Merge pull request #872 from kohya-ss/dev

fix make_captions_by_git, improve image generation scripts
update readme
2026-04-06 21:52:27 +00:00 · 2023-10-11 07:56:39 +09:00 · 2023-10-11 07:54:30 +09:00 · 2023-10-11 07:40:12 +09:00 · 2023-10-09 23:07:14 +09:00 · 2023-10-09 21:42:13 +09:00
42 changed files with 2080 additions and 789 deletions
--- a/.github/dependabot.yml
+++ b/.github/dependabot.yml
@@ -0,0 +1,7 @@
+---
+version: 2
+updates:
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      interval: "monthly"
--- a/.github/workflows/typos.yml
+++ b/.github/workflows/typos.yml
@@ -15,7 +15,7 @@ jobs:
    runs-on: ubuntu-latest

    steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4

      - name: typos-action
-        uses: crate-ci/typos@v1.13.10
+        uses: crate-ci/typos@v1.16.15
--- a/README-ja.md
+++ b/README-ja.md
@@ -1,3 +1,7 @@
+SDXLがサポートされました。sdxlブランチはmainブランチにマージされました。リポジトリを更新したときにはUpgradeの手順を実行してください。また accelerate のバージョンが上がっていますので、accelerate config を再度実行してください。
+
+SDXL学習については[こちら](./README.md#sdxl-training)をご覧ください（英語です）。
+
 ## リポジトリについて
 Stable Diffusionの学習、画像生成、その他のスクリプトを入れたリポジトリです。

@@ -9,13 +13,12 @@ GUIやPowerShellスクリプトなど、より使いやすくする機能が[bma

 * DreamBooth、U-NetおよびText Encoderの学習をサポート
 * fine-tuning、同上
+* LoRAの学習をサポート
 * 画像生成
 * モデル変換（Stable Diffision ckpt/safetensorsとDiffusersの相互変換）

 ## 使用法について

-当リポジトリ内およびnote.comに記事がありますのでそちらをご覧ください（将来的にはすべてこちらへ移すかもしれません）。
-
 * [学習について、共通編](./docs/train_README-ja.md) : データ整備やオプションなど
    * [データセット設定](./docs/config_README-ja.md)
 * [DreamBoothの学習について](./docs/train_db_README-ja.md)
@@ -41,11 +44,13 @@ PowerShellを使う場合、venvを使えるようにするためには以下の

 ## Windows環境でのインストール

-以下の例ではPyTorchは1.12.1／CUDA 11.6版をインストールします。CUDA 11.3版やPyTorch 1.13を使う場合は適宜書き換えください。
+スクリプトはPyTorch 2.0.1でテストしています。PyTorch 1.12.1でも動作すると思われます。
+
+以下の例ではPyTorchは2.0.1／CUDA 11.8版をインストールします。CUDA 11.6版やPyTorch 1.12.1を使う場合は適宜書き換えください。

 （なお、python -m venv～の行で「python」とだけ表示された場合、py -m venv～のようにpythonをpyに変更してください。）

-通常の（管理者ではない）PowerShellを開き以下を順に実行します。
+PowerShellを使う場合、通常の（管理者ではない）PowerShellを開き以下を順に実行します。

 ```powershell
 git clone https://github.com/kohya-ss/sd-scripts.git
@@ -54,43 +59,14 @@ cd sd-scripts
 python -m venv venv
 .\venv\Scripts\activate

-pip install torch==1.12.1+cu116 torchvision==0.13.1+cu116 --extra-index-url https://download.pytorch.org/whl/cu116
+pip install torch==2.0.1+cu118 torchvision==0.15.2+cu118 --index-url https://download.pytorch.org/whl/cu118
 pip install --upgrade -r requirements.txt
-pip install -U -I --no-deps https://github.com/C43H66N12O12S2/stable-diffusion-webui/releases/download/f/xformers-0.0.14.dev0-cp310-cp310-win_amd64.whl
-
-cp .\bitsandbytes_windows\*.dll .\venv\Lib\site-packages\bitsandbytes\
-cp .\bitsandbytes_windows\cextension.py .\venv\Lib\site-packages\bitsandbytes\cextension.py
-cp .\bitsandbytes_windows\main.py .\venv\Lib\site-packages\bitsandbytes\cuda_setup\main.py
+pip install xformers==0.0.20

 accelerate config
 ```

-<!-- 
-pip install torch==1.13.1+cu117 torchvision==0.14.1+cu117 --extra-index-url https://download.pytorch.org/whl/cu117
-pip install --use-pep517 --upgrade -r requirements.txt
-pip install -U -I --no-deps xformers==0.0.16
-->
-
-コマンドプロンプトでは以下になります。
-
-
-```bat
-git clone https://github.com/kohya-ss/sd-scripts.git
-cd sd-scripts
-
-python -m venv venv
-.\venv\Scripts\activate
-
-pip install torch==1.12.1+cu116 torchvision==0.13.1+cu116 --extra-index-url https://download.pytorch.org/whl/cu116
-pip install --upgrade -r requirements.txt
-pip install -U -I --no-deps https://github.com/C43H66N12O12S2/stable-diffusion-webui/releases/download/f/xformers-0.0.14.dev0-cp310-cp310-win_amd64.whl
-
-copy /y .\bitsandbytes_windows\*.dll .\venv\Lib\site-packages\bitsandbytes\
-copy /y .\bitsandbytes_windows\cextension.py .\venv\Lib\site-packages\bitsandbytes\cextension.py
-copy /y .\bitsandbytes_windows\main.py .\venv\Lib\site-packages\bitsandbytes\cuda_setup\main.py
-
-accelerate config
-```
+コマンドプロンプトでも同一です。

 （注:``python -m venv venv`` のほうが ``python -m venv --system-site-packages venv`` より安全そうなため書き換えました。globalなpythonにパッケージがインストールしてあると、後者だといろいろと問題が起きます。）

@@ -111,30 +87,41 @@ accelerate configの質問には以下のように答えてください。（bf1
 ※場合によって ``ValueError: fp16 mixed precision requires a GPU`` というエラーが出ることがあるようです。この場合、6番目の質問（
 ``What GPU(s) (by id) should be used for training on this machine as a comma-separated list? [all]:``）に「0」と答えてください。（id `0`のGPUが使われます。）

-### PyTorchとxformersのバージョンについて
+### オプション：`bitsandbytes`（8bit optimizer）を使う

-他のバージョンでは学習がうまくいかない場合があるようです。特に他の理由がなければ指定のバージョンをお使いください。
+`bitsandbytes`はオプションになりました。Linuxでは通常通りpipでインストールできます（0.41.1または以降のバージョンを推奨）。

-### オプション：Lion8bitを使う
+Windowsでは0.35.0または0.41.1を推奨します。

-Lion8bitを使う場合には`bitsandbytes`を0.38.0以降にアップグレードする必要があります。`bitsandbytes`をアンインストールし、Windows環境では例えば[こちら](https://github.com/jllllll/bitsandbytes-windows-webui)などからWindows版のwhlファイルをインストールしてください。たとえば以下のような手順になります。
+- `bitsandbytes` 0.35.0: 安定しているとみられるバージョンです。AdamW8bitは使用できますが、他のいくつかの8bit optimizer、学習時の`full_bf16`オプションは使用できません。
+- `bitsandbytes` 0.41.1: Lion8bit、PagedAdamW8bit、PagedLion8bitをサポートします。`full_bf16`が使用できます。
+
+注：`bitsandbytes` 0.35.0から0.41.0までのバージョンには問題があるようです。 https://github.com/TimDettmers/bitsandbytes/issues/659
+
+以下の手順に従い、`bitsandbytes`をインストールしてください。
+
+### 0.35.0を使う場合
+
+PowerShellの例です。コマンドプロンプトではcpの代わりにcopyを使ってください。

 ```powershell
-pip install https://github.com/jllllll/bitsandbytes-windows-webui/raw/main/bitsandbytes-0.38.1-py3-none-any.whl
+cd sd-scripts
+.\venv\Scripts\activate
+pip install bitsandbytes==0.35.0
+
+cp .\bitsandbytes_windows\*.dll .\venv\Lib\site-packages\bitsandbytes\
+cp .\bitsandbytes_windows\cextension.py .\venv\Lib\site-packages\bitsandbytes\cextension.py
+cp .\bitsandbytes_windows\main.py .\venv\Lib\site-packages\bitsandbytes\cuda_setup\main.py
 ```

-アップグレード時には`pip install .`でこのリポジトリを更新し、必要に応じて他のパッケージもアップグレードしてください。
+### 0.41.1を使う場合

-### オプション：PagedAdamW8bitとPagedLion8bitを使う
-
-PagedAdamW8bitとPagedLion8bitを使う場合には`bitsandbytes`を0.39.0以降にアップグレードする必要があります。`bitsandbytes`をアンインストールし、Windows環境では例えば[こちら](https://github.com/jllllll/bitsandbytes-windows-webui)などからWindows版のwhlファイルをインストールしてください。たとえば以下のような手順になります。
+jllllll氏の配布されている[こちら](https://github.com/jllllll/bitsandbytes-windows-webui) または他の場所から、Windows用のwhlファイルをインストールしてください。

 ```powershell
-pip install https://github.com/jllllll/bitsandbytes-windows-webui/releases/download/wheels/bitsandbytes-0.39.1-py3-none-win_amd64.whl
+python -m pip install bitsandbytes==0.41.1 --prefer-binary --extra-index-url=https://jllllll.github.io/bitsandbytes-windows-webui
 ```

-アップグレード時には`pip install .`でこのリポジトリを更新し、必要に応じて他のパッケージもアップグレードしてください。
-
 ## アップグレード

 新しいリリースがあった場合、以下のコマンドで更新できます。
--- a/README.md
+++ b/README.md
@@ -1,9 +1,11 @@
+__SDXL is now supported. The sdxl branch has been merged into the main branch. If you update the repository, please follow the upgrade instructions. Also, the version of accelerate has been updated, so please run accelerate config again.__ The documentation for SDXL training is [here](./README.md#sdxl-training).
+
 This repository contains training, generation and utility scripts for Stable Diffusion.

-[__Change History__](#change-history) is moved to the bottom of the page.
+[__Change History__](#change-history) is moved to the bottom of the page. 
 更新履歴は[ページ末尾](#change-history)に移しました。

-[日本語版README](./README-ja.md)
+[日本語版READMEはこちら](./README-ja.md)

 For easier use (GUI and PowerShell scripts etc...), please visit [the repository maintained by bmaltais](https://github.com/bmaltais/kohya_ss). Thanks to @bmaltais!

@@ -16,135 +18,13 @@ This repository contains the scripts for:
 * Image generation
 * Model conversion (supports 1.x and 2.x, Stable Diffision ckpt/safetensors and Diffusers)

-__Stable Diffusion web UI now seems to support LoRA trained by ``sd-scripts``.__ Thank you for great work!!! 
-
-## About SDXL training
-
-The feature of SDXL training is now available in sdxl branch as an experimental feature. 
-
-Sep 3, 2023: The feature will be merged into the main branch soon. Following are the changes from the previous version. 
-
- ControlNet-LLLite is added. See [documentation](./docs/train_lllite_README.md) for details.
- JPEG XL is supported. [#786](https://github.com/kohya-ss/sd-scripts/pull/786) 
- Peak memory usage is reduced. [#791](https://github.com/kohya-ss/sd-scripts/pull/791)
- Input perturbation noise is added. See [#798](https://github.com/kohya-ss/sd-scripts/pull/798) for details.
- Dataset subset now has `caption_prefix` and `caption_suffix` options. The strings are added to the beginning and the end of the captions before shuffling. You can specify the options in `.toml`.
- Other minor changes.
- Thanks for contributions from Isotr0py, vvern999, lansing  and others!
-
-Aug 13, 2023: 
-
- LoRA-FA is added experimentally. Specify `--network_module networks.lora_fa` option instead of `--network_module networks.lora`. The trained model can be used as a normal LoRA model.
-
-Aug 12, 2023: 
-
- The default value of noise offset when omitted has been changed to 0 from 0.0357.
- The different learning rates for each U-Net block are now supported. Specify with `--block_lr` option. Specify 23 values separated by commas like `--block_lr 1e-3,1e-3 ... 1e-3`.
-  - 23 values correspond to `0: time/label embed, 1-9: input blocks 0-8, 10-12: mid blocks 0-2, 13-21: output blocks 0-8, 22: out`.
-
-Aug 6, 2023: 
-
- [SAI Model Spec](https://github.com/Stability-AI/ModelSpec) metadata is now supported partially. `hash_sha256` is not supported yet.
-  - The main items are set automatically. 
-  - You can set title, author, description, license and tags with `--metadata_xxx` options in each training script.
-  - Merging scripts also support minimum SAI Model Spec metadata. See the help message for the usage.
-  - Metadata editor will be available soon.
- SDXL LoRA has `sdxl_base_v1-0` now  for `ss_base_model_version` metadata item, instead of `v0-9`.
-
-Aug 4, 2023: 
-
- `bitsandbytes` is now optional. Please install it if you want to use it. The insructions are in the later section.
- `albumentations` is not required anymore.
- An issue for pooled output for Textual Inversion training is fixed.
- `--v_pred_like_loss ratio` option is added. This option adds the loss like v-prediction loss in SDXL training. `0.1` means that the loss is added 10% of the v-prediction loss. The default value is None (disabled).
-  - In v-prediction, the loss is higher in the early timesteps (near the noise). This option can be used to increase the loss in the early timesteps.
- Arbitrary options can be used for Diffusers' schedulers. For example `--lr_scheduler_args "lr_end=1e-8"`.
- `sdxl_gen_imgs.py` supports batch size > 1.
- Fix ControlNet to work with attention couple and reginal LoRA in `gen_img_diffusers.py`.
-
-Summary of the feature:
-
- `tools/cache_latents.py` is added. This script can be used to cache the latents to disk in advance. 
-  - The options are almost the same as `sdxl_train.py'. See the help message for the usage.
-  - Please launch the script as follows:
-    `accelerate launch  --num_cpu_threads_per_process 1 tools/cache_latents.py ...`
-  - This script should work with multi-GPU, but it is not tested in my environment.
-
- `tools/cache_text_encoder_outputs.py` is added. This script can be used to cache the text encoder outputs to disk in advance. 
-  - The options are almost the same as `cache_latents.py' and `sdxl_train.py'. See the help message for the usage.
-
- `sdxl_train.py` is a script for SDXL fine-tuning. The usage is almost the same as `fine_tune.py`, but it also supports DreamBooth dataset.
-  - `--full_bf16` option is added. Thanks to KohakuBlueleaf!
-    - This option enables the full bfloat16 training (includes gradients). This option is useful to reduce the GPU memory usage. 
-    - However, bitsandbytes==0.35 doesn't seem to support this. Please use a newer version of bitsandbytes or another optimizer.
-    - I cannot find bitsandbytes>0.35.0 that works correctly on Windows.
-    - In addition, the full bfloat16 training might be unstable. Please use it at your own risk.
- `prepare_buckets_latents.py` now supports SDXL fine-tuning.
- `sdxl_train_network.py` is a script for LoRA training for SDXL. The usage is almost the same as `train_network.py`.
- Both scripts has following additional options:
-  - `--cache_text_encoder_outputs` and `--cache_text_encoder_outputs_to_disk`: Cache the outputs of the text encoders. This option is useful to reduce the GPU memory usage. This option cannot be used with options for shuffling or dropping the captions.
-  - `--no_half_vae`: Disable the half-precision (mixed-precision) VAE. VAE for SDXL seems to produce NaNs in some cases. This option is useful to avoid the NaNs.
- The image generation during training is now available. `--no_half_vae` option also works to avoid black images.
-
- `--weighted_captions` option is not supported yet for both scripts.
- `--min_timestep` and `--max_timestep` options are added to each training script. These options can be used to train U-Net with different timesteps. The default values are 0 and 1000.
-
- `sdxl_train_textual_inversion.py` is a script for Textual Inversion training for SDXL. The usage is almost the same as `train_textual_inversion.py`.
-  - `--cache_text_encoder_outputs` is not supported.
-  - `token_string` must be alphabet only currently, due to the limitation of the open-clip tokenizer.
-  - There are two options for captions:
-    1. Training with captions. All captions must include the token string. The token string is replaced with multiple tokens.
-    2. Use `--use_object_template` or `--use_style_template` option. The captions are generated from the template. The existing captions are ignored.
-  - See below for the format of the embeddings.
-  
- `sdxl_gen_img.py` is added. This script can be used to generate images with SDXL, including LoRA. See the help message for the usage.
-  - Textual Inversion is supported, but the name for the embeds in the caption becomes alphabet only. For example, `neg_hand_v1.safetensors` can be activated with `neghandv`.
-
-`requirements.txt` is updated to support SDXL training. 
-
-### Tips for SDXL training
-
- The default resolution of SDXL is 1024x1024.
- The fine-tuning can be done with 24GB GPU memory with the batch size of 1. For 24GB GPU, the following options are recommended __for the fine-tuning with 24GB GPU memory__:
-  - Train U-Net only.
-  - Use gradient checkpointing.
-  - Use `--cache_text_encoder_outputs` option and caching latents.
-  - Use Adafactor optimizer. RMSprop 8bit or Adagrad 8bit may work. AdamW 8bit doesn't seem to work.
- The LoRA training can be done with 8GB GPU memory (10GB recommended). For reducing the GPU memory usage, the following options are recommended:
-  - Train U-Net only.
-  - Use gradient checkpointing.
-  - Use `--cache_text_encoder_outputs` option and caching latents.
-  - Use one of 8bit optimizers or Adafactor optimizer.
-  - Use lower dim (-8 for 8GB GPU).
- `--network_train_unet_only` option is highly recommended for SDXL LoRA. Because SDXL has two text encoders, the result of the training will be unexpected.
- PyTorch 2 seems to use slightly less GPU memory than PyTorch 1.
- `--bucket_reso_steps` can be set to 32 instead of the default value 64. Smaller values than 32 will not work for SDXL training.
-
-Example of the optimizer settings for Adafactor with the fixed learning rate:
-```toml
-optimizer_type = "adafactor"
-optimizer_args = [ "scale_parameter=False", "relative_step=False", "warmup_init=False" ]
-lr_scheduler = "constant_with_warmup"
-lr_warmup_steps = 100
-learning_rate = 4e-7 # SDXL original learning rate
-```
-
-### Format of Textual Inversion embeddings
-
-```python
-from safetensors.torch import save_file
-
-state_dict = {"clip_g": embs_for_text_encoder_1280, "clip_l": embs_for_text_encoder_768}
-save_file(state_dict, file)
-```
-
 ## About requirements.txt

 These files do not contain requirements for PyTorch. Because the versions of them depend on your environment. Please install PyTorch at first (see installation guide below.) 

-The scripts are tested with PyTorch 1.12.1 and 2.0.1, Diffusers 0.18.2.
+The scripts are tested with Pytorch 2.0.1. 1.12.1 is not tested but should work.

-## Links to how-to-use documents
+## Links to usage documentation

 Most of the documents are written in Japanese.

@@ -184,9 +64,9 @@ cd sd-scripts
 python -m venv venv
 .\venv\Scripts\activate

-pip install torch==1.12.1+cu116 torchvision==0.13.1+cu116 --extra-index-url https://download.pytorch.org/whl/cu116
+pip install torch==2.0.1+cu118 torchvision==0.15.2+cu118 --index-url https://download.pytorch.org/whl/cu118
 pip install --upgrade -r requirements.txt
-pip install -U -I --no-deps https://github.com/C43H66N12O12S2/stable-diffusion-webui/releases/download/f/xformers-0.0.14.dev0-cp310-cp310-win_amd64.whl
+pip install xformers==0.0.20

