VictorMylle/Kohya-ss-sd-scripts
1
0
Fork 1
mirror of https://github.com/kohya-ss/sd-scripts.git synced 2026-04-09 06:45:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'dev' into dev_device_support

Browse Source
This commit is contained in:
Kohya S
2024-02-12 13:01:54 +09:00
parent e579648ce9 35c6053de3
commit 358ca205a3
62 changed files with 1387 additions and 993 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
networks/lora_interrogator.py
View File

@@ -12,6 +12,10 @@ init_ipex()
import library.model_util as model_util
import lora
from library.utils import setup_logging
setup_logging()
import logging
logger = logging.getLogger(__name__)
TOKENIZER_PATH = "openai/clip-vit-large-patch14"
V2_STABLE_DIFFUSION_PATH = "stabilityai/stable-diffusion-2" # ここからtokenizerだけ使う
@@ -23,12 +27,12 @@ def interrogate(args):
weights_dtype = torch.float16
# いろいろ準備する
print(f"loading SD model: {args.sd_model}")
logger.info(f"loading SD model: {args.sd_model}")
args.pretrained_model_name_or_path = args.sd_model
args.vae = None
text_encoder, vae, unet, _ = train_util._load_target_model(args,weights_dtype, DEVICE)
print(f"loading LoRA: {args.model}")
logger.info(f"loading LoRA: {args.model}")
network, weights_sd = lora.create_network_from_weights(1.0, args.model, vae, text_encoder, unet)
# text encoder向けの重みがあるかチェックする：本当はlora側でやるのがいい
@@ -38,11 +42,11 @@ def interrogate(args):
has_te_weight = True
break
if not has_te_weight:
print("This LoRA does not have modules for Text Encoder, cannot interrogate / このLoRAはText Encoder向けのモジュールがないため調査できません")
logger.error("This LoRA does not have modules for Text Encoder, cannot interrogate / このLoRAはText Encoder向けのモジュールがないため調査できません")
return
del vae
print("loading tokenizer")
logger.info("loading tokenizer")
if args.v2:
tokenizer: CLIPTokenizer = CLIPTokenizer.from_pretrained(V2_STABLE_DIFFUSION_PATH, subfolder="tokenizer")
else:
@@ -56,7 +60,7 @@ def interrogate(args):
# トークンをひとつひとつ当たっていく
token_id_start = 0
token_id_end = max(tokenizer.all_special_ids)
print(f"interrogate tokens are: {token_id_start} to {token_id_end}")
logger.info(f"interrogate tokens are: {token_id_start} to {token_id_end}")
def get_all_embeddings(text_encoder):
embs = []
@@ -82,24 +86,24 @@ def interrogate(args):
embs.extend(encoder_hidden_states)
return torch.stack(embs)
print("get original text encoder embeddings.")
logger.info("get original text encoder embeddings.")
orig_embs = get_all_embeddings(text_encoder)
network.apply_to(text_encoder, unet, True, len(network.unet_loras) > 0)
info = network.load_state_dict(weights_sd, strict=False)
print(f"Loading LoRA weights: {info}")
logger.info(f"Loading LoRA weights: {info}")
network.to(DEVICE, dtype=weights_dtype)
network.eval()
del unet
print("You can ignore warning messages start with '_IncompatibleKeys' (LoRA model does not have alpha because trained by older script) / '_IncompatibleKeys'の警告は無視して構いません（以前のスクリプトで学習されたLoRAモデルのためalphaの定義がありません）")
print("get text encoder embeddings with lora.")
logger.info("You can ignore warning messages start with '_IncompatibleKeys' (LoRA model does not have alpha because trained by older script) / '_IncompatibleKeys'の警告は無視して構いません（以前のスクリプトで学習されたLoRAモデルのためalphaの定義がありません）")
logger.info("get text encoder embeddings with lora.")
lora_embs = get_all_embeddings(text_encoder)
# 比べる：とりあえず単純に差分の絶対値で
print("comparing...")
logger.info("comparing...")
diffs = {}
for i, (orig_emb, lora_emb) in enumerate(zip(orig_embs, tqdm(lora_embs))):
diff = torch.mean(torch.abs(orig_emb - lora_emb))