1580158
/
kohya_ss
mirror of https://github.com/bmaltais/kohya_ss
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
1,966 Commits
16 Branches
178 Tags
68 MiB
Commit Graph

257 Commits (b9d5db8a423a4e321546eb609bfba0c572306d24)

Author SHA1 Message Date
ykume 4e25c8f78e fix to work with Diffusers 0.17.0 2023-06-11 16:57:17 +09:00
Kohya S c0a7df9ee1 fix eps value, enable xformers, etc. 2023-06-03 21:29:27 +09:00
Kohya S ec2efe52e4 scale v-pred loss like noise pred 2023-06-03 10:52:22 +09:00
ddPn08 1e3daa247b
fix bucketing 2023-06-01 21:58:45 +09:00
ddPn08 3bd00b88c2
support for controlnet in sample output 2023-06-01 20:48:30 +09:00
ddPn08 62d00b4520
add controlnet training 2023-06-01 20:48:25 +09:00
ddPn08 c8d209d36c
update diffusers to 1.16 | train_network 2023-06-01 20:39:26 +09:00
AI-Casanova 9c7237157d
Dropout and Max Norm Regularization for LoRA training (#545) 
* Instantiate max_norm

* minor

* Move to end of step

* argparse

* metadata

* phrasing

* Sqrt ratio and logging

* fix logging

* Dropout test

* Dropout Args

* Dropout changed to affect LoRA only

---------

Co-authored-by: Kohya S <52813779+kohya-ss@users.noreply.github.com>
 2023-06-01 14:58:38 +09:00
Kohya S 3a06968332 warn and continue if huggingface uploading failed 2023-05-31 20:48:33 +09:00
Kohya S 990ceddd14 show warning if no caption and no class token 2023-05-30 22:53:50 +09:00
Kohya S 2429ac73b2
Merge pull request #533 from TingTingin/main 
Added warning on training without captions
 2023-05-29 08:37:33 +09:00
TingTingin db756e9a34
Update train_util.py 
I removed the sleep since it triggers per subset and if someone had a lot of subsets it would trigger multiple times
 2023-05-26 08:08:34 -04:00
青龍聖者@bdsqlsz 5cdf4e34a1
support for dadapaption V3 (#530) 
* Update train_util.py for DAdaptLion

* Update train_README-zh.md for dadaptlion

* Update train_README-ja.md for DAdaptLion

* add DAdatpt V3

* Alignment

* Update train_util.py for experimental

* Update train_util.py V3

* Update train_README-zh.md

* Update train_README-ja.md

* Update train_util.py fix

* Update train_util.py

---------

Co-authored-by: Kohya S <52813779+kohya-ss@users.noreply.github.com>
 2023-05-25 21:52:36 +09:00
TingTingin 061e157191
Update train_util.py 2023-05-23 02:02:39 -04:00
TingTingin d859a3a925
Update train_util.py 
fix mistake
 2023-05-23 02:00:33 -04:00
TingTingin 5a1a14f9fc
Update train_util.py 
Added feature to add "." if missing in caption_extension
Added warning on training without captions
 2023-05-23 01:57:35 -04:00
Kohya S 02bb8e0ac3 use xformers in VAE in gen script 2023-05-21 12:59:01 +09:00
Kohya S bc909e8359
Merge pull request #521 from akshaal/fix/save_state 
fix: don't save state if no --save-state arg given
 2023-05-21 08:48:48 +09:00
Evgeny Chukreev 0c942106bf
fix: don't save state if no --save-state arg given 2023-05-18 20:09:06 +02:00
Fair c0c4d4ddc6 new line with print "generating sample images" 2023-05-17 10:59:06 +08:00
青龍聖者@bdsqlsz 7e5b6154d0
Update train_util.py 2023-05-16 00:09:53 +08:00
Kohya S 714846e1e1 revert perlin_noise 2023-05-15 23:12:11 +09:00
Kohya S 08d85d4013 Merge branch 'dev' of https://github.com/kohya-ss/sd-scripts into dev 2023-05-15 20:58:04 +09:00
Kohya S 0ec7743436 show loading model path 2023-05-15 20:57:53 +09:00
Kohya S a72d80aa85
Merge pull request #507 from HkingAuditore/main 
Added support for Perlin noise in Noise Offset
 2023-05-15 20:56:46 +09:00
hkinghuang bca6a44974 Perlin noise 2023-05-15 11:16:08 +08:00
Linaqruf 8ab5c8cb28 feat: added json support as well 2023-05-14 19:49:54 +07:00
Linaqruf 774c4059fb feat: added toml support for sample prompt 2023-05-14 19:38:44 +07:00
Kohya S 968bbd2f47
Merge pull request #480 from yanhuifair/main 
fix print "saving" and "epoch" in newline
 2023-05-11 21:05:37 +09:00
Kohya S 09c719c926 add adaptive noise scale 2023-05-07 18:09:08 +09:00
Kohya S e54b6311ef do not save cuda_rng_state if no cuda closes #390 2023-05-07 10:23:25 +09:00
Fair b08154dc36 fix print "saving" and "epoch" in newline 2023-05-07 02:51:01 +08:00
Kohya S 165fc43655 fix comment 2023-05-06 18:25:26 +09:00
Kohya S 2127907dd3 refactor selection and logging for DAdaptation 2023-05-06 18:14:16 +09:00
青龍聖者@bdsqlsz 164a1978de
Support for more Dadaptation (#455) 
* Update train_util.py for add DAdaptAdan and DAdaptSGD

