VictorMylle/Kohya-ss-sd-scripts
1
0
Fork 1
mirror of https://github.com/kohya-ss/sd-scripts.git synced 2026-04-08 14:34:23 +00:00
Code Issues Packages Projects Releases Wiki Activity

change block name doesn't contain '.' at end

Browse Source
This commit is contained in:
Kohya S
2023-09-24 12:10:56 +09:00
parent 1f169ee7fb
commit f2491ee0ac
1 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
networks/resize_lora.py
View File

@@ -219,7 +219,7 @@ def resize_lora_model(lora_sd, new_rank, save_dtype, device, dynamic_method, dyn
for key, value in tqdm(lora_sd.items()):
weight_name = None
if 'lora_down' in key:
block_down_name = key.rsplit('lora_down', 1)[0]
block_down_name = key.rsplit('.lora_down', 1)[0]
weight_name = key.rsplit(".", 1)[-1]
lora_down_weight = value
else:
@@ -227,8 +227,8 @@ def resize_lora_model(lora_sd, new_rank, save_dtype, device, dynamic_method, dyn
# find corresponding lora_up and alpha
block_up_name = block_down_name
lora_up_weight = lora_sd.get(block_up_name + 'lora_up.' + weight_name, None)
lora_alpha = lora_sd.get(block_down_name + 'alpha', None)
lora_up_weight = lora_sd.get(block_up_name + '.lora_up.' + weight_name, None)
lora_alpha = lora_sd.get(block_down_name + '.alpha', None)
weights_loaded = (lora_down_weight is not None and lora_up_weight is not None)
@@ -263,9 +263,9 @@ def resize_lora_model(lora_sd, new_rank, save_dtype, device, dynamic_method, dyn
verbose_str+=f"\n"
new_alpha = param_dict['new_alpha']
o_lora_sd[block_down_name + "lora_down.weight"] = param_dict["lora_down"].to(save_dtype).contiguous()
o_lora_sd[block_up_name + "lora_up.weight"] = param_dict["lora_up"].to(save_dtype).contiguous()
o_lora_sd[block_up_name + "alpha"] = torch.tensor(param_dict['new_alpha']).to(save_dtype)
o_lora_sd[block_down_name + "." + "lora_down.weight"] = param_dict["lora_down"].to(save_dtype).contiguous()
o_lora_sd[block_up_name + "." + "lora_up.weight"] = param_dict["lora_up"].to(save_dtype).contiguous()
o_lora_sd[block_up_name + "." "alpha"] = torch.tensor(param_dict['new_alpha']).to(save_dtype)
block_down_name = None
block_up_name = None