d5cc7c66f1
Merge pull request #100 from a-r-r-o-w/wandb-tracker-fix
...
wandb tracker in scheduling problems during the training initiation and training stages
2024-11-29 15:08:19 +05:30
d10963f707
Merge pull request #90 from a-r-r-o-w/mochi-1-tuner
...
[feat] add Mochi-1 trainer
2024-11-29 14:56:20 +05:30
7090bcb9de
Update README.md
2024-11-29 13:47:26 +05:30
dceded0a94
add a preview table
2024-11-29 13:40:02 +05:30
0011fa1406
better command
2024-11-29 13:34:05 +05:30
95775ba5d1
pyav pining.
2024-11-29 13:27:23 +05:30
8a32b139af
pin moviepy.
2024-11-29 13:22:23 +05:30
38f157c6d0
Apply suggestions from code review
...
Co-authored-by: Aryan <aryan@huggingface.co >
2024-11-29 13:11:08 +05:30
9c86706cbb
fix help message
2024-11-29 13:08:55 +05:30
e1866d844d
better example code.
2024-11-29 13:07:26 +05:30
4e3bb7ab5f
updates
2024-11-29 10:52:12 +05:30
ced8558eb0
updates
2024-11-29 10:48:54 +05:30
2fde026d30
updates
2024-11-28 12:33:25 +05:30
a9cf49ecea
wandb tracker in scheduling problems during the training initiation and training stages.
2024-11-27 17:14:28 +08:00
62a82a8647
Merge pull request #84 from zhipuch/main
...
sft with multigpu gradient_norm_before_clip error update
2024-11-27 17:02:28 +08:00
559ba47993
docs: fix help message in args.py ( #98 )
2024-11-27 08:27:41 +05:30
58a06320ba
updates
2024-11-26 12:28:44 +05:30
0b80dba1de
adaption for CogVideoX1.5 ( #92 )
...
* adaption for CogVideoX1.5
* add patch_size_t in full finetuning of T2V and lora finetuning of I2V
* Update training/args.py
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
2024-11-23 06:28:20 +05:30
2dbddd566e
updates
2024-11-22 23:34:17 +05:30
9a15eaee41
updates
2024-11-20 19:31:13 +05:30
7e203db9aa
updates
2024-11-20 10:39:52 +05:30
9eea656fa0
fixes
2024-11-19 18:38:05 +05:30
5208c59394
fix
2024-11-19 11:36:56 +05:30
a01592b25d
betterments.
2024-11-19 11:32:24 +05:30
440dc257c8
better reuse.
2024-11-19 10:51:00 +05:30
4216705510
Merge branch 'main' into mochi-1-tuner
2024-11-19 10:29:48 +05:30
316a705f6a
revert changes to training/dataset.py
2024-11-19 10:29:16 +05:30
c80bd177fd
nearest_frame_bucket.
2024-11-19 10:27:24 +05:30
1409d478d5
updates.
2024-11-19 10:21:28 +05:30
ac83c78a5a
updates
2024-11-18 15:32:04 +05:30
a8dd94ee1c
updates
2024-11-18 15:31:11 +05:30
4c05ea6669
updates
2024-11-18 15:30:23 +05:30
a40ccd2237
updates
2024-11-18 15:28:37 +05:30
9852c3d994
updates.
2024-11-18 15:26:52 +05:30
5ba510e061
updates
2024-11-17 10:18:20 +05:30
a9adf22386
dataprep.
2024-11-17 08:31:21 +05:30
ac5f417555
sft with multigpu
2024-11-12 15:20:41 +08:00
d63a826f37
I2V multiresolution finetuning by removing learned PEs ( #31 )
...
* i2v finetuning without learned pe
* update
* update
* update
* update readme
* refactor
* refactor
2024-11-08 07:59:12 +05:30
db9b295408
Windows support for T2V scripts ( #48 )
2024-10-21 02:40:05 +05:30
6c00cf094b
fix: resuming from a checkpoint when using deepspeed. ( #38 )
...
* fix: resuming from a checkpoint when using deepspeed.
* remove changes to prepare_dataset.py
* propagate to others.
* tackle gradnorm.
---------
Co-authored-by: Aryan <aryan@huggingface.co >
2024-10-20 02:56:32 +05:30
f0d9908bbd
fix: correct type in .py files ( #52 )
2024-10-20 01:22:07 +05:30
8da9638896
more dataset fixes from stashed changes ( #49 )
2024-10-19 06:30:26 +05:30
f2a1626de2
improve dataset preparation ( #43 )
...
* update
* update
* update
* update
2024-10-18 02:43:22 +05:30
feb2e26547
Improve dataset preparation support + multiresolution prep ( #39 )
...
* update
* make style
* renormalize correctly
* apply suggestions from review
* apply suggestions from review
* update
2024-10-17 16:48:52 +05:30
4f2744ee59
Update for windows compability ( #32 )
...
* replaced lambda statements with local functions to work with pickle in dataset.py and moved collate function outside of main and added its own class in cogvideox_image_to_video_lora.py
* Update cogvideox_image_to_video_lora.py
Remove the bug-fix related to image encoding since it's already present in #31
* Update cogvideox_image_to_video_lora.py
Revert to last commit (60c46827404cd4780ee0e7d3c9c56a953e2a04ca)
2024-10-15 19:10:50 +05:30
ee02219eac
readme updates + refactor ( #14 )
...
* refactor
* update
* update
* Update README.md
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
* Update README.md
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
* address review comments part i
* fix adapter name causing peft error
* prompts -> prompt
* update
* add video
* update
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
2024-10-10 18:03:45 +05:30
2a70b16d3c
add "max_sequence_length": model_config.max_text_seq_length, ( #15 )
...
Co-authored-by: --unset <--unset>
2024-10-10 17:08:40 +05:30
00f7519f0e
add VideoDatasetWithResizeAndRectangleCrop dataset resize crop ( #13 )
...
* fix issue 12: train dataset _preprocess_data crop video
* add_argument video_reshape_mode input videos are reshaped to this mode
* arg enable_model_cpu_offloading
* Add imageio-ffmpeg and imageio to requirements.txt
* apply changes to i2v lora script
* make style
---------
Co-authored-by: --unset <--unset>
Co-authored-by: Aryan <aryan@huggingface.co >
2024-10-10 01:55:12 +05:30
b1b72c0e38
CogVideoX I2V; CPU offloading; Model README descriptions ( #11 )
...
* update
* update
* update readme
* update
* update
* update model desc
* update
* Update training/prepare_dataset.py
2024-10-09 17:44:59 +05:30
cc1d2e759f
Multi-GPU parallel encoding support for training videos. ( #6 )
...
* Multi-GPU parallel encoding support for training videos.
* revert
* make style
* update
---------
Co-authored-by: Aryan <aryan@huggingface.co >
2024-10-09 04:19:59 +05:30
Sayak Paul