* Update train_util.py for DAdaptadam

* Update train_network.py for dadapt

* Update train_README-ja.md for DAdapt

* Update train_util.py for DAdapt

* Update train_network.py for DAdaptAdaGrad

* Update train_db.py for DAdapt

* Update fine_tune.py for DAdapt

* Update train_textual_inversion.py for DAdapt

* Update train_textual_inversion_XTI.py for DAdapt
 2023-05-06 17:30:09 +09:00
Kohya S 60bbe64489 raise error when both noise offset and multires 2023-05-03 20:58:12 +09:00
ykume f6556f7972 add ja help message for mutires noise 2023-05-03 11:31:13 +09:00
ykume 69579668bb Merge branch 'dev' of https://github.com/kohya-ss/sd-scripts into dev 2023-05-03 11:17:43 +09:00
Kohya S 2e688b7cd3
Merge pull request #471 from pamparamm/multires-noise 
Multi-Resolution Noise
 2023-05-03 11:17:21 +09:00
ykume 2fcbfec178 make transform_DDP more intuitive 2023-05-03 11:07:29 +09:00
Isotr0py e1143caf38
Fix DDP issues and Support DDP for all training scripts (#448) 
* Fix DDP bugs

* Fix DDP bugs for finetune and db

* refactor model loader

* fix DDP network

* try to fix DDP network in train unet only

* remove unuse DDP import

* refactor DDP transform

* refactor DDP transform

* fix sample images bugs

* change DDP tranform location

* add autocast to train_db

* support DDP in XTI

* Clear DDP import
 2023-05-03 10:37:47 +09:00
ykume a7485e4d9e Add error message if no Lion8bit 2023-05-03 10:35:47 +09:00
青龍聖者@bdsqlsz 335b2f960e
Support for Lion8bit (#447) 
* ADD libbitsandbytes.dll for 0.38.1

* Delete libbitsandbytes_cuda116.dll

* Delete cextension.py

* add main.py

* Update requirements.txt for bitsandbytes 0.38.1

* Update README.md for bitsandbytes-windows

* Update README-ja.md  for bitsandbytes 0.38.1

* Update main.py for return cuda118

* Update train_util.py for lion8bit

* Update train_README-ja.md for lion8bit

* Update train_util.py for add DAdaptAdan and DAdaptSGD

* Update train_util.py for DAdaptadam

* Update train_network.py for dadapt

* Update train_README-ja.md for DAdapt

* Update train_util.py for DAdapt

* Update train_network.py for DAdaptAdaGrad

* Update train_db.py for DAdapt

* Update fine_tune.py for DAdapt

* Update train_textual_inversion.py for DAdapt

* Update train_textual_inversion_XTI.py for DAdapt

* Revert "Merge branch 'qinglong' into main"

This reverts commit b65c023083d6d1e8a30eb42eddd603d1aac97650, reversing
changes made to f6fda20caf5e773d56bcfb5c4575c650bb85362b.

