diff options
Diffstat (limited to 'modules/sd_models.py')
| -rw-r--r-- | modules/sd_models.py | 76 |
1 files changed, 50 insertions, 26 deletions
diff --git a/modules/sd_models.py b/modules/sd_models.py index 5b37f3fe..76a89e88 100644 --- a/modules/sd_models.py +++ b/modules/sd_models.py @@ -20,7 +20,7 @@ from modules.sd_hijack_inpainting import do_inpainting_hijack, should_hijack_inp model_dir = "Stable-diffusion"
model_path = os.path.abspath(os.path.join(models_path, model_dir))
-CheckpointInfo = namedtuple("CheckpointInfo", ['filename', 'title', 'hash', 'model_name', 'config'])
+CheckpointInfo = namedtuple("CheckpointInfo", ['filename', 'title', 'hash', 'model_name'])
checkpoints_list = {}
checkpoints_loaded = collections.OrderedDict()
@@ -48,6 +48,14 @@ def checkpoint_tiles(): return sorted([x.title for x in checkpoints_list.values()], key = alphanumeric_key)
+def find_checkpoint_config(info):
+ config = os.path.splitext(info.filename)[0] + ".yaml"
+ if os.path.exists(config):
+ return config
+
+ return shared.cmd_opts.config
+
+
def list_models():
checkpoints_list.clear()
model_list = modelloader.load_models(model_path=model_path, command_path=shared.cmd_opts.ckpt_dir, ext_filter=[".ckpt", ".safetensors"])
@@ -73,7 +81,7 @@ def list_models(): if os.path.exists(cmd_ckpt):
h = model_hash(cmd_ckpt)
title, short_model_name = modeltitle(cmd_ckpt, h)
- checkpoints_list[title] = CheckpointInfo(cmd_ckpt, title, h, short_model_name, shared.cmd_opts.config)
+ checkpoints_list[title] = CheckpointInfo(cmd_ckpt, title, h, short_model_name)
shared.opts.data['sd_model_checkpoint'] = title
elif cmd_ckpt is not None and cmd_ckpt != shared.default_sd_model_file:
print(f"Checkpoint in --ckpt argument not found (Possible it was moved to {model_path}: {cmd_ckpt}", file=sys.stderr)
@@ -81,12 +89,7 @@ def list_models(): h = model_hash(filename)
title, short_model_name = modeltitle(filename, h)
- basename, _ = os.path.splitext(filename)
- config = basename + ".yaml"
- if not os.path.exists(config):
- config = shared.cmd_opts.config
-
- checkpoints_list[title] = CheckpointInfo(filename, title, h, short_model_name, config)
+ checkpoints_list[title] = CheckpointInfo(filename, title, h, short_model_name)
def get_closet_checkpoint_match(searchString):
@@ -111,18 +114,19 @@ def model_hash(filename): def select_checkpoint():
model_checkpoint = shared.opts.sd_model_checkpoint
+
checkpoint_info = checkpoints_list.get(model_checkpoint, None)
if checkpoint_info is not None:
return checkpoint_info
if len(checkpoints_list) == 0:
- print(f"No checkpoints found. When searching for checkpoints, looked at:", file=sys.stderr)
+ print("No checkpoints found. When searching for checkpoints, looked at:", file=sys.stderr)
if shared.cmd_opts.ckpt is not None:
print(f" - file {os.path.abspath(shared.cmd_opts.ckpt)}", file=sys.stderr)
print(f" - directory {model_path}", file=sys.stderr)
if shared.cmd_opts.ckpt_dir is not None:
print(f" - directory {os.path.abspath(shared.cmd_opts.ckpt_dir)}", file=sys.stderr)
- print(f"Can't run without a checkpoint. Find and place a .ckpt file into any of those locations. The program will exit.", file=sys.stderr)
+ print("Can't run without a checkpoint. Find and place a .ckpt file into any of those locations. The program will exit.", file=sys.stderr)
exit(1)
checkpoint_info = next(iter(checkpoints_list.values()))
@@ -167,7 +171,10 @@ def get_state_dict_from_checkpoint(pl_sd): def read_state_dict(checkpoint_file, print_global_state=False, map_location=None):
_, extension = os.path.splitext(checkpoint_file)
if extension.lower() == ".safetensors":
- pl_sd = safetensors.torch.load_file(checkpoint_file, device=map_location or shared.weight_load_location)
+ device = map_location or shared.weight_load_location
+ if device is None:
+ device = devices.get_cuda_device_string() if torch.cuda.is_available() else "cpu"
+ pl_sd = safetensors.torch.load_file(checkpoint_file, device=device)
else:
pl_sd = torch.load(checkpoint_file, map_location=map_location or shared.weight_load_location)
@@ -227,6 +234,8 @@ def load_model_weights(model, checkpoint_info, vae_file="auto"): model.sd_model_checkpoint = checkpoint_file
model.sd_checkpoint_info = checkpoint_info
+ model.logvar = model.logvar.to(devices.device) # fix for training
+
sd_vae.delete_base_vae()
sd_vae.clear_loaded_vae()
vae_file = sd_vae.resolve_vae(checkpoint_file, vae_file=vae_file)
@@ -275,12 +284,14 @@ def enable_midas_autodownload(): midas.api.load_model = load_model_wrapper
+
def load_model(checkpoint_info=None):
from modules import lowvram, sd_hijack
checkpoint_info = checkpoint_info or select_checkpoint()
+ checkpoint_config = find_checkpoint_config(checkpoint_info)
- if checkpoint_info.config != shared.cmd_opts.config:
- print(f"Loading config from: {checkpoint_info.config}")
+ if checkpoint_config != shared.cmd_opts.config:
+ print(f"Loading config from: {checkpoint_config}")
if shared.sd_model:
sd_hijack.model_hijack.undo_hijack(shared.sd_model)
@@ -288,17 +299,17 @@ def load_model(checkpoint_info=None): gc.collect()
devices.torch_gc()
- sd_config = OmegaConf.load(checkpoint_info.config)
+ sd_config = OmegaConf.load(checkpoint_config)
if should_hijack_inpainting(checkpoint_info):
