Commit Graph
81 Commits
Author SHA1 Message Date
LinB203 c521999f80 attn_mask with bf16
Former-commit-id: d9667a8c15a5344e3c777197fd6badec15dcddbd
2024-03-07 13:35:01 +08:00
lb203 fcc8009570 Merge pull request #64 from khan-yin/main
[feat]: add sit model

Former-commit-id: acacfd670282b709e6c769717b5295e230861724
2024-03-07 13:29:24 +08:00
lb203 5a7e9c8822 Update README.md
Former-commit-id: 29f24a819fe2d4593c48afe6ca990659784bd776
2024-03-07 12:58:47 +08:00
khan-yin c716b3281e feat: update sample
Former-commit-id: e7dc492b9b4764db28aa31cc1eede914052b4ca6
2024-03-07 04:13:57 +00:00
khan-yin 94b71a5ea0 Merge remote-tracking branch 'upstream/main' into main
Former-commit-id: 7ef9b5e76a91b444f87955c3e0e7ea3d72315396
2024-03-07 04:05:06 +00:00
khan-yin e90373b161 feat: Incorporating SiT sample
Former-commit-id: 91e2931e412f8c96ef748cb2322ee956309fb6aa
2024-03-07 03:40:24 +00:00
lb203 7cf1bcc93d Update README.md
Former-commit-id: b3f6bb7f252505efb0ffd8195003c79206fac313
2024-03-07 11:38:23 +08:00
LinB203 cca84f1c6c denorm fun
Former-commit-id: 05707a07b14295f5e1932d0c10bb84a20a0cd328
2024-03-07 10:40:06 +08:00
lb203 1ebf531e1e Update README.md
Former-commit-id: 1a7178a61547c33a744d032302fb76ee0a3c6671
2024-03-07 10:33:20 +08:00
lb203 c9adc3e9a5 Update README.md
Former-commit-id: d097e40998a3396f74c2bfbcf26d6ace8080d0b6
2024-03-07 10:20:51 +08:00
Kehan Yin 3f19169f55 Merge branch 'PKU-YuanGroup:main' into main
Former-commit-id: a2dea6886ce90634fb7d1c2e0fbfce95a500eefc
2024-03-07 10:04:08 +08:00
lb203 006ad5353a Merge pull request #70 from qqingzheng/videogpt
[refactor] Performed code abstraction on the VideoGPT code and enabled support for accelerate training.

Former-commit-id: 822c1a173f5bc792868fae9e6e4cd7defe2ea183
2024-03-07 09:23:28 +08:00
Chestnut 589da07fec Merge branch 'main' into videogpt
Former-commit-id: 714cc434cea5823fc09ef6fe502b1fe83170cae2
2024-03-07 01:25:58 +08:00
qqingzheng b409ef399a [refactor] added more methods in abstract classes
Former-commit-id: a535e3a06eb766bebe7baa3d9df64b04d04d40e6
2024-03-06 17:23:40 +00:00
lb203 8944956249 Update README.md
Former-commit-id: 604a29d1de43e0fcb90ada4dd97f5f080e37caea
2024-03-07 01:11:28 +08:00
lb203 921d927a27 Update train.sh
Former-commit-id: 66aa3709f80d89a30d0a57c5e033ce1ae46f998f
2024-03-07 01:06:20 +08:00
lb203 06423d785d Update train.sh
Former-commit-id: 6c4957d715eaa489691be21dd0abdb9309660e0d
2024-03-07 00:57:14 +08:00
LinB203 38bf4afdfa support accelerate training
Former-commit-id: b32671c9991b9a9db66cb148822ccb3e8be7dc11
2024-03-07 00:55:35 +08:00
qqingzheng 7a78ef2a97 [refactor] configuration default values
Former-commit-id: d7ca7685d6a8c6d015082dda241bc98ab156770d
2024-03-06 15:59:13 +00:00
qqingzheng 4549eeeab3 Merge branch 'videogpt' of github.com:qqingzheng/Open-Sora-Plan into videogpt
Former-commit-id: ca5f92952c7a0daae6deaca40f6dab1d00a0c0b2
2024-03-06 15:48:26 +00:00
qqingzheng da08e767db [refactor] rename
Former-commit-id: ec317605d24223527b8c2d4b4d2ee876f86abbaf
2024-03-06 15:48:19 +00:00
lb203 0118dd8f1d Merge pull request #69 from luo3300612/dev-rec-video
[fix]: disable gradient computation to save GPU memory when reconstructing a video

Former-commit-id: f9afcda4130e1709d6f201ed310adcc8335480d9
2024-03-06 23:46:10 +08:00
lb203 4efdbc2aca Update README.md
Former-commit-id: 59a595a1e31b156500a944f244afec40988e872d
2024-03-06 23:39:17 +08:00
Chestnut 140eb267d9 Merge branch 'main' into videogpt
Former-commit-id: 0a45d76fdcc3faf167239dfe5dc055b30f3f4e05
2024-03-06 23:36:14 +08:00
LinB203 1b99cf2789 add sample script
Former-commit-id: 9ce847651bb6dba28c51ce5807e222ae2902a78a
2024-03-06 23:23:17 +08:00
lb203 ae2646e213 Merge pull request #74 from Linzy19/2307
[update]: update super_resolution