 accelerate config
 ```
@@ -215,31 +95,6 @@ note: Some user reports ``ValueError: fp16 mixed precision requires a GPU`` is o

 (Single GPU with id `0` will be used.)

-### Experimental: Use PyTorch 2.0
-
-In this case, you need to install PyTorch 2.0 and xformers 0.0.20. Instead of the above, please type the following:
-
-```powershell
-git clone https://github.com/kohya-ss/sd-scripts.git
-cd sd-scripts
-
-python -m venv venv
-.\venv\Scripts\activate
-
-pip install torch==2.0.1+cu118 torchvision==0.15.2+cu118 --index-url https://download.pytorch.org/whl/cu118
-pip install --upgrade -r requirements.txt
-pip install xformers==0.0.20
-
-accelerate config
-```
-
-Answers to accelerate config should be the same as above.
-
-### about PyTorch and xformers
-
-Other versions of PyTorch and xformers seem to have problems with training.
-If there is no other reason, please install the specified version.
-
 ### Optional: Use `bitsandbytes` (8bit optimizer)

 For 8bit optimizer, you need to install `bitsandbytes`. For Linux, please install `bitsandbytes` as usual (0.41.1 or later is recommended.)
@@ -306,214 +161,169 @@ The majority of scripts is licensed under ASL 2.0 (including codes from Diffuser

 [BLIP](https://github.com/salesforce/BLIP): BSD-3-Clause

+
+## SDXL training
+
+The documentation in this section will be moved to a separate document later.
+
+### Training scripts for SDXL
+
+- `sdxl_train.py` is a script for SDXL fine-tuning. The usage is almost the same as `fine_tune.py`, but it also supports DreamBooth dataset.
+  - `--full_bf16` option is added. Thanks to KohakuBlueleaf!
+    - This option enables the full bfloat16 training (includes gradients). This option is useful to reduce the GPU memory usage. 
+    - The full bfloat16 training might be unstable. Please use it at your own risk.
+  - The different learning rates for each U-Net block are now supported in sdxl_train.py. Specify with `--block_lr` option. Specify 23 values separated by commas like `--block_lr 1e-3,1e-3 ... 1e-3`.
+    - 23 values correspond to `0: time/label embed, 1-9: input blocks 0-8, 10-12: mid blocks 0-2, 13-21: output blocks 0-8, 22: out`.
+- `prepare_buckets_latents.py` now supports SDXL fine-tuning.
+
+- `sdxl_train_network.py` is a script for LoRA training for SDXL. The usage is almost the same as `train_network.py`.
+
+- Both scripts has following additional options:
+  - `--cache_text_encoder_outputs` and `--cache_text_encoder_outputs_to_disk`: Cache the outputs of the text encoders. This option is useful to reduce the GPU memory usage. This option cannot be used with options for shuffling or dropping the captions.
+  - `--no_half_vae`: Disable the half-precision (mixed-precision) VAE. VAE for SDXL seems to produce NaNs in some cases. This option is useful to avoid the NaNs.
+
+- `--weighted_captions` option is not supported yet for both scripts.
+
+- `sdxl_train_textual_inversion.py` is a script for Textual Inversion training for SDXL. The usage is almost the same as `train_textual_inversion.py`.
+  - `--cache_text_encoder_outputs` is not supported.
+  - There are two options for captions:
+    1. Training with captions. All captions must include the token string. The token string is replaced with multiple tokens.
+    2. Use `--use_object_template` or `--use_style_template` option. The captions are generated from the template. The existing captions are ignored.
+  - See below for the format of the embeddings.
+
+- `--min_timestep` and `--max_timestep` options are added to each training script. These options can be used to train U-Net with different timesteps. The default values are 0 and 1000.
+
+### Utility scripts for SDXL
+
+- `tools/cache_latents.py` is added. This script can be used to cache the latents to disk in advance. 
+  - The options are almost the same as `sdxl_train.py'. See the help message for the usage.
+  - Please launch the script as follows:
+    `accelerate launch  --num_cpu_threads_per_process 1 tools/cache_latents.py ...`
+  - This script should work with multi-GPU, but it is not tested in my environment.
+
+- `tools/cache_text_encoder_outputs.py` is added. This script can be used to cache the text encoder outputs to disk in advance. 
+  - The options are almost the same as `cache_latents.py` and `sdxl_train.py`. See the help message for the usage.
+
+- `sdxl_gen_img.py` is added. This script can be used to generate images with SDXL, including LoRA, Textual Inversion and ControlNet-LLLite. See the help message for the usage.
+
+### Tips for SDXL training
+
+- The default resolution of SDXL is 1024x1024.
+- The fine-tuning can be done with 24GB GPU memory with the batch size of 1. For 24GB GPU, the following options are recommended __for the fine-tuning with 24GB GPU memory__:
+  - Train U-Net only.
+  - Use gradient checkpointing.
+  - Use `--cache_text_encoder_outputs` option and caching latents.
+  - Use Adafactor optimizer. RMSprop 8bit or Adagrad 8bit may work. AdamW 8bit doesn't seem to work.
+- The LoRA training can be done with 8GB GPU memory (10GB recommended). For reducing the GPU memory usage, the following options are recommended:
+  - Train U-Net only.
+  - Use gradient checkpointing.
+  - Use `--cache_text_encoder_outputs` option and caching latents.
+  - Use one of 8bit optimizers or Adafactor optimizer.
+  - Use lower dim (4 to 8 for 8GB GPU).
+- `--network_train_unet_only` option is highly recommended for SDXL LoRA. Because SDXL has two text encoders, the result of the training will be unexpected.
+- PyTorch 2 seems to use slightly less GPU memory than PyTorch 1.
+- `--bucket_reso_steps` can be set to 32 instead of the default value 64. Smaller values than 32 will not work for SDXL training.
+
+Example of the optimizer settings for Adafactor with the fixed learning rate:
+```toml
+optimizer_type = "adafactor"
+optimizer_args = [ "scale_parameter=False", "relative_step=False", "warmup_init=False" ]
+lr_scheduler = "constant_with_warmup"
+lr_warmup_steps = 100
+learning_rate = 4e-7 # SDXL original learning rate
+```
+
+### Format of Textual Inversion embeddings for SDXL
+
+```python
+from safetensors.torch import save_file
+
+state_dict = {"clip_g": embs_for_text_encoder_1280, "clip_l": embs_for_text_encoder_768}
+save_file(state_dict, file)
+```
+
+### ControlNet-LLLite
+
+ControlNet-LLLite, a novel method for ControlNet with SDXL, is added. See [documentation](./docs/train_lllite_README.md) for details.
+
+
 ## Change History

-### 15 Jun. 2023, 2023/06/15
+### Oct 11, 2023 / 2023/10/11
+- Fix to work `make_captions_by_git.py` with the latest version of transformers.
+- Improve `gen_img_diffusers.py` and `sdxl_gen_img.py`. Both scripts now support the following options:
+  - `--network_merge_n_models` option can be used to merge some of the models. The remaining models aren't merged, so the multiplier can be changed, and the regional LoRA also works.
+  - `--network_regional_mask_max_color_codes` is added. Now you can use up to 7 regions.
+    - When this option is specified, the mask of the regional LoRA is the color code based instead of the channel based. The value is the maximum number of the color codes (up to 7). 
+    - You can specify the mask for each LoRA by colors: 0x0000ff, 0x00ff00, 0x00ffff, 0xff0000, 0xff00ff, 0xffff00, 0xffffff.

- Prodigy optimizer is supported in each training script. It is a member of D-Adaptation and is effective for DyLoRA training. [PR #585](https://github.com/kohya-ss/sd-scripts/pull/585) Please see the PR for details. Thanks to sdbds!
-  - Install the package with `pip install prodigyopt`. Then specify the option like `--optimizer_type="prodigy"`.
- Arbitrary Dataset is supported in each training script (except XTI). You can use it by defining a Dataset class that returns images and captions.
-  - Prepare a Python script and define a class that inherits `train_util.MinimalDataset`. Then specify the option like `--dataset_class package.module.DatasetClass` in each training script.
-  - Please refer to `MinimalDataset` for implementation. I will prepare a sample later.
- The following features have been added to the generation script.
-  - Added an option `--highres_fix_disable_control_net` to disable ControlNet in the 2nd stage of Highres. Fix. Please try it if the image is disturbed by some ControlNet such as Canny.
-  - Added Variants similar to sd-dynamic-propmpts in the prompt.
-    - If you specify `{spring|summer|autumn|winter}`, one of them will be randomly selected.
-    - If you specify `{2$$chocolate|vanilla|strawberry}`, two of them will be randomly selected.
-    - If you specify `{1-2$$ and $$chocolate|vanilla|strawberry}`, one or two of them will be randomly selected and connected by ` and `.
-    - You can specify the number of candidates in the range `0-2`. You cannot omit one side like `-2` or `1-`.
-    - It can also be specified for the prompt option.
-    - If you specify `e` or `E`, all candidates will be selected and the prompt will be repeated multiple times (`--images_per_prompt` is ignored). It may be useful for creating X/Y plots.
-    - You can also specify `--am {e$$0.2|0.4|0.6|0.8|1.0},{e$$0.4|0.7|1.0} --d 1234`. In this case, 15 prompts will be generated with 5*3.
-    - There is no weighting function.
+- `make_captions_by_git.py` が最新の transformers で動作するように修正しました。
+- `gen_img_diffusers.py` と `sdxl_gen_img.py` を更新し、以下のオプションを追加しました。
+  - `--network_merge_n_models` オプションで一部のモデルのみマージできます。残りのモデルはマージされないため、重みを変更したり、領域別LoRAを使用したりできます。
+  - `--network_regional_mask_max_color_codes` を追加しました。最大7つの領域を使用できます。
+    - このオプションを指定すると、領域別LoRAのマスクはチャンネルベースではなくカラーコードベースになります。値はカラーコードの最大数（最大7）です。
+    - 各LoRAに対してマスクをカラーで指定できます：0x0000ff、0x00ff00、0x00ffff、0xff0000、0xff00ff、0xffff00、0xffffff。

- 各学習スクリプトでProdigyオプティマイザがサポートされました。D-Adaptationの仲間でDyLoRAの学習に有効とのことです。 [PR #585](https://github.com/kohya-ss/sd-scripts/pull/585)  詳細はPRをご覧ください。sdbds氏に感謝します。
-  - `pip install prodigyopt` としてパッケージをインストールしてください。また `--optimizer_type="prodigy"` のようにオプションを指定します。
- 各学習スクリプトで任意のDatasetをサポートしました（XTIを除く）。画像とキャプションを返すDatasetクラスを定義することで、学習スクリプトから利用できます。
-  - Pythonスクリプトを用意し、`train_util.MinimalDataset`を継承するクラスを定義してください。そして各学習スクリプトのオプションで `--dataset_class package.module.DatasetClass` のように指定してください。
-  - 実装方法は `MinimalDataset` を参考にしてください。のちほどサンプルを用意します。
- 生成スクリプトに以下の機能追加を行いました。
-  - Highres. Fixの2nd stageでControlNetを無効化するオプション `--highres_fix_disable_control_net` を追加しました。Canny等一部のControlNetで画像が乱れる場合にお試しください。
-  - プロンプトでsd-dynamic-propmptsに似たVariantをサポートしました。
-    - `{spring|summer|autumn|winter}` のように指定すると、いずれかがランダムに選択されます。
-    - `{2$$chocolate|vanilla|strawberry}` のように指定すると、いずれか2個がランダムに選択されます。
-    - `{1-2$$ and $$chocolate|vanilla|strawberry}` のように指定すると、1個か2個がランダムに選択され ` and ` で接続されます。
-    - 個数のレンジ指定では`0-2`のように0個も指定可能です。`-2`や`1-`のような片側の省略はできません。
-    - プロンプトオプションに対しても指定可能です。
-    - `{e$$chocolate|vanilla|strawberry}` のように`e`または`E`を指定すると、すべての候補が選択されプロンプトが複数回繰り返されます（`--images_per_prompt`は無視されます）。X/Y plotの作成に便利かもしれません。
-    - `--am {e$$0.2|0.4|0.6|0.8|1.0},{e$$0.4|0.7|1.0} --d 1234`のような指定も可能です。この場合、5*3で15回のプロンプトが生成されます。
-    - Weightingの機能はありません。
+### Oct 9. 2023 / 2023/10/9

-### 8 Jun. 2023, 2023/06/08
+- `tag_images_by_wd_14_tagger.py` now supports Onnx. If you use Onnx, TensorFlow is not required anymore. [#864](https://github.com/kohya-ss/sd-scripts/pull/864) Thanks to Isotr0py!
+  - `--onnx` option is added. If you use Onnx, specify `--onnx` option.
+  - Please install Onnx and other required packages. 
+    1. Uninstall TensorFlow.
+    1. `pip install tensorboard==2.14.1` This is required for the specified version of protobuf.
+    1. `pip install protobuf==3.20.3` This is required for Onnx.
+    1. `pip install onnx==1.14.1`
+    1. `pip install onnxruntime-gpu==1.16.0` or `pip install onnxruntime==1.16.0`
+- `--append_tags` option is added to `tag_images_by_wd_14_tagger.py`. This option appends the tags to the existing tags, instead of replacing them. [#858](https://github.com/kohya-ss/sd-scripts/pull/858) Thanks to a-l-e-x-d-s-9! 
+- [OFT](https://oft.wyliu.com/) is now supported.
+  - You can use `networks.oft` for the network module in `sdxl_train_network.py`.  The usage is the same as `networks.lora`. Some options are not supported.
+  - `sdxl_gen_img.py` also supports OFT as `--network_module`. 
+  - OFT only supports SDXL currently. Because current OFT tweaks Q/K/V and O in the transformer, and SD1/2 have extremely fewer transformers than SDXL.
+  - The implementation is heavily based on laksjdjf's [OFT implementation](https://github.com/laksjdjf/sd-trainer/blob/dev/networks/lora_modules.py). Thanks to laksjdjf!
+- Other bug fixes and improvements.

- Fixed a bug where clip skip did not work when training with weighted captions (`--weighted_captions` specified) and when generating sample images during training.
- 重みづけキャプションでの学習時（`--weighted_captions`指定時）および学習中のサンプル画像生成時にclip skipが機能しない不具合を修正しました。
+- `tag_images_by_wd_14_tagger.py` が Onnx をサポートしました。Onnx を使用する場合は TensorFlow は不要です。[#864](https://github.com/kohya-ss/sd-scripts/pull/864) Isotr0py氏に感謝します。
+  - Onnxを使用する場合は、`--onnx` オプションを指定してください。
+  - Onnx とその他の必要なパッケージをインストールしてください。
+    1. TensorFlow をアンインストールしてください。
+    1. `pip install tensorboard==2.14.1` protobufの指定バージョンにこれが必要。
+    1. `pip install protobuf==3.20.3` Onnxのために必要。
+    1. `pip install onnx==1.14.1`
+    1. `pip install onnxruntime-gpu==1.16.0` または `pip install onnxruntime==1.16.0`
+- `tag_images_by_wd_14_tagger.py` に `--append_tags` オプションが追加されました。このオプションを指定すると、既存のタグに上書きするのではなく、新しいタグのみが既存のタグに追加されます。 [#858](https://github.com/kohya-ss/sd-scripts/pull/858) a-l-e-x-d-s-9氏に感謝します。
+- [OFT](https://oft.wyliu.com/) をサポートしました。
+  - `sdxl_train_network.py` の`--network_module`に `networks.oft` を指定してください。使用方法は `networks.lora` と同様ですが一部のオプションは未サポートです。
+  - `sdxl_gen_img.py` でも同様に  OFT を指定できます。
+  - OFT は現在 SDXL のみサポートしています。OFT は現在 transformer の Q/K/V と O を変更しますが、SD1/2 は transformer の数が SDXL よりも極端に少ないためです。
+  - 実装は laksjdjf 氏の [OFT実装](https://github.com/laksjdjf/sd-trainer/blob/dev/networks/lora_modules.py) を多くの部分で参考にしています。laksjdjf 氏に感謝します。
+- その他のバグ修正と改善。

-### 6 Jun. 2023, 2023/06/06
+### Oct 1. 2023 / 2023/10/1

- Fix `train_network.py` to probably work with older versions of LyCORIS.
- `gen_img_diffusers.py` now supports `BREAK` syntax.
- `train_network.py`がLyCORISの以前のバージョンでも恐らく動作するよう修正しました。
- `gen_img_diffusers.py` で `BREAK` 構文をサポートしました。
+- SDXL training is now available in the main branch. The sdxl branch is merged into the main branch.

-### 3 Jun. 2023, 2023/06/03
+- [SAI Model Spec](https://github.com/Stability-AI/ModelSpec) metadata is now supported partially. `hash_sha256` is not supported yet.
+  - The main items are set automatically. 
+  - You can set title, author, description, license and tags with `--metadata_xxx` options in each training script.
+  - Merging scripts also support minimum SAI Model Spec metadata. See the help message for the usage.
+  - Metadata editor will be available soon.

- Max Norm Regularization is now available in `train_network.py`. [PR #545](https://github.com/kohya-ss/sd-scripts/pull/545) Thanks to AI-Casanova!
-  - Max Norm Regularization is a technique to stabilize network training by limiting the norm of network weights. It may be effective in suppressing overfitting of LoRA and improving stability when used with other LoRAs. See PR for details.
-  - Specify as `--scale_weight_norms=1.0`. It seems good to try from `1.0`.
-  - The networks other than LoRA in this repository (such as LyCORIS) do not support this option.
+- `bitsandbytes` is now optional. Please install it if you want to use it. The insructions are in the later section.

- Three types of dropout have been added to `train_network.py` and LoRA network.
-  - Dropout is a technique to suppress overfitting and improve network performance by randomly setting some of the network outputs to 0.
-  - `--network_dropout` is a normal dropout at the neuron level. In the case of LoRA, it is applied to the output of down. Proposed in [PR #545](https://github.com/kohya-ss/sd-scripts/pull/545) Thanks to AI-Casanova!
-    - `--network_dropout=0.1` specifies the dropout probability to `0.1`.
-    - Note that the specification method is different from LyCORIS.
-  - For LoRA network, `--network_args` can specify `rank_dropout` to dropout each rank with specified probability. Also `module_dropout` can be specified to dropout each module with specified probability.
-    - Specify as `--network_args "rank_dropout=0.2" "module_dropout=0.1"`.
-  - `--network_dropout`, `rank_dropout`, and `module_dropout` can be specified at the same time.
-  - Values of 0.1 to 0.3 may be good to try. Values greater than 0.5 should not be specified.
-  - `rank_dropout` and `module_dropout` are original techniques of this repository. Their effectiveness has not been verified yet.
-  - The networks other than LoRA in this repository (such as LyCORIS) do not support these options.
+- `albumentations` is not required anymore.

- Added an option `--scale_v_pred_loss_like_noise_pred` to scale v-prediction loss like noise prediction in each training script.
-  - By scaling the loss according to the time step, the weights of global noise prediction and local noise prediction become the same, and the improvement of details may be expected.
-  - See [this article](https://xrg.hatenablog.com/entry/2023/06/02/202418) by xrg for details (written in Japanese). Thanks to xrg for the great suggestion!
+- `--v_pred_like_loss ratio` option is added. This option adds the loss like v-prediction loss in SDXL training. `0.1` means that the loss is added 10% of the v-prediction loss. The default value is None (disabled).
+  - In v-prediction, the loss is higher in the early timesteps (near the noise). This option can be used to increase the loss in the early timesteps.

- Max Norm Regularizationが`train_network.py`で使えるようになりました。[PR #545](https://github.com/kohya-ss/sd-scripts/pull/545) AI-Casanova氏に感謝します。
-  - Max Norm Regularizationは、ネットワークの重みのノルムを制限することで、ネットワークの学習を安定させる手法です。LoRAの過学習の抑制、他のLoRAと併用した時の安定性の向上が期待できるかもしれません。詳細はPRを参照してください。
-  - `--scale_weight_norms=1.0`のように `--scale_weight_norms` で指定してください。`1.0`から試すと良いようです。
-  - LyCORIS等、当リポジトリ以外のネットワークは現時点では未対応です。
+- Arbitrary options can be used for Diffusers' schedulers. For example `--lr_scheduler_args "lr_end=1e-8"`.

- `train_network.py` およびLoRAに計三種類のdropoutを追加しました。
-  - dropoutはネットワークの一部の出力をランダムに0にすることで、過学習の抑制、ネットワークの性能向上等を図る手法です。
-  - `--network_dropout` はニューロン単位の通常のdropoutです。LoRAの場合、downの出力に対して適用されます。[PR #545](https://github.com/kohya-ss/sd-scripts/pull/545) で提案されました。AI-Casanova氏に感謝します。
-    - `--network_dropout=0.1` などとすることで、dropoutの確率を指定できます。
-    - LyCORISとは指定方法が異なりますのでご注意ください。
-  - LoRAの場合、`--network_args`に`rank_dropout`を指定することで各rankを指定確率でdropoutします。また同じくLoRAの場合、`--network_args`に`module_dropout`を指定することで各モジュールを指定確率でdropoutします。
-    - `--network_args "rank_dropout=0.2" "module_dropout=0.1"` のように指定します。
-  - `--network_dropout`、`rank_dropout` 、 `module_dropout` は同時に指定できます。
-  - それぞれの値は0.1~0.3程度から試してみると良いかもしれません。0.5を超える値は指定しない方が良いでしょう。
-  - `rank_dropout`および`module_dropout`は当リポジトリ独自の手法です。有効性の検証はまだ行っていません。
-  - これらのdropoutはLyCORIS等、当リポジトリ以外のネットワークは現時点では未対応です。
+- LoRA-FA is added experimentally. Specify `--network_module networks.lora_fa` option instead of `--network_module networks.lora`. The trained model can be used as a normal LoRA model.
+- JPEG XL is supported. [#786](https://github.com/kohya-ss/sd-scripts/pull/786) 
+- Input perturbation noise is added. See [#798](https://github.com/kohya-ss/sd-scripts/pull/798) for details.
+- Dataset subset now has `caption_prefix` and `caption_suffix` options. The strings are added to the beginning and the end of the captions before shuffling. You can specify the options in `.toml`.
+- Intel ARC support with IPEX is added. [#825](https://github.com/kohya-ss/sd-scripts/pull/825)
+- Other bug fixes and improvements.

- 各学習スクリプトにv-prediction lossをnoise predictionと同様の値にスケールするオプション`--scale_v_pred_loss_like_noise_pred`を追加しました。
-  - タイムステップに応じてlossをスケールすることで、 大域的なノイズの予測と局所的なノイズの予測の重みが同じになり、ディテールの改善が期待できるかもしれません。
-  - 詳細はxrg氏のこちらの記事をご参照ください：[noise_predictionモデルとv_predictionモデルの損失 - 勾配降下党青年局](https://xrg.hatenablog.com/entry/2023/06/02/202418) xrg氏の素晴らしい記事に感謝します。
-
-### 31 May 2023, 2023/05/31
-
- Show warning when image caption file does not exist during training. [PR #533](https://github.com/kohya-ss/sd-scripts/pull/533) Thanks to TingTingin!
-  - Warning is also displayed when using class+identifier dataset. Please ignore if it is intended.
- `train_network.py` now supports merging network weights before training. [PR #542](https://github.com/kohya-ss/sd-scripts/pull/542) Thanks to u-haru!
-  - `--base_weights` option specifies LoRA or other model files (multiple files are allowed) to merge.
-  - `--base_weights_multiplier` option specifies multiplier of the weights to merge (multiple values are allowed). If omitted or less than `base_weights`, 1.0 is used.
-  - This is useful for incremental learning. See PR for details.
- Show warning and continue training when uploading to HuggingFace fails.
-
- 学習時に画像のキャプションファイルが存在しない場合、警告が表示されるようになりました。 [PR #533](https://github.com/kohya-ss/sd-scripts/pull/533) TingTingin氏に感謝します。
-  - class+identifier方式のデータセットを利用している場合も警告が表示されます。意図している通りの場合は無視してください。
- `train_network.py` に学習前にモデルにnetworkの重みをマージする機能が追加されました。 [PR #542](https://github.com/kohya-ss/sd-scripts/pull/542) u-haru氏に感謝します。
-  - `--base_weights` オプションでLoRA等のモデルファイル（複数可）を指定すると、それらの重みをマージします。
-  - `--base_weights_multiplier` オプションでマージする重みの倍率（複数可）を指定できます。省略時または`base_weights`よりも数が少ない場合は1.0になります。
-  - 差分追加学習などにご利用ください。詳細はPRをご覧ください。
- HuggingFaceへのアップロードに失敗した場合、警告を表示しそのまま学習を続行するよう変更しました。
-
-### 25 May 2023, 2023/05/25
-
- [D-Adaptation v3.0](https://github.com/facebookresearch/dadaptation) is now supported. [PR #530](https://github.com/kohya-ss/sd-scripts/pull/530) Thanks to sdbds!
-  - `--optimizer_type` now accepts `DAdaptAdamPreprint`, `DAdaptAdanIP`, and `DAdaptLion`.
-  - `DAdaptAdam` is now new. The old `DAdaptAdam` is available with `DAdaptAdamPreprint`.
-  - Simply specifying `DAdaptation` will use `DAdaptAdamPreprint` (same behavior as before).
-  - You need to install D-Adaptation v3.0. After activating venv, please do `pip install -U dadaptation`.
-  - See PR and D-Adaptation documentation for details.
- [D-Adaptation v3.0](https://github.com/facebookresearch/dadaptation)がサポートされました。 [PR #530](https://github.com/kohya-ss/sd-scripts/pull/530)  sdbds氏に感謝します。
-  - `--optimizer_type`に`DAdaptAdamPreprint`、`DAdaptAdanIP`、`DAdaptLion` が追加されました。
-  - `DAdaptAdam`が新しくなりました。今までの`DAdaptAdam`は`DAdaptAdamPreprint`で使用できます。
-  - 単に `DAdaptation` を指定すると`DAdaptAdamPreprint`が使用されます（今までと同じ動き）。
-  - D-Adaptation v3.0のインストールが必要です。venvを有効にした後 `pip install -U dadaptation` としてください。
-  - 詳細はPRおよびD-Adaptationのドキュメントを参照してください。
-
-### 22 May 2023, 2023/05/22
-
- Fixed several bugs.
-  - The state is saved even when the `--save_state` option is not specified in `fine_tune.py` and `train_db.py`. [PR #521](https://github.com/kohya-ss/sd-scripts/pull/521) Thanks to akshaal!
-  - Cannot load LoRA without `alpha`. [PR #527](https://github.com/kohya-ss/sd-scripts/pull/527) Thanks to Manjiz!
-  - Minor changes to console output during sample generation. [PR #515](https://github.com/kohya-ss/sd-scripts/pull/515) Thanks to yanhuifair!
- The generation script now uses xformers for VAE as well.
- いくつかのバグ修正を行いました。
-  -  `fine_tune.py`と`train_db.py`で`--save_state`オプション未指定時にもstateが保存される。 [PR #521](https://github.com/kohya-ss/sd-scripts/pull/521) akshaal氏に感謝します。
-  - `alpha`を持たないLoRAを読み込めない。[PR #527](https://github.com/kohya-ss/sd-scripts/pull/527) Manjiz氏に感謝します。
-  - サンプル生成時のコンソール出力の軽微な変更。[PR #515](https://github.com/kohya-ss/sd-scripts/pull/515) yanhuifair氏に感謝します。
- 生成スクリプトでVAEについてもxformersを使うようにしました。
-
-### 16 May 2023, 2023/05/16
-
- Fixed an issue where an error would occur if the encoding of the prompt file was different from the default. [PR #510](https://github.com/kohya-ss/sd-scripts/pull/510) Thanks to sdbds!
-  - Please save the prompt file in UTF-8.
- プロンプトファイルのエンコーディングがデフォルトと異なる場合にエラーが発生する問題を修正しました。 [PR #510](https://github.com/kohya-ss/sd-scripts/pull/510) sdbds氏に感謝します。
-  - プロンプトファイルはUTF-8で保存してください。
-
-### 15 May 2023, 2023/05/15
-
- Added [English translation of documents](https://github.com/darkstorm2150/sd-scripts#links-to-usage-documentation) by darkstorm2150. Thank you very much!
- The prompt for sample generation during training can now be specified in `.toml` or `.json`. [PR #504](https://github.com/kohya-ss/sd-scripts/pull/504) Thanks to Linaqruf!
-  - For details on prompt description, please see the PR.
-
- darkstorm2150氏に[ドキュメント類を英訳](https://github.com/darkstorm2150/sd-scripts#links-to-usage-documentation)していただきました。ありがとうございます！
- 学習中のサンプル生成のプロンプトを`.toml`または`.json`で指定可能になりました。 [PR #504](https://github.com/kohya-ss/sd-scripts/pull/504) Linaqruf氏に感謝します。
-  - プロンプト記述の詳細は当該PRをご覧ください。
-
-### 11 May 2023, 2023/05/11
-
- Added an option `--dim_from_weights` to `train_network.py` to automatically determine the dim(rank) from the weight file. [PR #491](https://github.com/kohya-ss/sd-scripts/pull/491) Thanks to AI-Casanova!
-  - It is useful in combination with `resize_lora.py`. Please see the PR for details.
- Fixed a bug where the noise resolution was incorrect with Multires noise. [PR #489](https://github.com/kohya-ss/sd-scripts/pull/489) Thanks to sdbds!
-  - Please see the PR for details.
- The image generation scripts can now use img2img and highres fix at the same time.
- Fixed a bug where the hint image of ControlNet was incorrectly BGR instead of RGB in the image generation scripts.
- Added a feature to the image generation scripts to use the memory-efficient VAE.
-  - If you specify a number with the `--vae_slices` option, the memory-efficient VAE will be used. The maximum output size will be larger, but it will be slower. Please specify a value of about `16` or `32`.
-  - The implementation of the VAE is in `library/slicing_vae.py`.
-
- `train_network.py`にdim(rank)を重みファイルから自動決定するオプション`--dim_from_weights`が追加されました。 [PR #491](https://github.com/kohya-ss/sd-scripts/pull/491) AI-Casanova氏に感謝します。
-  - `resize_lora.py`と組み合わせると有用です。詳細はPRもご参照ください。
- Multires noiseでノイズ解像度が正しくない不具合が修正されました。 [PR #489](https://github.com/kohya-ss/sd-scripts/pull/489)  sdbds氏に感謝します。
-  - 詳細は当該PRをご参照ください。
- 生成スクリプトでimg2imgとhighres fixを同時に使用できるようにしました。
- 生成スクリプトでControlNetのhint画像が誤ってBGRだったのをRGBに修正しました。
- 生成スクリプトで省メモリ化VAEを使えるよう機能追加しました。
-  - `--vae_slices`オプションに数値を指定すると、省メモリ化VAEを用います。出力可能な最大サイズが大きくなりますが、遅くなります。`16`または`32`程度の値を指定してください。
-  - VAEの実装は`library/slicing_vae.py`にあります。
-
-### 7 May 2023, 2023/05/07
-
- The documentation has been moved to the `docs` folder. If you have links, please change them.
- Removed `gradio` from `requirements.txt`.
- DAdaptAdaGrad, DAdaptAdan, and DAdaptSGD are now supported by DAdaptation. [PR#455](https://github.com/kohya-ss/sd-scripts/pull/455) Thanks to sdbds!
-  - DAdaptation needs to be installed. Also, depending on the optimizer, DAdaptation may need to be updated. Please update with `pip install --upgrade dadaptation`.
- Added support for pre-calculation of LoRA weights in image generation scripts. Specify `--network_pre_calc`.
-  - The prompt option `--am` is available. Also, it is disabled when Regional LoRA is used.
- Added Adaptive noise scale to each training script. Specify a number with `--adaptive_noise_scale` to enable it.
-  - __Experimental option. It may be removed or changed in the future.__
-  - This is an original implementation that automatically adjusts the value of the noise offset according to the absolute value of the mean of each channel of the latents. It is expected that appropriate noise offsets will be set for bright and dark images, respectively.
-  - Specify it together with `--noise_offset`.
-  - The actual value of the noise offset is calculated as `noise_offset + abs(mean(latents, dim=(2,3))) * adaptive_noise_scale`. Since the latent is close to a normal distribution, it may be a good idea to specify a value of about 1/10 to the same as the noise offset.
-  - Negative values can also be specified, in which case the noise offset will be clipped to 0 or more.
- Other minor fixes.
-
- ドキュメントを`docs`フォルダに移動しました。リンク等を張られている場合は変更をお願いいたします。
- `requirements.txt`から`gradio`を削除しました。
- DAdaptationで新しくDAdaptAdaGrad、DAdaptAdan、DAdaptSGDがサポートされました。[PR#455](https://github.com/kohya-ss/sd-scripts/pull/455) sdbds氏に感謝します。
-  - dadaptationのインストールが必要です。またオプティマイザによってはdadaptationの更新が必要です。`pip install --upgrade dadaptation`で更新してください。
- 画像生成スクリプトでLoRAの重みの事前計算をサポートしました。`--network_pre_calc`を指定してください。
-  - プロンプトオプションの`--am`が利用できます。またRegional LoRA使用時には無効になります。
- 各学習スクリプトにAdaptive noise scaleを追加しました。`--adaptive_noise_scale`で数値を指定すると有効になります。
-  - __実験的オプションです。将来的に削除、仕様変更される可能性があります。__
-  - Noise offsetの値を、latentsの各チャネルの平均値の絶対値に応じて自動調整するオプションです。独自の実装で、明るい画像、暗い画像に対してそれぞれ適切なnoise offsetが設定されることが期待されます。
-  - `--noise_offset` と同時に指定してください。
-  - 実際のNoise offsetの値は `noise_offset + abs(mean(latents, dim=(2,3))) * adaptive_noise_scale` で計算されます。 latentは正規分布に近いためnoise_offsetの1/10～同程度の値を指定するとよいかもしれません。
-  - 負の値も指定でき、その場合はnoise offsetは0以上にclipされます。
- その他の細かい修正を行いました。

 Please read [Releases](https://github.com/kohya-ss/sd-scripts/releases) for recent updates.
 最近の更新情報は [Release](https://github.com/kohya-ss/sd-scripts/releases) をご覧ください。
--- a/XTI_hijack.py
+++ b/XTI_hijack.py
@@ -1,4 +1,11 @@
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from typing import Union, List, Optional, Dict, Any, Tuple
 from diffusers.models.unet_2d_condition import UNet2DConditionOutput

--- a/_typos.toml
+++ b/_typos.toml
@@ -9,7 +9,25 @@ parms="parms"
 nin="nin"
 extention="extention" # Intentionally left
 nd="nd"
+shs="shs"
+sts="sts"
+scs="scs"
+cpc="cpc"
+coc="coc"
+cic="cic"
+msm="msm"
+usu="usu"
+ici="ici"
+lvl="lvl"
+dii="dii"
+muk="muk"
+ori="ori"
+hru="hru"
+rik="rik"
+koo="koo"
+yos="yos"
+wn="wn"


 [files]
-extend-exclude = ["_typos.toml"]
+extend-exclude = ["_typos.toml", "venv"]
--- a/docs/train_network_README-ja.md
+++ b/docs/train_network_README-ja.md
@@ -181,6 +181,8 @@ python networks\extract_lora_from_dylora.py --model "foldername/dylora-model.saf

 詳細は[PR #355](https://github.com/kohya-ss/sd-scripts/pull/355) をご覧ください。

+SDXLは現在サポートしていません。
+
 フルモデルの25個のブロックの重みを指定できます。最初のブロックに該当するLoRAは存在しませんが、階層別LoRA適用等との互換性のために25個としています。またconv2d3x3に拡張しない場合も一部のブロックにはLoRAが存在しませんが、記述を統一するため常に25個の値を指定してください。

 `--network_args` で以下の引数を指定してください。
@@ -246,6 +248,8 @@ network_args = [ "block_dims=2,4,4,4,8,8,8,8,12,12,12,12,16,12,12,12,12,8,8,8,8,

 merge_lora.pyでStable DiffusionのモデルにLoRAの学習結果をマージしたり、複数のLoRAモデルをマージしたりできます。

+SDXL向けにはsdxl_merge_lora.pyを用意しています。オプション等は同一ですので、以下のmerge_lora.pyを読み替えてください。
+
 ### Stable DiffusionのモデルにLoRAのモデルをマージする

 マージ後のモデルは通常のStable Diffusionのckptと同様に扱えます。たとえば以下のようなコマンドラインになります。
@@ -276,29 +280,29 @@ python networks\merge_lora.py --sd_model ..\model\model.ckpt

 ### 複数のLoRAのモデルをマージする

-__複数のLoRAをマージする場合は原則として `svd_merge_lora.py` を使用してください。__ 単純なup同士やdown同士のマージでは、計算結果が正しくなくなるためです。
-
-`merge_lora.py` によるマージは差分抽出法でLoRAを生成する場合等、ごく限られた場合でのみ有効です。
+--concatオプションを指定すると、複数のLoRAを単純に結合して新しいLoRAモデルを作成できます。ファイルサイズ（およびdim/rank）は指定したLoRAの合計サイズになります（マージ時にdim (rank)を変更する場合は `svd_merge_lora.py` を使用してください）。

 たとえば以下のようなコマンドラインになります。

