Kohya S.
29983f5ac3
Update anima_minimal_inference.py
...
Remove unused code.
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2026-02-11 22:37:10 +09:00
Kohya S.
07df0e54a0
Update library/safetensors_utils.py
...
Fix duplicated new_key for concat_hook.
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2026-02-11 22:36:19 +09:00
Kohya S.
82f963650b
Update library/strategy_anima.py
...
Use torch.all instead of all.
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2026-02-11 22:28:00 +09:00
Kohya S
59267d19f3
feat: optimize RMSNorm forward method and remove unused torch_attention_op
2026-02-11 22:06:47 +09:00
Kohya S
4b2283491e
feat: add VAE chunking and caching options to reduce memory usage
2026-02-11 21:32:00 +09:00
Kohya S
a7cd38dcaf
feat: split CFG processing in do_sample function to reduce memory usage
2026-02-11 18:00:10 +09:00
Kohya S
9349c91c89
fix: remove unused code
2026-02-11 17:23:58 +09:00
Kohya S
b67cc5a457
feat: update Anima SAI model spec metadata handling
2026-02-11 14:56:23 +09:00
Kohya S
90725eba64
feat: support LORA for Qwen3
2026-02-11 14:21:33 +09:00
Kohya S
f355a97a32
feat: update Anima training guide with new argument specifications and regex-based module selection. Thank you Claude!
2026-02-10 22:11:34 +09:00
Kohya S
dbb40ae4c0
fix: wait all block trasfer before siwtching offloader mode
2026-02-10 21:46:58 +09:00
Kohya S
4992aae311
feat: optimize VAE processing by adjusting tensor dimensions and data types
2026-02-10 21:35:25 +09:00
Kohya S
8d2d286a13
feat: refactor Anima training script
2026-02-10 21:27:22 +09:00
Kohya S
6d08c93b23
feat: enhance block swap functionality for inference and training in Anima model
2026-02-10 21:26:50 +09:00
Kohya S
02a75944b3
fix: keep latents 4D except DiT call
2026-02-10 21:26:20 +09:00
Kohya S
6a4e392445
fix: update logging message for regex match in LoRANetwork
2026-02-10 21:21:29 +09:00
Kohya S
f3b6e59900
fix: improve regex matching for module selection and learning rates in LoRANetwork
2026-02-10 08:25:20 +09:00
Kohya S
3d24736238
feat: add regex-based learning rates and dimensions handling in create_network
2026-02-10 08:17:43 +09:00
Kohya S
58db77a488
fix: update to work fp8_scaled option
2026-02-09 23:31:33 +09:00
Kohya S
0f413974b7
feat: support LoRA format without net. prefix
2026-02-09 23:21:04 +09:00
kohya-ss
8f5b298906
feat: simplify Anima training script and update empty caption handling
2026-02-09 12:46:53 +09:00
kohya-ss
06dcb30016
fix: update argument names in anima_train_utils to align with other archtectures
2026-02-09 12:46:04 +09:00
kohya-ss
f320c1b964
feat: update loss calculation to support 5d tensor
2026-02-09 12:44:47 +09:00
kohya-ss
bf9806648c
feat: support 5d tensor in get_noisy_model_input_and_timesteps
2026-02-09 12:43:37 +09:00
kohya-ss
2774e7757b
feat: add dtype property and all-zero mask handling in cross-attention in LLMAdapterTransformerBlock
2026-02-09 12:43:11 +09:00
kohya-ss
a1e3d02259
fix: use strategy instead of using tokenizers directly
2026-02-09 12:42:16 +09:00
kohya-ss
35161b044c
feat: update train_llm_adapter and verbose default values to string type
2026-02-09 12:41:36 +09:00
kohya-ss
4f6511bf28
feat: kept caption dropout rate in cache and handle in training script
2026-02-08 15:35:53 +09:00
kohya-ss
c3556d455f
feat: simplify target module selection by regular expression patterns
2026-02-08 13:45:52 +09:00
kohya-ss
d992037984
format: format
2026-02-08 12:22:54 +09:00
kohya-ss
648c045cb0
feat: add anima minimal inference script (WIP)
2026-02-08 12:17:18 +09:00
kohya-ss
5c4643516c
feat: loading with dynamic fp8 optimization and LoRA support
2026-02-08 12:17:02 +09:00
kohya-ss
44b8d79577
feat: use unified attention module, add wrapper for state dict compatibility
2026-02-08 12:16:00 +09:00
kohya-ss
10445ff660
feat: simplify encode_tokens
2026-02-08 12:06:12 +09:00
kohya-ss
7b0ed3269a
feat: add Qwen-Image VAE
2026-02-08 11:13:09 +09:00
kohya-ss
9e2b4dc8a0
fix: update default value for --discrete_flow_shift in anima training guide
2026-02-08 10:56:34 +09:00
kohya-ss
3f3e2684c7
feat: add entry for 'temperal' in extend-words section of _typos.toml for Qwen-Image VAE
2026-02-08 10:41:13 +09:00
kohya-ss
65637517e8
fix: exclude anima tests from pytest
2026-02-08 10:36:33 +09:00
kohya-ss
402c468082
fix: update extend-exclude list in _typos.toml to include configs
2026-02-08 10:22:24 +09:00
duongve13112002
e21a7736f8
Support Anima model ( #2260 )
...