# Hardcoded config for now...
sd_config.model.target = "ldm.models.diffusion.ddpm.LatentInpaintDiffusion"
- sd_config.model.params.use_ema = False
sd_config.model.params.conditioning_key = "hybrid"
sd_config.model.params.unet_config.params.in_channels = 9
+ sd_config.model.params.finetune_keys = None
- # Create a "fake" config with a different name so that we know to unload it when switching models.
- checkpoint_info = checkpoint_info._replace(config=checkpoint_info.config.replace(".yaml", "-inpainting.yaml"))
+ if not hasattr(sd_config.model.params, "use_ema"):
+ sd_config.model.params.use_ema = False
do_inpainting_hijack()
@@ -306,6 +317,7 @@ def load_model(checkpoint_info=None): sd_config.model.params.unet_config.params.use_fp16 = False
sd_model = instantiate_from_config(sd_config.model)
+
load_model_weights(sd_model, checkpoint_info)
if shared.cmd_opts.lowvram or shared.cmd_opts.medvram:
@@ -318,23 +330,29 @@ def load_model(checkpoint_info=None): sd_model.eval()
shared.sd_model = sd_model
+ sd_hijack.model_hijack.embedding_db.load_textual_inversion_embeddings(force_reload=True) # Reload embeddings after model load as they may or may not fit the model
+
script_callbacks.model_loaded_callback(sd_model)
- print(f"Model loaded.")
+ print("Model loaded.")
+
return sd_model
def reload_model_weights(sd_model=None, info=None):
from modules import lowvram, devices, sd_hijack
checkpoint_info = info or select_checkpoint()
-
+
if not sd_model:
sd_model = shared.sd_model
+ current_checkpoint_info = sd_model.sd_checkpoint_info
+ checkpoint_config = find_checkpoint_config(current_checkpoint_info)
+
if sd_model.sd_model_checkpoint == checkpoint_info.filename:
return
- if sd_model.sd_checkpoint_info.config != checkpoint_info.config or should_hijack_inpainting(checkpoint_info) != should_hijack_inpainting(sd_model.sd_checkpoint_info):
+ if checkpoint_config != find_checkpoint_config(checkpoint_info) or should_hijack_inpainting(checkpoint_info) != should_hijack_inpainting(sd_model.sd_checkpoint_info):
del sd_model
checkpoints_loaded.clear()
load_model(checkpoint_info)
@@ -347,13 +365,19 @@ def reload_model_weights(sd_model=None, info=None): sd_hijack.model_hijack.undo_hijack(sd_model)
- load_model_weights(sd_model, checkpoint_info)
+ try:
+ load_model_weights(sd_model, checkpoint_info)
+ except Exception as e:
+ print("Failed to load checkpoint, restoring previous")
+ load_model_weights(sd_model, current_checkpoint_info)
+ raise
+ finally:
+ sd_hijack.model_hijack.hijack(sd_model)
+ script_callbacks.model_loaded_callback(sd_model)
- sd_hijack.model_hijack.hijack(sd_model)
- script_callbacks.model_loaded_callback(sd_model)
+ if not shared.cmd_opts.lowvram and not shared.cmd_opts.medvram:
+ sd_model.to(devices.device)
- if not shared.cmd_opts.lowvram and not shared.cmd_opts.medvram:
- sd_model.to(devices.device)
+ print("Weights loaded.")
- print(f"Weights loaded.")
return sd_model
|