Former-commit-id: a2586e672bd5d60f1a5e937a796d5f6b6ac1da4d
2024-03-06 23:19:44 +08:00
ZongyingLin 36e0f8813a update
add RGT


Former-commit-id: a491111e48cec588095381eb70ffca94600ca4f7
2024-03-06 23:10:26 +08:00
qqingzheng 0622ea11dc [fix] fix a bug when using ucf101_stride4x4x4
Former-commit-id: f541e723d65ea44ae4560c68908123c7e6781bbc
2024-03-06 15:08:36 +00:00
qqingzheng df625f66cc [refactor] remove test.ipynb
Former-commit-id: e73c3a9e0d41e51a0fd755e57739424213a76312
2024-03-06 15:07:11 +00:00
qqingzheng 107f140b56 [fix] fix a bug when using ucf101_stride4x4x4
Former-commit-id: 9911403c688321db731450d9383bc02923a4be2d
2024-03-06 14:51:35 +00:00
Chestnut 2fd71909a4 Merge branch 'main' into videogpt
Former-commit-id: f8de10e627ce27f1511ddbbc5d6b29f3daf094ab
2024-03-06 22:13:26 +08:00
qqingzheng 2b272863b0 [refactor] adapt to old training code
Former-commit-id: 358d362ecbc100b21ea02895cb0426cf5b386a72
2024-03-06 13:58:13 +00:00
qqingzheng 1bed11aee3 [refactor] reformat videogpt, support training videogpt on accelerate
Former-commit-id: 75385ed47fe60008b61b1bcf89f61ea696d095bc
2024-03-06 13:45:16 +00:00
lb203 5dc0b3cb41 Update README.md
Former-commit-id: 673659495b7503ba402fbcb6c46204499761b4cf
2024-03-06 21:24:58 +08:00
root 80d394d64a [fix]: disable gradient computation to save GPU memory when reconstructing a video
Former-commit-id: b7bf763e7f8800959ed82bb89f41684db216dd8e
2024-03-06 21:04:29 +08:00
lb203 f58d14ba24 Update README.md
Former-commit-id: 398cfcd7b19cf1ba827eb49fc11223f72fa93a85
2024-03-06 20:46:10 +08:00
lb203 d3acf63c4e Update README.md
Former-commit-id: 9fd88eb8ac5122b0999ccb0071f195aaeb0979a0
2024-03-06 20:30:57 +08:00
LinB203 193d7ef7a0 support latte training
Former-commit-id: d471a9414842f422d1200c2fd3f3ddfcf9e97fd6
2024-03-06 20:29:31 +08:00
khan-yin e47f894f69 fix: rename sit model filename
Former-commit-id: f3b2cd7dcf644f680616eee0571db90478fede23
2024-03-06 12:08:04 +00:00
khan-yin 7cee5443e4 fix: remove replicated diffusion module in sit
Former-commit-id: 997cb049185421d303b14d59ab6c0603f1d1a66f
2024-03-06 12:06:55 +00:00
Kehan Yin edfe14e9fa Merge branch 'PKU-YuanGroup:main' into main
Former-commit-id: c7d4f9c2cfaa61e3babc711f13a095ed237978b1
2024-03-06 19:46:26 +08:00
lb203 31f5ce25d4 Update README.md
Former-commit-id: 898d35f15ce0fd124cd48078f0a8af2764bf55d1
2024-03-06 19:33:06 +08:00
LinB203 5a80da7ec1 add vae and fix bug in dit
Former-commit-id: 2dd4254c22ce671d5cd5e7d22a2386b7660fbdc4
2024-03-06 19:32:20 +08:00
lb203 ec0dd141e1 Merge pull request #63 from kabachuha/alternative-attentions
Support for less VRAM heavy alternative attentions (ReBased and Ring attention)

Former-commit-id: c90f9d3c780843edccc72d29f68f5ed7c655a721
2024-03-06 19:28:59 +08:00
khan-yin 54d3069702 [feat]: add sit model
Former-commit-id: 2eda854d6c25254b71d0105a3ed092953a9d6f68
2024-03-06 09:52:27 +00:00
kabachuha abffb358f9 support for ring attention
Former-commit-id: 0533c73b99217e88da121bd0f68e0949bc88016e
2024-03-06 12:24:10 +03:00
kabachuha e1459ff3d0 option to use rebased linear attention
Former-commit-id: f1b39b34df5fbdef14ef3f6a0e6d0972a8570445
2024-03-06 12:18:03 +03:00
lb203 29409555f3 Update README.md
Former-commit-id: 10ee64c122f28efea39fc1a4542d278683a58ce1
2024-03-06 16:27:06 +08:00
lb203 d878709456 Update README.md
Former-commit-id: 0f2bdd2bcd819d528f2a41a96ba93f5c3c04dc8c
2024-03-06 16:23:53 +08:00
qqingzheng 3810227fd6 first commit
Former-commit-id: ec3481bc35a8120d519082dc76585428f0896470
2024-03-06 08:16:58 +00:00