernel_center, + percentile_min=0.9, percentile_max=1, min_width=1) + converted_mask = 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 = settings.composite_difference_threshold * mask_scalar + converted_mask = converted_mask / half_weighted_distance + + converted_mask = 1 / (1 + converted_mask ** settings.composite_difference_contrast) + converted_mask = 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.append(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') + + return masks_for_overlay + + +def apply_masks( + settings, + nmask, + overlay_images, + width, height, + paste_to): + import torch + 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_(settings.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) + + masks_for_overlay = [] + + 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') + + return masks_for_overlay + + +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 + + +# ------------------- Constants ------------------- + + +default = SoftInpaintingSettings(1, 0.5, 4, 0, 0.5, 2) + +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", + "Mask influence", + "Difference threshold", + "Difference contrast") + +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.", + "How strongly the original mask should bias the difference threshold.", + "How much an image region can change before the original pixels are not blended in anymore.", + "How sharp the transition should be between blended and not blended.") + +gen_param_labels = SoftInpaintingSettings( + "Soft inpainting schedule bias", + "Soft inpainting preservation strength", + "Soft inpainting transition contrast boost", + "Soft inpainting mask influence", + "Soft inpainting difference threshold", + "Soft inpainting difference contrast") + +el_ids = SoftInpaintingSettings( + "mask_blend_power", + "mask_blend_scale", + "inpaint_detail_preservation", + "composite_mask_influence", + "composite_difference_threshold", + "composite_difference_contrast") + + +# ------------------- Script ------------------- + + +class Script(scripts.Script): + def __init__(self): + self.section = "inpaint" + self.masks_for_overlay = None + self.overlay_images = None + + def title(self): + return "Soft Inpainting" + + def show(self, is_img2img): + return scripts.AlwaysVisible if is_img2img else False + + def ui(self, is_img2img): + if not is_img2img: + return + + 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! + """) + + power = \ + 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) + scale = \ + 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) + detail = \ + 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) + + gr.Markdown( + """ + ### Pixel Composite Settings + """) + + mask_inf = \ + gr.Slider(label=ui_labels.composite_mask_influence, + info=ui_info.composite_mask_influence, + minimum=0, + maximum=1, + step=0.05, + value=default.composite_mask_influence, + elem_id=el_ids.composite_mask_influence) + + dif_thresh = \ + gr.Slider(label=ui_labels.composite_difference_threshold, + info=ui_info.composite_difference_threshold, + minimum=0, + maximum=8, + step=0.25, + value=default.composite_difference_threshold, + elem_id=el_ids.composite_difference_threshold) + + dif_contr = \ + gr.Slider(label=ui_labels.composite_difference_contrast, + info=ui_info.composite_difference_contrast, + minimum=0, + maximum=8, + step=0.25, + value=default.composite_difference_contrast, + elem_id=el_ids.composite_difference_contrast) + + 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. + """) + + gr.Markdown( + """ + ## Pixel Composite Settings + + Masks are generated based on how much a part of the image changed after denoising. + These masks are used to blend the original and final images together. + If the difference is low, the original pixels are used instead of the pixels returned by the inpainting process. + """) + + gr.Markdown( + f""" + ### {ui_labels.composite_mask_influence} + + This parameter controls how much the mask should bias this sensitivity to difference. + + - **0**: Ignore the mask, only consider differences in image content. + - **1**: Follow the mask closely despite image content changes. + """) + + gr.Markdown( + f""" + ### {ui_labels.composite_difference_threshold} + + This value represents the difference at which the original pixels will have less than 50% opacity. + + - **Low values**: Two images patches must be almost the same in order to retain original pixels. + - **High values**: Two images patches can be very different and still retain original pixels. + """) + + gr.Markdown( + f""" + ### {ui_labels.composite_difference_contrast} + + This value represents the contrast between the opacity of the original and inpainted content. + + - **Low values**: The blend will be more gradual and have longer transitions, but may cause