* Revert "Update requirements.txt for bitsandbytes 0.38.1"

This reverts commit 83abc60dfaddb26845f54228425b98dd67997528.

* Revert "Delete cextension.py"

This reverts commit 3ba4dfe046874393f2a022a4cbef3628ada35391.

* Revert "Update README.md for bitsandbytes-windows"

This reverts commit 4642c52086b5e9791233007e2fdfd97f832cd897.

* Revert "Update README-ja.md  for bitsandbytes 0.38.1"

This reverts commit fa6d7485ac067ebc49e6f381afdb8dd2f12caa8f.

* Revert "ADD libbitsandbytes.dll for 0.38.1"

This reverts commit bee1e6f731d2428dacb34b61997f06143c69c278.

* Revert "Delete libbitsandbytes_cuda116.dll"

This reverts commit 891c7e92623dab92f3767663982627cca6a26724.

* reverse main.py

* Reverse main.py
 2023-05-03 10:22:40 +09:00
Pam b18d099291 Multi-Resolution Noise 2023-05-02 09:42:17 +05:00
Kohya S 1890535d1b enable `cache_latents` when `_to_disk` #438 2023-04-25 08:08:49 +09:00
Kohya S 74008ce487 add `save_every_n_steps` option 2023-04-24 23:22:24 +09:00
Kohya S 46cbae088e fix to log with logging_dir without log_with 2023-04-23 19:15:48 +09:00
Kohya S 9ba4c3edca update readme/comment 2023-04-22 20:18:25 +09:00
Linaqruf e9a641bde7 Merge branch 'main' of https://github.com/Linaqruf/sd-scripts 2023-04-22 16:17:22 +07:00
Linaqruf ae3965a2a7 feat: add arguments to set \--wandb_api_key\ before training 2023-04-22 16:14:14 +07:00
Kohya S 66edc5af7b invert condition for checking log_with 2023-04-22 18:05:19 +09:00
saltacc dc37fd2ff6
fix no logging command line arg 2023-04-22 01:26:31 -07:00
Kohya S 884e6bff5d fix face_crop_aug not working on finetune method, prepare upscaler 2023-04-22 10:41:36 +09:00
Kohya S 220436244c some minor fixes 2023-04-22 09:55:04 +09:00
Kohya S c430cf481a
Merge pull request #428 from p1atdev/dev 
Add WandB logging support
 2023-04-22 09:39:01 +09:00
tsukimiya e746829b5f おそらくlibgtk2がインストールされていない環境でcv2.waitKey() および cv2.destroyAllWindows() が動作しないので除外 2023-04-20 06:20:02 +09:00
Plat a69b24a069 fix: tensorboard not working 2023-04-20 05:33:32 +09:00
Plat 8090daca40 fix: wandb not working without logging_dir 2023-04-20 05:14:28 +09:00
Plat 27ffd9fe3d feat: support wandb logging 2023-04-20 01:41:12 +09:00
Kohya S 423e6c229c support metadata json+.npz caching (no prepare) 2023-04-13 22:12:13 +09:00
Kohya S a8632b7329 fix latents disk cache 2023-04-13 21:14:39 +09:00
Kohya S 2e9f7b5f91 cache latents to disk in dreambooth method 2023-04-12 23:10:39 +09:00
AI-Casanova 0d54609435
Merge branch 'kohya-ss:main' into weighted_captions 2023-04-07 14:55:40 -05:00
AI-Casanova 7527436549
Merge branch 'kohya-ss:main' into weighted_captions 2023-04-05 17:07:15 -05:00
Kohya S 541539a144 change method name, repo is private in default etc 2023-04-05 23:16:49 +09:00
Kohya S 74220bb52c
Merge pull request #348 from ddPn08/dev 
Added a function to upload to Huggingface and resume from Huggingface.
 2023-04-05 21:47:36 +09:00
