Kohya S aff4a2e732 feat: Enhance LoHa and LoKr modules with Tucker decomposition support ...

- Added Tucker decomposition functionality to LoHa and LoKr modules.
- Implemented new methods for weight rebuilding using Tucker decomposition.
- Updated initialization and weight handling for Conv2d 3x3+ layers.
- Modified get_diff_weight methods to accommodate Tucker and non-Tucker modes.
- Enhanced network base to include unet_conv_target_modules for architecture detection.

2026-02-15 23:35:17 +09:00

..

check_lora_weights.py

improve OFT implementation closes #944

2024-09-13 22:37:11 +09:00

control_net_lllite_for_train.py

make forward/backward pathes same ref #1363

2024-06-09 19:26:36 +09:00

control_net_lllite.py

Replace print with logger if they are logs (#905)

2024-02-04 18:14:34 +09:00

convert_anima_lora_to_comfy.py

Add/modify some implementation for anima (#2261)

2026-02-13 08:15:06 +09:00

convert_flux_lora.py

add workaround for 'Some tensors share memory' error #1614

2024-09-28 20:57:27 +09:00

convert_hunyuan_image_lora_to_comfy.py

feat: add conversion script for LoRA models to ComfyUI format with reverse option

2025-09-16 21:48:47 +09:00

dylora.py

Fix [occasionally] missing text encoder attn modules

2025-03-01 20:35:45 +03:00

extract_lora_from_dylora.py

Replace print with logger if they are logs (#905)

2024-02-04 18:14:34 +09:00

extract_lora_from_models.py

Replace print with logger if they are logs (#905)

2024-02-04 18:14:34 +09:00

flux_extract_lora.py

fix: update metadata construction to include model_config for flux

2025-09-24 21:15:37 +09:00

flux_merge_lora.py

fix: update metadata construction to include model_config for flux

2025-09-24 21:15:37 +09:00

loha.py

feat: Enhance LoHa and LoKr modules with Tucker decomposition support

2026-02-15 23:35:17 +09:00

lokr.py

feat: Enhance LoHa and LoKr modules with Tucker decomposition support

2026-02-15 23:35:17 +09:00

lora_anima.py

feat: Add LoHa/LoKr network support for SDXL and Anima

2026-02-15 21:50:50 +09:00

lora_diffusers.py

Fix [occasionally] missing text encoder attn modules

2025-03-01 20:35:45 +03:00

lora_fa.py

Fix [occasionally] missing text encoder attn modules

2025-03-01 20:35:45 +03:00

lora_flux.py

feat: block swap for inference and initial impl for HunyuanImage LoRA (not working)

2025-09-11 22:15:22 +09:00

lora_hunyuan_image.py

feat: HunyuanImage LoRA training

2025-09-12 21:40:42 +09:00

lora_interrogator.py

Merge branch 'dev' into dev_device_support

2024-02-12 13:01:54 +09:00

lora_lumina.py

fix: support LoRA w/o TE for create_network_from_weights

2025-07-13 20:46:24 +09:00

lora_sd3.py

Fix SD3 trained lora loading and merging

2024-11-07 09:54:12 +00:00

lora.py

Fix [occasionally] missing text encoder attn modules

2025-03-01 20:35:45 +03:00

merge_lora_old.py

Replace print with logger if they are logs (#905)

2024-02-04 18:14:34 +09:00

merge_lora.py

Replace print with logger if they are logs (#905)

2024-02-04 18:14:34 +09:00

network_base.py

feat: Enhance LoHa and LoKr modules with Tucker decomposition support

2026-02-15 23:35:17 +09:00

oft_flux.py

OFT for FLUX.1

2024-09-14 15:48:16 +09:00

oft.py

OFT for FLUX.1

2024-09-14 15:48:16 +09:00

resize_lora.py

Refactor to avoid mutable global variable

2025-08-15 11:24:51 +08:00

sdxl_merge_lora.py

Fix LoRA metadata hash calculation bug in svd_merge_lora.py, sdxl_merge_lora.py, and resize_lora.py closes #1722

2024-10-26 22:01:10 +09:00

svd_merge_lora.py

reduce VRAM usage, instead of increasing main RAM usage

2024-10-27 10:19:05 +09:00