ghosting. + - **High values**: Ghosting will be less common, but transitions may be very sudden. + """) + + self.infotext_fields = [(soft_inpainting_enabled, enabled_gen_param_label), + (power, gen_param_labels.mask_blend_power), + (scale, gen_param_labels.mask_blend_scale), + (detail, gen_param_labels.inpaint_detail_preservation), + (mask_inf, gen_param_labels.composite_mask_influence), + (dif_thresh, gen_param_labels.composite_difference_threshold), + (dif_contr, gen_param_labels.composite_difference_contrast)] + + self.paste_field_names = [] + for _, field_name in self.infotext_fields: + self.paste_field_names.append(field_name) + + return [soft_inpainting_enabled, + power, + scale, + detail, + mask_inf, + dif_thresh, + dif_contr] + + def process(self, p, enabled, power, scale, detail_preservation, mask_inf, dif_thresh, dif_contr): + if not enabled: + return + + if not processing_uses_inpainting(p): + return + + # Shut off the rounding it normally does. + p.mask_round = False + + settings = SoftInpaintingSettings(power, scale, detail_preservation, mask_inf, dif_thresh, dif_contr) + + # p.extra_generation_params["Mask rounding"] = False + settings.add_generation_params(p.extra_generation_params) + + def on_mask_blend(self, p, mba: scripts.MaskBlendArgs, enabled, power, scale, detail_preservation, mask_inf, + dif_thresh, dif_contr): + if not enabled: + return + + if not processing_uses_inpainting(p): + return + + if mba.is_final_blend: + mba.blended_latent = mba.current_latent + return + + settings = SoftInpaintingSettings(power, scale, detail_preservation, mask_inf, dif_thresh, dif_contr) + + # todo: Why is sigma 2D? Both values are the same. + mba.blended_latent = latent_blend(settings, + mba.init_latent, + mba.current_latent, + get_modified_nmask(settings, mba.nmask, mba.sigma[0])) + + def post_sample(self, p, ps: scripts.PostSampleArgs, enabled, power, scale, detail_preservation, mask_inf, + dif_thresh, dif_contr): + if not enabled: + return + + if not processing_uses_inpainting(p): + return + + nmask = getattr(p, "nmask", None) + if nmask is None: + return + + from modules import images + from modules.shared import opts + + settings = SoftInpaintingSettings(power, scale, detail_preservation, mask_inf, dif_thresh, dif_contr) + + # since the original code puts holes in the existing overlay images, + # we have to rebuild them. + self.overlay_images = [] + for img in p.init_images: + + image = images.flatten(img, opts.img2img_background_color) + + if p.paste_to is None and p.resize_mode != 3: + image = images.resize_image(p.resize_mode, image, p.width, p.height) + + self.overlay_images.append(image.convert('RGBA')) + + if len(p.init_images) == 1: + self.overlay_images = self.overlay_images * p.batch_size + + if getattr(ps.samples, 'already_decoded', False): + self.masks_for_overlay = apply_masks(settings=settings, + nmask=nmask, + overlay_images=self.overlay_images, + width=p.width, + height=p.height, + paste_to=p.paste_to) + else: + self.masks_for_overlay = apply_adaptive_masks(settings=settings, + nmask=nmask, + latent_orig=p.init_latent, + latent_processed=ps.samples, + overlay_images=self.overlay_images, + width=p.width, + height=p.height, + paste_to=p.paste_to) + + def postprocess_maskoverlay(self, p, ppmo: scripts.PostProcessMaskOverlayArgs, enabled, power, scale, + detail_preservation, mask_inf, dif_thresh, dif_contr): + if not enabled: + return + + if not processing_uses_inpainting(p): + return + + if self.masks_for_overlay is None: + return + + if self.overlay_images is None: + return + + ppmo.mask_for_overlay = self.masks_for_overlay[ppmo.index] + ppmo.overlay_image = self.overlay_images[ppmo.index] diff --git a/modules/api/api.py b/modules/api/api.py index b3d74e51..2918f785 100644 --- a/modules/api/api.py +++ b/modules/api/api.py @@ -31,7 +31,7 @@ from typing import 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: @@ -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 @@ -335,11 +340,86 @@ 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, *, 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 {} + + 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) + + def get_field_value(field, params): + value = field.function(params) if field.function else params.get(field.label) + if value is None: + 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 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 + + 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) + 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: + value = get_field_value(field, params) + + if value is None: + continue + + mentioned_script_args[index] = value + + return params + def text2imgapi(self, txt2imgreq: models.StableDiffusionTxt2ImgProcessingAPI): + 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() + + 