v23.0.6

What's Changed

• Fix issue with PYTHON path that have "spaces" in them

Full Changelog: v23.0.5...v23.0.6

v23.0.5

What's Changed

• Move cudnn out of requirements by @bmaltais in #2066
• Update python module verification

Full Changelog: 23.0.4...v23.0.5

v23.0.4

What's Changed

• Update README by @bmaltais in #2059
• Update bitsandbytes to 0.43.0
• Add packaging to runpod setup

Full Changelog: v23.0.3...23.0.4

v23.0.3

What's Changed

• Fix bug with setup
• Enforce proper python version before running the GUI to prevent issues with execution of the GUI.

Full Changelog: v23.0.2...v23.0.3

v23.0.2

THIS RELEASE MAY CONTAIN BUGS... BUT CONTAIN A COOL NEW GUI. PLEASE TEST IT AND REPORT BUGS.

What's Changed

• Bump RunPod requirements to match the Linux ones by @ashleykleynhans in #2054
• Improve validation of path provided by users before running training

New Contributors

• @ashleykleynhans made their first contribution in #2054

Full Changelog: v23.0.1...v23.0.2

v23.0.1

• Update bitsandbytes module to 0.43.0 as it provide native windows support
• Minor fixes to code

What's Changed

• add missing nonlocal scope keywords by @wkpark in #2042
• Merge master to des by @bmaltais in #2045
• v23.0.1 by @bmaltais in #2048

Full Changelog: v23.0.0...v23.0.1

v23.0.0

WARNING: THIS RELEASE IS BUGGY... Stay in the v22 series for now unless you like leaving on the edge

• Use sd-scripts release 0.8.4 post commit fccbee27277d65a8dcbdeeb81787ed4116b92e0b
• Major code refactoring thanks to @wkpark , This will make updating sd-script cleaner by keeping sd-scripts files separate from the GUI files. This will also make configuration more streamlined with fewer tabs and more accordion elements. Hope you like the new style.
• This new release is implementing a significant structure change, moving all of the sd-scripts written by kohya under a folder called sd-scripts in the root of this project. This folder is a submodule that will be populated during setup or gui execution.

What's Changed

• Bump crate-ci/typos from 1.17.2 to 1.18.2 by @dependabot in #2022
• refactor kohya_gui parts by @wkpark in #2017
• Gui fix by @wkpark in #2025
• Fix block_lr issue by @storuky in #2031
• Dev pure by @bmaltais in #2039
• v23.0.0 by @bmaltais in #2040

New Contributors

• @storuky made their first contribution in #2031

Full Changelog: v22.6.2...v23.0.0

v22.6.2

• 2024/02/24 (v22.6.2)

• Fix issue with Lora Extract GUI
  • Fix syntax issue where parameter lora_network_weights is actually called network_weights
• Merge sd-scripts v0.8.4 code update
  • Fixed a bug that the VRAM usage without Text Encoder training is larger than before in training scripts for LoRA etc (train_network.py, sdxl_train_network.py).
    • Text Encoders were not moved to CPU.
  • Fixed typos. Thanks to akx! PR #1053
  • The log output has been improved. PR #905 Thanks to shirayu!
    • The log is formatted by default. The rich library is required. Please see Upgrade and update the library.
    • If rich is not installed, the log output will be the same as before.
    • The following options are available in each training script:
    • --console_log_simple option can be used to switch to the previous log output.
    • --console_log_level option can be used to specify the log level. The default is INFO.
    • --console_log_file option can be used to output the log to a file. The default is None (output to the console).
  • The sample image generation during multi-GPU training is now done with multiple GPUs. PR #1061 Thanks to DKnight54!
  • The support for mps devices is improved. PR #1054 Thanks to akx! If mps device exists instead of CUDA, the mps device is used automatically.
  • The --new_conv_rank option to specify the new rank of Conv2d is added to networks/resize_lora.py. PR #1102 Thanks to mgz-dev!
  • An option --highvram to disable the optimization for environments with little VRAM is added to the training scripts. If you specify it when there is enough VRAM, the operation will be faster.
    • Currently, only the cache part of latents is optimized.
  • The IPEX support is improved. PR #1086 Thanks to Disty0!
  • Fixed a bug that svd_merge_lora.py crashes in some cases. PR #1087 Thanks to mgz-dev!
  • DyLoRA is fixed to work with SDXL. PR #1126 Thanks to tamlog06!
  • The common image generation script gen_img.py for SD 1/2 and SDXL is added. The basic functions are the same as the scripts for SD 1/2 and SDXL, but some new features are added.
    • External scripts to generate prompts can be supported. It can be called with --from_module option. (The documentation will be added later)
    • The normalization method after prompt weighting can be specified with --emb_normalize_mode option. original is the original method, abs is the normalization with the average of the absolute values, none is no normalization.
  • Gradual Latent Hires fix is added to each generation script. See here for details.

