Commit Graph
423 Commits
Author SHA1 Message Date
JerryZhou54 92dfbfe4fc Refactor ode trajectory generation
Add env variable to enforce running flash attn
2025-09-26 01:26:03 +00:00
JerryZhou54 d31695b736 Test 2025-09-22 22:33:18 +00:00
JerryZhou54 357eaa4f1e small change 2025-09-20 05:02:06 +00:00
JerryZhou54 ba02d0ae7e Numerical Debugging 2025-09-20 05:02:06 +00:00
SolitaryThinker c499783f74 fix ode_init 2025-09-20 05:02:06 +00:00
JerryZhou54 586013575c Resolve timestep mismatch between dit forward and pred_noise_to_pred_video 2025-09-20 05:02:05 +00:00
SolitaryThinker 97e33f9097 debug 2025-09-20 05:02:05 +00:00
SolitaryThinker 8674224232 comment out vmoba 2025-09-20 05:02:04 +00:00
JerryZhou54 24d35b5512 Fix test for forward_train 2025-09-20 05:01:27 +00:00
JerryZhou54 3d2af0823c Ensure 0 numerical diff for forward_train 2025-09-20 05:01:27 +00:00
JerryZhou54 27bac993ac Add test for _forward_train 2025-09-20 05:01:27 +00:00
JerryZhou54 15a89d3f27 Fix numerical diff between causal_wanvideo.py and SF's causal wan 2025-09-20 05:01:25 +00:00
RandNMR73 ddf5e76ceb text preprocessing ready 2025-09-20 05:00:53 +00:00
JerryZhou54 1f78d8114e checkpoint 2025-09-20 05:00:06 +00:00
JerryZhou54 8cc2c113b8 Enable timestep warping & using SelfForcing scheduler 2025-09-20 04:59:33 +00:00
JerryZhou54 0b1519db19 Stop backprop through kv_cache 2025-09-20 04:59:33 +00:00
RandNMR73 055e7dc47b new branch 2025-09-20 04:59:31 +00:00
RandNMR73 0892c2c448 inference works after changes added 2025-09-20 04:57:54 +00:00
SolitaryThinker c501a484e0 lmdb datasets 2025-09-20 04:57:54 +00:00
SolitaryThinker d965f578c9 t2v to i2v finetune
checkpoint ode