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) @@ -356,12 +436,15 @@ 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) + 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) + 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 @@ -371,12 +454,14 @@ 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() @@ -386,6 +471,8 @@ class Api: return models.TextToImageResponse(images=b64images, parameters=vars(txt2imgreq), info=processed.js()) def img2imgapi(self, img2imgreq: models.StableDiffusionImg2ImgProcessingAPI): + task_id = img2imgreq.force_task_id or create_task_id("img2img") + init_images = img2imgreq.init_images if init_images is None: raise HTTPException(status_code=404, detail="Init image not found") @@ -398,6 +485,10 @@ 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) @@ -417,11 +508,13 @@ 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) + 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] @@ -432,12 +525,14 @@ 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() @@ -511,7 +606,7 @@ 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 33894b3e..16edf11c 100644 --- a/modules/api/models.py +++ b/modules/api/models.py @@ -107,6 +107,8 @@ 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}, + {"key": "infotext", "type": str, "default": None}, ] ).generate_model() @@ -124,6 +126,8 @@ 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}, + {"key": "infotext", "type": str, "default": None}, ] ).generate_model() diff --git a/modules/devices.py b/modules/devices.py index ea1f712f..c956207f 100644 --- a/modules/devices.py +++ b/modules/devices.py @@ -23,6 +23,23 @@ 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}" @@ -73,8 +90,7 @@ 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 @@ -84,6 +100,7 @@ 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 @@ -104,12 +121,51 @@ def cond_cast_float(input): nv_rng = None +patch_module_list = [ + torch.nn.Linear, + torch.nn.Conv2d, + torch.nn.MultiheadAttention, + torch.nn.GroupNorm, + 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_cast(): + 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): if disable: return contextlib.nullcontext() + 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": return contextlib.nullcontext() diff --git a/modules/errors.py b/modules/errors.py index eb234a83..48aa13a1 100644 --- a/modules/errors.py +++ b/modules/errors.py @@ -107,8 +107,8 @@ def check_versions(): import torch
import gradio
- expected_torch_version = "2.0.0"
- expected_xformers_version = "0.0.20"
+ 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/generation_parameters_copypaste.py b/modules/generation_parameters_copypaste.py index 4efe53e0..86a36c32 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
@@ -314,6 +333,12 @@ 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}
@@ -365,6 +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:
@@ -406,29 +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))
diff --git a/modules/images.py b/modules/images.py index daf4eebe..16f9ae7c 100644 --- a/modules/images.py +++ b/modules/images.py @@ -791,3 +791,4 @@ def flatten(img, bgcolor): img = background
return img.convert('RGB')
+
diff --git a/modules/initialize_util.py b/modules/initialize_util.py index 2e9b6d89..b6767138 100644 --- a/modules/initialize_util.py +++ b/modules/initialize_util.py @@ -177,6 +177,8 @@ 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)
+ 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")
diff --git a/modules/launch_utils.py b/modules/launch_utils.py index 29506f24..dabef0f5 100644 --- a/modules/launch_utils.py +++ b/modules/launch_utils.py @@ -314,8 +314,8 @@ 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==2.1.2 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==0.0.20')
+ 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")
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/processing.py b/modules/processing.py index 6f01c95f..7789f9a4 100644 --- a/modules/processing.py +++ b/modules/processing.py @@ -62,18 +62,22 @@ def apply_color_correction(correction, original_image): return image.convert('RGB')
-def apply_overlay(image, paste_loc, index, overlays):
- if overlays is None or index >= len(overlays):
- return image
+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
- overlay = overlays[index]
+
+def apply_overlay(image, paste_loc, overlay):
+ if overlay is None:
+ return image
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)
@@ -81,9 +85,12 @@ 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").point(lambda x: 255 if x > 128 else 0)
+ 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
@@ -106,6 +113,21 @@ 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 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)
+
+ 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.