What's Changed

• Fix Lora Extract GUI issue by @bmaltais in #1975
• Update options.md by @mikeboensel in #1978
• Update options.md by @mikeboensel in #1979
• Update options.md by @mikeboensel in #1980
• Optimize the Dockerfile to reduce the size of the image. (50.76 GB to 19.04 GB) by @jim60105 in #1976
• Label clarifications by @mikeboensel in #1981
• v22.6.2 by @bmaltais in #1995

New Contributors

• @mikeboensel made their first contribution in #1978

Full Changelog: v22.6.1...v22.6.2

v22.6.1

• 2024/02/15 (v22.6.1)

This release contain some serious code refactoring... There might be bugs I have not seen... so be ready to return to the previous release if you discover some...

• Add support for multi-gpu parameters in the GUI under the "Parameters > Advanced" tab.
• Significant rewrite of how parameters are created in the code. I hope I did not break anything in the process... Will make the code easier to update.
• Update TW locallisation
• Update gradio module version to latest 3.x

What's Changed

• Bump crate-ci/typos from 1.16.26 to 1.17.2 by @dependabot in #1922
• Update localization for zh-TW by @hinablue in #1966
• v22.6.1 by @bmaltais in #1967

Full Changelog: v22.6.0...v22.6.1

v22.6.0

• 2024/01/27 (v22.6.0)

• Merge sd-scripts v0.8.3 code update

  • Fixed a bug that the training crashes when --fp8_base is specified with --save_state. PR #1079 Thanks to feffy380!
    • safetensors is updated. Please see Upgrade and update the library.
  • Fixed a bug that the training crashes when network_multiplier is specified with multi-GPU training. PR #1084 Thanks to fireicewolf!
  • Fixed a bug that the training crashes when training ControlNet-LLLite.

• Merge sd-scripts v0.8.2 code update

  • [Experimental] The --fp8_base option is added to the training scripts for LoRA etc. The base model (U-Net, and Text Encoder when training modules for Text Encoder) can be trained with fp8. PR #1057 Thanks to KohakuBlueleaf!

    • Please specify --fp8_base in train_network.py or sdxl_train_network.py.
    • PyTorch 2.1 or later is required.
    • If you use xformers with PyTorch 2.1, please see xformers repository and install the appropriate version according to your CUDA version.
    • The sample image generation during training consumes a lot of memory. It is recommended to turn it off.

  • [Experimental] The network multiplier can be specified for each dataset in the training scripts for LoRA etc.

    • This is an experimental option and may be removed or changed in the future.
    • For example, if you train with state A as 1.0 and state B as -1.0, you may be able to generate by switching between state A and B depending on the LoRA application rate.
    • Also, if you prepare five states and train them as 0.2, 0.4, 0.6, 0.8, and 1.0, you may be able to generate by switching the states smoothly depending on the application rate.
    • Please specify network_multiplier in [[datasets]] in .toml file.

  • Some options are added to networks/extract_lora_from_models.py to reduce the memory usage.

    • --load_precision option can be used to specify the precision when loading the model. If the model is saved in fp16, you can reduce the memory usage by specifying --load_precision fp16 without losing precision.
    • --load_original_model_to option can be used to specify the device to load the original model. --load_tuned_model_to option can be used to specify the device to load the derived model. The default is cpu for both options, but you can specify cuda etc. You can reduce the memory usage by loading one of them to GPU. This option is available only for SDXL.

  • The gradient synchronization in LoRA training with multi-GPU is improved. PR #1064 Thanks to KohakuBlueleaf!

  • The code for Intel IPEX support is improved. PR #1060 Thanks to akx!

  • Fixed a bug in multi-GPU Textual Inversion training.

  • .toml example for network multiplier

    [general]
    [[datasets]]
    resolution = 512
    batch_size = 8
    network_multiplier = 1.0
    
    ... subset settings ...
    
    [[datasets]]
    resolution = 512
    batch_size = 8
    network_multiplier = -1.0
    
    ... subset settings ...

• Merge sd-scripts v0.8.1 code update

  • Fixed a bug that the VRAM usage without Text Encoder training is larger than before in training scripts for LoRA etc (train_network.py, sdxl_train_network.py).

    • Text Encoders were not moved to CPU.

  • Fixed typos. Thanks to akx! PR #1053

What's Changed

• Update Chinese Documentation by @boombbo in #1896
• Change cudann to cuDNN by @EugeoSynthesisThirtyTwo in #1902
• v22.6.0 by @bmaltais in #1907

New Contributors

• @EugeoSynthesisThirtyTwo made their first contribution in #1902

Full Changelog: v22.5.0...v22.6.0