 ```
-python networks\merge_lora.py 
+python networks\merge_lora.py --save_precision bf16 
    --save_to ..\lora_train1\model-char1-style1-merged.safetensors 
-    --models ..\lora_train1\last.safetensors ..\lora_train2\last.safetensors --ratios 0.6 0.4
+    --models ..\lora_train1\last.safetensors ..\lora_train2\last.safetensors 
+    --ratios 1.0 -1.0 --concat --shuffle
 ```

--sd_modelオプションは指定不要です。
+--concatオプションを指定します。
+
+また--shuffleオプションを追加し、重みをシャッフルします。シャッフルしないとマージ後のLoRAから元のLoRAを取り出せるため、コピー機学習などの場合には学習元データが明らかになります。ご注意ください。

 --save_toオプションにマージ後のLoRAモデルの保存先を指定します（.ckptまたは.safetensors、拡張子で自動判定）。

 --modelsに学習したLoRAのモデルファイルを指定します。三つ以上も指定可能です。

--ratiosにそれぞれのモデルの比率（どのくらい重みを元モデルに反映するか）を0~1.0の数値で指定します。二つのモデルを一対一でマージす場合は、「0.5 0.5」になります。「1.0 1.0」では合計の重みが大きくなりすぎて、恐らく結果はあまり望ましくないものになると思われます。
+--ratiosにそれぞれのモデルの比率（どのくらい重みを元モデルに反映するか）を0~1.0の数値で指定します。二つのモデルを一対一でマージする場合は、「0.5 0.5」になります。「1.0 1.0」では合計の重みが大きくなりすぎて、恐らく結果はあまり望ましくないものになると思われます。

 v1で学習したLoRAとv2で学習したLoRA、rank（次元数）の異なるLoRAはマージできません。U-NetだけのLoRAとU-Net+Text EncoderのLoRAはマージできるはずですが、結果は未知数です。

-
 ### その他のオプション

 * precision
@@ -306,6 +310,7 @@ v1で学習したLoRAとv2で学習したLoRA、rank（次元数）の異なるL
 * save_precision
  * モデル保存時の精度をfloat、fp16、bf16から指定できます。省略時はprecisionと同じ精度になります。

+他にもいくつかのオプションがありますので、--helpで確認してください。

 ## 複数のrankが異なるLoRAのモデルをマージする

--- a/fine_tune.py
+++ b/fine_tune.py
@@ -10,6 +10,13 @@ import toml

 from tqdm import tqdm
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from accelerate.utils import set_seed
 from diffusers import DDPMScheduler

@@ -73,8 +80,8 @@ def train(args):

    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    if args.debug_dataset:
        train_util.debug_dataset(train_dataset_group)
@@ -201,7 +208,7 @@ def train(args):
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
--- a/finetune/make_captions_by_git.py
+++ b/finetune/make_captions_by_git.py
@@ -52,6 +52,9 @@ def collate_fn_remove_corrupted(batch):


 def main(args):
+    r"""
+    transformers 4.30.2で、バッチサイズ>1でも動くようになったので、以下コメントアウト
+
    # GITにバッチサイズが1より大きくても動くようにパッチを当てる: transformers 4.26.0用
    org_prepare_input_ids_for_generation = GenerationMixin._prepare_input_ids_for_generation
    curr_batch_size = [args.batch_size]  # ループの最後で件数がbatch_size未満になるので入れ替えられるように
@@ -65,6 +68,7 @@ def main(args):
        return input_ids

    GenerationMixin._prepare_input_ids_for_generation = _prepare_input_ids_for_generation_patch
+    """

    print(f"load images from {args.train_data_dir}")
    train_data_dir_path = Path(args.train_data_dir)
@@ -81,7 +85,7 @@ def main(args):
    def run_batch(path_imgs):
        imgs = [im for _, im in path_imgs]

-        curr_batch_size[0] = len(path_imgs)
+        # curr_batch_size[0] = len(path_imgs)
        inputs = git_processor(images=imgs, return_tensors="pt").to(DEVICE)  # 画像はpil形式
        generated_ids = git_model.generate(pixel_values=inputs.pixel_values, max_length=args.max_length)
        captions = git_processor.batch_decode(generated_ids, skip_special_tokens=True)
--- a/finetune/tag_images_by_wd14_tagger.py
+++ b/finetune/tag_images_by_wd14_tagger.py
@@ -1,17 +1,15 @@
 import argparse
 import csv
-import glob
 import os
-
-from PIL import Image
-import cv2
-from tqdm import tqdm
-import numpy as np
-from tensorflow.keras.models import load_model
-from huggingface_hub import hf_hub_download
-import torch
 from pathlib import Path

+import cv2
+import numpy as np
+import torch
+from huggingface_hub import hf_hub_download
+from PIL import Image
+from tqdm import tqdm
+
 import library.train_util as train_util

 # from wd14 tagger
@@ -20,6 +18,7 @@ IMAGE_SIZE = 448
 # wd-v1-4-swinv2-tagger-v2 / wd-v1-4-vit-tagger / wd-v1-4-vit-tagger-v2/ wd-v1-4-convnext-tagger / wd-v1-4-convnext-tagger-v2
 DEFAULT_WD14_TAGGER_REPO = "SmilingWolf/wd-v1-4-convnext-tagger-v2"
 FILES = ["keras_metadata.pb", "saved_model.pb", "selected_tags.csv"]
+FILES_ONNX = ["model.onnx"]
 SUB_DIR = "variables"
 SUB_DIR_FILES = ["variables.data-00000-of-00001", "variables.index"]
 CSV_FILE = FILES[-1]
@@ -81,7 +80,10 @@ def main(args):
    # https://github.com/toriato/stable-diffusion-webui-wd14-tagger/issues/22
    if not os.path.exists(args.model_dir) or args.force_download:
        print(f"downloading wd14 tagger model from hf_hub. id: {args.repo_id}")
-        for file in FILES:
+        files = FILES
+        if args.onnx:
+            files += FILES_ONNX
+        for file in files:
            hf_hub_download(args.repo_id, file, cache_dir=args.model_dir, force_download=True, force_filename=file)
        for file in SUB_DIR_FILES:
            hf_hub_download(
@@ -96,7 +98,46 @@ def main(args):
        print("using existing wd14 tagger model")

    # 画像を読み込む
-    model = load_model(args.model_dir)
+    if args.onnx:
+        import onnx
+        import onnxruntime as ort
+
+        onnx_path = f"{args.model_dir}/model.onnx"
+        print("Running wd14 tagger with onnx")
+        print(f"loading onnx model: {onnx_path}")
+
+        if not os.path.exists(onnx_path):
+            raise Exception(
+                f"onnx model not found: {onnx_path}, please redownload the model with --force_download"
+                + " / onnxモデルが見つかりませんでした。--force_downloadで再ダウンロードしてください"
+            )
+
+        model = onnx.load(onnx_path)
+        input_name = model.graph.input[0].name
+        try:
+            batch_size = model.graph.input[0].type.tensor_type.shape.dim[0].dim_value
+        except:
+            batch_size = model.graph.input[0].type.tensor_type.shape.dim[0].dim_param
+
+        if args.batch_size != batch_size and type(batch_size) != str:
+            # some rebatch model may use 'N' as dynamic axes
+            print(
+                f"Batch size {args.batch_size} doesn't match onnx model batch size {batch_size}, use model batch size {batch_size}"
+            )
+            args.batch_size = batch_size
+
+        del model
+
+        ort_sess = ort.InferenceSession(
+            onnx_path,
+            providers=["CUDAExecutionProvider"]
+            if "CUDAExecutionProvider" in ort.get_available_providers()
+            else ["CPUExecutionProvider"],
+        )
+    else:
+        from tensorflow.keras.models import load_model
+
+        model = load_model(f"{args.model_dir}")

    # label_names = pd.read_csv("2022_0000_0899_6549/selected_tags.csv")
    # 依存ライブラリを増やしたくないので自力で読むよ
@@ -124,8 +165,14 @@ def main(args):
    def run_batch(path_imgs):
        imgs = np.array([im for _, im in path_imgs])

-        probs = model(imgs, training=False)
-        probs = probs.numpy()
+        if args.onnx:
+            if len(imgs) < args.batch_size:
+                imgs = np.concatenate([imgs, np.zeros((args.batch_size - len(imgs), IMAGE_SIZE, IMAGE_SIZE, 3))], axis=0)
+            probs = ort_sess.run(None, {input_name: imgs})[0]  # onnx output numpy
+            probs = probs[: len(path_imgs)]
+        else:
+            probs = model(imgs, training=False)
+            probs = probs.numpy()

        for (image_path, _), prob in zip(path_imgs, probs):
            # 最初の4つはratingなので無視する
@@ -165,9 +212,27 @@ def main(args):
            if len(character_tag_text) > 0:
                character_tag_text = character_tag_text[2:]

+            caption_file = os.path.splitext(image_path)[0] + args.caption_extension
+
            tag_text = ", ".join(combined_tags)

-            with open(os.path.splitext(image_path)[0] + args.caption_extension, "wt", encoding="utf-8") as f:
+            if args.append_tags:
+                # Check if file exists
+                if os.path.exists(caption_file):
+                    with open(caption_file, "rt", encoding="utf-8") as f:
+                        # Read file and remove new lines
+                        existing_content = f.read().strip("\n")  # Remove newlines
+
+                    # Split the content into tags and store them in a list
+                    existing_tags = [tag.strip() for tag in existing_content.split(",") if tag.strip()]
+
+                    # Check and remove repeating tags in tag_text
+                    new_tags = [tag for tag in combined_tags if tag not in existing_tags]
+
+                    # Create new tag_text
+                    tag_text = ", ".join(existing_tags + new_tags)
+
+            with open(caption_file, "wt", encoding="utf-8") as f:
                f.write(tag_text + "\n")
                if args.debug:
                    print(f"\n{image_path}:\n  Character tags: {character_tag_text}\n  General tags: {general_tag_text}")
@@ -283,12 +348,15 @@ def setup_parser() -> argparse.ArgumentParser:
        help="comma-separated list of undesired tags to remove from the output / 出力から除外したいタグのカンマ区切りのリスト",
    )
    parser.add_argument("--frequency_tags", action="store_true", help="Show frequency of tags for images / 画像ごとのタグの出現頻度を表示する")
+    parser.add_argument("--onnx", action="store_true", help="use onnx model for inference / onnxモデルを推論に使用する")
+    parser.add_argument("--append_tags", action="store_true", help="Append captions instead of overwriting / 上書きではなくキャプションを追記する")

    return parser

+
 if __name__ == "__main__":
    parser = setup_parser()
-    
+
    args = parser.parse_args()

    # スペルミスしていたオプションを復元する
--- a/gen_img_diffusers.py
+++ b/gen_img_diffusers.py
@@ -65,6 +65,16 @@ import re
 import diffusers
 import numpy as np
 import torch
+
+try:
+    import intel_extension_for_pytorch as ipex
+
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+
+        ipex_init()
+except Exception:
+    pass
 import torchvision
 from diffusers import (
    AutoencoderKL,
@@ -947,7 +957,7 @@ class PipelineLike:
            text_emb_last = torch.stack(text_emb_last)
        else:
            text_emb_last = text_embeddings
-            
+
        for i, t in enumerate(tqdm(timesteps)):
            # expand the latents if we are doing classifier free guidance
            latent_model_input = latents.repeat((num_latent_input, 1, 1, 1))
@@ -2356,12 +2366,19 @@ def main(args):
        network_default_muls = []
        network_pre_calc = args.network_pre_calc

+        # merge関連の引数を統合する
+        if args.network_merge:
+            network_merge = len(args.network_module)  # all networks are merged
+        elif args.network_merge_n_models:
+            network_merge = args.network_merge_n_models
+        else:
+            network_merge = 0
+
        for i, network_module in enumerate(args.network_module):
            print("import network module:", network_module)
            imported_module = importlib.import_module(network_module)

            network_mul = 1.0 if args.network_mul is None or len(args.network_mul) <= i else args.network_mul[i]
-            network_default_muls.append(network_mul)

            net_kwargs = {}
            if args.network_args and i < len(args.network_args):
@@ -2372,31 +2389,32 @@ def main(args):
                    key, value = net_arg.split("=")
                    net_kwargs[key] = value

-            if args.network_weights and i < len(args.network_weights):
-                network_weight = args.network_weights[i]
-                print("load network weights from:", network_weight)
-
-                if model_util.is_safetensors(network_weight) and args.network_show_meta:
-                    from safetensors.torch import safe_open
-
-                    with safe_open(network_weight, framework="pt") as f:
-                        metadata = f.metadata()
-                    if metadata is not None:
-                        print(f"metadata for: {network_weight}: {metadata}")
-
-                network, weights_sd = imported_module.create_network_from_weights(
-                    network_mul, network_weight, vae, text_encoder, unet, for_inference=True, **net_kwargs
-                )
-            else:
+            if args.network_weights is None or len(args.network_weights) <= i:
                raise ValueError("No weight. Weight is required.")
+
+            network_weight = args.network_weights[i]
+            print("load network weights from:", network_weight)
+
+            if model_util.is_safetensors(network_weight) and args.network_show_meta:
+                from safetensors.torch import safe_open
+
+                with safe_open(network_weight, framework="pt") as f:
+                    metadata = f.metadata()
+                if metadata is not None:
+                    print(f"metadata for: {network_weight}: {metadata}")
+
+            network, weights_sd = imported_module.create_network_from_weights(
+                network_mul, network_weight, vae, text_encoder, unet, for_inference=True, **net_kwargs
+            )
            if network is None:
                return

            mergeable = network.is_mergeable()
-            if args.network_merge and not mergeable:
+            if network_merge and not mergeable:
                print("network is not mergiable. ignore merge option.")

-            if not args.network_merge or not mergeable:
+            if not mergeable or i >= network_merge:
+                # not merging
                network.apply_to(text_encoder, unet)
                info = network.load_state_dict(weights_sd, False)  # network.load_weightsを使うようにするとよい
                print(f"weights are loaded: {info}")
@@ -2410,6 +2428,7 @@ def main(args):
                    network.backup_weights()

                networks.append(network)
+                network_default_muls.append(network_mul)
            else:
                network.merge_to(text_encoder, unet, weights_sd, dtype, device)

@@ -2705,9 +2724,18 @@ def main(args):

        size = None
        for i, network in enumerate(networks):
-            if i < 3:
+            if (i < 3 and args.network_regional_mask_max_color_codes is None) or i < args.network_regional_mask_max_color_codes:
                np_mask = np.array(mask_images[0])
-                np_mask = np_mask[:, :, i]
+
+                if args.network_regional_mask_max_color_codes:
+                    # カラーコードでマスクを指定する
+                    ch0 = (i + 1) & 1
+                    ch1 = ((i + 1) >> 1) & 1
+                    ch2 = ((i + 1) >> 2) & 1
+                    np_mask = np.all(np_mask == np.array([ch0, ch1, ch2]) * 255, axis=2)
+                    np_mask = np_mask.astype(np.uint8) * 255
+                else:
+                    np_mask = np_mask[:, :, i]
                size = np_mask.shape
            else:
                np_mask = np.full(size, 255, dtype=np.uint8)
@@ -3145,7 +3173,7 @@ def main(args):
                            print("predefined seeds are exhausted")
                            seed = None
                    elif args.iter_same_seed:
-                        seeds = iter_seed
+                        seed = iter_seed
                    else:
                        seed = None  # 前のを消す

@@ -3357,13 +3385,22 @@ def setup_parser() -> argparse.ArgumentParser:
    )
    parser.add_argument("--network_mul", type=float, default=None, nargs="*", help="additional network multiplier / 追加ネットワークの効果の倍率")
    parser.add_argument(
-        "--network_args", type=str, default=None, nargs="*", help="additional argmuments for network (key=value) / ネットワークへの追加の引数"
+        "--network_args", type=str, default=None, nargs="*", help="additional arguments for network (key=value) / ネットワークへの追加の引数"
    )
    parser.add_argument("--network_show_meta", action="store_true", help="show metadata of network model / ネットワークモデルのメタデータを表示する")
+    parser.add_argument(
+        "--network_merge_n_models", type=int, default=None, help="merge this number of networks / この数だけネットワークをマージする"
+    )
    parser.add_argument("--network_merge", action="store_true", help="merge network weights to original model / ネットワークの重みをマージする")
    parser.add_argument(
        "--network_pre_calc", action="store_true", help="pre-calculate network for generation / ネットワークのあらかじめ計算して生成する"
    )
+    parser.add_argument(
+        "--network_regional_mask_max_color_codes",
+        type=int,
+        default=None,
+        help="max color codes for regional mask (default is None, mask by channel) / regional maskの最大色数（デフォルトはNoneでチャンネルごとのマスク）",
+    )
    parser.add_argument(
        "--textual_inversion_embeddings",
        type=str,
@@ -3383,7 +3420,7 @@ def setup_parser() -> argparse.ArgumentParser:
        "--max_embeddings_multiples",
        type=int,
        default=None,
-        help="max embeding multiples, max token length is 75 * multiples / トークン長をデフォルトの何倍とするか 75*この値 がトークン長となる",
+        help="max embedding multiples, max token length is 75 * multiples / トークン長をデフォルトの何倍とするか 75*この値 がトークン長となる",
    )
    parser.add_argument(
        "--clip_guidance_scale",
@@ -3442,7 +3479,7 @@ def setup_parser() -> argparse.ArgumentParser:
        "--highres_fix_upscaler_args",
        type=str,
        default=None,
-        help="additional argmuments for upscaler (key=value) / upscalerへの追加の引数",
+        help="additional arguments for upscaler (key=value) / upscalerへの追加の引数",
    )
    parser.add_argument(
        "--highres_fix_disable_control_net",
--- a/library/config_util.py
+++ b/library/config_util.py
@@ -547,17 +547,13 @@ def generate_controlnet_subsets_config_by_subdirs(train_data_dir: Optional[str]
      return []

    subsets_config = []
-    for subdir in base_dir.iterdir():
-      if not subdir.is_dir():
-        continue
-
-      subset_config = {"image_dir": str(subdir), "conditioning_data_dir": conditioning_data_dir, "caption_extension": caption_extension, "num_repeats": 1}
-      subsets_config.append(subset_config)
+    subset_config = {"image_dir": train_data_dir, "conditioning_data_dir": conditioning_data_dir, "caption_extension": caption_extension, "num_repeats": 1}
+    subsets_config.append(subset_config)

    return subsets_config

  subsets_config = []
-  subsets_config += generate(train_data_dir, False)
+  subsets_config += generate(train_data_dir)

  return subsets_config

--- a/library/ipex/init.py
+++ b/library/ipex/init.py
@@ -0,0 +1,175 @@
+import os
+import sys
+import contextlib
+import torch
+import intel_extension_for_pytorch as ipex # pylint: disable=import-error, unused-import
+from .hijacks import ipex_hijacks
+from .attention import attention_init
+
+# pylint: disable=protected-access, missing-function-docstring, line-too-long
+
+def ipex_init(): # pylint: disable=too-many-statements
+    try:
+        #Replace cuda with xpu:
+        torch.cuda.current_device = torch.xpu.current_device
+        torch.cuda.current_stream = torch.xpu.current_stream
+        torch.cuda.device = torch.xpu.device
+        torch.cuda.device_count = torch.xpu.device_count
+        torch.cuda.device_of = torch.xpu.device_of
+        torch.cuda.get_device_name = torch.xpu.get_device_name
+        torch.cuda.get_device_properties = torch.xpu.get_device_properties
+        torch.cuda.init = torch.xpu.init
+        torch.cuda.is_available = torch.xpu.is_available
+        torch.cuda.is_initialized = torch.xpu.is_initialized
+        torch.cuda.is_current_stream_capturing = lambda: False
+        torch.cuda.set_device = torch.xpu.set_device
+        torch.cuda.stream = torch.xpu.stream
+        torch.cuda.synchronize = torch.xpu.synchronize
+        torch.cuda.Event = torch.xpu.Event
+        torch.cuda.Stream = torch.xpu.Stream
+        torch.cuda.FloatTensor = torch.xpu.FloatTensor
+        torch.Tensor.cuda = torch.Tensor.xpu
+        torch.Tensor.is_cuda = torch.Tensor.is_xpu
+        torch.cuda._initialization_lock = torch.xpu.lazy_init._initialization_lock
+        torch.cuda._initialized = torch.xpu.lazy_init._initialized
+        torch.cuda._lazy_seed_tracker = torch.xpu.lazy_init._lazy_seed_tracker
+        torch.cuda._queued_calls = torch.xpu.lazy_init._queued_calls
+        torch.cuda._tls = torch.xpu.lazy_init._tls
+        torch.cuda.threading = torch.xpu.lazy_init.threading
+        torch.cuda.traceback = torch.xpu.lazy_init.traceback
+        torch.cuda.Optional = torch.xpu.Optional
+        torch.cuda.__cached__ = torch.xpu.__cached__
+        torch.cuda.__loader__ = torch.xpu.__loader__
+        torch.cuda.ComplexFloatStorage = torch.xpu.ComplexFloatStorage
+        torch.cuda.Tuple = torch.xpu.Tuple
+        torch.cuda.streams = torch.xpu.streams
+        torch.cuda._lazy_new = torch.xpu._lazy_new
+        torch.cuda.FloatStorage = torch.xpu.FloatStorage
+        torch.cuda.Any = torch.xpu.Any
+        torch.cuda.__doc__ = torch.xpu.__doc__
+        torch.cuda.default_generators = torch.xpu.default_generators
+        torch.cuda.HalfTensor = torch.xpu.HalfTensor
+        torch.cuda._get_device_index = torch.xpu._get_device_index
+        torch.cuda.__path__ = torch.xpu.__path__
+        torch.cuda.Device = torch.xpu.Device
+        torch.cuda.IntTensor = torch.xpu.IntTensor
+        torch.cuda.ByteStorage = torch.xpu.ByteStorage
+        torch.cuda.set_stream = torch.xpu.set_stream
+        torch.cuda.BoolStorage = torch.xpu.BoolStorage
+        torch.cuda.os = torch.xpu.os
+        torch.cuda.torch = torch.xpu.torch
+        torch.cuda.BFloat16Storage = torch.xpu.BFloat16Storage
+        torch.cuda.Union = torch.xpu.Union
+        torch.cuda.DoubleTensor = torch.xpu.DoubleTensor
+        torch.cuda.ShortTensor = torch.xpu.ShortTensor
+        torch.cuda.LongTensor = torch.xpu.LongTensor
+        torch.cuda.IntStorage = torch.xpu.IntStorage
+        torch.cuda.LongStorage = torch.xpu.LongStorage
+        torch.cuda.__annotations__ = torch.xpu.__annotations__
+        torch.cuda.__package__ = torch.xpu.__package__
+        torch.cuda.__builtins__ = torch.xpu.__builtins__
+        torch.cuda.CharTensor = torch.xpu.CharTensor
+        torch.cuda.List = torch.xpu.List
+        torch.cuda._lazy_init = torch.xpu._lazy_init
+        torch.cuda.BFloat16Tensor = torch.xpu.BFloat16Tensor
+        torch.cuda.DoubleStorage = torch.xpu.DoubleStorage
+        torch.cuda.ByteTensor = torch.xpu.ByteTensor
+        torch.cuda.StreamContext = torch.xpu.StreamContext
+        torch.cuda.ComplexDoubleStorage = torch.xpu.ComplexDoubleStorage
+        torch.cuda.ShortStorage = torch.xpu.ShortStorage
+        torch.cuda._lazy_call = torch.xpu._lazy_call
+        torch.cuda.HalfStorage = torch.xpu.HalfStorage
+        torch.cuda.random = torch.xpu.random
+        torch.cuda._device = torch.xpu._device
+        torch.cuda.classproperty = torch.xpu.classproperty
+        torch.cuda.__name__ = torch.xpu.__name__
+        torch.cuda._device_t = torch.xpu._device_t
+        torch.cuda.warnings = torch.xpu.warnings
+        torch.cuda.__spec__ = torch.xpu.__spec__
+        torch.cuda.BoolTensor = torch.xpu.BoolTensor
+        torch.cuda.CharStorage = torch.xpu.CharStorage
+        torch.cuda.__file__ = torch.xpu.__file__
+        torch.cuda._is_in_bad_fork = torch.xpu.lazy_init._is_in_bad_fork
+        #torch.cuda.is_current_stream_capturing = torch.xpu.is_current_stream_capturing
+
+        #Memory:
+        torch.cuda.memory = torch.xpu.memory
+        if 'linux' in sys.platform and "WSL2" in os.popen("uname -a").read():
+            torch.xpu.empty_cache = lambda: None
+        torch.cuda.empty_cache = torch.xpu.empty_cache
+        torch.cuda.memory_stats = torch.xpu.memory_stats
+        torch.cuda.memory_summary = torch.xpu.memory_summary
+        torch.cuda.memory_snapshot = torch.xpu.memory_snapshot
+        torch.cuda.memory_allocated = torch.xpu.memory_allocated
+        torch.cuda.max_memory_allocated = torch.xpu.max_memory_allocated
+        torch.cuda.memory_reserved = torch.xpu.memory_reserved
+        torch.cuda.memory_cached = torch.xpu.memory_reserved
