6b54867c3c
fixes
2024-12-25 21:34:56 +05:30
5aef4a6c80
Merge pull request #148 from a-r-r-o-w/deepspeed-changes
...
Grad Norm tracking in DeepSpeed
2024-12-25 20:08:09 +05:30
47e4aaac2f
track grad norm for deepspeed
2024-12-25 15:04:16 +01:00
e2178deea0
fix deadlock; make style
2024-12-25 14:16:31 +01:00
d7aa50917f
fixes
2024-12-25 12:07:50 +05:30
69b585a992
fixes
2024-12-25 11:35:09 +05:30
875cb72f63
address review feedback.
2024-12-24 21:20:37 +05:30
6329e9ed07
support DeepSpeed.
2024-12-24 10:59:21 +05:30
cf9be17819
Merge pull request #129 from a-r-r-o-w/condition-precomputation
...
Precomputation of conditions and latents
2024-12-23 20:00:34 +05:30
2858346b27
address review
2024-12-23 15:11:03 +01:00
1c8f33967c
add note about grad ckpt in readme
2024-12-23 14:59:05 +01:00
8c0c28c971
create markdown tables
2024-12-23 14:16:47 +01:00
78712f60a5
update comment
2024-12-23 14:01:39 +01:00
75a854f9df
add coauthor
...
Co-Authored-By: Sayak Paul <spsayakpaul@gmail.com >
2024-12-23 13:59:45 +01:00
5261bed284
remove pipeline hooks
2024-12-23 13:58:37 +01:00
86e1d1888a
other condtions -> text conditions
2024-12-23 13:48:58 +01:00
363841ae3f
address some reviews; change hunyuan checkpoint repo-id
2024-12-23 12:54:52 +01:00
b045037850
make style
2024-12-21 15:19:03 +01:00
0db7bf28aa
update
2024-12-21 15:18:46 +01:00
2da8cb5798
update
2024-12-21 12:24:33 +01:00
66189c804b
update
2024-12-20 10:03:39 +01:00
223add1a59
Hunyuan Video LoRA ( #126 )
...
* add hunyuan-video lora support
* minor fixes; make style
* update readme
* update
* update
* Update README.md
* Update README.md
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
* update
* update
* change move train scripts to internal directory
* update
---------
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com >
2024-12-20 08:02:08 +05:30
9ef58e2f3a
LTX Video ( #123 )
...
* rename files
* ltx finetuning
* update
* update
* improvements
* make style
* gradient clipping
* update
* fix distributed inference
* update
* update
2024-12-19 04:27:36 +05:30
80d1150a0e
add I2V sft and fix an error ( #97 )
...
* adaption for CogVideoX1.5
* add patch_size_t in full finetuning of T2V and lora finetuning of I2V
* add I2V sft and add ofs to adapt cogvideox 1.5
2024-12-06 01:59:57 +05:30
1c7d796629
Merge pull request #112 from a-r-r-o-w/sayakpaul-patch-1
...
Update README.md to include mochi-1 trainer
2024-12-05 15:11:31 +05:30
aa8612fc06
Update README.md to include mochi-1 trainer
2024-12-05 12:59:13 +05:30
6c41984d8a
RoPE fixes for 1.5, bfloat16 support in prepare_dataset, gradient_accumulation grad norm undefined fix ( #107 )
...
* rope scaling fixes for cog 1.5
* grad norm log fix
* bf16 prepare dataset fix
* update
* style
2024-12-02 15:12:12 +05:30
76a9f2b911
Merge pull request #106 from a-r-r-o-w/resume-mochi-1
...
feat: support checkpointing saving and loading
2024-12-01 20:43:25 +05:30
cca97bfd40
readme
2024-12-01 20:35:51 +05:30
79b546b264
feat: support checkpointing.
2024-12-01 20:34:35 +05:30
ec3ec47891
Unbound fix ( #105 )
...
* is some error
```
[rank1]: File "/mnt/ceph/develop/jiawei/cogvideox-distillation/training/cogvideox_image_to_video_lora.py", line 1005, in <module> [rank1]: main(args) [rank1]: File "/mnt/ceph/develop/jiawei/cogvideox-distillation/training/cogvideox_image_to_video_lora.py", line 884, in main [rank1]: "gradient_norm_before_clip": gradient_norm_before_clip, [rank1]: UnboundLocalError: local variable 'gradient_norm_before_clip' referenced before assignment
```
ref1 :https://github.com/a-r-r-o-w/cogvideox-factory/pull/84
ref2: https://github.com/a-r-r-o-w/cogvideox-factory/pull/100
It seems that there is a bug in the accelerator.is_main_process task and accelerator.distributed_type. Is_main_process has scheduling problems during the training initiation and training stages.
* fix
2024-12-01 01:58:27 +05:30
099c9c35c4
Merge pull request #104 from a-r-r-o-w/fix-format-specifier
...
fix format specifier.
2024-11-29 16:29:58 +05:30
c8a118cefd
fix format specifier.
2024-11-29 16:23:14 +05:30
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
sayakpaul