* Support Anima model
* Update document and fix bug
* Fix latent normlization
* Fix typo
* Fix cache embedding
* fix typo in tests/test_anima_cache.py
* Remove redundant argument apply_t5_attn_mask
* Improving caching with argument caption_dropout_rate
* Fix W&B logging bugs
* Fix discrete_flow_shift default value
2026-02-08 10:18:55 +09:00
Kohya S.
b996440c5f
Doc update sd3 branch documentation ( #2253 )
...
* doc: move sample prompt file documentation, and remove history for branch
* doc: remove outdated FLUX.1 and SD3 training information from README
* doc: update README and training documentation for clarity and structure
2026-01-19 21:38:46 +09:00
Kohya S.
a9af52692a
feat: add pyramid noise and noise offset options to generation script ( #2252 )
...
* feat: add pyramid noise and noise offset options to generation script
* fix: fix to work with SD1.5 models
* doc: update to match with latest gen_img.py
* doc: update README to clarify script capabilities and remove deprecated sections
2026-01-18 16:56:48 +09:00
Kohya S.
c6bc632ec6
fix: metadata dataset degradation and make it work ( #2186 )
...
* fix: support dataset with metadata
* feat: support another tagger model
* fix: improve handling of image size and caption/tag processing in FineTuningDataset
* fix: enhance metadata loading to support JSONL format in FineTuningDataset
* feat: enhance image loading and processing in ImageLoadingPrepDataset with batch support and output options
* fix: improve image path handling and memory management in dataset classes
* Update finetune/tag_images_by_wd14_tagger.py
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
* fix: add return type annotation for process_tag_replacement function and ensure tags are returned
* feat: add artist category threshold for tagging
* doc: add comment for clarification
---------
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com >
2026-01-18 15:17:07 +09:00
Kohya S.
f7f971f50d
Merge pull request #2251 from kohya-ss/fix-pytest-for-lumina
...
fix(tests): add ip_noise_gamma args for MockArgs in pytest
2026-01-18 15:09:47 +09:00
Kohya S
c4be615f69
fix(tests): add ip_noise_gamma args for MockArgs in pytest
2026-01-18 15:05:57 +09:00
Kohya S.
e06e063970
Merge pull request #2225 from urlesistiana/sd3_lumina2_ts_fix
...
fix: lumina 2 timesteps handling
2026-01-18 14:39:04 +09:00
Kohya S.
94e3dbebea
Merge pull request #2246 from kozistr/deps/pytorch-optimizer
...
Bump `pytorch-optimizer` version to v3.9.0
2025-12-21 22:51:32 +09:00
kozistr
95a65b89a5
build(deps): bump pytorch-optimizer to v3.9.0
2025-12-21 15:53:47 +09:00
Kohya S.
a5a162044c
Merge pull request #2226 from kohya-ss/fix-hunyuan-image-batch-gen-error
...
fix: error on batch generation closes #2209
2025-10-15 21:57:45 +09:00
Kohya S
a33cad714e
fix: error on batch generation closes #2209
2025-10-15 21:57:11 +09:00