+        torch.cuda.max_memory_reserved = torch.xpu.max_memory_reserved
+        torch.cuda.max_memory_cached = torch.xpu.max_memory_reserved
+        torch.cuda.reset_peak_memory_stats = torch.xpu.reset_peak_memory_stats
+        torch.cuda.reset_max_memory_cached = torch.xpu.reset_peak_memory_stats
+        torch.cuda.reset_max_memory_allocated = torch.xpu.reset_peak_memory_stats
+        torch.cuda.memory_stats_as_nested_dict = torch.xpu.memory_stats_as_nested_dict
+        torch.cuda.reset_accumulated_memory_stats = torch.xpu.reset_accumulated_memory_stats
+
+        #RNG:
+        torch.cuda.get_rng_state = torch.xpu.get_rng_state
+        torch.cuda.get_rng_state_all = torch.xpu.get_rng_state_all
+        torch.cuda.set_rng_state = torch.xpu.set_rng_state
+        torch.cuda.set_rng_state_all = torch.xpu.set_rng_state_all
+        torch.cuda.manual_seed = torch.xpu.manual_seed
+        torch.cuda.manual_seed_all = torch.xpu.manual_seed_all
+        torch.cuda.seed = torch.xpu.seed
+        torch.cuda.seed_all = torch.xpu.seed_all
+        torch.cuda.initial_seed = torch.xpu.initial_seed
+
+        #AMP:
+        torch.cuda.amp = torch.xpu.amp
+        if not hasattr(torch.cuda.amp, "common"):
+            torch.cuda.amp.common = contextlib.nullcontext()
+        torch.cuda.amp.common.amp_definitely_not_available = lambda: False
+        try:
+            torch.cuda.amp.GradScaler = torch.xpu.amp.GradScaler
+        except Exception: # pylint: disable=broad-exception-caught
+            try:
+                from .gradscaler import gradscaler_init # pylint: disable=import-outside-toplevel, import-error
+                gradscaler_init()
+                torch.cuda.amp.GradScaler = torch.xpu.amp.GradScaler
+            except Exception: # pylint: disable=broad-exception-caught
+                torch.cuda.amp.GradScaler = ipex.cpu.autocast._grad_scaler.GradScaler
+
+        #C
+        torch._C._cuda_getCurrentRawStream = ipex._C._getCurrentStream
+        ipex._C._DeviceProperties.major = 2023
+        ipex._C._DeviceProperties.minor = 2
+
+        #Fix functions with ipex:
+        torch.cuda.mem_get_info = lambda device=None: [(torch.xpu.get_device_properties(device).total_memory - torch.xpu.memory_reserved(device)), torch.xpu.get_device_properties(device).total_memory]
+        torch._utils._get_available_device_type = lambda: "xpu"
+        torch.has_cuda = True
+        torch.cuda.has_half = True
+        torch.cuda.is_bf16_supported = lambda *args, **kwargs: True
+        torch.cuda.is_fp16_supported = lambda *args, **kwargs: True
+        torch.version.cuda = "11.7"
+        torch.cuda.get_device_capability = lambda *args, **kwargs: [11,7]
+        torch.cuda.get_device_properties.major = 11
+        torch.cuda.get_device_properties.minor = 7
+        torch.cuda.ipc_collect = lambda *args, **kwargs: None
+        torch.cuda.utilization = lambda *args, **kwargs: 0
+        if hasattr(torch.xpu, 'getDeviceIdListForCard'):
+            torch.cuda.getDeviceIdListForCard = torch.xpu.getDeviceIdListForCard
+            torch.cuda.get_device_id_list_per_card = torch.xpu.getDeviceIdListForCard
+        else:
+            torch.cuda.getDeviceIdListForCard = torch.xpu.get_device_id_list_per_card
+            torch.cuda.get_device_id_list_per_card = torch.xpu.get_device_id_list_per_card
+
+        ipex_hijacks()
+        attention_init()
+        try:
+            from .diffusers import ipex_diffusers
+            ipex_diffusers()
+        except Exception: # pylint: disable=broad-exception-caught
+            pass
+    except Exception as e:
+        return False, e
+    return True, None
--- a/library/ipex/attention.py
+++ b/library/ipex/attention.py
@@ -0,0 +1,157 @@
+import torch
+import intel_extension_for_pytorch as ipex # pylint: disable=import-error, unused-import
+
+# pylint: disable=protected-access, missing-function-docstring, line-too-long
+
+original_torch_bmm = torch.bmm
+def torch_bmm(input, mat2, *, out=None):
+    if input.dtype != mat2.dtype:
+        mat2 = mat2.to(input.dtype)
+
+    #ARC GPUs can't allocate more than 4GB to a single block, Slice it:
+    batch_size_attention, input_tokens, mat2_shape = input.shape[0], input.shape[1], mat2.shape[2]
+    block_multiply = input.element_size()
+    slice_block_size = input_tokens * mat2_shape / 1024 / 1024 * block_multiply
+    block_size = batch_size_attention * slice_block_size
+
+    split_slice_size = batch_size_attention
+    if block_size > 4:
+        do_split = True
+        #Find something divisible with the input_tokens
+        while (split_slice_size * slice_block_size) > 4:
+            split_slice_size = split_slice_size // 2
+            if split_slice_size <= 1:
+                split_slice_size = 1
+                break
+    else:
+        do_split = False
+
+    split_2_slice_size = input_tokens
+    if split_slice_size * slice_block_size > 4:
+        slice_block_size2 = split_slice_size * mat2_shape / 1024 / 1024 * block_multiply
+        do_split_2 = True
+        #Find something divisible with the input_tokens
+        while (split_2_slice_size * slice_block_size2) > 4:
+            split_2_slice_size = split_2_slice_size // 2
+            if split_2_slice_size <= 1:
+                split_2_slice_size = 1
+                break
+    else:
+        do_split_2 = False
+
+    if do_split:
+        hidden_states = torch.zeros(input.shape[0], input.shape[1], mat2.shape[2], device=input.device, dtype=input.dtype)
+        for i in range(batch_size_attention // split_slice_size):
+            start_idx = i * split_slice_size
+            end_idx = (i + 1) * split_slice_size
+            if do_split_2:
+                for i2 in range(input_tokens // split_2_slice_size): # pylint: disable=invalid-name
+                    start_idx_2 = i2 * split_2_slice_size
+                    end_idx_2 = (i2 + 1) * split_2_slice_size
+                    hidden_states[start_idx:end_idx, start_idx_2:end_idx_2] = original_torch_bmm(
+                        input[start_idx:end_idx, start_idx_2:end_idx_2],
+                        mat2[start_idx:end_idx, start_idx_2:end_idx_2],
+                        out=out
+                    )
+            else:
+                hidden_states[start_idx:end_idx] = original_torch_bmm(
+                    input[start_idx:end_idx],
+                    mat2[start_idx:end_idx],
+                    out=out
+                )
+    else:
+        return original_torch_bmm(input, mat2, out=out)
+    return hidden_states
+
+original_scaled_dot_product_attention = torch.nn.functional.scaled_dot_product_attention
+def scaled_dot_product_attention(query, key, value, attn_mask=None, dropout_p=0.0, is_causal=False):
+    #ARC GPUs can't allocate more than 4GB to a single block, Slice it:
+    if len(query.shape) == 3:
+        batch_size_attention, query_tokens, shape_four = query.shape
+        shape_one = 1
+        no_shape_one = True
+    else:
+        shape_one, batch_size_attention, query_tokens, shape_four = query.shape
+        no_shape_one = False
+
+    block_multiply = query.element_size()
+    slice_block_size = shape_one * query_tokens * shape_four / 1024 / 1024 * block_multiply
+    block_size = batch_size_attention * slice_block_size
+
+    split_slice_size = batch_size_attention
+    if block_size > 4:
+        do_split = True
+        #Find something divisible with the shape_one
+        while (split_slice_size * slice_block_size) > 4:
+            split_slice_size = split_slice_size // 2
+            if split_slice_size <= 1:
+                split_slice_size = 1
+                break
+    else:
+        do_split = False
+
+    split_2_slice_size = query_tokens
+    if split_slice_size * slice_block_size > 4:
+        slice_block_size2 = shape_one * split_slice_size * shape_four / 1024 / 1024 * block_multiply
+        do_split_2 = True
+        #Find something divisible with the batch_size_attention
+        while (split_2_slice_size * slice_block_size2) > 4:
+            split_2_slice_size = split_2_slice_size // 2
+            if split_2_slice_size <= 1:
+                split_2_slice_size = 1
+                break
+    else:
+        do_split_2 = False
+
+    if do_split:
+        hidden_states = torch.zeros(query.shape, device=query.device, dtype=query.dtype)
+        for i in range(batch_size_attention // split_slice_size):
+            start_idx = i * split_slice_size
+            end_idx = (i + 1) * split_slice_size
+            if do_split_2:
+                for i2 in range(query_tokens // split_2_slice_size): # pylint: disable=invalid-name
+                    start_idx_2 = i2 * split_2_slice_size
+                    end_idx_2 = (i2 + 1) * split_2_slice_size
+                    if no_shape_one:
+                        hidden_states[start_idx:end_idx, start_idx_2:end_idx_2] = original_scaled_dot_product_attention(
+                            query[start_idx:end_idx, start_idx_2:end_idx_2],
+                            key[start_idx:end_idx, start_idx_2:end_idx_2],
+                            value[start_idx:end_idx, start_idx_2:end_idx_2],
+                            attn_mask=attn_mask[start_idx:end_idx, start_idx_2:end_idx_2] if attn_mask is not None else attn_mask,
+                            dropout_p=dropout_p, is_causal=is_causal
+                        )
+                    else:
+                        hidden_states[:, start_idx:end_idx, start_idx_2:end_idx_2] = original_scaled_dot_product_attention(
+                            query[:, start_idx:end_idx, start_idx_2:end_idx_2],
+                            key[:, start_idx:end_idx, start_idx_2:end_idx_2],
+                            value[:, start_idx:end_idx, start_idx_2:end_idx_2],
+                            attn_mask=attn_mask[:, start_idx:end_idx, start_idx_2:end_idx_2] if attn_mask is not None else attn_mask,
+                            dropout_p=dropout_p, is_causal=is_causal
+                        )
+            else:
+                if no_shape_one:
+                    hidden_states[start_idx:end_idx] = original_scaled_dot_product_attention(
+                        query[start_idx:end_idx],
+                        key[start_idx:end_idx],
+                        value[start_idx:end_idx],
+                        attn_mask=attn_mask[start_idx:end_idx] if attn_mask is not None else attn_mask,
+                        dropout_p=dropout_p, is_causal=is_causal
+                    )
+                else:
+                    hidden_states[:, start_idx:end_idx] = original_scaled_dot_product_attention(
+                        query[:, start_idx:end_idx],
+                        key[:, start_idx:end_idx],
+                        value[:, start_idx:end_idx],
+                        attn_mask=attn_mask[:, start_idx:end_idx] if attn_mask is not None else attn_mask,
+                        dropout_p=dropout_p, is_causal=is_causal
+                    )
+    else:
+        return original_scaled_dot_product_attention(
+            query, key, value, attn_mask=attn_mask, dropout_p=dropout_p, is_causal=is_causal
+        )
+    return hidden_states
+
+def attention_init():
+    #ARC GPUs can't allocate more than 4GB to a single block:
+    torch.bmm = torch_bmm
+    torch.nn.functional.scaled_dot_product_attention = scaled_dot_product_attention
--- a/library/ipex/diffusers.py
+++ b/library/ipex/diffusers.py
@@ -0,0 +1,120 @@
+import torch
+import intel_extension_for_pytorch as ipex # pylint: disable=import-error, unused-import
+import diffusers #0.21.1 # pylint: disable=import-error
+from diffusers.models.attention_processor import Attention
+
+# pylint: disable=protected-access, missing-function-docstring, line-too-long
+
+class SlicedAttnProcessor: # pylint: disable=too-few-public-methods
+    r"""
+    Processor for implementing sliced attention.
+
+    Args:
+        slice_size (`int`, *optional*):
+            The number of steps to compute attention. Uses as many slices as `attention_head_dim // slice_size`, and
+            `attention_head_dim` must be a multiple of the `slice_size`.
+    """
+
+    def __init__(self, slice_size):
+        self.slice_size = slice_size
+
+    def __call__(self, attn: Attention, hidden_states, encoder_hidden_states=None, attention_mask=None): # pylint: disable=too-many-statements, too-many-locals, too-many-branches
+        residual = hidden_states
+
+        input_ndim = hidden_states.ndim
+
+        if input_ndim == 4:
+            batch_size, channel, height, width = hidden_states.shape
+            hidden_states = hidden_states.view(batch_size, channel, height * width).transpose(1, 2)
+
+        batch_size, sequence_length, _ = (
+            hidden_states.shape if encoder_hidden_states is None else encoder_hidden_states.shape
+        )
+        attention_mask = attn.prepare_attention_mask(attention_mask, sequence_length, batch_size)
+
+        if attn.group_norm is not None:
+            hidden_states = attn.group_norm(hidden_states.transpose(1, 2)).transpose(1, 2)
+
+        query = attn.to_q(hidden_states)
+        dim = query.shape[-1]
+        query = attn.head_to_batch_dim(query)
+
+        if encoder_hidden_states is None:
+            encoder_hidden_states = hidden_states
+        elif attn.norm_cross:
+            encoder_hidden_states = attn.norm_encoder_hidden_states(encoder_hidden_states)
+
+        key = attn.to_k(encoder_hidden_states)
+        value = attn.to_v(encoder_hidden_states)
+        key = attn.head_to_batch_dim(key)
+        value = attn.head_to_batch_dim(value)
+
+        batch_size_attention, query_tokens, shape_three = query.shape
+        hidden_states = torch.zeros(
+            (batch_size_attention, query_tokens, dim // attn.heads), device=query.device, dtype=query.dtype
+        )
+
+        #ARC GPUs can't allocate more than 4GB to a single block, Slice it:
+        block_multiply = query.element_size()
+        slice_block_size = self.slice_size * shape_three / 1024 / 1024 * block_multiply
+        block_size = query_tokens * slice_block_size
+        split_2_slice_size = query_tokens
+        if block_size > 4:
+            do_split_2 = True
+            #Find something divisible with the query_tokens
+            while (split_2_slice_size * slice_block_size) > 4:
+                split_2_slice_size = split_2_slice_size // 2
+                if split_2_slice_size <= 1:
+                    split_2_slice_size = 1
+                    break
+        else:
+            do_split_2 = False
+
+        for i in range(batch_size_attention // self.slice_size):
+            start_idx = i * self.slice_size
+            end_idx = (i + 1) * self.slice_size
+
+            if do_split_2:
+                for i2 in range(query_tokens // split_2_slice_size): # pylint: disable=invalid-name
+                    start_idx_2 = i2 * split_2_slice_size
+                    end_idx_2 = (i2 + 1) * split_2_slice_size
+
+                    query_slice = query[start_idx:end_idx, start_idx_2:end_idx_2]
+                    key_slice = key[start_idx:end_idx, start_idx_2:end_idx_2]
+                    attn_mask_slice = attention_mask[start_idx:end_idx, start_idx_2:end_idx_2] if attention_mask is not None else None
+
+                    attn_slice = attn.get_attention_scores(query_slice, key_slice, attn_mask_slice)
+                    attn_slice = torch.bmm(attn_slice, value[start_idx:end_idx, start_idx_2:end_idx_2])
+
+                    hidden_states[start_idx:end_idx, start_idx_2:end_idx_2] = attn_slice
+            else:
+                query_slice = query[start_idx:end_idx]
+                key_slice = key[start_idx:end_idx]
+                attn_mask_slice = attention_mask[start_idx:end_idx] if attention_mask is not None else None
+
+                attn_slice = attn.get_attention_scores(query_slice, key_slice, attn_mask_slice)
+
+                attn_slice = torch.bmm(attn_slice, value[start_idx:end_idx])
+
+                hidden_states[start_idx:end_idx] = attn_slice
+
+        hidden_states = attn.batch_to_head_dim(hidden_states)
+
+        # linear proj
+        hidden_states = attn.to_out[0](hidden_states)
+        # dropout
+        hidden_states = attn.to_out[1](hidden_states)
+
+        if input_ndim == 4:
+            hidden_states = hidden_states.transpose(-1, -2).reshape(batch_size, channel, height, width)
+
+        if attn.residual_connection:
+            hidden_states = hidden_states + residual
+
+        hidden_states = hidden_states / attn.rescale_output_factor
+
+        return hidden_states
+
+def ipex_diffusers():
+    #ARC GPUs can't allocate more than 4GB to a single block:
+    diffusers.models.attention_processor.SlicedAttnProcessor = SlicedAttnProcessor
--- a/library/ipex/gradscaler.py
+++ b/library/ipex/gradscaler.py
@@ -0,0 +1,179 @@
+from collections import defaultdict
+import torch
+import intel_extension_for_pytorch as ipex # pylint: disable=import-error, unused-import
+import intel_extension_for_pytorch._C as core # pylint: disable=import-error, unused-import
+
+# pylint: disable=protected-access, missing-function-docstring, line-too-long
+
+OptState = ipex.cpu.autocast._grad_scaler.OptState
+_MultiDeviceReplicator = ipex.cpu.autocast._grad_scaler._MultiDeviceReplicator
+_refresh_per_optimizer_state = ipex.cpu.autocast._grad_scaler._refresh_per_optimizer_state
+
+def _unscale_grads_(self, optimizer, inv_scale, found_inf, allow_fp16): # pylint: disable=unused-argument
+    per_device_inv_scale = _MultiDeviceReplicator(inv_scale)
+    per_device_found_inf = _MultiDeviceReplicator(found_inf)
+
+    # To set up _amp_foreach_non_finite_check_and_unscale_, split grads by device and dtype.
+    # There could be hundreds of grads, so we'd like to iterate through them just once.
+    # However, we don't know their devices or dtypes in advance.
+
+    # https://stackoverflow.com/questions/5029934/defaultdict-of-defaultdict
+    # Google says mypy struggles with defaultdicts type annotations.
+    per_device_and_dtype_grads = defaultdict(lambda: defaultdict(list))  # type: ignore[var-annotated]
+    # sync grad to master weight
+    if hasattr(optimizer, "sync_grad"):
+        optimizer.sync_grad()
+    with torch.no_grad():
+        for group in optimizer.param_groups:
+            for param in group["params"]:
+                if param.grad is None:
+                    continue
+                if (not allow_fp16) and param.grad.dtype == torch.float16:
+                    raise ValueError("Attempting to unscale FP16 gradients.")
+                if param.grad.is_sparse:
+                    # is_coalesced() == False means the sparse grad has values with duplicate indices.
+                    # coalesce() deduplicates indices and adds all values that have the same index.
+                    # For scaled fp16 values, there's a good chance coalescing will cause overflow,
+                    # so we should check the coalesced _values().
+                    if param.grad.dtype is torch.float16:
+                        param.grad = param.grad.coalesce()
+                    to_unscale = param.grad._values()
+                else:
+                    to_unscale = param.grad
+
+                # -: is there a way to split by device and dtype without appending in the inner loop?
+                to_unscale = to_unscale.to("cpu")
+                per_device_and_dtype_grads[to_unscale.device][
+                    to_unscale.dtype
+                ].append(to_unscale)
+
+        for _, per_dtype_grads in per_device_and_dtype_grads.items():
+            for grads in per_dtype_grads.values():
+                core._amp_foreach_non_finite_check_and_unscale_(
+                    grads,
+                    per_device_found_inf.get("cpu"),
+                    per_device_inv_scale.get("cpu"),
+                )
+
+    return per_device_found_inf._per_device_tensors
+
+def unscale_(self, optimizer):
+    """
+    Divides ("unscales") the optimizer's gradient tensors by the scale factor.
+    :meth:`unscale_` is optional, serving cases where you need to
+    :ref:`modify or inspect gradients<working-with-unscaled-gradients>`
+    between the backward pass(es) and :meth:`step`.
+    If :meth:`unscale_` is not called explicitly,  gradients will be unscaled  automatically during :meth:`step`.
+    Simple example, using :meth:`unscale_` to enable clipping of unscaled gradients::
+        ...
+        scaler.scale(loss).backward()
+        scaler.unscale_(optimizer)
+        torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm)
+        scaler.step(optimizer)
+        scaler.update()
+    Args:
+        optimizer (torch.optim.Optimizer):  Optimizer that owns the gradients to be unscaled.
+    .. warning::
+        :meth:`unscale_` should only be called once per optimizer per :meth:`step` call,
+        and only after all gradients for that optimizer's assigned parameters have been accumulated.
+        Calling :meth:`unscale_` twice for a given optimizer between each :meth:`step` triggers a RuntimeError.
+    .. warning::
+        :meth:`unscale_` may unscale sparse gradients out of place, replacing the ``.grad`` attribute.
+    """
+    if not self._enabled:
+        return
+
+    self._check_scale_growth_tracker("unscale_")
+
+    optimizer_state = self._per_optimizer_states[id(optimizer)]
+
+    if optimizer_state["stage"] is OptState.UNSCALED: # pylint: disable=no-else-raise
+        raise RuntimeError(
+            "unscale_() has already been called on this optimizer since the last update()."
+        )
+    elif optimizer_state["stage"] is OptState.STEPPED:
+        raise RuntimeError("unscale_() is being called after step().")
+
+    # FP32 division can be imprecise for certain compile options, so we carry out the reciprocal in FP64.
+    assert self._scale is not None
+    inv_scale = self._scale.to("cpu").double().reciprocal().float().to(self._scale.device)
+    found_inf = torch.full(
+        (1,), 0.0, dtype=torch.float32, device=self._scale.device
+    )
+
+    optimizer_state["found_inf_per_device"] = self._unscale_grads_(
+        optimizer, inv_scale, found_inf, False
+    )
+    optimizer_state["stage"] = OptState.UNSCALED
+
+def update(self, new_scale=None):
+    """
+    Updates the scale factor.
+    If any optimizer steps were skipped the scale is multiplied by ``backoff_factor``
+    to reduce it. If ``growth_interval`` unskipped iterations occurred consecutively,
+    the scale is multiplied by ``growth_factor`` to increase it.
+    Passing ``new_scale`` sets the new scale value manually. (``new_scale`` is not
+    used directly, it's used to fill GradScaler's internal scale tensor. So if
+    ``new_scale`` was a tensor, later in-place changes to that tensor will not further
+    affect the scale GradScaler uses internally.)
+    Args:
+        new_scale (float or :class:`torch.FloatTensor`, optional, default=None):  New scale factor.
+    .. warning::
+        :meth:`update` should only be called at the end of the iteration, after ``scaler.step(optimizer)`` has
+        been invoked for all optimizers used this iteration.
+    """
+    if not self._enabled:
+        return
+
+    _scale, _growth_tracker = self._check_scale_growth_tracker("update")
+
+    if new_scale is not None:
+        # Accept a new user-defined scale.
+        if isinstance(new_scale, float):
+            self._scale.fill_(new_scale)  # type: ignore[union-attr]
+        else:
+            reason = "new_scale should be a float or a 1-element torch.FloatTensor with requires_grad=False."
