Public/automatic
1
0
Fork 0
mirror of https://github.com/vladmandic/automatic synced 2026-09-19 09:14:35 +02:00
Code Issues Packages Projects Releases Wiki Activity

Flux-LoRa first commit

Browse Source
This commit is contained in:
AI-Casanova
2024-09-26 19:15:06 -05:00
parent bcb704a4ac
commit 314e3339d2
3 changed files with 20 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files
extensions-builtin/Lora/network_lora.py
+1 -1
View File
@@ -27,7 +27,7 @@ class NetworkModuleLora(network.NetworkModule):
if weight is None and none_ok:
return None
linear_modules = [torch.nn.Linear, torch.nn.modules.linear.NonDynamicallyQuantizableLinear, torch.nn.MultiheadAttention, diffusers_lora.LoRACompatibleLinear]
is_linear = type(self.sd_module) in linear_modules or self.sd_module.__class__.__name__ in {"NNCFLinear", "QLinear"}
is_linear = type(self.sd_module) in linear_modules or self.sd_module.__class__.__name__ in {"NNCFLinear", "QLinear", "Linear4bit"}
is_conv = type(self.sd_module) in [torch.nn.Conv2d, diffusers_lora.LoRACompatibleConv] or self.sd_module.__class__.__name__ in {"NNCFConv2d", "QConv2d"}
if is_linear:
weight = weight.reshape(weight.shape[0], -1)
extensions-builtin/Lora/network_overrides.py
-1
View File
@@ -30,7 +30,6 @@ force_models = [ # forced always
'kandinsky',
'hunyuandit',
'auraflow',
'f1',
]
force_classes = [ # forced always
extensions-builtin/Lora/networks.py
+19 -13
View File
@@ -50,23 +50,29 @@ convert_diffusers_name_to_compvis = lora_convert.convert_diffusers_name_to_compv
def assign_network_names_to_compvis_modules(sd_model):
network_layer_mapping = {}
if shared.native:
if not hasattr(shared.sd_model, 'text_encoder') or not hasattr(shared.sd_model, 'unet'):
sd_model.network_layer_mapping = {}
return
for name, module in shared.sd_model.text_encoder.named_modules():
prefix = "lora_te1_" if shared.sd_model_type == "sdxl" else "lora_te_"
network_name = prefix + name.replace(".", "_")
network_layer_mapping[network_name] = module
module.network_layer_name = network_name
if shared.sd_model_type == "sdxl":
if hasattr(shared.sd_model, 'text_encoder'):
for name, module in shared.sd_model.text_encoder.named_modules():
prefix = "lora_te1_" if hasattr(shared.sd_model, 'text_encoder_2') else "lora_te_"
network_name = prefix + name.replace(".", "_")
network_layer_mapping[network_name] = module
module.network_layer_name = network_name
if hasattr(shared.sd_model, 'text_encoder_2'):
for name, module in shared.sd_model.text_encoder_2.named_modules():
network_name = "lora_te2_" + name.replace(".", "_")
network_layer_mapping[network_name] = module
module.network_layer_name = network_name
for name, module in shared.sd_model.unet.named_modules():
network_name = "lora_unet_" + name.replace(".", "_")
network_layer_mapping[network_name] = module
module.network_layer_name = network_name
if hasattr(shared.sd_model, 'unet'):
for name, module in shared.sd_model.unet.named_modules():
network_name = "lora_unet_" + name.replace(".", "_")
network_layer_mapping[network_name] = module
module.network_layer_name = network_name
if hasattr(shared.sd_model, 'transformer'):
for name, module in shared.sd_model.transformer.named_modules():
network_name = "lora_transformer_" + name.replace(".", "_")
network_layer_mapping[network_name] = module
if "norm" in network_name and "linear" not in network_name:
continue
module.network_layer_name = network_name
else:
if not hasattr(shared.sd_model, 'cond_stage_model'):
sd_model.network_layer_mapping = {}