forked from huggingface/diffusers
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Support convert LoRA safetensors into diffusers format (huggingface#2403
) * add lora convertor * Update convert_lora_safetensor_to_diffusers.py * Update README.md * Update convert_lora_safetensor_to_diffusers.py
- Loading branch information
1 parent
9920c33
commit 63805f8
Showing
2 changed files
with
134 additions
and
4 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,130 @@ | ||
# coding=utf-8 | ||
# Copyright 2023, Haofan Wang, Qixun Wang, All rights reserved. | ||
# | ||
# Licensed under the Apache License, Version 2.0 (the "License"); | ||
# you may not use this file except in compliance with the License. | ||
# You may obtain a copy of the License at | ||
# | ||
# http://www.apache.org/licenses/LICENSE-2.0 | ||
# | ||
# Unless required by applicable law or agreed to in writing, software | ||
# distributed under the License is distributed on an "AS IS" BASIS, | ||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. | ||
# See the License for the specific language governing permissions and | ||
# limitations under the License. | ||
|
||
""" Conversion script for the LoRA's safetensors checkpoints. """ | ||
|
||
import argparse | ||
|
||
import torch | ||
from safetensors.torch import load_file | ||
|
||
from diffusers import StableDiffusionPipeline | ||
|
||
|
||
def convert(base_model_path, checkpoint_path, LORA_PREFIX_UNET, LORA_PREFIX_TEXT_ENCODER, alpha): | ||
|
||
# load base model | ||
pipeline = StableDiffusionPipeline.from_pretrained(base_model_path, torch_dtype=torch.float32) | ||
|
||
# load LoRA weight from .safetensors | ||
state_dict = load_file(checkpoint_path) | ||
|
||
visited = [] | ||
|
||
# directly update weight in diffusers model | ||
for key in state_dict: | ||
|
||
# it is suggested to print out the key, it usually will be something like below | ||
# "lora_te_text_model_encoder_layers_0_self_attn_k_proj.lora_down.weight" | ||
|
||
# as we have set the alpha beforehand, so just skip | ||
if ".alpha" in key or key in visited: | ||
continue | ||
|
||
if "text" in key: | ||
layer_infos = key.split(".")[0].split(LORA_PREFIX_TEXT_ENCODER + "_")[-1].split("_") | ||
curr_layer = pipeline.text_encoder | ||
else: | ||
layer_infos = key.split(".")[0].split(LORA_PREFIX_UNET + "_")[-1].split("_") | ||
curr_layer = pipeline.unet | ||
|
||
# find the target layer | ||
temp_name = layer_infos.pop(0) | ||
while len(layer_infos) > -1: | ||
try: | ||
curr_layer = curr_layer.__getattr__(temp_name) | ||
if len(layer_infos) > 0: | ||
temp_name = layer_infos.pop(0) | ||
elif len(layer_infos) == 0: | ||
break | ||
except Exception: | ||
if len(temp_name) > 0: | ||
temp_name += "_" + layer_infos.pop(0) | ||
else: | ||
temp_name = layer_infos.pop(0) | ||
|
||
pair_keys = [] | ||
if "lora_down" in key: | ||
pair_keys.append(key.replace("lora_down", "lora_up")) | ||
pair_keys.append(key) | ||
else: | ||
pair_keys.append(key) | ||
pair_keys.append(key.replace("lora_up", "lora_down")) | ||
|
||
# update weight | ||
if len(state_dict[pair_keys[0]].shape) == 4: | ||
weight_up = state_dict[pair_keys[0]].squeeze(3).squeeze(2).to(torch.float32) | ||
weight_down = state_dict[pair_keys[1]].squeeze(3).squeeze(2).to(torch.float32) | ||
curr_layer.weight.data += alpha * torch.mm(weight_up, weight_down).unsqueeze(2).unsqueeze(3) | ||
else: | ||
weight_up = state_dict[pair_keys[0]].to(torch.float32) | ||
weight_down = state_dict[pair_keys[1]].to(torch.float32) | ||
curr_layer.weight.data += alpha * torch.mm(weight_up, weight_down) | ||
|
||
# update visited list | ||
for item in pair_keys: | ||
visited.append(item) | ||
|
||
return pipeline | ||
|
||
|
||
if __name__ == "__main__": | ||
parser = argparse.ArgumentParser() | ||
|
||
parser.add_argument( | ||
"--base_model_path", default=None, type=str, required=True, help="Path to the base model in diffusers format." | ||
) | ||
parser.add_argument( | ||
"--checkpoint_path", default=None, type=str, required=True, help="Path to the checkpoint to convert." | ||
) | ||
parser.add_argument("--dump_path", default=None, type=str, required=True, help="Path to the output model.") | ||
parser.add_argument( | ||
"--lora_prefix_unet", default="lora_unet", type=str, help="The prefix of UNet weight in safetensors" | ||
) | ||
parser.add_argument( | ||
"--lora_prefix_text_encoder", | ||
default="lora_te", | ||
type=str, | ||
help="The prefix of text encoder weight in safetensors", | ||
) | ||
parser.add_argument("--alpha", default=0.75, type=float, help="The merging ratio in W = W0 + alpha * deltaW") | ||
parser.add_argument( | ||
"--to_safetensors", action="store_true", help="Whether to store pipeline in safetensors format or not." | ||
) | ||
parser.add_argument("--device", type=str, help="Device to use (e.g. cpu, cuda:0, cuda:1, etc.)") | ||
|
||
args = parser.parse_args() | ||
|
||
base_model_path = args.base_model_path | ||
checkpoint_path = args.checkpoint_path | ||
dump_path = args.dump_path | ||
lora_prefix_unet = args.lora_prefix_unet | ||
lora_prefix_text_encoder = args.lora_prefix_text_encoder | ||
alpha = args.alpha | ||
|
||
pipe = convert(base_model_path, checkpoint_path, lora_prefix_unet, lora_prefix_text_encoder, alpha) | ||
|
||
pipe = pipe.to(args.device) | ||
pipe.save_pretrained(args.dump_path, safe_serialization=args.to_safetensors) |