Commit Graph

503 Commits

Author SHA1 Message Date
Isotr0py
db84530074 Fix gradients synchronization for multi-GPUs training (#989)
* delete DDP wrapper

* fix train_db vae and train_network

* fix train_db vae and train_network unwrap

* network grad sync

---------

Co-authored-by: Kohya S <52813779+kohya-ss@users.noreply.github.com>
2023-12-07 22:01:42 +09:00
Kohya S
81a440c8e8 Merge pull request #955 from xzuyn/paged_adamw
Add PagedAdamW
2023-12-03 21:22:38 +09:00
Kohya S
f24a3b5282 show seed in generating samples 2023-12-03 21:15:30 +09:00
Kohya S
383b4a2c3e Merge pull request #907 from shirayu/add_option_sample_at_first
Add option --sample_at_first
2023-12-03 21:00:32 +09:00
Kohya S
df59822a27 Merge pull request #906 from shirayu/accept_scheduler_designation_in_training
Accept sampler designation in sampling of training
2023-12-03 20:46:16 +09:00
Furqanil Taqwa
1bdd83a85f remove unnecessary debug print 2023-11-28 17:26:27 +07:00
Furqanil Taqwa
1624c239c2 added keep_tokens_separator to dynamically keep token for being shuffled 2023-11-28 17:23:55 +07:00
Kohya S
f312522cef Merge pull request #913 from KohakuBlueleaf/custom-seperator
Add custom seperator for shuffle caption
2023-11-19 21:32:01 +09:00
xzuyn
da5a144589 Add PagedAdamW 2023-11-18 07:47:27 -05:00
rockerBOO
9c591bdb12 Remove unnecessary subset line from collate 2023-11-05 16:58:20 -05:00
rockerBOO
3de9e6c443 Add validation split of datasets 2023-11-05 12:37:44 -05:00
rockerBOO
5b19bda85c Add validation loss 2023-11-05 12:35:46 -05:00
Kohya S
6231aa91e2 common lr logging, set default None to ddp_timeout 2023-11-05 19:09:17 +09:00
Kohaku-Blueleaf
583e2b2d01 Fix typo 2023-10-30 20:02:04 +08:00
Kohaku-Blueleaf
5dc2a0d3fd Add custom seperator 2023-10-30 19:55:30 +08:00
Yuta Hayashibe
fea810b437 Added --sample_at_first to generate sample images before training 2023-10-29 21:44:57 +09:00
Yuta Hayashibe
40d917b0fe Removed incorrect comments 2023-10-29 21:02:44 +09:00
Yuta Hayashibe
cf876fcdb4 Accept --ss to set sample_sampler dynamically 2023-10-29 20:15:04 +09:00
Yuta Hayashibe
291c29caaf Added a function line_to_prompt_dict() and removed duplicated initializations 2023-10-29 19:57:25 +09:00
Yuta Hayashibe
01e00ac1b0 Make a function get_my_scheduler() 2023-10-29 19:46:02 +09:00
Kohya S
a9ed4ed8a8 Merge pull request #900 from xzuyn/paged_adamw_32bit
Add PagedAdamW32bit
2023-10-29 15:01:55 +09:00
Kohya S
9d6a5a0c79 Merge pull request #899 from shirayu/use_moving_average
Show moving average loss in the progress bar
2023-10-29 14:37:58 +09:00
Kohya S
fb97a7aab1 Merge pull request #898 from shirayu/update_repare_buckets_latents
Fix a typo and add assertions in making buckets
2023-10-29 14:29:53 +09:00
Kohaku-Blueleaf
1cefb2a753 Better implementation for te autocast (#895)
* Better implementation for te

* Fix some misunderstanding

* as same as unet, add explicit convert

* Better cache TE and TE lr

* Fix with list

* Add timeout settings

* Fix arg style
2023-10-28 15:49:59 +09:00
xzuyn
d8f68674fb Update train_util.py 2023-10-27 07:05:53 -04:00
Yuta Hayashibe
0d21925bdf Use @property 2023-10-27 18:14:27 +09:00
Yuta Hayashibe
3d2bb1a8f1 Add LossRecorder and use moving average in all places 2023-10-27 17:49:49 +09:00
Yuta Hayashibe
837a4dddb8 Added assertions 2023-10-26 13:34:36 +09:00
Isotr0py
592014923f Support JPEG-XL on windows 2023-10-04 21:48:25 +08:00
Yuta Hayashibe
27f9b6ffeb updated typos to v1.16.15 and fix typos 2023-10-01 21:51:24 +09:00
Kohya S
360af27749 fix ControlNetDataset not working 2023-09-03 12:27:58 +09:00
Kohya S
0ee75fd75d fix typos, add comments etc. 2023-09-03 12:24:15 +09:00
Kohya S
2eae9b66d0 Merge pull request #798 from vvern999/vvern999-patch-1
add input perturbation noise
2023-09-03 10:51:23 +09:00
Kohya S
948cf17499 add caption_prefix/suffix to dataset 2023-09-02 16:17:12 +09:00
Kohya S
497051c14b Merge pull request #786 from Isotr0py/jxl
Support JPEG XL
2023-09-02 15:30:07 +09:00
vvern999
e0beb6a999 add input perturbation noise
from https://arxiv.org/abs/2301.11706
2023-09-02 07:33:27 +03:00
Kohya S
7e850f3b7e Merge branch 'main' into sdxl 2023-09-01 07:59:26 +09:00
Isotr0py
5d88351bb5 support jpeg xl 2023-08-25 11:07:02 +08:00
Kohya S
1161a5c6da fix debug_dataset for controlnet dataset 2023-08-20 17:39:48 +09:00
Kohya S
e191892824 fix bucketing doesn't work in controlnet training 2023-08-20 12:24:40 +09:00
Kohya S
3f7235c36f add lora controlnet train/gen temporarily 2023-08-17 10:08:02 +09:00
Kohya S
3307ccb2dc revert default noise offset to 0 (None) in sdxl 2023-08-11 20:35:46 +09:00
Kohya S
6889ee2b85 add warning for bucket_reso_steps with SDXL 2023-08-11 19:02:36 +09:00
Kohya S
e73d103eca fix sample gen failed in sdxl training 2023-08-11 16:58:52 +09:00
Kohya S
daad50e384 fix to work when input_ids has multiple EOS tokens 2023-08-10 20:13:59 +09:00
Kohya S
6f80fe17fc fix crashing in saving lora with clipskip 2023-08-08 21:03:16 +09:00
Kohya S
c142dadb46 support sai model spec 2023-08-06 21:50:05 +09:00
Kohya S
e5f9772a35 fix training textencoder in sdxl not working 2023-08-05 21:22:50 +09:00
reid3333
a02056c566 fix: load may fail if symbolic link points to relative path 2023-08-05 17:47:43 +09:00
Kohya S
9d855091bf make bitsandbytes optional 2023-08-04 22:29:14 +09:00