@@ -308,7 +330,7 @@ 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
@@ -320,8 +342,10 @@ class StableDiffusionProcessing: 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)
+ 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:])
@@ -345,7 +369,7 @@ 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
@@ -357,11 +381,17 @@ 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)
+ 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 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)
@@ -422,6 +452,8 @@ 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):
@@ -679,6 +711,8 @@ 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])),
@@ -867,6 +901,11 @@ 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 = ps.samples
+
if getattr(samples_ddim, 'already_decoded', False):
x_samples_ddim = samples_ddim
else:
@@ -922,13 +961,31 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: pp = scripts.PostprocessImageArgs(image)
p.scripts.postprocess_image(p, pp)
image = pp.image
+
+ 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)
+ p.scripts.postprocess_maskoverlay(p, ppmo)
+ 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:
- 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)
- image = apply_overlay(image, p.paste_to, i, p.overlay_images)
+ # 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()
+
+ if p.paste_to is not None:
+ 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)
if save_samples:
images.save_image(image, p.outpath_samples, "", p.seeds[i], p.prompts[i], opts.samples_format, info=infotext(i), p=p)
@@ -938,16 +995,17 @@ def process_images_inner(p: StableDiffusionProcessing) -> Processed: if opts.enable_pnginfo:
image.info["parameters"] = text
output_images.append(image)
- if hasattr(p, 'mask_for_overlay') and p.mask_for_overlay:
+
+ if mask_for_overlay is not None:
if opts.return_mask or opts.save_mask:
- image_mask = p.mask_for_overlay.convert('RGB')
+ image_mask = mask_for_overlay.convert('RGB')
if save_samples and 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.return_mask:
output_images.append(image_mask)
if opts.return_mask_composite or opts.save_mask_composite:
- 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')
+ image_mask_composite = Image.composite(original_denoised_image.convert('RGBA').convert('RGBa'), Image.new('RGBa', image.size), images.resize_image(2, mask_for_overlay, image.width, image.height).convert('L')).convert('RGBA')
if save_samples and 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_composite:
@@ -1025,6 +1083,7 @@ 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]
@@ -1097,7 +1156,7 @@ 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:
@@ -1351,12 +1410,14 @@ class StableDiffusionProcessingImg2Img(StableDiffusionProcessing): mask_blur_x: int = 4
mask_blur_y: int = 4
mask_blur: int = None
+ mask_round: bool = True
inpainting_fill: int = 0
inpaint_full_res: bool = True
inpaint_full_res_padding: int = 0
inpainting_mask_invert: int = 0
initial_noise_multiplier: float = None
latent_mask: Image = None
+ force_task_id: str = None
image_mask: Any = field(default=None, init=False)
@@ -1396,7 +1457,7 @@ 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.mask_round)
if self.inpainting_mask_invert:
image_mask = ImageOps.invert(image_mask)
@@ -1442,7 +1503,7 @@ 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)
@@ -1503,7 +1564,8 @@ 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]
- latmask = np.around(latmask)
+ if self.mask_round:
+ 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)
@@ -1515,7 +1577,7 @@ 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.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()
@@ -1527,7 +1589,14 @@ 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:
- 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(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
del x
devices.torch_gc()
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/progress.py b/modules/progress.py index 69921de7..85255e82 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")
@@ -63,9 +74,16 @@ class ProgressResponse(BaseModel): def setup_progress_api(app):
+ 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)
+ return PendingTasksResponse(size=pending_len, tasks=pending_tasks_ids)
+
+
def progressapi(req: ProgressRequest):
active = req.id_task == current_task
queued = req.id_task in pending_tasks
diff --git a/modules/scripts.py b/modules/scripts.py index 7f9454eb..3a766911 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_latent, denoiser=None, sigma=None):
+ self.current_latent = current_latent
+ self.nmask = nmask
+ self.init_latent = init_latent
+ self.mask = mask
+ self.blended_latent = blended_latent
+
+ 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.
@@ -520,7 +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