+            assert isinstance(new_scale, torch.FloatTensor), reason  # type: ignore[attr-defined]
+            assert new_scale.numel() == 1, reason
+            assert new_scale.requires_grad is False, reason
+            self._scale.copy_(new_scale)  # type: ignore[union-attr]
+    else:
+        # Consume shared inf/nan data collected from optimizers to update the scale.
+        # If all found_inf tensors are on the same device as self._scale, this operation is asynchronous.
+        found_infs = [
+            found_inf.to(device="cpu", non_blocking=True)
+            for state in self._per_optimizer_states.values()
+            for found_inf in state["found_inf_per_device"].values()
+        ]
+
+        assert len(found_infs) > 0, "No inf checks were recorded prior to update."
+
+        found_inf_combined = found_infs[0]
+        if len(found_infs) > 1:
+            for i in range(1, len(found_infs)):
+                found_inf_combined += found_infs[i]
+
+        to_device = _scale.device
+        _scale = _scale.to("cpu")
+        _growth_tracker = _growth_tracker.to("cpu")
+
+        core._amp_update_scale_(
+            _scale,
+            _growth_tracker,
+            found_inf_combined,
+            self._growth_factor,
+            self._backoff_factor,
+            self._growth_interval,
+        )
+
+        _scale = _scale.to(to_device)
+        _growth_tracker = _growth_tracker.to(to_device)
+    # To prepare for next iteration, clear the data collected from optimizers this iteration.
+    self._per_optimizer_states = defaultdict(_refresh_per_optimizer_state)
+
+def gradscaler_init():
+    torch.xpu.amp.GradScaler = ipex.cpu.autocast._grad_scaler.GradScaler
+    torch.xpu.amp.GradScaler._unscale_grads_ = _unscale_grads_
+    torch.xpu.amp.GradScaler.unscale_ = unscale_
+    torch.xpu.amp.GradScaler.update = update
+    return torch.xpu.amp.GradScaler
--- a/library/ipex/hijacks.py
+++ b/library/ipex/hijacks.py
@@ -0,0 +1,196 @@
+import contextlib
+import importlib
+import torch
+import intel_extension_for_pytorch as ipex # pylint: disable=import-error, unused-import
+
+# pylint: disable=protected-access, missing-function-docstring, line-too-long, unnecessary-lambda, no-else-return
+
+class CondFunc: # pylint: disable=missing-class-docstring
+    def __new__(cls, orig_func, sub_func, cond_func):
+        self = super(CondFunc, cls).__new__(cls)
+        if isinstance(orig_func, str):
+            func_path = orig_func.split('.')
+            for i in range(len(func_path)-1, -1, -1):
+                try:
+                    resolved_obj = importlib.import_module('.'.join(func_path[:i]))
+                    break
+                except ImportError:
+                    pass
+            for attr_name in func_path[i:-1]:
+                resolved_obj = getattr(resolved_obj, attr_name)
+            orig_func = getattr(resolved_obj, func_path[-1])
+            setattr(resolved_obj, func_path[-1], lambda *args, **kwargs: self(*args, **kwargs))
+        self.__init__(orig_func, sub_func, cond_func)
+        return lambda *args, **kwargs: self(*args, **kwargs)
+    def __init__(self, orig_func, sub_func, cond_func):
+        self.__orig_func = orig_func
+        self.__sub_func = sub_func
+        self.__cond_func = cond_func
+    def __call__(self, *args, **kwargs):
+        if not self.__cond_func or self.__cond_func(self.__orig_func, *args, **kwargs):
+            return self.__sub_func(self.__orig_func, *args, **kwargs)
+        else:
+            return self.__orig_func(*args, **kwargs)
+
+_utils = torch.utils.data._utils
+def _shutdown_workers(self):
+    if torch.utils.data._utils is None or torch.utils.data._utils.python_exit_status is True or torch.utils.data._utils.python_exit_status is None:
+        return
+    if hasattr(self, "_shutdown") and not self._shutdown:
+        self._shutdown = True
+        try:
+            if hasattr(self, '_pin_memory_thread'):
+                self._pin_memory_thread_done_event.set()
+                self._worker_result_queue.put((None, None))
+                self._pin_memory_thread.join()
+                self._worker_result_queue.cancel_join_thread()
+                self._worker_result_queue.close()
+            self._workers_done_event.set()
+            for worker_id in range(len(self._workers)):
+                if self._persistent_workers or self._workers_status[worker_id]:
+                    self._mark_worker_as_unavailable(worker_id, shutdown=True)
+            for w in self._workers: # pylint: disable=invalid-name
+                w.join(timeout=torch.utils.data._utils.MP_STATUS_CHECK_INTERVAL)
+            for q in self._index_queues: # pylint: disable=invalid-name
+                q.cancel_join_thread()
+                q.close()
+        finally:
+            if self._worker_pids_set:
+                torch.utils.data._utils.signal_handling._remove_worker_pids(id(self))
+                self._worker_pids_set = False
+            for w in self._workers: # pylint: disable=invalid-name
+                if w.is_alive():
+                    w.terminate()
+
+class DummyDataParallel(torch.nn.Module): # pylint: disable=missing-class-docstring, unused-argument, too-few-public-methods
+    def __new__(cls, module, device_ids=None, output_device=None, dim=0): # pylint: disable=unused-argument
+        if isinstance(device_ids, list) and len(device_ids) > 1:
+            print("IPEX backend doesn't support DataParallel on multiple XPU devices")
+        return module.to("xpu")
+
+def return_null_context(*args, **kwargs): # pylint: disable=unused-argument
+    return contextlib.nullcontext()
+
+def check_device(device):
+    return bool((isinstance(device, torch.device) and device.type == "cuda") or (isinstance(device, str) and "cuda" in device) or isinstance(device, int))
+
+def return_xpu(device):
+    return f"xpu:{device.split(':')[-1]}" if isinstance(device, str) and ":" in device else f"xpu:{device}" if isinstance(device, int) else torch.device("xpu") if isinstance(device, torch.device) else "xpu"
+
+def ipex_no_cuda(orig_func, *args, **kwargs):
+    torch.cuda.is_available = lambda: False
+    orig_func(*args, **kwargs)
+    torch.cuda.is_available = torch.xpu.is_available
+
+original_autocast = torch.autocast
+def ipex_autocast(*args, **kwargs):
+    if len(args) > 0 and args[0] == "cuda":
+        return original_autocast("xpu", *args[1:], **kwargs)
+    else:
+        return original_autocast(*args, **kwargs)
+
+original_torch_cat = torch.cat
+def torch_cat(tensor, *args, **kwargs):
+    if len(tensor) == 3 and (tensor[0].dtype != tensor[1].dtype or tensor[2].dtype != tensor[1].dtype):
+        return original_torch_cat([tensor[0].to(tensor[1].dtype), tensor[1], tensor[2].to(tensor[1].dtype)], *args, **kwargs)
+    else:
+        return original_torch_cat(tensor, *args, **kwargs)
+
+original_interpolate = torch.nn.functional.interpolate
+def interpolate(tensor, size=None, scale_factor=None, mode='nearest', align_corners=None, recompute_scale_factor=None, antialias=False): # pylint: disable=too-many-arguments
+    if antialias or align_corners is not None:
+        return_device = tensor.device
+        return_dtype = tensor.dtype
+        return original_interpolate(tensor.to("cpu", dtype=torch.float32), size=size, scale_factor=scale_factor, mode=mode,
+        align_corners=align_corners, recompute_scale_factor=recompute_scale_factor, antialias=antialias).to(return_device, dtype=return_dtype)
+    else:
+        return original_interpolate(tensor, size=size, scale_factor=scale_factor, mode=mode,
+        align_corners=align_corners, recompute_scale_factor=recompute_scale_factor, antialias=antialias)
+
+original_linalg_solve = torch.linalg.solve
+def linalg_solve(A, B, *args, **kwargs): # pylint: disable=invalid-name
+    if A.device != torch.device("cpu") or B.device != torch.device("cpu"):
+        return_device = A.device
+        return original_linalg_solve(A.to("cpu"), B.to("cpu"), *args, **kwargs).to(return_device)
+    else:
+        return original_linalg_solve(A, B, *args, **kwargs)
+
+def ipex_hijacks():
+    CondFunc('torch.Tensor.to',
+        lambda orig_func, self, device=None, *args, **kwargs: orig_func(self, return_xpu(device), *args, **kwargs),
+        lambda orig_func, self, device=None, *args, **kwargs: check_device(device))
+    CondFunc('torch.Tensor.cuda',
+        lambda orig_func, self, device=None, *args, **kwargs: orig_func(self, return_xpu(device), *args, **kwargs),
+        lambda orig_func, self, device=None, *args, **kwargs: check_device(device))
+    CondFunc('torch.empty',
+        lambda orig_func, *args, device=None, **kwargs: orig_func(*args, device=return_xpu(device), **kwargs),
+        lambda orig_func, *args, device=None, **kwargs: check_device(device))
+    CondFunc('torch.load',
+        lambda orig_func, *args, map_location=None, **kwargs: orig_func(*args, return_xpu(map_location), **kwargs),
+        lambda orig_func, *args, map_location=None, **kwargs: map_location is None or check_device(map_location))
+    CondFunc('torch.randn',
+        lambda orig_func, *args, device=None, **kwargs: orig_func(*args, device=return_xpu(device), **kwargs),
+        lambda orig_func, *args, device=None, **kwargs: check_device(device))
+    CondFunc('torch.ones',
+        lambda orig_func, *args, device=None, **kwargs: orig_func(*args, device=return_xpu(device), **kwargs),
+        lambda orig_func, *args, device=None, **kwargs: check_device(device))
+    CondFunc('torch.zeros',
+        lambda orig_func, *args, device=None, **kwargs: orig_func(*args, device=return_xpu(device), **kwargs),
+        lambda orig_func, *args, device=None, **kwargs: check_device(device))
+    CondFunc('torch.tensor',
+        lambda orig_func, *args, device=None, **kwargs: orig_func(*args, device=return_xpu(device), **kwargs),
+        lambda orig_func, *args, device=None, **kwargs: check_device(device))
+    CondFunc('torch.linspace',
+        lambda orig_func, *args, device=None, **kwargs: orig_func(*args, device=return_xpu(device), **kwargs),
+        lambda orig_func, *args, device=None, **kwargs: check_device(device))
+
+    CondFunc('torch.Generator',
+        lambda orig_func, device=None: torch.xpu.Generator(device),
+        lambda orig_func, device=None: device is not None and device != torch.device("cpu") and device != "cpu")
+
+    CondFunc('torch.batch_norm',
+        lambda orig_func, input, weight, bias, *args, **kwargs: orig_func(input,
+        weight if weight is not None else torch.ones(input.size()[1], device=input.device),
+        bias if bias is not None else torch.zeros(input.size()[1], device=input.device), *args, **kwargs),
+        lambda orig_func, input, *args, **kwargs: input.device != torch.device("cpu"))
+    CondFunc('torch.instance_norm',
+        lambda orig_func, input, weight, bias, *args, **kwargs: orig_func(input,
+        weight if weight is not None else torch.ones(input.size()[1], device=input.device),
+        bias if bias is not None else torch.zeros(input.size()[1], device=input.device), *args, **kwargs),
+        lambda orig_func, input, *args, **kwargs: input.device != torch.device("cpu"))
+
+    #Functions with dtype errors:
+    CondFunc('torch.nn.modules.GroupNorm.forward',
+        lambda orig_func, self, input: orig_func(self, input.to(self.weight.data.dtype)),
+        lambda orig_func, self, input: input.dtype != self.weight.data.dtype)
+    CondFunc('torch.nn.modules.linear.Linear.forward',
+        lambda orig_func, self, input: orig_func(self, input.to(self.weight.data.dtype)),
+        lambda orig_func, self, input: input.dtype != self.weight.data.dtype)
+    CondFunc('torch.nn.modules.conv.Conv2d.forward',
+        lambda orig_func, self, input: orig_func(self, input.to(self.weight.data.dtype)),
+        lambda orig_func, self, input: input.dtype != self.weight.data.dtype)
+    CondFunc('torch.nn.functional.layer_norm',
+        lambda orig_func, input, normalized_shape=None, weight=None, *args, **kwargs:
+        orig_func(input.to(weight.data.dtype), normalized_shape, weight, *args, **kwargs),
+        lambda orig_func, input, normalized_shape=None, weight=None, *args, **kwargs:
+        weight is not None and input.dtype != weight.data.dtype)
+
+    #Diffusers Float64 (ARC GPUs doesn't support double or Float64):
+    if not torch.xpu.has_fp64_dtype():
+        CondFunc('torch.from_numpy',
+        lambda orig_func, ndarray: orig_func(ndarray.astype('float32')),
+        lambda orig_func, ndarray: ndarray.dtype == float)
+
+    #Broken functions when torch.cuda.is_available is True:
+    CondFunc('torch.utils.data.dataloader._BaseDataLoaderIter.__init__',
+        lambda orig_func, *args, **kwargs: ipex_no_cuda(orig_func, *args, **kwargs),
+        lambda orig_func, *args, **kwargs: True)
+
+    #Functions that make compile mad with CondFunc:
+    torch.utils.data.dataloader._MultiProcessingDataLoaderIter._shutdown_workers = _shutdown_workers
+    torch.nn.DataParallel = DummyDataParallel
+    torch.autocast = ipex_autocast
+    torch.cat = torch_cat
+    torch.linalg.solve = linalg_solve
+    torch.nn.functional.interpolate = interpolate
+    torch.backends.cuda.sdp_kernel = return_null_context
--- a/library/model_util.py
+++ b/library/model_util.py
@@ -4,6 +4,13 @@
 import math
 import os
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 import diffusers
 from transformers import CLIPTextModel, CLIPTokenizer, CLIPTextConfig, logging
 from diffusers import AutoencoderKL, DDIMScheduler, StableDiffusionPipeline  # , UNet2DConditionModel
--- a/library/original_unet.py
+++ b/library/original_unet.py
@@ -131,7 +131,7 @@ DOWN_BLOCK_TYPES = ["CrossAttnDownBlock2D", "CrossAttnDownBlock2D", "CrossAttnDo
 UP_BLOCK_TYPES = ["UpBlock2D", "CrossAttnUpBlock2D", "CrossAttnUpBlock2D", "CrossAttnUpBlock2D"]


-# region memory effcient attention
+# region memory efficient attention

 # FlashAttentionを使うCrossAttention
 # based on https://github.com/lucidrains/memory-efficient-attention-pytorch/blob/main/memory_efficient_attention_pytorch/flash_attention.py
--- a/library/sdxl_original_unet.py
+++ b/library/sdxl_original_unet.py
@@ -41,7 +41,7 @@ TIME_EMBED_DIM = 320 * 4

 USE_REENTRANT = True

-# region memory effcient attention
+# region memory efficient attention

 # FlashAttentionを使うCrossAttention
 # based on https://github.com/lucidrains/memory-efficient-attention-pytorch/blob/main/memory_efficient_attention_pytorch/flash_attention.py
@@ -996,109 +996,6 @@ class SdxlUNet2DConditionModel(nn.Module):
            [GroupNorm32(32, self.model_channels), nn.SiLU(), nn.Conv2d(self.model_channels, self.out_channels, 3, padding=1)]
        )

-        # FreeU
-        self.freeU = False
-        self.freeUB1 = 1.0
-        self.freeUB2 = 1.0
-        self.freeUS1 = 1.0
-        self.freeUS2 = 1.0
-        self.freeURThres = 1
-
-    # implementation of FreeU
-    # FreeU: Free Lunch in Diffusion U-Net  https://arxiv.org/abs/2309.11497
-
-    def set_free_u_enabled(self, enabled: bool, b1=1.0, b2=1.0, s1=1.0, s2=1.0, rthresh=1):
-        print(f"FreeU: {enabled}, b1={b1}, b2={b2}, s1={s1}, s2={s2}, rthresh={rthresh}")
-        self.freeU = enabled
-        self.freeUB1 = b1
-        self.freeUB2 = b2
-        self.freeUS1 = s1
-        self.freeUS2 = s2
-        self.freeURThres = rthresh
-
-    def spectral_modulation(self, skip_feature, sl=1.0, rthresh=1):
-        """
-        スキップ特徴を周波数領域で修正する関数
-
-        :param skip_feature: スキップ特徴のテンソル [b, c, H, W]
-        :param sl: スケーリング係数
-        :param rthresh: 周波数の閾値
-        :return: 修正されたスキップ特徴
-        """
-
-        import torch.fft
-
-        r""" 
-        # 論文に従った実装
-
-        org_dtype = skip_feature.dtype
-        if org_dtype == torch.bfloat16:
-            skip_feature = skip_feature.to(torch.float32)
-
-        # FFTを計算
-        F = torch.fft.fftn(skip_feature, dim=(2, 3))
-
-        # 周波数領域での座標を計算
-        freq_x = torch.fft.fftfreq(skip_feature.size(2), d=1 / skip_feature.size(2)).to(skip_feature.device)
-        freq_y = torch.fft.fftfreq(skip_feature.size(3), d=1 / skip_feature.size(3)).to(skip_feature.device)
-
-        # 2Dグリッドを作成
-        freq_x = freq_x[:, None]  # [H, 1]
-        freq_y = freq_y[None, :]  # [1, W]
-
-        # ラジアス（距離）を計算
-        r = torch.sqrt(freq_x**2 + freq_y**2)
-        # 32,32: tensor(0., device='cuda:0') tensor(22.6274, device='cuda:0') tensor(12.2521, device='cuda:0')
-        # 64,64: tensor(0., device='cuda:0') tensor(45.2548, device='cuda:0') tensor(24.4908, device='cuda:0')
-        # 128,128: tensor(0., device='cuda:0') tensor(90.5097, device='cuda:0') tensor(48.9748, device='cuda:0')
-
-        # マスクを作成
-        mask = torch.ones_like(r)
-        mask[r < rthresh] = sl
-
-        # b,c,H,Wの形状にブロードキャスト
-        # TODO shapeごとに同じなのでキャッシュすると良さそう
-        mask = mask[None, None, :, :]
-
-        # 周波数領域での要素ごとの乗算
-        F_prime = F * mask
-
-        # 逆FFTを計算
-        modified_skip_feature = torch.fft.ifftn(F_prime, dim=(2, 3))
-
-        modified_skip_feature = modified_skip_feature.real  # 実部のみを取得
-        """
-
-        # 公式リポジトリの実装
-
-        org_dtype = skip_feature.dtype
-
-        x = skip_feature
-        threshold = rthresh
-        scale = sl
-
-        # FFT
-        x_freq = torch.fft.fftn(x.float(), dim=(-2, -1))
-        x_freq = torch.fft.fftshift(x_freq, dim=(-2, -1))
-
-        B, C, H, W = x_freq.shape
-        mask = torch.ones((B, C, H, W), device=x.device)
-
-        crow, ccol = H // 2, W // 2
-        mask[..., crow - threshold : crow + threshold, ccol - threshold : ccol + threshold] = scale
-        x_freq = x_freq * mask
-
-        # IFFT
-        x_freq = torch.fft.ifftshift(x_freq, dim=(-2, -1))
-        x_filtered = torch.fft.ifftn(x_freq, dim=(-2, -1)).real
-
-        modified_skip_feature = x_filtered
-
-        # if org_dtype == torch.bfloat16:
-        modified_skip_feature = modified_skip_feature.to(org_dtype)
-
-        return modified_skip_feature
-
    # region diffusers compatibility
    def prepare_config(self):
        self.config = SimpleNamespace()
@@ -1182,30 +1079,11 @@ class SdxlUNet2DConditionModel(nn.Module):
        h = x
        for module in self.input_blocks:
            h = call_module(module, h, emb, context)
-
-            if self.freeU:
-                ch = h.shape[1]
-                s = self.freeUS1 if ch == 1280 else (self.freeUS2 if ch == 640 else 1.0)
-                if s == 1.0:
-                    h_mod = h
-                else:
-                    h_mod = self.spectral_modulation(h, s, self.freeURThres)
-                hs.append(h_mod)
-            else:
-                hs.append(h)
+            hs.append(h)

        h = call_module(self.middle_block, h, emb, context)

        for module in self.output_blocks:
-            if self.freeU:
-                ch = h.shape[1]
-                if ch == 1280:
-                    h[:, : ch // 2] = h[:, : ch // 2] * self.freeUB1
-                elif ch == 640:
-                    h[:, : ch // 2] = h[:, : ch // 2] * self.freeUB2
-                # else:
-                #     print(f"disable freeU: {ch}")
-
            h = torch.cat([h, hs.pop()], dim=1)
            h = call_module(module, h, emb, context)

--- a/library/train_util.py
+++ b/library/train_util.py
@@ -96,6 +96,7 @@ try:
 except:
    pass

+# JPEG-XL on Linux
 try:
    from jxlpy import JXLImagePlugin

@@ -103,6 +104,14 @@ try:
 except:
    pass

+# JPEG-XL on Windows
+try:
+    import pillow_jxl
+
+    IMAGE_EXTENSIONS.extend([".jxl", ".JXL"])
+except:
+    pass
+
 IMAGE_TRANSFORMS = transforms.Compose(
    [
        transforms.ToTensor(),
@@ -4658,7 +4667,7 @@ class ImageLoadingDataset(torch.utils.data.Dataset):


 # collate_fn用 epoch,stepはmultiprocessing.Value
-class collater_class:
+class collator_class:
    def __init__(self, epoch, step, dataset):
        self.current_epoch = epoch
        self.current_step = step
--- a/networks/extract_lora_from_models.py
+++ b/networks/extract_lora_from_models.py
@@ -14,7 +14,7 @@ import lora


 CLAMP_QUANTILE = 0.99
-MIN_DIFF = 1e-4
+MIN_DIFF = 1e-1


 def save_to_file(file_name, model, state_dict, dtype):
@@ -200,7 +200,7 @@ def svd(args):
    if not args.no_metadata:
        title = os.path.splitext(os.path.basename(args.save_to))[0]
        sai_metadata = sai_model_spec.build_metadata(
-            None, args.v2, args.v_parameterization, False, True, False, time.time(), title=title
+            None, args.v2, args.v_parameterization, args.sdxl, True, False, time.time(), title=title
        )
        metadata.update(sai_metadata)

--- a/networks/lora_diffusers.py
+++ b/networks/lora_diffusers.py
@@ -117,7 +117,7 @@ class LoRAModule(torch.nn.Module):
        super().__init__()
        self.lora_name = lora_name

-        if org_module.__class__.__name__ == "Conv2d":
+        if org_module.__class__.__name__ == "Conv2d" or org_module.__class__.__name__ == "LoRACompatibleConv":
            in_dim = org_module.in_channels
            out_dim = org_module.out_channels
        else:
@@ -126,7 +126,7 @@ class LoRAModule(torch.nn.Module):

        self.lora_dim = lora_dim

-        if org_module.__class__.__name__ == "Conv2d":
+        if org_module.__class__.__name__ == "Conv2d" or org_module.__class__.__name__ == "LoRACompatibleConv":
            kernel_size = org_module.kernel_size
            stride = org_module.stride
            padding = org_module.padding
@@ -166,7 +166,8 @@ class LoRAModule(torch.nn.Module):
            self.org_module[0].forward = self.org_forward

    # forward with lora
-    def forward(self, x):
+    # scale is used LoRACompatibleConv, but we ignore it because we have multiplier
+    def forward(self, x, scale=1.0):
        if not self.enabled:
            return self.org_forward(x)
        return self.org_forward(x) + self.lora_up(self.lora_down(x)) * self.multiplier * self.scale
@@ -318,8 +319,12 @@ class LoRANetwork(torch.nn.Module):
            for name, module in root_module.named_modules():
                if module.__class__.__name__ in target_replace_modules:
                    for child_name, child_module in module.named_modules():
-                        is_linear = child_module.__class__.__name__ == "Linear"
-                        is_conv2d = child_module.__class__.__name__ == "Conv2d"
+                        is_linear = (
+                            child_module.__class__.__name__ == "Linear" or child_module.__class__.__name__ == "LoRACompatibleLinear"
+                        )
+                        is_conv2d = (
+                            child_module.__class__.__name__ == "Conv2d" or child_module.__class__.__name__ == "LoRACompatibleConv"
+                        )

                        if is_linear or is_conv2d:
                            lora_name = prefix + "." + name + "." + child_name
@@ -359,7 +364,7 @@ class LoRANetwork(torch.nn.Module):
            skipped_te += skipped
        print(f"create LoRA for Text Encoder: {len(self.text_encoder_loras)} modules.")
        if len(skipped_te) > 0:
-            print(f"skipped {len(skipped_te)} modules because of missing weight.")
+            print(f"skipped {len(skipped_te)} modules because of missing weight for text encoder.")

        # extend U-Net target modules to include Conv2d 3x3
        target_modules = LoRANetwork.UNET_TARGET_REPLACE_MODULE + LoRANetwork.UNET_TARGET_REPLACE_MODULE_CONV2D_3X3
@@ -368,7 +373,7 @@ class LoRANetwork(torch.nn.Module):
        self.unet_loras, skipped_un = create_modules(True, None, unet, target_modules)
