VictorMylle/Kohya-ss-sd-scripts
1
0
Fork 1
mirror of https://github.com/kohya-ss/sd-scripts.git synced 2026-04-08 22:35:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

init

Browse Source
This commit is contained in:
sdbds
2024-09-25 19:38:52 +08:00
parent 29177d2f03
commit ab7b231870
2 changed files with 19 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
library/train_util.py
View File

@@ -2994,7 +2994,7 @@ def add_optimizer_arguments(parser: argparse.ArgumentParser):
"--optimizer_type", "--optimizer_type",
type=str, type=str,
default="", default="",
help="Optimizer to use / オプティマイザの種類: AdamW (default), AdamW8bit, PagedAdamW, PagedAdamW8bit, PagedAdamW32bit, Lion8bit, PagedLion8bit, Lion, SGDNesterov, SGDNesterov8bit, DAdaptation(DAdaptAdamPreprint), DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptAdanIP, DAdaptLion, DAdaptSGD, AdaFactor", help="Optimizer to use / オプティマイザの種類: AdamW (default), AdamW8bit, PagedAdamW, PagedAdamW8bit, PagedAdamW32bit, Lion8bit, PagedLion8bit, Lion, AdEMAMix8bit, PagedAdEMAMix8bit, SGDNesterov, SGDNesterov8bit, DAdaptation(DAdaptAdamPreprint), DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptAdanIP, DAdaptLion, DAdaptSGD, AdaFactor",
) )
# backward compatibility # backward compatibility
@@ -4032,7 +4032,7 @@ def resume_from_local_or_hf_if_specified(accelerator, args):
def get_optimizer(args, trainable_params): def get_optimizer(args, trainable_params):
# "Optimizer to use: AdamW, AdamW8bit, Lion, SGDNesterov, SGDNesterov8bit, PagedAdamW, PagedAdamW8bit, PagedAdamW32bit, Lion8bit, PagedLion8bit, DAdaptation(DAdaptAdamPreprint), DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptAdanIP, DAdaptLion, DAdaptSGD, Adafactor" # "Optimizer to use: AdamW, AdamW8bit, Lion, SGDNesterov, SGDNesterov8bit, PagedAdamW, PagedAdamW8bit, PagedAdamW32bit, Lion8bit, PagedLion8bit, AdEMAMix8bit, PagedAdEMAMix8bit, DAdaptation(DAdaptAdamPreprint), DAdaptAdaGrad, DAdaptAdam, DAdaptAdan, DAdaptAdanIP, DAdaptLion, DAdaptSGD, Adafactor"
optimizer_type = args.optimizer_type optimizer_type = args.optimizer_type
if args.use_8bit_adam: if args.use_8bit_adam:
@@ -4141,7 +4141,22 @@ def get_optimizer(args, trainable_params):
raise AttributeError( raise AttributeError(
"No PagedLion8bit. The version of bitsandbytes installed seems to be old. Please install 0.39.0 or later. / PagedLion8bitが定義されていません。インストールされているbitsandbytesのバージョンが古いようです。0.39.0以上をインストールしてください" "No PagedLion8bit. The version of bitsandbytes installed seems to be old. Please install 0.39.0 or later. / PagedLion8bitが定義されていません。インストールされているbitsandbytesのバージョンが古いようです。0.39.0以上をインストールしてください"
) )
elif optimizer_type == "Ademamix8bit".lower():
logger.info(f"use 8-bit Ademamix optimizer | {optimizer_kwargs}")
try:
optimizer_class = bnb.optim.AdEMAMix8bit
except AttributeError:
raise AttributeError(
"No Ademamix8bit. The version of bitsandbytes installed seems to be old. Please install 0.44.0 or later. / Ademamix8bitが定義されていません。インストールされているbitsandbytesのバージョンが古いようです。0.39.0以上をインストールしてください"
)
elif optimizer_type == "PagedAdemamix8bit".lower():
logger.info(f"use 8-bit PagedAdemamix optimizer | {optimizer_kwargs}")
try:
optimizer_class = bnb.optim.PagedAdEMAMix8bit
except AttributeError:
raise AttributeError(
"No PagedAdemamix8bit. The version of bitsandbytes installed seems to be old. Please install 0.44.0 or later. / PagedAdemamix8bitが定義されていません。インストールされているbitsandbytesのバージョンが古いようです。0.39.0以上をインストールしてください"
)
optimizer = optimizer_class(trainable_params, lr=lr, **optimizer_kwargs) optimizer = optimizer_class(trainable_params, lr=lr, **optimizer_kwargs)
elif optimizer_type == "PagedAdamW".lower(): elif optimizer_type == "PagedAdamW".lower():

2
requirements.txt
View File

@@ -6,7 +6,7 @@ ftfy==6.1.1
opencv-python==4.8.1.78 opencv-python==4.8.1.78
einops==0.7.0 einops==0.7.0
pytorch-lightning==1.9.0 pytorch-lightning==1.9.0
bitsandbytes==0.43.0 bitsandbytes==0.44.0
prodigyopt==1.0 prodigyopt==1.0
lion-pytorch==0.0.6 lion-pytorch==0.0.6
tensorboard tensorboard