@@ -767,6 +818,22 @@ 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 +842,14 @@ 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_models.py b/modules/sd_models.py index 9355f1e1..d0046f88 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -348,10 +348,28 @@ 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 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
@@ -404,6 +422,28 @@ 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, torch.nn.Linear)):
+ if shared.opts.cache_fp16_weight:
+ module.fp16_weight = module.weight.data.clone().cpu().half()
+ if module.bias is not None:
+ module.fp16_bias = module.bias.data.clone().cpu().half()
+ module.to(torch.float8_e4m3fn)
+ model.first_stage_model = first_stage
+ 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
model.first_stage_model.to(devices.dtype_vae)
@@ -746,7 +786,7 @@ 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()
@@ -758,11 +798,14 @@ 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 and not forced_reload:
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/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..1de31b0d 100644 --- a/modules/sd_models_xl.py +++ b/modules/sd_models_xl.py @@ -34,6 +34,12 @@ 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 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)
@@ -93,7 +99,7 @@ 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()
diff --git a/modules/sd_samplers_cfg_denoiser.py b/modules/sd_samplers_cfg_denoiser.py index b8101d38..eb9d5daf 100644 --- a/modules/sd_samplers_cfg_denoiser.py +++ b/modules/sd_samplers_cfg_denoiser.py @@ -56,6 +56,9 @@ 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
@@ -105,8 +108,21 @@ 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(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:
- x = self.init_latent * self.mask + self.nmask * x
+ x = apply_blend(x)
batch_size = len(conds_list)
repeats = [len(conds_list[i]) for i in range(batch_size)]
@@ -207,8 +223,9 @@ 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 = 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/shared_options.py b/modules/shared_options.py index d2e86ff1..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"), {
@@ -206,6 +207,8 @@ 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.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."),
}))
options_templates.update(options_section(('compatibility', "Compatibility", "sd"), {
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)
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():
diff --git a/modules/ui_toprow.py b/modules/ui_toprow.py index 88838f97..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 (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, 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 (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, Esc to interrupt)", elem_classes=["prompt"])
self.prompt_img.change(
fn=modules.images.image_data,
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
@@ -121,16 +121,22 @@ document.addEventListener("DOMContentLoaded", function() { }); /** - * Add a ctrl+enter as a shortcut to start a generation + * Add keyboard shortcuts: + * Ctrl+Enter to start/restart a generation + * Alt/Option+Enter to skip a generation + * Esc to interrupt a generation */ document.addEventListener('keydown', function(e) { const isEnter = e.key === 'Enter' || e.keyCode === 13; - const isModifierKey = e.metaKey || e.ctrlKey || e.altKey; + const isCtrlKey = e.metaKey || e.ctrlKey; + const isAltKey = e.altKey; + const isEsc = e.key === 'Escape'; - const interruptButton = get_uiCurrentTabContent().querySelector('button[id$=_interrupt]'); const generateButton = get_uiCurrentTabContent().querySelector('button[id$=_generate]'); + const interruptButton = get_uiCurrentTabContent().querySelector('button[id$=_interrupt]'); + const skipButton = get_uiCurrentTabContent().querySelector('button[id$=_skip]'); - if (isEnter && isModifierKey) { + if (isCtrlKey && isEnter) { if (interruptButton.style.display === 'block') { interruptButton.click(); const callback = (mutationList) => { @@ -150,6 +156,21 @@ document.addEventListener('keydown', function(e) { } e.preventDefault(); } + + if (isAltKey && isEnter) { + skipButton.click(); + e.preventDefault(); + } + + if (isEsc) { + const globalPopup = document.querySelector('.global-popup'); + const lightboxModal = document.querySelector('#lightboxModal'); + if (!globalPopup || globalPopup.style.display === 'none') { + if (document.activeElement === lightboxModal) return; + interruptButton.click(); + e.preventDefault(); + } + } }); /** diff --git a/scripts/postprocessing_caption.py b/scripts/postprocessing_caption.py index 243e3ad9..9482a03c 100644 --- a/scripts/postprocessing_caption.py +++ b/scripts/postprocessing_caption.py @@ -25,6 +25,6 @@ class ScriptPostprocessingCeption(scripts_postprocessing.ScriptPostprocessing): captions.append(deepbooru.model.tag(pp.image))
if "BLIP" in option:
- captions.append(shared.interrogator.generate_caption(pp.image))
+ captions.append(shared.interrogator.interrogate(pp.image.convert("RGB")))
pp.caption = ", ".join([x for x in captions if x])
diff --git a/scripts/xyz_grid.py b/scripts/xyz_grid.py index 0dc255bc..e5083874 100644 --- a/scripts/xyz_grid.py +++ b/scripts/xyz_grid.py @@ -270,6 +270,7 @@ axis_options = [ AxisOption("Refiner checkpoint", str, apply_field('refiner_checkpoint'), format_value=format_remove_path, confirm=confirm_checkpoints_or_none, cost=1.0, choices=lambda: ['None'] + sorted(sd_models.checkpoints_list, key=str.casefold)),
AxisOption("Refiner switch at", float, apply_field('refiner_switch_at')),
AxisOption("RNG source", str, apply_override("randn_source"), choices=lambda: ["GPU", "CPU", "NV"]),
+ AxisOption("FP8 mode", str, apply_override("fp8_storage"), cost=0.9, choices=lambda: ["Disable", "Enable for SDXL", "Enable"]),
]
@@ -475,6 +476,8 @@ class Script(scripts.Script): fill_z_button.click(fn=fill, inputs=[z_type, csv_mode], outputs=[z_values, z_values_dropdown])
def select_axis(axis_type, axis_values, axis_values_dropdown, csv_mode):
+ axis_type = axis_type or 0 # if axle type is None set to 0
+
choices = self.current_axis_options[axis_type].choices
has_choices = choices is not None
@@ -525,6 +528,8 @@ class Script(scripts.Script): return [x_type, x_values, x_values_dropdown, y_type, y_values, y_values_dropdown, z_type, z_values, z_values_dropdown, draw_legend, include_lone_images, include_sub_grids, no_fixed_seeds, margin_size, csv_mode]
def run(self, p, x_type, x_values, x_values_dropdown, y_type, y_values, y_values_dropdown, z_type, z_values, z_values_dropdown, draw_legend, include_lone_images, include_sub_grids, no_fixed_seeds, margin_size, csv_mode):
+ x_type, y_type, z_type = x_type or 0, y_type or 0, z_type or 0 # if axle type is None set to 0
+
if not no_fixed_seeds:
modules.processing.fix_seed(p)
diff --git a/webui-macos-env.sh b/webui-macos-env.sh index 24bc5c42..db7e8b1a 100644 --- a/webui-macos-env.sh +++ b/webui-macos-env.sh @@ -11,7 +11,7 @@ fi export install_dir="$HOME" export COMMANDLINE_ARGS="--skip-torch-cuda-test --upcast-sampling --no-half-vae --use-cpu interrogate" -export TORCH_COMMAND="pip install torch==2.0.1 torchvision==0.15.2" +export TORCH_COMMAND="pip install torch==2.1.0 torchvision==0.16.0" export PYTORCH_ENABLE_MPS_FALLBACK=1 #################################################################### @@ -133,7 +133,7 @@ case "$gpu_info" in if [[ $(bc <<< "$pyv <= 3.10") -eq 1 ]] then # Navi users will still use torch 1.13 because 2.0 does not seem to work. - export TORCH_COMMAND="pip install torch==1.13.1+rocm5.2 torchvision==0.14.1+rocm5.2 --index-url https://download.pytorch.org/whl/rocm5.2" + export TORCH_COMMAND="pip install --pre torch torchvision --index-url https://download.pytorch.org/whl/nightly/rocm5.6" else printf "\e[1m\e[31mERROR: RX 5000 series GPUs must be using at max python 3.10, aborting...\e[0m" exit 1 @@ -143,8 +143,7 @@ case "$gpu_info" in *"Navi 2"*) export HSA_OVERRIDE_GFX_VERSION=10.3.0 ;; *"Navi 3"*) [[ -z "${TORCH_COMMAND}" ]] && \ - export TORCH_COMMAND="pip install torch torchvision --index-url https://download.pytorch.org/whl/test/rocm5.6" - # Navi 3 needs at least 5.5 which is only on the torch 2.1.0 release candidates right now + export TORCH_COMMAND="pip install --pre torch torchvision --index-url https://download.pytorch.org/whl/nightly/rocm5.7" ;; *"Renoir"*) export HSA_OVERRIDE_GFX_VERSION=9.0.0 printf "\n%s\n" "${delimiter}" @@ -223,13 +222,30 @@ fi # Try using TCMalloc on Linux prepare_tcmalloc() { if [[ "${OSTYPE}" == "linux"* ]] && [[ -z "${NO_TCMALLOC}" ]] && [[ -z "${LD_PRELOAD}" ]]; then - TCMALLOC="$(PATH=/sbin:$PATH ldconfig -p | grep -Po "libtcmalloc(_minimal|)\.so\.\d" | head -n 1)" - if [[ ! -z "${TCMALLOC}" ]]; then - echo "Using TCMalloc: ${TCMALLOC}" - export LD_PRELOAD="${TCMALLOC}" - else - printf "\e[1m\e[31mCannot locate TCMalloc (improves CPU memory usage)\e[0m\n" - fi + # Define Tcmalloc Libs arrays + TCMALLOC_LIBS=("libtcmalloc(_minimal|)\.so\.\d" "libtcmalloc\.so\.\d") + + # Traversal array + for lib in "${TCMALLOC_LIBS[@]}" + do + #Determine which type of tcmalloc library the library supports + TCMALLOC="$(PATH=/usr/sbin:$PATH ldconfig -p | grep -P $lib | head -n 1)" + TC_INFO=(${TCMALLOC//=>/}) + if [[ ! -z "${TC_INFO}" ]]; then + echo "Using TCMalloc: ${TC_INFO}" + #Determine if the library is linked to libptthread and resolve undefined symbol: ptthread_Key_Create + if ldd ${TC_INFO[2]} | grep -q 'libpthread'; then + echo "$TC_INFO is linked with libpthread,execute LD_PRELOAD=${TC_INFO}" + export LD_PRELOAD="${TC_INFO}" + break + else + echo "$TC_INFO is not linked with libpthreadand will trigger undefined symbol: ptthread_Key_Create error" + fi + else + printf "\e[1m\e[31mCannot locate TCMalloc (improves CPU memory usage)\e[0m\n" + fi + done + fi } |