        print(f"create LoRA for U-Net: {len(self.unet_loras)} modules.")
        if len(skipped_un) > 0:
-            print(f"skipped {len(skipped_un)} modules because of missing weight.")
+            print(f"skipped {len(skipped_un)} modules because of missing weight for U-Net.")

        # assertion
        names = set()
--- a/networks/merge_lora.py
+++ b/networks/merge_lora.py
@@ -110,7 +110,7 @@ def merge_to_sd_model(text_encoder, unet, models, ratios, merge_dtype):
                module.weight = torch.nn.Parameter(weight)


-def merge_lora_models(models, ratios, merge_dtype):
+def merge_lora_models(models, ratios, merge_dtype, concat=False, shuffle=False):
    base_alphas = {}  # alpha for merged model
    base_dims = {}

@@ -158,6 +158,12 @@ def merge_lora_models(models, ratios, merge_dtype):
        for key in lora_sd.keys():
            if "alpha" in key:
                continue
+            if "lora_up" in key and concat:
+                concat_dim = 1
+            elif "lora_down" in key and concat:
+                concat_dim = 0
+            else:
+                concat_dim = None

            lora_module_name = key[: key.rfind(".lora_")]

@@ -165,12 +171,16 @@ def merge_lora_models(models, ratios, merge_dtype):
            alpha = alphas[lora_module_name]

            scale = math.sqrt(alpha / base_alpha) * ratio
+            scale = abs(scale) if "lora_up" in key else scale # マイナスの重みに対応する。

            if key in merged_sd:
                assert (
-                    merged_sd[key].size() == lora_sd[key].size()
+                    merged_sd[key].size() == lora_sd[key].size() or concat_dim is not None
                ), f"weights shape mismatch merging v1 and v2, different dims? / 重みのサイズが合いません。v1とv2、または次元数の異なるモデルはマージできません"
-                merged_sd[key] = merged_sd[key] + lora_sd[key] * scale
+                if concat_dim is not None:
+                    merged_sd[key] = torch.cat([merged_sd[key], lora_sd[key] * scale], dim=concat_dim)
+                else:
+                    merged_sd[key] = merged_sd[key] + lora_sd[key] * scale
            else:
                merged_sd[key] = lora_sd[key] * scale

@@ -178,6 +188,13 @@ def merge_lora_models(models, ratios, merge_dtype):
    for lora_module_name, alpha in base_alphas.items():
        key = lora_module_name + ".alpha"
        merged_sd[key] = torch.tensor(alpha)
+        if shuffle:
+            key_down = lora_module_name + ".lora_down.weight"
+            key_up = lora_module_name + ".lora_up.weight"
+            dim = merged_sd[key_down].shape[0]
+            perm = torch.randperm(dim)
+            merged_sd[key_down] = merged_sd[key_down][perm]
+            merged_sd[key_up] = merged_sd[key_up][:,perm]

    print("merged model")
    print(f"dim: {list(set(base_dims.values()))}, alpha: {list(set(base_alphas.values()))}")
@@ -256,7 +273,7 @@ def merge(args):
            args.v2, args.save_to, text_encoder, unet, args.sd_model, 0, 0, sai_metadata, save_dtype, vae
        )
    else:
-        state_dict, metadata, v2 = merge_lora_models(args.models, args.ratios, merge_dtype)
+        state_dict, metadata, v2 = merge_lora_models(args.models, args.ratios, merge_dtype, args.concat, args.shuffle)

        print(f"calculating hashes and creating metadata...")

@@ -317,7 +334,19 @@ def setup_parser() -> argparse.ArgumentParser:
        help="do not save sai modelspec metadata (minimum ss_metadata for LoRA is saved) / "
        + "sai modelspecのメタデータを保存しない（LoRAの最低限のss_metadataは保存される）",
    )
-
+    parser.add_argument(
+        "--concat",
+        action="store_true",
+        help="concat lora instead of merge (The dim(rank) of the output LoRA is the sum of the input dims) / "
+        + "マージの代わりに結合する（LoRAのdim(rank)は入力dimの合計になる）",
+    )
+    parser.add_argument(
+        "--shuffle",
+        action="store_true",
+        help="shuffle lora weight./ "
+        + "LoRAの重みをシャッフルする",
+    )
+    
    return parser


--- a/networks/oft.py
+++ b/networks/oft.py
@@ -0,0 +1,430 @@
+# OFT network module
+
+import math
+import os
+from typing import Dict, List, Optional, Tuple, Type, Union
+from diffusers import AutoencoderKL
+from transformers import CLIPTextModel
+import numpy as np
+import torch
+import re
+
+
+RE_UPDOWN = re.compile(r"(up|down)_blocks_(\d+)_(resnets|upsamplers|downsamplers|attentions)_(\d+)_")
+
+
+class OFTModule(torch.nn.Module):
+    """
+    replaces forward method of the original Linear, instead of replacing the original Linear module.
+    """
+
+    def __init__(
+        self,
+        oft_name,
+        org_module: torch.nn.Module,
+        multiplier=1.0,
+        dim=4,
+        alpha=1,
+    ):
+        """
+        dim -> num blocks
+        alpha -> constraint
+        """
+        super().__init__()
+        self.oft_name = oft_name
+
+        self.num_blocks = dim
+
+        if "Linear" in org_module.__class__.__name__:
+            out_dim = org_module.out_features
+        elif "Conv" in org_module.__class__.__name__:
+            out_dim = org_module.out_channels
+
+        if type(alpha) == torch.Tensor:
+            alpha = alpha.detach().numpy()
+        self.constraint = alpha * out_dim
+        self.register_buffer("alpha", torch.tensor(alpha))
+
+        self.block_size = out_dim // self.num_blocks
+        self.oft_blocks = torch.nn.Parameter(torch.zeros(self.num_blocks, self.block_size, self.block_size))
+
+        self.out_dim = out_dim
+        self.shape = org_module.weight.shape
+
+        self.multiplier = multiplier
+        self.org_module = [org_module]  # moduleにならないようにlistに入れる
+
+    def apply_to(self):
+        self.org_forward = self.org_module[0].forward
+        self.org_module[0].forward = self.forward
+
+    def get_weight(self, multiplier=None):
+        if multiplier is None:
+            multiplier = self.multiplier
+
+        block_Q = self.oft_blocks - self.oft_blocks.transpose(1, 2)
+        norm_Q = torch.norm(block_Q.flatten())
+        new_norm_Q = torch.clamp(norm_Q, max=self.constraint)
+        block_Q = block_Q * ((new_norm_Q + 1e-8) / (norm_Q + 1e-8))
+        I = torch.eye(self.block_size, device=self.oft_blocks.device).unsqueeze(0).repeat(self.num_blocks, 1, 1)
+        block_R = torch.matmul(I + block_Q, (I - block_Q).inverse())
+
+        block_R_weighted = self.multiplier * block_R + (1 - self.multiplier) * I
+        R = torch.block_diag(*block_R_weighted)
+
+        return R
+
+    def forward(self, x, scale=None):
+        x = self.org_forward(x)
+        if self.multiplier == 0.0:
+            return x
+
+        R = self.get_weight().to(x.device, dtype=x.dtype)
+        if x.dim() == 4:
+            x = x.permute(0, 2, 3, 1)
+            x = torch.matmul(x, R)
+            x = x.permute(0, 3, 1, 2)
+        else:
+            x = torch.matmul(x, R)
+        return x
+
+
+class OFTInfModule(OFTModule):
+    def __init__(
+        self,
+        oft_name,
+        org_module: torch.nn.Module,
+        multiplier=1.0,
+        dim=4,
+        alpha=1,
+        **kwargs,
+    ):
+        # no dropout for inference
+        super().__init__(oft_name, org_module, multiplier, dim, alpha)
+        self.enabled = True
+        self.network: OFTNetwork = None
+
+    def set_network(self, network):
+        self.network = network
+
+    def forward(self, x, scale=None):
+        if not self.enabled:
+            return self.org_forward(x)
+        return super().forward(x, scale)
+
+    def merge_to(self, multiplier=None, sign=1):
+        R = self.get_weight(multiplier) * sign
+
+        # get org weight
+        org_sd = self.org_module[0].state_dict()
+        org_weight = org_sd["weight"]
+        R = R.to(org_weight.device, dtype=org_weight.dtype)
+
+        if org_weight.dim() == 4:
+            weight = torch.einsum("oihw, op -> pihw", org_weight, R)
+        else:
+            weight = torch.einsum("oi, op -> pi", org_weight, R)
+
+        # set weight to org_module
+        org_sd["weight"] = weight
+        self.org_module[0].load_state_dict(org_sd)
+
+
+def create_network(
+    multiplier: float,
+    network_dim: Optional[int],
+    network_alpha: Optional[float],
+    vae: AutoencoderKL,
+    text_encoder: Union[CLIPTextModel, List[CLIPTextModel]],
+    unet,
+    neuron_dropout: Optional[float] = None,
+    **kwargs,
+):
+    if network_dim is None:
+        network_dim = 4  # default
+    if network_alpha is None:
+        network_alpha = 1.0
+
+    enable_all_linear = kwargs.get("enable_all_linear", None)
+    enable_conv = kwargs.get("enable_conv", None)
+    if enable_all_linear is not None:
+        enable_all_linear = bool(enable_all_linear)
+    if enable_conv is not None:
+        enable_conv = bool(enable_conv)
+
+    network = OFTNetwork(
+        text_encoder,
+        unet,
+        multiplier=multiplier,
+        dim=network_dim,
+        alpha=network_alpha,
+        enable_all_linear=enable_all_linear,
+        enable_conv=enable_conv,
+        varbose=True,
+    )
+    return network
+
+
+# Create network from weights for inference, weights are not loaded here (because can be merged)
+def create_network_from_weights(multiplier, file, vae, text_encoder, unet, weights_sd=None, for_inference=False, **kwargs):
+    if weights_sd is None:
+        if os.path.splitext(file)[1] == ".safetensors":
+            from safetensors.torch import load_file, safe_open
+
+            weights_sd = load_file(file)
+        else:
+            weights_sd = torch.load(file, map_location="cpu")
+
+    # check dim, alpha and if weights have for conv2d
+    dim = None
+    alpha = None
+    has_conv2d = None
+    all_linear = None
+    for name, param in weights_sd.items():
+        if name.endswith(".alpha"):
+            if alpha is None:
+                alpha = param.item()
+        else:
+            if dim is None:
+                dim = param.size()[0]
+            if has_conv2d is None and param.dim() == 4:
+                has_conv2d = True
+            if all_linear is None:
+                if param.dim() == 3 and "attn" not in name:
+                    all_linear = True
+        if dim is not None and alpha is not None and has_conv2d is not None:
+            break
+    if has_conv2d is None:
+        has_conv2d = False
+    if all_linear is None:
+        all_linear = False
+
+    module_class = OFTInfModule if for_inference else OFTModule
+    network = OFTNetwork(
+        text_encoder,
+        unet,
+        multiplier=multiplier,
+        dim=dim,
+        alpha=alpha,
+        enable_all_linear=all_linear,
+        enable_conv=has_conv2d,
+        module_class=module_class,
+    )
+    return network, weights_sd
+
+
+class OFTNetwork(torch.nn.Module):
+    UNET_TARGET_REPLACE_MODULE_ATTN_ONLY = ["CrossAttention"]
+    UNET_TARGET_REPLACE_MODULE_ALL_LINEAR = ["Transformer2DModel"]
+    UNET_TARGET_REPLACE_MODULE_CONV2D_3X3 = ["ResnetBlock2D", "Downsample2D", "Upsample2D"]
+    OFT_PREFIX_UNET = "oft_unet"  # これ変えないほうがいいかな
+
+    def __init__(
+        self,
+        text_encoder: Union[List[CLIPTextModel], CLIPTextModel],
+        unet,
+        multiplier: float = 1.0,
+        dim: int = 4,
+        alpha: float = 1,
+        enable_all_linear: Optional[bool] = False,
+        enable_conv: Optional[bool] = False,
+        module_class: Type[object] = OFTModule,
+        varbose: Optional[bool] = False,
+    ) -> None:
+        super().__init__()
+        self.multiplier = multiplier
+
+        self.dim = dim
+        self.alpha = alpha
+
+        print(
+            f"create OFT network. num blocks: {self.dim}, constraint: {self.alpha}, multiplier: {self.multiplier}, enable_conv: {enable_conv}"
+        )
+
+        # create module instances
+        def create_modules(
+            root_module: torch.nn.Module,
+            target_replace_modules: List[torch.nn.Module],
+        ) -> List[OFTModule]:
+            prefix = self.OFT_PREFIX_UNET
+            ofts = []
+            for name, module in root_module.named_modules():
+                if module.__class__.__name__ in target_replace_modules:
+                    for child_name, child_module in module.named_modules():
+                        is_linear = "Linear" in child_module.__class__.__name__
+                        is_conv2d = "Conv2d" in child_module.__class__.__name__
+                        is_conv2d_1x1 = is_conv2d and child_module.kernel_size == (1, 1)
+
+                        if is_linear or is_conv2d_1x1 or (is_conv2d and enable_conv):
+                            oft_name = prefix + "." + name + "." + child_name
+                            oft_name = oft_name.replace(".", "_")
+                            # print(oft_name)
+
+                            oft = module_class(
+                                oft_name,
+                                child_module,
+                                self.multiplier,
+                                dim,
+                                alpha,
+                            )
+                            ofts.append(oft)
+            return ofts
+
+        # extend U-Net target modules if conv2d 3x3 is enabled, or load from weights
+        if enable_all_linear:
+            target_modules = OFTNetwork.UNET_TARGET_REPLACE_MODULE_ALL_LINEAR
+        else:
+            target_modules = OFTNetwork.UNET_TARGET_REPLACE_MODULE_ATTN_ONLY
+        if enable_conv:
+            target_modules += OFTNetwork.UNET_TARGET_REPLACE_MODULE_CONV2D_3X3
+
+        self.unet_ofts: List[OFTModule] = create_modules(unet, target_modules)
+        print(f"create OFT for U-Net: {len(self.unet_ofts)} modules.")
+
+        # assertion
+        names = set()
+        for oft in self.unet_ofts:
+            assert oft.oft_name not in names, f"duplicated oft name: {oft.oft_name}"
+            names.add(oft.oft_name)
+
+    def set_multiplier(self, multiplier):
+        self.multiplier = multiplier
+        for oft in self.unet_ofts:
+            oft.multiplier = self.multiplier
+
+    def load_weights(self, file):
+        if os.path.splitext(file)[1] == ".safetensors":
+            from safetensors.torch import load_file
+
+            weights_sd = load_file(file)
+        else:
+            weights_sd = torch.load(file, map_location="cpu")
+
+        info = self.load_state_dict(weights_sd, False)
+        return info
+
+    def apply_to(self, text_encoder, unet, apply_text_encoder=True, apply_unet=True):
+        assert apply_unet, "apply_unet must be True"
+
+        for oft in self.unet_ofts:
+            oft.apply_to()
+            self.add_module(oft.oft_name, oft)
+
+    # マージできるかどうかを返す
+    def is_mergeable(self):
+        return True
+
+    # TODO refactor to common function with apply_to
+    def merge_to(self, text_encoder, unet, weights_sd, dtype, device):
+        print("enable OFT for U-Net")
+
+        for oft in self.unet_ofts:
+            sd_for_lora = {}
+            for key in weights_sd.keys():
+                if key.startswith(oft.oft_name):
+                    sd_for_lora[key[len(oft.oft_name) + 1 :]] = weights_sd[key]
+            oft.load_state_dict(sd_for_lora, False)
+            oft.merge_to()
+
+        print(f"weights are merged")
+
+    # 二つのText Encoderに別々の学習率を設定できるようにするといいかも
+    def prepare_optimizer_params(self, text_encoder_lr, unet_lr, default_lr):
+        self.requires_grad_(True)
+        all_params = []
+
+        def enumerate_params(ofts):
+            params = []
+            for oft in ofts:
+                params.extend(oft.parameters())
+
+            # print num of params
+            num_params = 0
+            for p in params:
+                num_params += p.numel()
+            print(f"OFT params: {num_params}")
+            return params
+
+        param_data = {"params": enumerate_params(self.unet_ofts)}
+        if unet_lr is not None:
+            param_data["lr"] = unet_lr
+        all_params.append(param_data)
+
+        return all_params
+
+    def enable_gradient_checkpointing(self):
+        # not supported
+        pass
+
+    def prepare_grad_etc(self, text_encoder, unet):
+        self.requires_grad_(True)
+
+    def on_epoch_start(self, text_encoder, unet):
+        self.train()
+
+    def get_trainable_params(self):
+        return self.parameters()
+
+    def save_weights(self, file, dtype, metadata):
+        if metadata is not None and len(metadata) == 0:
+            metadata = None
+
+        state_dict = self.state_dict()
+
+        if dtype is not None:
+            for key in list(state_dict.keys()):
+                v = state_dict[key]
+                v = v.detach().clone().to("cpu").to(dtype)
+                state_dict[key] = v
+
+        if os.path.splitext(file)[1] == ".safetensors":
+            from safetensors.torch import save_file
+            from library import train_util
+
+            # Precalculate model hashes to save time on indexing
+            if metadata is None:
+                metadata = {}
+            model_hash, legacy_hash = train_util.precalculate_safetensors_hashes(state_dict, metadata)
+            metadata["sshs_model_hash"] = model_hash
+            metadata["sshs_legacy_hash"] = legacy_hash
+
+            save_file(state_dict, file, metadata)
+        else:
+            torch.save(state_dict, file)
+
+    def backup_weights(self):
+        # 重みのバックアップを行う
+        ofts: List[OFTInfModule] = self.unet_ofts
+        for oft in ofts:
+            org_module = oft.org_module[0]
+            if not hasattr(org_module, "_lora_org_weight"):
+                sd = org_module.state_dict()
+                org_module._lora_org_weight = sd["weight"].detach().clone()
+                org_module._lora_restored = True
+
+    def restore_weights(self):
+        # 重みのリストアを行う
+        ofts: List[OFTInfModule] = self.unet_ofts
+        for oft in ofts:
+            org_module = oft.org_module[0]
+            if not org_module._lora_restored:
+                sd = org_module.state_dict()
+                sd["weight"] = org_module._lora_org_weight
+                org_module.load_state_dict(sd)
+                org_module._lora_restored = True
+
+    def pre_calculation(self):
+        # 事前計算を行う
+        ofts: List[OFTInfModule] = self.unet_ofts
+        for oft in ofts:
+            org_module = oft.org_module[0]
+            oft.merge_to()
+            # sd = org_module.state_dict()
+            # org_weight = sd["weight"]
+            # lora_weight = oft.get_weight().to(org_weight.device, dtype=org_weight.dtype)
+            # sd["weight"] = org_weight + lora_weight
+            # assert sd["weight"].shape == org_weight.shape
+            # org_module.load_state_dict(sd)
+
+            org_module._lora_restored = False
+            oft.enabled = False
--- a/networks/resize_lora.py
+++ b/networks/resize_lora.py
@@ -219,8 +219,8 @@ 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.split(".")[0]
-        weight_name = key.split(".")[-1]
+        block_down_name = key.rsplit('.lora_down', 1)[0]
+        weight_name = key.rsplit(".", 1)[-1]
        lora_down_weight = value
      else:
        continue
@@ -283,7 +283,10 @@ def resize_lora_model(lora_sd, new_rank, save_dtype, device, dynamic_method, dyn


 def resize(args):
+  if args.save_to is None or not (args.save_to.endswith('.ckpt') or args.save_to.endswith('.pt') or args.save_to.endswith('.pth') or args.save_to.endswith('.safetensors')):
+    raise Exception("The --save_to argument must be specified and must be a .ckpt , .pt, .pth or .safetensors file.")

+    
  def str_to_dtype(p):
    if p == 'float':
      return torch.float
--- a/networks/sdxl_merge_lora.py
+++ b/networks/sdxl_merge_lora.py
@@ -113,7 +113,7 @@ def merge_to_sd_model(text_encoder1, text_encoder2, unet, models, ratios, merge_
                module.weight = torch.nn.Parameter(weight)


-def merge_lora_models(models, ratios, merge_dtype):
+def merge_lora_models(models, ratios, merge_dtype, concat=False, shuffle=False):
    base_alphas = {}  # alpha for merged model
    base_dims = {}

@@ -161,6 +161,13 @@ def merge_lora_models(models, ratios, merge_dtype):
        for key in tqdm(lora_sd.keys()):
            if "alpha" in key:
                continue
+            
+            if "lora_up" in key and concat:
+                concat_dim = 1
+            elif "lora_down" in key and concat:
+                concat_dim = 0
+            else:
+                concat_dim = None

            lora_module_name = key[: key.rfind(".lora_")]

@@ -168,12 +175,16 @@ def merge_lora_models(models, ratios, merge_dtype):
            alpha = alphas[lora_module_name]

            scale = math.sqrt(alpha / base_alpha) * ratio
-
+            scale = abs(scale) if "lora_up" in key else scale # マイナスの重みに対応する。
+            
            if key in merged_sd:
                assert (
-                    merged_sd[key].size() == lora_sd[key].size()
+                    merged_sd[key].size() == lora_sd[key].size() or concat_dim is not None
                ), f"weights shape mismatch merging v1 and v2, different dims? / 重みのサイズが合いません。v1とv2、または次元数の異なるモデルはマージできません"
-                merged_sd[key] = merged_sd[key] + lora_sd[key] * scale
+                if concat_dim is not None:
+                    merged_sd[key] = torch.cat([merged_sd[key], lora_sd[key] * scale], dim=concat_dim)
+                else:
+                    merged_sd[key] = merged_sd[key] + lora_sd[key] * scale
            else:
                merged_sd[key] = lora_sd[key] * scale

@@ -181,6 +192,13 @@ def merge_lora_models(models, ratios, merge_dtype):
    for lora_module_name, alpha in base_alphas.items():
        key = lora_module_name + ".alpha"
        merged_sd[key] = torch.tensor(alpha)
+        if shuffle:
+            key_down = lora_module_name + ".lora_down.weight"
+            key_up = lora_module_name + ".lora_up.weight"
+            dim = merged_sd[key_down].shape[0]
+            perm = torch.randperm(dim)
+            merged_sd[key_down] = merged_sd[key_down][perm]
+            merged_sd[key_up] = merged_sd[key_up][:,perm]

    print("merged model")
    print(f"dim: {list(set(base_dims.values()))}, alpha: {list(set(base_alphas.values()))}")
@@ -252,7 +270,7 @@ def merge(args):
            args.save_to, text_model1, text_model2, unet, 0, 0, ckpt_info, vae, logit_scale, sai_metadata, save_dtype
        )
    else:
-        state_dict, metadata = merge_lora_models(args.models, args.ratios, merge_dtype)
+        state_dict, metadata = merge_lora_models(args.models, args.ratios, merge_dtype, args.concat, args.shuffle)

        print(f"calculating hashes and creating metadata...")

@@ -307,6 +325,18 @@ def setup_parser() -> argparse.ArgumentParser:
        help="do not save sai modelspec metadata (minimum ss_metadata for LoRA is saved) / "
        + "sai modelspecのメタデータを保存しない（LoRAの最低限のss_metadataは保存される）",
    )
+    parser.add_argument(
+        "--concat",
+        action="store_true",
+        help="concat lora instead of merge (The dim(rank) of the output LoRA is the sum of the input dims) / "
+        + "マージの代わりに結合する（LoRAのdim(rank)は入力dimの合計になる）",
+    )
+    parser.add_argument(
+        "--shuffle",
+        action="store_true",
+        help="shuffle lora weight./ "
+        + "LoRAの重みをシャッフルする",
+    )

    return parser

--- a/requirements.txt
+++ b/requirements.txt
@@ -1,6 +1,6 @@
-accelerate==0.19.0
+accelerate==0.23.0
 transformers==4.30.2
-diffusers[torch]==0.18.2
+diffusers[torch]==0.21.2
 ftfy==6.1.1
 # albumentations==1.3.0
 opencv-python==4.7.0.68
@@ -15,14 +15,18 @@ easygui==0.98.3
 toml==0.10.2
 voluptuous==0.13.1