checkpoint

fix t2v to i2v

lint

chekpt

checkpoint

hacked but working

ode_init scripts

WIP fixing time embedding

WIP fixing time embedding

checkpoint

update

fix

revert

revert

revert

update

update

visualize
2025-09-20 04:57:52 +00:00
William Lin c5f9ea53b2 [self-forcing] [4/n] Preprocessing for collecting ODE trajectory (#788) 2025-09-15 17:54:42 -07:00
William Lin d32a7184da [bugfix] Wan2.2 Boundary ratio (#804) 2025-09-15 11:17:35 -07:00
Wenxuan Tanandgemini-code-assist[bot] 2930abe456 [Bugfix] Fix VMoba requirements (#802)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-09-14 18:28:52 -07:00
William Lin b93ef4289d [bugfix] Fix empty PipelineConfigs for Wan2.2 A14B (#800) 2025-09-13 17:31:38 -07:00
401bdbd316 [self-forcing] [3/n] Text embed only preprocessing (#797)
Co-authored-by: RandNMR73 <notomatthew31@gmail.com>
Co-authored-by: JerryZhou54 <zhouw.jerry2017@outlook.com>
Co-authored-by: kevin314 <kevin.lin.cs1@gmail.com>
2025-09-13 14:03:53 -07:00
William Lin 1048d79cf8 [bugfix] pin gradio version and set current_vsa_sparsity in TrainingPipeline (#798) 2025-09-11 17:04:47 -07:00
1e8406162d [bugfix] Fix delta calculation (#796)
Co-authored-by: zbchu2 <zbchu2@iflytek.com>
Co-authored-by: William Lin <SolitaryThinker@users.noreply.github.com>
2025-09-11 16:31:23 -07:00
William Lin 03edd35c83 [preprocessing] [self-forcing] [2/n] Improve preprocessing and add ode trajectory dataset schema (#794) 2025-09-10 17:33:57 -07:00
William LinandRandNMR73 ac11127397 [Self-forcing] [1/n] Handle extra dim in time embedding and add timestep warping (#792)
Co-authored-by: RandNMR73 <notomatthew31@gmail.com>
2025-09-09 02:52:02 -07:00
Eric LiangandEricLiang e028dcc7c0 [Backend][Vmoba] Add implementation of VMoba (#778)
Co-authored-by: EricLiang <https://github.com/EricLina>
2025-09-08 23:53:25 -07:00
Wenxuan Tanandgemini-code-assist[bot] 076f45c1ee [Feature] Support Lora for DMD (#755)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-09-08 14:18:21 -07:00
85eb7265db fix: lora_B init zeros (#781)
Co-authored-by: zbchu2 <zbchu2@iflytek.com>
Co-authored-by: Wenxuan Tan <wenxuan.tan@wisc.edu>
2025-09-05 22:56:52 -07:00
William Lin d3ceb67e66 [misc] Update Slack invite link (#786) 2025-09-05 12:16:18 -07:00
Zhang Peiyuan 7ac153a5ca Update WeChat Link 2025-09-05 11:40:47 -07:00
William Lin d1e7aa0abd [CI] Add ssim test for causal inference (#784) 2025-09-05 01:23:01 -07:00
William Lin 2d846c55a1 [misc] Improve text encoding stage (#774) 2025-09-04 17:51:27 -07:00
Jinzhe Pan b318063c0a [Preprocess][Fix] video quality issue (#773) 2025-09-03 20:47:33 -07:00
Jinzhe Pan 4aa307be55 [Preprocess][Feat] support torchvision to load video in new preprocessing (#761) 2025-09-01 23:37:01 -07:00
William Lin 055e52e5ea [misc] [VSA] [STA] fix tk_root in setup.py for VSA and STA (#772) 2025-08-29 01:13:37 -07:00
William Lin 7d2069596b [bugfix] [VSA] [STA] Fix MANIFEST.in for VSA and STA; Move tk into both directories (#771) 2025-08-29 00:51:05 -07:00
William Lin c45009c9a4 [bugfix] fix STA install setup.py import (#770) 2025-08-28 23:02:53 -07:00
William LinandPeiyuan Zhang b91020b407 [VSA] [STA] Fix directory structure for pypi publishing (#769)
Co-authored-by: Peiyuan Zhang <a1286225768@gmail.com>
2025-08-28 22:34:03 -07:00
William Lin 2dcc5ea4f6 [chore] Release 0.1.6 (#768) v0.1.6 2025-08-28 20:56:21 -07:00
Wei ZhouandSolitaryThinker 359151d9a0 [Feature] Add wan2.2 5b i2v (#760)
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com>
2025-08-28 18:15:59 -07:00
Wei ZhouandSolitaryThinker ce67cd3729 [Feat] Support Self-Forcing's Causal Inference for Wan2.1 T2V 1.3B (#766)
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com>
2025-08-28 16:47:49 -07:00
Zhang Peiyuan 7c554e5da8 Update Community Link (#765) 2025-08-27 16:12:47 -07:00
William Lin 663ea33ff1 [bugfix] Fix wrong HF model string for FastWan2.2 5B (#763) 2025-08-26 22:05:40 -07:00
William Lin 3ef04f1654 [misc] [docs] Various fixes for logging and docs (#758) 2025-08-23 21:13:50 -07:00
Jinzhe Pan 0eced76a41 [Feat][Preprocess] support multi-gpus (#753) 2025-08-23 11:34:42 +08:00
Jinzhe Pan 3ab6470d1a [Feat][Preprocess] support merged dataset (#752) 2025-08-22 15:29:33 -07:00