Commit Graph
86 Commits
Author SHA1 Message Date
hkunzhe 3f9a13ff63 update reward training 2025-01-10 16:47:57 +08:00
hkunzhe dd01a9bfee update train_reward_lora.py 2024-12-27 10:24:21 +08:00
bubbliiiing 1ca8119b1f Merge branch 'main' into v5.1 2024-12-26 07:16:54 +00:00
bubbliiiing fece877980 Fix bug in t2v 2024-12-26 07:15:55 +00:00
hkunzhe a4443d40ef add image recaptioning 2024-12-25 11:15:51 +08:00
bubbliiiing da09cc4983 Update Flow 2024-12-24 09:10:25 +00:00
bubbliiiing 938aac0c5f Update Flow 2024-12-24 09:09:49 +00:00
bubbliiiing 2ac1956384 Update Flow 2024-12-24 07:25:59 +00:00
2eb809a0c8 Add lora cache mode in comfyui && Use cuda in merge_lora func for speeding up. (#172)
* fix lora precision problem during multiple merge unmerge, keep the original weights on cpu

* add comments

* update

* Update Lora cache

---------

Co-authored-by: mengli.cml <mengli.cml@alibaba-inc.com>
Co-authored-by: bubbliiiing <3323290568@qq.com>
2024-12-24 10:02:57 +08:00
hkz 58b27cee1c fix caption rewrite (#169) 2024-12-12 18:21:58 +08:00
hkz 485e771acf Fix train reward lora and video caption (#165) 2024-12-10 14:23:33 +08:00
Bubbliiiing ccbdce4492 Update Readme (#161)
* Update Readme

* Fix bug in training

* Update Readme
2024-12-02 11:27:57 +08:00
hkzandbubbliiiing 17c1f02ba8 Add reward LoRA training (#160)
* Add reward lora

* Fix bug in lora utils && Update Readme

---------

Co-authored-by: bubbliiiing <3323290568@qq.com>
2024-11-28 13:59:53 +08:00
Bubbliiiing f419bf850b Update EasyAnimateV5-7B and reduce vram in vae decoding (#158) 2024-11-22 14:20:29 +08:00
Bubbliiiing 6d15f2799f Update readme (#155) 2024-11-20 16:03:55 +08:00
Bubbliiiing 94f420c257 Low ram support. (#154)
* Fix bug in install comfyui

* support model load in 30GB RAM

* Update requirement

* Update requirement

* Update VAE outlier_penalty_loss

* Fix bug in ui

* Fix bug in ui

* Fix bug in ui
2024-11-20 15:56:28 +08:00
Bubbliiiing 9d8fed9cec Fix bug in install comfyui (#148) 2024-11-13 19:49:18 +08:00
hkz 30682ba3c5 Update video caption (#129) 2024-11-13 17:43:21 +08:00
Bubbliiiing 5dde5c9a7e Update Reference (#143)
* Fix bug in comfyui t2v

* Fix bug in textbox

* Update Reference
2024-11-12 12:57:30 +08:00
Bubbliiiing 59bb7f0169 Bug fix in TextBox (#141)
* Fix bug in comfyui t2v

* Fix bug in textbox
2024-11-12 10:59:18 +08:00
Bubbliiiing 2f76bace4d Fix bug in comfyui t2v (#140) 2024-11-12 10:28:13 +08:00
yunkchen 3e3e4756ce Update README.md (#139) 2024-11-12 10:25:00 +08:00
Ikko Eltociear Ashimine 95f11a5403 docs: add Japanese README file (#138)
I created Japanese translated README.
2024-11-12 10:19:27 +08:00
Bubbliiiing 9da6b79ece Update Prompt && Update Readme && Fix control bug (#131)
* Update Readme

* Update prompt tips

* Fix bug in ui

* Fix control bug and Update readme

* Update prompt again
2024-11-11 13:48:53 +08:00
Bubbliiiing 44539fe77b Update Readme (#130) 2024-11-08 22:38:17 +08:00
Bubbliiiing 62de94e2f1 Update V5 (#128)
Update V5
2024-11-08 19:46:50 +08:00
yunkchen abcf42bce7 Update requirements.txt (#124)
Ensure gradio version.
2024-10-18 17:04:56 +08:00
Bubbliiiing d3b8bbbd14 Fix bug in v3 training (#98) 2024-08-22 16:03:14 +08:00
hkz 5ea1bf2450 Update Video Caption (#93)
* Add VILA1.5 in video caption