 huggingface-hub==0.15.1
-# for loading Diffusers' SDXL
-invisible-watermark==0.2.0 
 # for BLIP captioning
 # requests==2.28.2
 # timm==0.6.12
 # fairscale==0.4.13
-# for WD14 captioning
+# for WD14 captioning (tensorflow)
 # tensorflow==2.10.1
+# for WD14 captioning (onnx)
+# onnx==1.14.1
+# onnxruntime-gpu==1.16.0
+# onnxruntime==1.16.0
+# this is for onnx: 
+# protobuf==3.20.3
 # open clip for SDXL
 open-clip-torch==2.20.0
 # for kohya_ss library
--- a/sdxl_gen_img.py
+++ b/sdxl_gen_img.py
@@ -17,6 +17,16 @@ import re
 import diffusers
 import numpy as np
 import torch
+
+try:
+    import intel_extension_for_pytorch as ipex
+
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+
+        ipex_init()
+except Exception:
+    pass
 import torchvision
 from diffusers import (
    AutoencoderKL,
@@ -1521,22 +1531,26 @@ def main(args):
    text_encoder2.to(dtype).to(device)
    unet.to(dtype).to(device)

-    # freeU
-    # unet.set_free_u_enabled(False, 1.0, 1.0, 0)
-    unet.set_free_u_enabled(True, 1.1, 1.2, 0.9, 0.2)
-
    # networkを組み込む
    if args.network_module:
        networks = []
        network_default_muls = []
        network_pre_calc = args.network_pre_calc

+        # merge関連の引数を統合する
+        if args.network_merge:
+            network_merge = len(args.network_module)  # all networks are merged
+        elif args.network_merge_n_models:
+            network_merge = args.network_merge_n_models
+        else:
+            network_merge = 0
+        print(f"network_merge: {network_merge}")
+
        for i, network_module in enumerate(args.network_module):
            print("import network module:", network_module)
            imported_module = importlib.import_module(network_module)

            network_mul = 1.0 if args.network_mul is None or len(args.network_mul) <= i else args.network_mul[i]
-            network_default_muls.append(network_mul)

            net_kwargs = {}
            if args.network_args and i < len(args.network_args):
@@ -1547,31 +1561,32 @@ def main(args):
                    key, value = net_arg.split("=")
                    net_kwargs[key] = value

-            if args.network_weights and i < len(args.network_weights):
-                network_weight = args.network_weights[i]
-                print("load network weights from:", network_weight)
-
-                if model_util.is_safetensors(network_weight) and args.network_show_meta:
-                    from safetensors.torch import safe_open
-
-                    with safe_open(network_weight, framework="pt") as f:
-                        metadata = f.metadata()
-                    if metadata is not None:
-                        print(f"metadata for: {network_weight}: {metadata}")
-
-                network, weights_sd = imported_module.create_network_from_weights(
-                    network_mul, network_weight, vae, [text_encoder1, text_encoder2], unet, for_inference=True, **net_kwargs
-                )
-            else:
+            if args.network_weights is None or len(args.network_weights) <= i:
                raise ValueError("No weight. Weight is required.")
+
+            network_weight = args.network_weights[i]
+            print("load network weights from:", network_weight)
+
+            if model_util.is_safetensors(network_weight) and args.network_show_meta:
+                from safetensors.torch import safe_open
+
+                with safe_open(network_weight, framework="pt") as f:
+                    metadata = f.metadata()
+                if metadata is not None:
+                    print(f"metadata for: {network_weight}: {metadata}")
+
+            network, weights_sd = imported_module.create_network_from_weights(
+                network_mul, network_weight, vae, [text_encoder1, text_encoder2], unet, for_inference=True, **net_kwargs
+            )
            if network is None:
                return

            mergeable = network.is_mergeable()
-            if args.network_merge and not mergeable:
+            if network_merge and not mergeable:
                print("network is not mergiable. ignore merge option.")

-            if not args.network_merge or not mergeable:
+            if not mergeable or i >= network_merge:
+                # not merging
                network.apply_to([text_encoder1, text_encoder2], unet)
                info = network.load_state_dict(weights_sd, False)  # network.load_weightsを使うようにするとよい
                print(f"weights are loaded: {info}")
@@ -1585,6 +1600,7 @@ def main(args):
                    network.backup_weights()

                networks.append(network)
+                network_default_muls.append(network_mul)
            else:
                network.merge_to([text_encoder1, text_encoder2], unet, weights_sd, dtype, device)

@@ -1861,9 +1877,18 @@ def main(args):

        size = None
        for i, network in enumerate(networks):
-            if i < 3:
+            if (i < 3 and args.network_regional_mask_max_color_codes is None) or i < args.network_regional_mask_max_color_codes:
                np_mask = np.array(mask_images[0])
-                np_mask = np_mask[:, :, i]
+
+                if args.network_regional_mask_max_color_codes:
+                    # カラーコードでマスクを指定する
+                    ch0 = (i + 1) & 1
+                    ch1 = ((i + 1) >> 1) & 1
+                    ch2 = ((i + 1) >> 2) & 1
+                    np_mask = np.all(np_mask == np.array([ch0, ch1, ch2]) * 255, axis=2)
+                    np_mask = np_mask.astype(np.uint8) * 255
+                else:
+                    np_mask = np_mask[:, :, i]
                size = np_mask.shape
            else:
                np_mask = np.full(size, 255, dtype=np.uint8)
@@ -2609,13 +2634,22 @@ def setup_parser() -> argparse.ArgumentParser:
    )
    parser.add_argument("--network_mul", type=float, default=None, nargs="*", help="additional network multiplier / 追加ネットワークの効果の倍率")
    parser.add_argument(
-        "--network_args", type=str, default=None, nargs="*", help="additional argmuments for network (key=value) / ネットワークへの追加の引数"
+        "--network_args", type=str, default=None, nargs="*", help="additional arguments for network (key=value) / ネットワークへの追加の引数"
    )
    parser.add_argument("--network_show_meta", action="store_true", help="show metadata of network model / ネットワークモデルのメタデータを表示する")
+    parser.add_argument(
+        "--network_merge_n_models", type=int, default=None, help="merge this number of networks / この数だけネットワークをマージする"
+    )
    parser.add_argument("--network_merge", action="store_true", help="merge network weights to original model / ネットワークの重みをマージする")
    parser.add_argument(
        "--network_pre_calc", action="store_true", help="pre-calculate network for generation / ネットワークのあらかじめ計算して生成する"
    )
+    parser.add_argument(
+        "--network_regional_mask_max_color_codes",
+        type=int,
+        default=None,
+        help="max color codes for regional mask (default is None, mask by channel) / regional maskの最大色数（デフォルトはNoneでチャンネルごとのマスク）",
+    )
    parser.add_argument(
        "--textual_inversion_embeddings",
        type=str,
@@ -2628,7 +2662,7 @@ def setup_parser() -> argparse.ArgumentParser:
        "--max_embeddings_multiples",
        type=int,
        default=None,
-        help="max embeding multiples, max token length is 75 * multiples / トークン長をデフォルトの何倍とするか 75*この値 がトークン長となる",
+        help="max embedding multiples, max token length is 75 * multiples / トークン長をデフォルトの何倍とするか 75*この値 がトークン長となる",
    )
    parser.add_argument(
        "--guide_image_path", type=str, default=None, nargs="*", help="image to CLIP guidance / CLIP guided SDでガイドに使う画像"
@@ -2663,7 +2697,7 @@ def setup_parser() -> argparse.ArgumentParser:
        "--highres_fix_upscaler_args",
        type=str,
        default=None,
-        help="additional argmuments for upscaler (key=value) / upscalerへの追加の引数",
+        help="additional arguments for upscaler (key=value) / upscalerへの追加の引数",
    )
    parser.add_argument(
        "--highres_fix_disable_control_net",
--- a/sdxl_minimal_inference.py
+++ b/sdxl_minimal_inference.py
@@ -9,6 +9,13 @@ import random
 from einops import repeat
 import numpy as np
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from tqdm import tqdm
 from transformers import CLIPTokenizer
 from diffusers import EulerDiscreteScheduler
@@ -94,7 +101,7 @@ if __name__ == "__main__":
        type=str,
        nargs="*",
        default=[],
-        help="LoRA weights, only supports networks.lora, each arguement is a `path;multiplier` (semi-colon separated)",
+        help="LoRA weights, only supports networks.lora, each argument is a `path;multiplier` (semi-colon separated)",
    )
    parser.add_argument("--interactive", action="store_true")
    args = parser.parse_args()
--- a/sdxl_train.py
+++ b/sdxl_train.py
@@ -10,6 +10,13 @@ import toml

 from tqdm import tqdm
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from accelerate.utils import set_seed
 from diffusers import DDPMScheduler
 from library import sdxl_model_util
@@ -165,8 +172,8 @@ def train(args):

    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    train_dataset_group.verify_bucket_reso_steps(32)

@@ -341,7 +348,7 @@ def train(args):
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
--- a/sdxl_train_control_net_lllite.py
+++ b/sdxl_train_control_net_lllite.py
@@ -1,3 +1,6 @@
+# cond_imageをU-Netのforwardで渡すバージョンのControlNet-LLLite検証用学習コード
+# training code for ControlNet-LLLite with passing cond_image to U-Net's forward
+
 import argparse
 import gc
 import json
@@ -11,8 +14,16 @@ import toml

 from tqdm import tqdm
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from torch.nn.parallel import DistributedDataParallel as DDP
 from accelerate.utils import set_seed
+import accelerate
 from diffusers import DDPMScheduler, ControlNetModel
 from safetensors.torch import load_file
 from library import sai_model_spec, sdxl_model_util, sdxl_original_unet, sdxl_train_util
@@ -34,7 +45,7 @@ from library.custom_train_functions import (
    apply_noise_offset,
    scale_v_prediction_loss_like_noise_prediction,
 )
-import networks.control_net_lllite as control_net_lllite
+import networks.control_net_lllite_for_train as control_net_lllite_for_train


 # TODO 他のスクリプトと共通化する
@@ -95,8 +106,8 @@ def train(args):

    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    train_dataset_group.verify_bucket_reso_steps(32)

@@ -141,9 +152,6 @@ def train(args):
        ckpt_info,
    ) = sdxl_train_util.load_target_model(args, accelerator, sdxl_model_util.MODEL_VERSION_SDXL_BASE_V1_0, weight_dtype)

-    # モデルに xformers とか memory efficient attention を組み込む
-    train_util.replace_unet_modules(unet, args.mem_eff_attn, args.xformers, args.sdpa)
-
    # 学習を準備する
    if cache_latents:
        vae.to(accelerator.device, dtype=vae_dtype)
@@ -177,22 +185,53 @@ def train(args):
            )
        accelerator.wait_for_everyone()

-    # prepare ControlNet
-    network = control_net_lllite.ControlNetLLLite(unet, args.cond_emb_dim, args.network_dim, args.network_dropout)
-    network.apply_to()
+    # prepare ControlNet-LLLite
+    control_net_lllite_for_train.replace_unet_linear_and_conv2d()

    if args.network_weights is not None:
-        info = network.load_weights(args.network_weights)
-        accelerator.print(f"load ControlNet weights from {args.network_weights}: {info}")
+        accelerator.print(f"initialize U-Net with ControlNet-LLLite")
+        with accelerate.init_empty_weights():
+            unet_lllite = control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite()
+        unet_lllite.to(accelerator.device, dtype=weight_dtype)
+
+        unet_sd = unet.state_dict()
+        info = unet_lllite.load_lllite_weights(args.network_weights, unet_sd)
+        accelerator.print(f"load ControlNet-LLLite weights from {args.network_weights}: {info}")
+    else:
+        # cosumes large memory, so send to GPU before creating the LLLite model
+        accelerator.print("sending U-Net to GPU")
+        unet.to(accelerator.device, dtype=weight_dtype)
+        unet_sd = unet.state_dict()
+
+        # init LLLite weights
+        accelerator.print(f"initialize U-Net with ControlNet-LLLite")
+
+        if args.lowram:
+            with accelerate.init_on_device(accelerator.device):
+                unet_lllite = control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite()
+        else:
+            unet_lllite = control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite()
+        unet_lllite.to(weight_dtype)
+
+        info = unet_lllite.load_lllite_weights(None, unet_sd)
+        accelerator.print(f"init U-Net with ControlNet-LLLite weights: {info}")
+    del unet_sd, unet
+
+    unet: control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite = unet_lllite
+    del unet_lllite
+
+    unet.apply_lllite(args.cond_emb_dim, args.network_dim, args.network_dropout)
+
+    # モデルに xformers とか memory efficient attention を組み込む
+    train_util.replace_unet_modules(unet, args.mem_eff_attn, args.xformers, args.sdpa)

    if args.gradient_checkpointing:
        unet.enable_gradient_checkpointing()
-        network.enable_gradient_checkpointing()  # may have no effect

    # 学習に必要なクラスを準備する
    accelerator.print("prepare optimizer, data loader etc.")

-    trainable_params = list(network.prepare_optimizer_params())
+    trainable_params = list(unet.prepare_params())
    print(f"trainable params count: {len(trainable_params)}")
    print(f"number of trainable parameters: {sum(p.numel() for p in trainable_params if p.requires_grad)}")

@@ -206,7 +245,7 @@ def train(args):
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
@@ -225,37 +264,32 @@ def train(args):
    lr_scheduler = train_util.get_scheduler_fix(args, optimizer, accelerator.num_processes)

    # 実験的機能：勾配も含めたfp16/bf16学習を行う　モデル全体をfp16/bf16にする
-    if args.full_fp16:
-        assert (
-            args.mixed_precision == "fp16"
-        ), "full_fp16 requires mixed precision='fp16' / full_fp16を使う場合はmixed_precision='fp16'を指定してください。"
-        accelerator.print("enable full fp16 training.")
-        unet.to(weight_dtype)
-        network.to(weight_dtype)
-    elif args.full_bf16:
-        assert (
-            args.mixed_precision == "bf16"
-        ), "full_bf16 requires mixed precision='bf16' / full_bf16を使う場合はmixed_precision='bf16'を指定してください。"
-        accelerator.print("enable full bf16 training.")
-        unet.to(weight_dtype)
-        network.to(weight_dtype)
+    # if args.full_fp16:
+    #     assert (
+    #         args.mixed_precision == "fp16"
+    #     ), "full_fp16 requires mixed precision='fp16' / full_fp16を使う場合はmixed_precision='fp16'を指定してください。"
+    #     accelerator.print("enable full fp16 training.")
+    #     unet.to(weight_dtype)
+    # elif args.full_bf16:
+    #     assert (
+    #         args.mixed_precision == "bf16"
+    #     ), "full_bf16 requires mixed precision='bf16' / full_bf16を使う場合はmixed_precision='bf16'を指定してください。"
+    #     accelerator.print("enable full bf16 training.")
+    #     unet.to(weight_dtype)
+
+    unet.to(weight_dtype)

    # acceleratorがなんかよろしくやってくれるらしい
-    unet, network, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(
-        unet, network, optimizer, train_dataloader, lr_scheduler
-    )
-    network: control_net_lllite.ControlNetLLLite
+    unet, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(unet, optimizer, train_dataloader, lr_scheduler)

    # transform DDP after prepare (train_network here only)
-    unet, network = train_util.transform_models_if_DDP([unet, network])
+    unet = train_util.transform_models_if_DDP([unet])[0]

    if args.gradient_checkpointing:
        unet.train()  # according to TI example in Diffusers, train is required -> これオリジナルのU-Netしたので本当は外せる
    else:
        unet.eval()

-    network.prepare_grad_etc()
-
    # TextEncoderの出力をキャッシュするときにはCPUへ移動する
    if args.cache_text_encoder_outputs:
        # move Text Encoders for sampling images. Text Encoder doesn't work on CPU with fp16
@@ -321,7 +355,13 @@ def train(args):
    del train_dataset_group

    # function for saving/removing
-    def save_model(ckpt_name, unwrapped_nw, steps, epoch_no, force_sync_upload=False):
+    def save_model(
+        ckpt_name,
+        unwrapped_nw: control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite,
+        steps,
+        epoch_no,
+        force_sync_upload=False,
+    ):
        os.makedirs(args.output_dir, exist_ok=True)
        ckpt_file = os.path.join(args.output_dir, ckpt_name)

@@ -329,7 +369,7 @@ def train(args):
        sai_metadata = train_util.get_sai_model_spec(None, args, True, True, False)
        sai_metadata["modelspec.architecture"] = sai_model_spec.ARCH_SD_XL_V1_BASE + "/control-net-lllite"

-        unwrapped_nw.save_weights(ckpt_file, save_dtype, sai_metadata)
+        unwrapped_nw.save_lllite_weights(ckpt_file, save_dtype, sai_metadata)
        if args.huggingface_repo_id is not None:
            huggingface_util.upload(args, ckpt_file, "/" + ckpt_name, force_sync_upload=force_sync_upload)

@@ -344,11 +384,9 @@ def train(args):
        accelerator.print(f"\nepoch {epoch+1}/{num_train_epochs}")
        current_epoch.value = epoch + 1

-        network.on_epoch_start()  # train()
-
        for step, batch in enumerate(train_dataloader):
            current_step.value = global_step
-            with accelerator.accumulate(network):
+            with accelerator.accumulate(unet):
                with torch.no_grad():
                    if "latents" in batch and batch["latents"] is not None:
                        latents = batch["latents"].to(accelerator.device)
@@ -405,10 +443,9 @@ def train(args):
                with accelerator.autocast():
                    # conditioning imageをControlNetに渡す / pass conditioning image to ControlNet
                    # 内部でcond_embに変換される / it will be converted to cond_emb inside
-                    network.set_cond_image(controlnet_image)

                    # それらの値を使いつつ、U-Netでノイズを予測する / predict noise with U-Net using those values
-                    noise_pred = unet(noisy_latents, timesteps, text_embedding, vector_embedding)
+                    noise_pred = unet(noisy_latents, timesteps, text_embedding, vector_embedding, controlnet_image)

                if args.v_parameterization:
                    # v-parameterization training
@@ -433,7 +470,7 @@ def train(args):

                accelerator.backward(loss)
                if accelerator.sync_gradients and args.max_grad_norm != 0.0:
-                    params_to_clip = network.get_trainable_params()
+                    params_to_clip = unet.get_trainable_params()
                    accelerator.clip_grad_norm_(params_to_clip, args.max_grad_norm)

                optimizer.step()
@@ -452,7 +489,7 @@ def train(args):
                    accelerator.wait_for_everyone()
                    if accelerator.is_main_process:
                        ckpt_name = train_util.get_step_ckpt_name(args, "." + args.save_model_as, global_step)
-                        save_model(ckpt_name, accelerator.unwrap_model(network), global_step, epoch)
+                        save_model(ckpt_name, accelerator.unwrap_model(unet), global_step, epoch)

                        if args.save_state:
                            train_util.save_and_remove_state_stepwise(args, accelerator, global_step)
@@ -491,7 +528,7 @@ def train(args):
            saving = (epoch + 1) % args.save_every_n_epochs == 0 and (epoch + 1) < num_train_epochs
            if is_main_process and saving:
                ckpt_name = train_util.get_epoch_ckpt_name(args, "." + args.save_model_as, epoch + 1)
-                save_model(ckpt_name, accelerator.unwrap_model(network), global_step, epoch + 1)
+                save_model(ckpt_name, accelerator.unwrap_model(unet), global_step, epoch + 1)

                remove_epoch_no = train_util.get_remove_epoch_no(args, epoch + 1)
                if remove_epoch_no is not None:
@@ -506,7 +543,7 @@ def train(args):
        # end of epoch

    if is_main_process:
-        network = accelerator.unwrap_model(network)
+        unet = accelerator.unwrap_model(unet)

    accelerator.end_training()

@@ -515,7 +552,7 @@ def train(args):

    if is_main_process:
        ckpt_name = train_util.get_last_ckpt_name(args, "." + args.save_model_as)
-        save_model(ckpt_name, network, global_step, num_train_epochs, force_sync_upload=True)
+        save_model(ckpt_name, unet, global_step, num_train_epochs, force_sync_upload=True)

        print("model saved.")

--- a/sdxl_train_control_net_lllite_old.py
+++ b/sdxl_train_control_net_lllite_old.py
@@ -1,6 +1,3 @@
-# cond_imageをU-Netのforwardで渡すバージョンのControlNet-LLLite検証用学習コード
-# training code for ControlNet-LLLite with passing cond_image to U-Net's forward
-
 import argparse
 import gc
 import json
@@ -14,9 +11,15 @@ import toml

 from tqdm import tqdm
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from torch.nn.parallel import DistributedDataParallel as DDP
 from accelerate.utils import set_seed
-import accelerate
 from diffusers import DDPMScheduler, ControlNetModel
 from safetensors.torch import load_file
 from library import sai_model_spec, sdxl_model_util, sdxl_original_unet, sdxl_train_util
@@ -38,7 +41,7 @@ from library.custom_train_functions import (
    apply_noise_offset,
    scale_v_prediction_loss_like_noise_prediction,
 )
-import networks.control_net_lllite_for_train as control_net_lllite_for_train
+import networks.control_net_lllite as control_net_lllite


 # TODO 他のスクリプトと共通化する
@@ -99,8 +102,8 @@ def train(args):

    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    train_dataset_group.verify_bucket_reso_steps(32)

@@ -145,6 +148,9 @@ def train(args):
        ckpt_info,
    ) = sdxl_train_util.load_target_model(args, accelerator, sdxl_model_util.MODEL_VERSION_SDXL_BASE_V1_0, weight_dtype)

+    # モデルに xformers とか memory efficient attention を組み込む
+    train_util.replace_unet_modules(unet, args.mem_eff_attn, args.xformers, args.sdpa)
+
    # 学習を準備する
    if cache_latents:
        vae.to(accelerator.device, dtype=vae_dtype)
@@ -178,53 +184,22 @@ def train(args):
            )
        accelerator.wait_for_everyone()

-    # prepare ControlNet-LLLite
-    control_net_lllite_for_train.replace_unet_linear_and_conv2d()
+    # prepare ControlNet
+    network = control_net_lllite.ControlNetLLLite(unet, args.cond_emb_dim, args.network_dim, args.network_dropout)
+    network.apply_to()

    if args.network_weights is not None:
-        accelerator.print(f"initialize U-Net with ControlNet-LLLite")
-        with accelerate.init_empty_weights():
-            unet_lllite = control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite()
-        unet_lllite.to(accelerator.device, dtype=weight_dtype)
-
-        unet_sd = unet.state_dict()
-        info = unet_lllite.load_lllite_weights(args.network_weights, unet_sd)
-        accelerator.print(f"load ControlNet-LLLite weights from {args.network_weights}: {info}")
-    else:
-        # cosumes large memory, so send to GPU before creating the LLLite model
-        accelerator.print("sending U-Net to GPU")
-        unet.to(accelerator.device, dtype=weight_dtype)
-        unet_sd = unet.state_dict()
-
-        # init LLLite weights
-        accelerator.print(f"initialize U-Net with ControlNet-LLLite")
-
-        if args.lowram:
-            with accelerate.init_on_device(accelerator.device):
-                unet_lllite = control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite()
-        else:
-            unet_lllite = control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite()
-        unet_lllite.to(weight_dtype)
-
-        info = unet_lllite.load_lllite_weights(None, unet_sd)
-        accelerator.print(f"init U-Net with ControlNet-LLLite weights: {info}")
-    del unet_sd, unet
-
-    unet: control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite = unet_lllite
-    del unet_lllite
-
-    unet.apply_lllite(args.cond_emb_dim, args.network_dim, args.network_dropout)
-
-    # モデルに xformers とか memory efficient attention を組み込む
-    train_util.replace_unet_modules(unet, args.mem_eff_attn, args.xformers, args.sdpa)
+        info = network.load_weights(args.network_weights)
+        accelerator.print(f"load ControlNet weights from {args.network_weights}: {info}")

    if args.gradient_checkpointing:
        unet.enable_gradient_checkpointing()
+        network.enable_gradient_checkpointing()  # may have no effect

    # 学習に必要なクラスを準備する
    accelerator.print("prepare optimizer, data loader etc.")

-    trainable_params = list(unet.prepare_params())
+    trainable_params = list(network.prepare_optimizer_params())
    print(f"trainable params count: {len(trainable_params)}")
    print(f"number of trainable parameters: {sum(p.numel() for p in trainable_params if p.requires_grad)}")

@@ -238,7 +213,7 @@ def train(args):
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
@@ -257,32 +232,37 @@ def train(args):
    lr_scheduler = train_util.get_scheduler_fix(args, optimizer, accelerator.num_processes)

    # 実験的機能：勾配も含めたfp16/bf16学習を行う　モデル全体をfp16/bf16にする
-    # if args.full_fp16:
-    #     assert (
-    #         args.mixed_precision == "fp16"
-    #     ), "full_fp16 requires mixed precision='fp16' / full_fp16を使う場合はmixed_precision='fp16'を指定してください。"
-    #     accelerator.print("enable full fp16 training.")
-    #     unet.to(weight_dtype)
-    # elif args.full_bf16:
-    #     assert (
-    #         args.mixed_precision == "bf16"
-    #     ), "full_bf16 requires mixed precision='bf16' / full_bf16を使う場合はmixed_precision='bf16'を指定してください。"
-    #     accelerator.print("enable full bf16 training.")
-    #     unet.to(weight_dtype)
-
-    unet.to(weight_dtype)
+    if args.full_fp16:
+        assert (
+            args.mixed_precision == "fp16"
+        ), "full_fp16 requires mixed precision='fp16' / full_fp16を使う場合はmixed_precision='fp16'を指定してください。"
+        accelerator.print("enable full fp16 training.")
+        unet.to(weight_dtype)
+        network.to(weight_dtype)
+    elif args.full_bf16:
+        assert (
+            args.mixed_precision == "bf16"
+        ), "full_bf16 requires mixed precision='bf16' / full_bf16を使う場合はmixed_precision='bf16'を指定してください。"
+        accelerator.print("enable full bf16 training.")
+        unet.to(weight_dtype)
+        network.to(weight_dtype)

    # acceleratorがなんかよろしくやってくれるらしい
-    unet, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(unet, optimizer, train_dataloader, lr_scheduler)
+    unet, network, optimizer, train_dataloader, lr_scheduler = accelerator.prepare(
+        unet, network, optimizer, train_dataloader, lr_scheduler
+    )
+    network: control_net_lllite.ControlNetLLLite

    # transform DDP after prepare (train_network here only)
-    unet = train_util.transform_models_if_DDP([unet])[0]
+    unet, network = train_util.transform_models_if_DDP([unet, network])

    if args.gradient_checkpointing:
        unet.train()  # according to TI example in Diffusers, train is required -> これオリジナルのU-Netしたので本当は外せる
    else:
        unet.eval()

+    network.prepare_grad_etc()
+
    # TextEncoderの出力をキャッシュするときにはCPUへ移動する
    if args.cache_text_encoder_outputs:
        # move Text Encoders for sampling images. Text Encoder doesn't work on CPU with fp16
@@ -348,13 +328,7 @@ def train(args):
    del train_dataset_group

    # function for saving/removing
-    def save_model(
-        ckpt_name,
-        unwrapped_nw: control_net_lllite_for_train.SdxlUNet2DConditionModelControlNetLLLite,
-        steps,
-        epoch_no,
-        force_sync_upload=False,
-    ):
+    def save_model(ckpt_name, unwrapped_nw, steps, epoch_no, force_sync_upload=False):
        os.makedirs(args.output_dir, exist_ok=True)
        ckpt_file = os.path.join(args.output_dir, ckpt_name)

@@ -362,7 +336,7 @@ def train(args):
        sai_metadata = train_util.get_sai_model_spec(None, args, True, True, False)
        sai_metadata["modelspec.architecture"] = sai_model_spec.ARCH_SD_XL_V1_BASE + "/control-net-lllite"

-        unwrapped_nw.save_lllite_weights(ckpt_file, save_dtype, sai_metadata)
+        unwrapped_nw.save_weights(ckpt_file, save_dtype, sai_metadata)
        if args.huggingface_repo_id is not None:
            huggingface_util.upload(args, ckpt_file, "/" + ckpt_name, force_sync_upload=force_sync_upload)

@@ -377,9 +351,11 @@ def train(args):
        accelerator.print(f"\nepoch {epoch+1}/{num_train_epochs}")
        current_epoch.value = epoch + 1

+        network.on_epoch_start()  # train()
+
        for step, batch in enumerate(train_dataloader):
            current_step.value = global_step
-            with accelerator.accumulate(unet):
+            with accelerator.accumulate(network):
                with torch.no_grad():
                    if "latents" in batch and batch["latents"] is not None:
                        latents = batch["latents"].to(accelerator.device)
@@ -436,9 +412,10 @@ def train(args):
                with accelerator.autocast():
                    # conditioning imageをControlNetに渡す / pass conditioning image to ControlNet
                    # 内部でcond_embに変換される / it will be converted to cond_emb inside
+                    network.set_cond_image(controlnet_image)

                    # それらの値を使いつつ、U-Netでノイズを予測する / predict noise with U-Net using those values
-                    noise_pred = unet(noisy_latents, timesteps, text_embedding, vector_embedding, controlnet_image)
+                    noise_pred = unet(noisy_latents, timesteps, text_embedding, vector_embedding)

                if args.v_parameterization:
                    # v-parameterization training
@@ -463,7 +440,7 @@ def train(args):

                accelerator.backward(loss)
                if accelerator.sync_gradients and args.max_grad_norm != 0.0:
-                    params_to_clip = unet.get_trainable_params()
+                    params_to_clip = network.get_trainable_params()
                    accelerator.clip_grad_norm_(params_to_clip, args.max_grad_norm)

                optimizer.step()
@@ -482,7 +459,7 @@ def train(args):
                    accelerator.wait_for_everyone()
                    if accelerator.is_main_process:
                        ckpt_name = train_util.get_step_ckpt_name(args, "." + args.save_model_as, global_step)
-                        save_model(ckpt_name, accelerator.unwrap_model(unet), global_step, epoch)
+                        save_model(ckpt_name, accelerator.unwrap_model(network), global_step, epoch)

                        if args.save_state:
                            train_util.save_and_remove_state_stepwise(args, accelerator, global_step)
@@ -521,7 +498,7 @@ def train(args):
            saving = (epoch + 1) % args.save_every_n_epochs == 0 and (epoch + 1) < num_train_epochs
            if is_main_process and saving:
                ckpt_name = train_util.get_epoch_ckpt_name(args, "." + args.save_model_as, epoch + 1)
-                save_model(ckpt_name, accelerator.unwrap_model(unet), global_step, epoch + 1)
+                save_model(ckpt_name, accelerator.unwrap_model(network), global_step, epoch + 1)

                remove_epoch_no = train_util.get_remove_epoch_no(args, epoch + 1)
                if remove_epoch_no is not None:
@@ -536,7 +513,7 @@ def train(args):
        # end of epoch

    if is_main_process:
-        unet = accelerator.unwrap_model(unet)
+        network = accelerator.unwrap_model(network)

    accelerator.end_training()

@@ -545,7 +522,7 @@ def train(args):

    if is_main_process:
        ckpt_name = train_util.get_last_ckpt_name(args, "." + args.save_model_as)
-        save_model(ckpt_name, unet, global_step, num_train_epochs, force_sync_upload=True)
+        save_model(ckpt_name, network, global_step, num_train_epochs, force_sync_upload=True)

        print("model saved.")

--- a/sdxl_train_network.py
+++ b/sdxl_train_network.py
@@ -1,5 +1,12 @@
 import argparse
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from library import sdxl_model_util, sdxl_train_util, train_util
 import train_network

--- a/sdxl_train_textual_inversion.py
+++ b/sdxl_train_textual_inversion.py
@@ -3,6 +3,13 @@ import os

 import regex
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 import open_clip
 from library import sdxl_model_util, sdxl_train_util, train_util

--- a/tools/cache_latents.py
+++ b/tools/cache_latents.py
@@ -86,8 +86,8 @@ def cache_to_disk(args: argparse.Namespace) -> None:

    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    # acceleratorを準備する
    print("prepare accelerator")
@@ -120,7 +120,7 @@ def cache_to_disk(args: argparse.Namespace) -> None:
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
--- a/tools/cache_text_encoder_outputs.py
+++ b/tools/cache_text_encoder_outputs.py
@@ -91,8 +91,8 @@ def cache_to_disk(args: argparse.Namespace) -> None:

    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    # acceleratorを準備する
    print("prepare accelerator")
@@ -125,7 +125,7 @@ def cache_to_disk(args: argparse.Namespace) -> None:
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
--- a/train_controlnet.py
+++ b/train_controlnet.py
@@ -11,6 +11,13 @@ import toml

 from tqdm import tqdm
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from torch.nn.parallel import DistributedDataParallel as DDP
 from accelerate.utils import set_seed
 from diffusers import DDPMScheduler, ControlNetModel
@@ -91,8 +98,8 @@ def train(args):

    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    if args.debug_dataset:
        train_util.debug_dataset(train_dataset_group)
@@ -238,7 +245,7 @@ def train(args):
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
--- a/train_db.py
+++ b/train_db.py
@@ -11,6 +11,13 @@ import toml

 from tqdm import tqdm
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from accelerate.utils import set_seed
 from diffusers import DDPMScheduler

@@ -71,8 +78,8 @@ def train(args):

    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    if args.no_token_padding:
        train_dataset_group.disable_token_padding()
@@ -170,7 +177,7 @@ def train(args):
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
--- a/train_network.py
+++ b/train_network.py
@@ -12,6 +12,16 @@ import toml

 from tqdm import tqdm
 import torch
+
+try:
+    import intel_extension_for_pytorch as ipex
+
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+
+        ipex_init()
+except Exception:
+    pass
 from accelerate.utils import set_seed
 from diffusers import DDPMScheduler
 from library import model_util
@@ -182,8 +192,8 @@ class NetworkTrainer:

        current_epoch = Value("i", 0)
        current_step = Value("i", 0)
-        ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-        collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+        ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+        collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

        if args.debug_dataset:
            train_util.debug_dataset(train_dataset_group)
@@ -273,7 +283,10 @@ class NetworkTrainer:
        if args.dim_from_weights:
            network, _ = network_module.create_network_from_weights(1, args.network_weights, vae, text_encoder, unet, **net_kwargs)
        else:
-            # LyCORIS will work with this...
+            if "dropout" not in net_kwargs:
+                # workaround for LyCORIS (;^ω^)
+                net_kwargs["dropout"] = args.network_dropout
+
            network = network_module.create_network(
                1.0,
                args.network_dim,
@@ -332,7 +345,7 @@ class NetworkTrainer:
            train_dataset_group,
            batch_size=1,
            shuffle=True,
-            collate_fn=collater,
+            collate_fn=collator,
            num_workers=n_workers,
            persistent_workers=args.persistent_data_loader_workers,
        )
@@ -419,7 +432,12 @@ class NetworkTrainer:
                t_enc.train()

                # set top parameter requires_grad = True for gradient checkpointing works
-                t_enc.text_model.embeddings.requires_grad_(True)
+                if train_text_encoder:
+                    t_enc.text_model.embeddings.requires_grad_(True)
+
+            # set top parameter requires_grad = True for gradient checkpointing works
+            if not train_text_encoder:  # train U-Net only
+                unet.parameters().__next__().requires_grad_(True)
        else:
            unet.eval()
            for t_enc in text_encoders:
@@ -509,6 +527,7 @@ class NetworkTrainer:
            "ss_prior_loss_weight": args.prior_loss_weight,
            "ss_min_snr_gamma": args.min_snr_gamma,
            "ss_scale_weight_norms": args.scale_weight_norms,
+            "ss_ip_noise_gamma": args.ip_noise_gamma,
        }

        if use_user_config:
@@ -938,7 +957,7 @@ def setup_parser() -> argparse.ArgumentParser:
        help="Drops neurons out of training every step (0 or None is default behavior (no dropout), 1 would drop all neurons) / 訓練時に毎ステップでニューロンをdropする（0またはNoneはdropoutなし、1は全ニューロンをdropout）",
    )
    parser.add_argument(
-        "--network_args", type=str, default=None, nargs="*", help="additional argmuments for network (key=value) / ネットワークへの追加の引数"
+        "--network_args", type=str, default=None, nargs="*", help="additional arguments for network (key=value) / ネットワークへの追加の引数"
    )
    parser.add_argument("--network_train_unet_only", action="store_true", help="only training U-Net part / U-Net関連部分のみ学習する")
    parser.add_argument(
--- a/train_textual_inversion.py
+++ b/train_textual_inversion.py
@@ -7,6 +7,13 @@ import toml

 from tqdm import tqdm
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from accelerate.utils import set_seed
 from diffusers import DDPMScheduler
 from transformers import CLIPTokenizer
@@ -305,8 +312,8 @@ class TextualInversionTrainer:

        current_epoch = Value("i", 0)
        current_step = Value("i", 0)
-        ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-        collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+        ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+        collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

        # make captions: tokenstring tokenstring1 tokenstring2 ...tokenstringn という文字列に書き換える超乱暴な実装
        if use_template:
@@ -382,7 +389,7 @@ class TextualInversionTrainer:
            train_dataset_group,
            batch_size=1,
            shuffle=True,
-            collate_fn=collater,
+            collate_fn=collator,
            num_workers=n_workers,
            persistent_workers=args.persistent_data_loader_workers,
        )
--- a/train_textual_inversion_XTI.py
+++ b/train_textual_inversion_XTI.py
@@ -8,6 +8,13 @@ from multiprocessing import Value

 from tqdm import tqdm
 import torch
+try:
+    import intel_extension_for_pytorch as ipex
+    if torch.xpu.is_available():
+        from library.ipex import ipex_init
+        ipex_init()
+except Exception:
+    pass
 from accelerate.utils import set_seed
 import diffusers
 from diffusers import DDPMScheduler
@@ -229,8 +236,8 @@ def train(args):
    train_dataset_group.enable_XTI(XTI_layers, token_strings=token_strings)
    current_epoch = Value("i", 0)
    current_step = Value("i", 0)
-    ds_for_collater = train_dataset_group if args.max_data_loader_n_workers == 0 else None
-    collater = train_util.collater_class(current_epoch, current_step, ds_for_collater)
+    ds_for_collator = train_dataset_group if args.max_data_loader_n_workers == 0 else None
+    collator = train_util.collator_class(current_epoch, current_step, ds_for_collator)

    # make captions: tokenstring tokenstring1 tokenstring2 ...tokenstringn という文字列に書き換える超乱暴な実装
    if use_template:
@@ -302,7 +309,7 @@ def train(args):
        train_dataset_group,
        batch_size=1,
        shuffle=True,
-        collate_fn=collater,
+        collate_fn=collator,
        num_workers=n_workers,
        persistent_workers=args.persistent_data_loader_workers,
    )
Author	SHA1	Message	Date
Kohya S	2a23713f71	Merge pull request #872 from kohya-ss/dev fix make_captions_by_git, improve image generation scripts	2023-10-11 07:56:39 +09:00
Kohya S	681034d001	update readme	2023-10-11 07:54:30 +09:00
Kohya S	17813ff5b4	remove workaround for transfomers bs>1 close #869	2023-10-11 07:40:12 +09:00
Kohya S	3e81bd6b67	fix network_merge, add regional mask as color code	2023-10-09 23:07:14 +09:00
Kohya S	23ae358e0f	Merge branch 'main' into dev	2023-10-09 21:42:13 +09:00
Kohya S	f611726364	add network_merge_n_models option	2023-10-09 21:41:50 +09:00
Kohya S	33ee0acd35	Merge pull request #867 from kohya-ss/dev onnx support in wd14 tagger, OFT	2023-10-09 18:04:17 +09:00
Kohya S	8b79e3b06c	fix typos	2023-10-09 18:00:45 +09:00
Kohya S	cf49e912fc	update readme	2023-10-09 17:59:31 +09:00
Kohya S	66741c035c	add OFT	2023-10-09 17:59:24 +09:00
Kohya S	406511c333	add error message if model.onnx doesn't exist	2023-10-09 17:08:58 +09:00
Kohya S	8a2d68d63e	Merge pull request #864 from Isotr0py/onnx Add `--onnx` to wd14 tagger	2023-10-09 15:14:11 +09:00
Kohya S	07d297fdbe	Merge branch 'dev' into onnx	2023-10-09 15:13:40 +09:00
Kohya S	0d4e8b50d0	change option to append_tags, minor update	2023-10-09 15:09:54 +09:00
Kohya S	1d7c5c2a98	Merge pull request #858 from a-l-e-x-d-s-9/main Add append_captions feature to wd14 tagger	2023-10-09 14:31:54 +09:00
Kohya S	0faa350175	Merge pull request #865 from kohya-ss/dev Support JPEG-XL on windows, dropout for LyCORIS	2023-10-09 14:11:49 +09:00
Kohya S	8a7509db75	Merge branch 'dev' of https://github.com/kohya-ss/sd-scripts into dev	2023-10-09 14:07:02 +09:00
Kohya S	025368f51c	may work dropout in LyCORIS #859	2023-10-09 14:06:58 +09:00
Kohya S	5fe52ed322	Merge pull request #856 from Isotr0py/jxl Fix JPEG-XL support	2023-10-09 13:55:03 +09:00
Kohya S	8b247a330b	Merge pull request #851 from kohya-ss/dependabot/github_actions/actions/checkout-4 Bump actions/checkout from 3 to 4	2023-10-09 11:45:47 +09:00
Isotr0py	d6f458fcb3	fix dependency	2023-10-08 23:51:18 +08:00
Isotr0py	b8b84021e5	fix a typo	2023-10-08 20:49:03 +08:00
Isotr0py	70fe7e18be	add onnx to wd14 tagger	2023-10-08 20:31:10 +08:00
alexds9	9378da3c82	Fix comment	2023-10-05 21:29:46 +03:00
alexds9	a4857fa764	Add append_captions feature to wd14 tagger This feature allows for appending new tags to the existing content of caption files. If the caption file for an image already exists, the tags generated from the current run are appended to the existing ones. Duplicate tags are checked and avoided.	2023-10-05 21:26:09 +03:00
Isotr0py	592014923f	Support JPEG-XL on windows	2023-10-04 21:48:25 +08:00
dependabot[bot]	6d06b215bf	Bump actions/checkout from 3 to 4 Bumps [actions/checkout](https://github.com/actions/checkout) from 3 to 4. - [Release notes](https://github.com/actions/checkout/releases) - [Changelog](https://github.com/actions/checkout/blob/main/CHANGELOG.md) - [Commits](https://github.com/actions/checkout/compare/v3...v4) --- updated-dependencies: - dependency-name: actions/checkout dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>	2023-10-01 22:51:32 +00:00
Kohya S	2d87bb648f	Merge pull request #850 from kohya-ss/dev fix typos	2023-10-02 07:51:05 +09:00
Kohya S	56ebef35b0	Merge pull request #848 from shirayu/update_typos Update typos to the latest version and add dependabot.yml	2023-10-02 07:45:29 +09:00
Yuta Hayashibe	13d8b22d25	Add dependabot	2023-10-01 21:52:16 +09:00
Yuta Hayashibe	27f9b6ffeb	updated typos to v1.16.15 and fix typos	2023-10-01 21:51:24 +09:00
Yuta Hayashibe	c8fcfd4581	Add "venv" to extend-exclude	2023-10-01 21:48:50 +09:00
Kohya S	49c24285c7	Merge pull request #847 from kohya-ss/sdxl merge sdxl into main	2023-10-01 20:40:33 +09:00
Kohya S	c918489259	update readme	2023-10-01 20:34:12 +09:00
Kohya S	93155242fa	Merge pull request #846 from kohya-ss/dev Fix to work training U-Net only LoRA for SD1/2	2023-10-01 16:44:13 +09:00
Kohya S	4cc919607a	fix placing of requires_grad_ of U-Net	2023-10-01 16:41:48 +09:00
Kohya S	81419f7f32	Fix to work training U-Net only LoRA for SD1/2	2023-10-01 16:37:23 +09:00
Kohya S	6bd6cd9c51	update doc	2023-10-01 12:17:54 +09:00
Kohya S	35a1d68eb6	Merge pull request #844 from kohya-ss/dev IPEX update, concat LoRA	2023-10-01 12:06:36 +09:00
Kohya S	365a06bdb6	Merge pull request #839 from laksjdjf/sdxl Support concat LoRA	2023-10-01 11:16:46 +09:00
Kohya S	8e117f9f92	Merge pull request #841 from Disty0/dev IPEX Attention optimizations	2023-10-01 10:58:19 +09:00
Disty0	209eafb631	IPEX attention optimizations	2023-09-28 14:02:25 +03:00
laksjdjf	14aa2923cf	Support concat LoRA	2023-09-28 14:39:32 +09:00
Kohya S	1e395ed285	Merge branch 'main' into sdxl	2023-09-24 17:51:08 +09:00
Kohya S	98615166b0	Merge pull request #831 from kohya-ss/dev update versions of accelerate and diffusers	2023-09-24 17:50:40 +09:00
Kohya S	28272de97a	update readme	2023-09-24 17:48:51 +09:00
Kohya S	7e736da30c	update versions of accelerate and diffusers	2023-09-24 17:46:57 +09:00
Kohya S	20e929e27e	fix to work iter_same_seed	2023-09-24 16:04:50 +09:00
Kohya S	477b5260aa	fix sai metadata for sdxl closes #824	2023-09-24 14:47:13 +09:00
Kohya S	d39f1a3427	Merge pull request #808 from rockerBOO/metadata Add ip_noise_gamma metadata	2023-09-24 14:35:18 +09:00
Kohya S	3757855231	rename train_lllite_alt to train_lllite	2023-09-24 14:34:31 +09:00
Kohya S	d846431015	Merge branch 'dev' into sdxl	2023-09-24 14:30:16 +09:00
Kohya S	624edf428f	Merge pull request #825 from Disty0/dev Intel ARC support with IPEX	2023-09-24 14:29:03 +09:00
Kohya S	54500b861d	Merge pull request #830 from kohya-ss/dev2 add extension checking for resize_lora.py	2023-09-24 12:12:32 +09:00
Kohya S	f2491ee0ac	change block name doesn't contain '.' at end	2023-09-24 12:10:56 +09:00
Kohya S	1f169ee7fb	Merge pull request #760 from Symbiomatrix/bugfix1 Update resize_lora.py	2023-09-24 11:59:18 +09:00
Kohya S	66817992c1	revert formatting	2023-09-24 11:50:44 +09:00
Kohya S	8052bcd5cd	format by black	2023-09-24 11:26:28 +09:00
Kohya S	55886a0116	add .pt and .pth for available extension	2023-09-24 11:25:54 +09:00
Kohya S	33e90cc6a0	Merge pull request #815 from jvkap/patch-1 Update resize_lora.py	2023-09-24 11:02:12 +09:00
青龍聖者@bdsqlsz	d5be8125b0	update bitsandbytes for 0.41.1 and fixed bugs with generate_controlnet_subsets_config for training (#823 ) * update for bnb 0.41.1 * fixed generate_controlnet_subsets_config for training * Revert "update for bnb 0.41.1" This reverts commit `70bd3612d8`.	2023-09-24 10:51:47 +09:00
Disty0	b99cd2a920	Update getDeviceIdListForCard	2023-09-20 17:16:06 +03:00
Disty0	b64389c8a9	Intel ARC support with IPEX	2023-09-19 18:05:05 +03:00
jvkap	a0e05fa291	Update resize_lora.py	2023-09-11 11:41:33 -03:00
jvkap	e33c007cd0	Update resize_lora.py	2023-09-11 11:29:06 -03:00
rockerBOO	80aca1ccc7	Add ip_noise_gamma metadata	2023-09-05 15:20:15 -04:00
Kohya S	6b3a580ee5	Merge pull request #804 from kohya-ss/dev fix to work regional LoRA	2023-09-03 17:52:23 +09:00
Kohya S	74561dbdac	update readme (#803 ) * update readme * update readme * fix typo	2023-09-03 12:51:09 +09:00
Symbiomatrix	9d678a6f41	Update resize_lora.py	2023-08-16 00:08:09 +03:00