AI-Casanova 1892c82a60 Reinstantiate weighted captions after a necessary revert to Main 2023-04-02 19:43:34 +00:00
ddPn08 16ba1cec69
change async uploading to optional 2023-04-02 17:45:26 +09:00
ddPn08 8bfa50e283
small fix 2023-04-02 17:39:23 +09:00
ddPn08 c4a11e5a5a
fix help 2023-04-02 17:39:23 +09:00
ddPn08 3cc4939dd3
Implement huggingface upload for all scripts 2023-04-02 17:39:22 +09:00
ddPn08 b5ff4e816f
resume from huggingface repository 2023-04-02 17:39:21 +09:00
ddPn08 d42431d73a
Added feature to upload to huggingface 2023-04-02 17:39:10 +09:00
Yuta Hayashibe 9577a9f38d Check needless num_warmup_steps 2023-04-01 20:33:20 +09:00
Kohya S 31069e1dc5 add comments about debice for clarify 2023-03-30 21:44:40 +09:00
Kohya S 6c28dfb417
Merge pull request #332 from guaneec/ddp-lowram 
Reduce peak RAM usage
 2023-03-30 21:37:37 +09:00
Jakaline-dev b0c33a4294 Merge remote-tracking branch 'upstream/main' 2023-03-30 01:35:38 +09:00
Kohya S 4f70e5dca6 fix to work with num_workers=0 2023-03-28 19:42:47 +09:00
Kohya S 238f01bc9c fix images are used twice, update debug dataset 2023-03-27 20:48:21 +09:00
guaneec 3cdae0cbd2
Reduce peak RAM usage 2023-03-27 14:34:17 +08:00
Kohya S 14891523ce fix seed for each dataset to make shuffling same 2023-03-26 22:17:03 +09:00
Kohya S 6732df93e2
Merge branch 'dev' into min-SNR 2023-03-26 17:10:53 +09:00
Kohya S 4f42f759ea
Merge pull request #322 from u-haru/feature/token_warmup 
タグ数を徐々に増やしながら学習するオプションの追加、persistent_workersに関する軽微なバグ修正
 2023-03-26 17:05:59 +09:00
Jakaline-dev a35d7ef227 Implement XTI 2023-03-26 05:26:10 +09:00
u-haru a4b34a9c3c blueprint_args_conflictは不要なため削除、shuffleが毎回行われる不具合修正 2023-03-26 03:26:55 +09:00
u-haru 5a3d564a30 print削除 2023-03-26 02:26:08 +09:00
u-haru 4dc1124f93 lora以外も対応 2023-03-26 02:19:55 +09:00
u-haru 292cdb8379 データセットにepoch、stepが通達されないバグ修正 2023-03-26 01:44:25 +09:00
u-haru 1b89b2a10e シャッフル前にタグを切り詰めるように変更 2023-03-24 13:44:30 +09:00
u-haru 447c56bf50 typo修正、stepをglobal_stepに修正、バグ修正 2023-03-23 09:53:14 +09:00
u-haru a9b26b73e0 implement token warmup 2023-03-23 07:37:14 +09:00
AI-Casanova 64c923230e Min-SNR Weighting Strategy: Refactored and added to all trainers 2023-03-22 01:27:29 +00:00
AI-Casanova 795a6bd2d8
Merge branch 'kohya-ss:main' into min-SNR 2023-03-21 13:19:15 -05:00
Kohya S 7b324bcc3b support extensions of image files with uppercases 2023-03-21 21:10:34 +09:00
Kohya S 6d9f3bc0b2 fix different reso in batch 2023-03-21 18:33:46 +09:00
Kohya S 1816ac3271 add vae_batch_size option for faster caching 2023-03-21 18:15:57 +09:00
Kohya S cb08fa0379 fix no npz with full path 2023-03-21 15:05:25 +09:00
AI-Casanova a265225972 Min-SNR Weighting Strategy 2023-03-20 22:51:38 +00:00
Kohya S de95431895 support win with diffusers, fix extra args eval 2023-03-19 22:09:36 +09:00
Kohya S 48c1be34f3
Merge branch 'dev' into main 2023-03-19 21:58:41 +09:00