* update VILA1.5 in video_caption

* add get_video_path_list back & update the accelerate version & auto load the vila quant_path & download models in the main process

* add exception handling & add force_ascii=False

* fix the NCCL timeout & use logger

* update utils

* Update video splitting

* Update pre-filtering

* Update video caption

* Add caption_rewrite.py

* Add VideoCLIPXL

* Add beautiful prompt demo

* fix test

* Update Dockerfile.ds

* update VideoCLIPXL & add filter_meta_train.py

* update stage3

* update README.md

* update README_zh-CN.md

* update README

* update requirements

* fix stage_3_video_recaptioning.sh

* update Beautiful Prompt

* update README

* update vila_video_recaptioning.py & fix the empty gather result
2024-08-21 16:40:54 +08:00
Bubbliiiing f2f0cbccc9 Bug fix/encode prompt (#96)
* update readme

* fix bug in encode_prompt
2024-08-19 16:53:05 +08:00
Bubbliiiing aa3d2afb34 update readme (#95) 2024-08-19 12:02:43 +08:00
Bubbliiiing c7c2fc1cd0 update readme (#94) 2024-08-19 11:26:41 +08:00
e4f1a4fe97 Update V4 version (#92)
* update train_lora && update deepspeed && update training with max token length

* fix bug in train.py

* fix bug in training_with_video_token_length

* update v2v && update v2v api

* add rope2d embedding precomputation; move text encoder to dataloader to reduce gpu memory consumpution

* add cuda multi-stream to speedup vae encode

* update new vae && new comfyui

* fix some bug in training code

* Add lcm lora (#89)

Co-authored-by: xuanyuan.lb <xuanyuan.lb@alibaba-inc.com>

* Update Training Code and fix bug in low vram mode

* fix bug in low vram mode

* update report

* update cfg

* actual text clip

---------

Co-authored-by: mengli.cml <mengli.cml@alibaba-inc.com>
Co-authored-by: liubo0902 <38622806+liubo0902@users.noreply.github.com>
Co-authored-by: xuanyuan.lb <xuanyuan.lb@alibaba-inc.com>
2024-08-19 11:22:17 +08:00
Bubbliiiing b54412ceb0 support float16 in comfyui (#75) 2024-07-24 14:01:35 +08:00
bubbliiiing 039d67acf1 support float16 && add reference && fix bug in training 2024-07-24 13:19:03 +08:00
bubbliiiing 1a5bab2234 fix bug in no inpaint model 2024-07-18 21:09:10 +08:00
bubbliiiing 2eef3f9f78 update v4 2024-07-18 16:33:56 +08:00
Bubbliiiing 616a35425d Fix the issue where Lora training cannot load state, fix the issue where training cannot eval (#66)
* rename the comfyui files and new readme

* fix bug in eval and load lora state dict
2024-07-18 15:47:51 +08:00
Bubbliiiing 8b7722463e rename the comfyui files and new readme (#53) 2024-07-13 14:57:13 +08:00
Bubbliiiing 883c0a20a8 update readme (#52) 2024-07-12 18:03:48 +08:00
Bubbliiiing e6ec6f0e04 Update init and install in comfyui (#51)
Update init and install in comfyui
2024-07-12 17:58:12 +08:00
Bubbliiiing fbbfc818ea Update pipeline (#50) 2024-07-12 17:47:47 +08:00
yunkchen 3cc15c456a Merge pull request #49 from aigc-apps/comfyui
Support for comfyui
2024-07-12 16:54:06 +08:00
bubbliiiing deaa46ff42 update readme for comfyui 2024-07-12 16:31:50 +08:00
bubbliiiing 8fd346a9af Merge branch 'main' into comfyui 2024-07-12 16:26:34 +08:00
bubbliiiing c1d7e503d3 update comfyui 2024-07-12 16:24:46 +08:00
yunkchenandbubbliiiiing 047712c8bd Add Discord (#48)
Add Discord
---------

Co-authored-by: bubbliiiiing <3323290568@qq.com>
2024-07-12 11:41:24 +08:00
Bubbliiiing 6ed8619ad4 Rename the files (#45)
* add update readme

* update demo show and prompt

* rename the train code
2024-07-10 17:46:02 +08:00
bubbliiiing 4f6bd0f6a6 rename the train code 2024-07-10 17:42:12 +08:00
yunkchen 19b38f674e Merge pull request #44 from wangqiang9/main
Added exception handling for video read bucket_sampler.py
2024-07-10 16:35:09 +08:00