Commit Graph
81 Commits
Author SHA1 Message Date
RandNMR73 d016119f2e fix wan 2.2 self forcing script 2025-10-02 03:42:50 +00:00
RandNMR73 a547fc1005 clean up 2025-10-02 01:11:58 +00:00
RandNMR73 f27e28ad87 fix datatype? 2025-10-01 21:26:44 +00:00
RandNMR73andgemini-code-assist[bot] 03747310c6 Update examples/distill/SFWan2.2-A14B/distill_dmd.sh
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-09-30 22:21:46 -07:00
RandNMR73 a444862561 cleanup + torch.compile training support 2025-09-27 01:40:57 +00:00
RandNMR73 6ff7686a9d make EMA and training utils MoE compatible 2025-09-26 20:49:07 +00:00
RandNMR73 92d3ce1612 training works with gradient checkpointing but config has issues 2025-09-26 11:48:07 +00:00
RandNMR73 2b9d8e91b5 gradient checkpointing bug 2025-09-26 08:49:34 +00:00
RandNMR73 2df82f0ee4 training not broken 2025-09-26 04:05:14 +00:00
RandNMR73 4f45b6e515 fix hang 2025-09-25 21:01:13 +00:00
SolitaryThinker 4c11512cde debug 2025-09-25 04:11:51 +00:00
SolitaryThinker 32a820c735 fix val 2025-09-24 04:44:22 +00:00
SolitaryThinker d6165981b4 offload dit in training 2025-09-24 00:51:41 +00:00
RandNMR73 aabf1fe135 inference works after changes added
new branch

Stop backprop through kv_cache

Enable timestep warping & using SelfForcing scheduler

checkpoint

text preprocessing ready

fix rotary embedding

Change Wan DiT to have 0 numerical diff with SF's Wan

wan2.2 training doesn't crash

training hangs
2025-09-23 22:20:29 +00:00
William Lin 404cbf4f3c [self-forcing] [6/n] Add Ode Init training (#811) 2025-09-22 17:58:19 -07:00
William Lin 958ffec844 [bugfix] Update learning rates for sparse distillation recipe (#812) 2025-09-22 12:07:03 -07:00
31f000d1cc [self-forcing] [5/n] Add Self-Forcing distillation pipeline (#808)
Co-authored-by: RandNMR73 <notomatthew31@gmail.com>
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com>
2025-09-20 19:32:10 -07:00
Yongqi Chen cd32b3e02f Update example files and readme (#809) 2025-09-20 18:15:59 -07:00
William Lin c5f9ea53b2 [self-forcing] [4/n] Preprocessing for collecting ODE trajectory (#788) 2025-09-15 17:54:42 -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
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
Jinzhe Pan 4aa307be55 [Preprocess][Feat] support torchvision to load video in new preprocessing (#761) 2025-09-01 23:37:01 -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
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
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
Wenxuan Tan 989a03532c Optionally use unmerged weights for inference (#745) 2025-08-22 15:20:31 -07:00
Jinzhe Pan 911a6a6a35 [Feat][Preprocessing] i2v preprocessing workflow (#737) 2025-08-14 20:47:25 -07:00
William LinandRandNMR73 8cae5e4708 [feature] add Gradio live serving demo code (#727)
Co-authored-by: RandNMR73 <notomatthew31@gmail.com>
2025-08-10 15:34:03 -07:00
Wenxuan Tan 5357f63327 Fix LoRA load from training checkpoint (#719) 2025-08-09 20:46:00 -05:00
Wenxuan Tan 6c6bcd914d Remove all empty_cache (#713) 2025-08-07 22:50:38 -07:00
Jinzhe Pan 86604a684b [3/3][Preprocess] add preprocessing workflows (#645) 2025-08-07 01:49:07 -07:00
Wei ZhouandSolitaryThinker c41305ad18 [Feat] Add Wan2.2 14B MoE (#688)
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com>
2025-08-06 20:31:03 -07:00
Yongqi Chen 1d018acb3e [Feature]Add Data-free distillation readme (#710) 2025-08-05 14:27:39 -04:00
Wenxuan Tan 4b7266349a [misc] Remove allow_tf32 in scripts (#705) 2025-08-04 15:37:56 -05:00
Yongqi ChenandSolitaryThinker b65f0316d1 [Feature] Add Wan2.2 DMD example files; Update lr scheduler (#694)
Co-authored-by: SolitaryThinker <wlsaidhi@gmail.com>
2025-08-03 22:55:01 -04:00
Wenxuan Tanandgemini-code-assist[bot] 31200fbb83 [Misc] Fix training scripts (#683)
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2025-08-01 15:54:31 -05:00
Zhang Peiyuan 516ecd374a [Misc] Update examples/ and other misc (#672) 2025-07-30 19:10:27 -07:00
Wei Zhou 3b1b54a74d Modify args to make sure the scripts are runnable on 4090 (#671) 2025-07-30 14:55:08 -07:00
Yongqi Chen f759980a58 [Feature]Add VSA slurm training example scripts (#666) 2025-07-30 01:27:54 -04:00
Zhang Peiyuan 37e0f8c236 [BUG] Fix distillation + vsa (#665) 2025-07-29 19:48:11 -07:00
William LinandJerryZhou54 6375223b16 [Feature] Add wan2.2 5B T2V (#658)
Co-authored-by: JerryZhou54 <zhouw.jerry2017@outlook.com>
2025-07-29 17:16:03 -07:00
Yongqi Chen 4cb046768d [Feature]Add DMD distillation training resume checkpoint; Update DMD CI test (#662) 2025-07-29 19:06:05 -04:00
Yongqi Chen 3322542444 [Feature] Add DMD CI test (#661) 2025-07-29 03:14:00 -04:00
Zhang Peiyuan cbc3a6bb9d [Feat] Support VSA with any resolution. (#650) 2025-07-28 20:14:40 -07:00
Yongqi Chen 2fa8d4ae6d [Feature][Distill]Add 14B 480p T2V distill example scripts (#655) 2025-07-28 18:36:31 -04:00
Yongqi Chen 6284eaa363 [Feature][Distill]Add DMD+VSA joint training example (#654) 2025-07-27 18:18:01 -04:00
William Lin 81c9df629c [core] Add offloading for vae and image encoder and rename offloading args (#643) 2025-07-25 17:55:03 -07:00