Commit Graph

118 Commits

Author SHA1 Message Date
hiyouga
23b881bff1 support mllm hf inference
Former-commit-id: 2c7c01282acd7ddabbb17ce3246b8dae4bc4b8cf
2024-04-26 05:34:58 +08:00
hoshi-hiyouga
2eede9ffd6 Update workflow.py
Former-commit-id: 5b8b5b975716d539ae2fae8536f79e106aa0b566
2024-04-26 03:29:12 +08:00
BUAADreamer
514ffafc12 modify some style
Former-commit-id: 053062abc007014a7fde95c5ae9f4d859893d8ad
2024-04-25 22:04:09 +08:00
BUAADreamer
8b2a735c14 modify some style
Former-commit-id: b016e6a671a2f228f0bdd9b8d5995b4669609655
2024-04-25 21:58:18 +08:00
BUAADreamer
110c2ce2a5 modify style
Former-commit-id: 3bffc1e1b8bcc4582cebea06d35e5146163c7bec
2024-04-25 21:27:48 +08:00
BUAADreamer
c425436676 modify style
Former-commit-id: 54b713d0c4ffdfc6a7faeb14471b58bb1cd8acf5
2024-04-25 21:15:16 +08:00
BUAADreamer
dbd905438b add some
Former-commit-id: 8d035a849c4a441d457791aab073861adf69a09f
2024-04-25 21:08:32 +08:00
BUAADreamer
9b210cf4b3 rm some
Former-commit-id: 2c85b4fabbebd8b51eee53f5d29184d4a6e97569
2024-04-25 20:09:43 +08:00
BUAADreamer
3c792174db merge data part to the text stream
Former-commit-id: 7ee20286d9bcc2d5378bfd6bb02cd3648396d873
2024-04-25 19:19:59 +08:00
BUAADreamer
00e2a272ef merge model part to the text stream
Former-commit-id: b6fcb832ddaed4647d6f2b926f3dfccd47f3ea84
2024-04-25 08:20:41 +08:00
BUAADreamer
6c1db2d012 remove conflicts
Former-commit-id: f8b637eb76cba7ec229e2978068805ad1cca8adb
2024-04-25 00:34:22 +08:00
BUAADreamer
12c51655ce add llava and instructblip
Former-commit-id: 142fb6f4541a1acfefe66ff2574dabde53b00c06
2024-04-25 00:22:43 +08:00
hiyouga
5420905a2e support unsloth generate
Former-commit-id: 0ef1ad9f505dba71db9342f524cc3a7565e5e09e
2024-04-24 04:46:53 +08:00
BUAADreamer
ab6dc0ea30 add multimodal LLM BLIP-2 and InstructBLIP
Former-commit-id: a730f89a972f1a9d37c718c716f199cb8d4903b2
2024-04-23 18:45:43 +08:00
hiyouga
233e167f68 fix optimizers
Former-commit-id: f811eee2fa12a89a55a9c5d3a05a1521b4347727
2024-04-21 20:40:54 +08:00
hiyouga
d16561e7a4 fix bug in galore optimizer
Former-commit-id: c05ac23261a5a8ba893c2918a43dc7777307407b
2024-04-21 18:53:22 +08:00
hiyouga
a4167fd925 support badam for all stages
Former-commit-id: 7a1380646119bfe6855f73dd90570defcea05281
2024-04-16 17:44:48 +08:00
hoshi-hiyouga
9d23f5dc89 Update utils.py
Former-commit-id: 01147536b2bb507e87e033fa696e9eb39fe96bbe
2024-04-16 17:30:12 +08:00
hoshi-hiyouga
5978427ae0 Update trainer.py
Former-commit-id: c6163be1444c00dd000f288e2f834968bd932981
2024-04-16 17:29:52 +08:00
Jonery
6dd6b3e396 resolve gradient checkpointing issue.
Former-commit-id: 6df9135d063bb6102f0cbcdf0d702076f5febbae
2024-04-16 12:05:27 +08:00
Jonery
d4d471450f Feature BAdam
Former-commit-id: d8d2807fbcf587c37f7fd34a23e9397d2775ceed
2024-04-15 23:15:27 +08:00
hiyouga
276f2cb24e update examples
Former-commit-id: 369294b31c8a03a1cafcee83eb31a817007d3c49
2024-04-15 22:14:34 +08:00
hiyouga
5486ea09e3 fix model card
Former-commit-id: 920e7149bf2b559c9829aa4b11cfb6d00bbb2f9e
2024-04-12 17:11:59 +08:00
hiyouga
31bbbb6d13 fix #3238
Former-commit-id: 4d7e81ab4722d13bec6ca1af141f94bdc74d0883
2024-04-12 14:28:11 +08:00
hiyouga
1348f7d860 fix resize vocab at inference #3022
Former-commit-id: c243720b89eec0af2872fa3c7980a0026d893f4d
2024-04-03 18:14:24 +08:00
hiyouga
f6530222f7 fix #3116
Former-commit-id: b7256aa33d761280751518c20f29f9b8ea3fb025
2024-04-03 14:47:59 +08:00
hiyouga
b12176d818 simplify readme
Former-commit-id: 0da6ec2d516326fe9c7583ba71cd1778eb838178
2024-04-02 20:07:43 +08:00
hiyouga
1dc963caa6 fix #3083
Former-commit-id: ff9a3f73961a362d0ddc22079f80a85465fffda8
2024-04-01 22:53:52 +08:00
hiyouga
be0a807e8c fix ORPO loss
Former-commit-id: 5544ddde9087f00f9e20b78d0079f20c2f5d1604
2024-04-01 14:42:41 +08:00
hiyouga
52d402e2a9 fix IPO and ORPO loss
Former-commit-id: fc27955732aedbb12003faf19b760e2768b228f2
2024-04-01 14:37:53 +08:00
hiyouga
c5a46f9113 fix plots
Former-commit-id: 81355671296b84d438967463bb2a92934ff31aae
2024-03-31 19:43:48 +08:00
hiyouga
00e17a377c use log1p in orpo loss
https://github.com/huggingface/trl/pull/1491

Former-commit-id: 3b15d495264b00a4f8716bafea334778874963d7
2024-03-31 19:27:08 +08:00
hiyouga
d764cd8736 support ORPO
Former-commit-id: f44a4c27e2461cdaa1b16865f597a31033c0e6d9
2024-03-31 18:29:50 +08:00
marko1616
a6858a36c0 Fix Llama model save for full param train
Former-commit-id: ca17b5db4f97c3ec9fe2004877f150e8f51ab4b5
2024-03-30 23:45:04 +08:00
hiyouga
fbd0584391 release v0.6.1
Former-commit-id: a59d823f554505b2e649e6e111b9dee8306d3ad8
2024-03-29 11:36:08 +08:00
hiyouga
9408366a36 fix #2982
Former-commit-id: e5e6a0c50c7a1c0052ed6b459450b9735ff2c9a1
2024-03-28 20:22:31 +08:00
hiyouga
59e6ebf039 update trainers
Former-commit-id: d0dd6eefed0b86895ed00a7cafb331e5193db645
2024-03-28 18:16:27 +08:00
hoshi-hiyouga
dc540dfaa8 fix ds optimizer
Former-commit-id: 2675127070a1e7584e71039a11c1ebac54ddd1db
2024-03-26 23:39:56 +08:00
hiyouga
3336422760 fix #2961
Former-commit-id: 616917bb3be7f71073b56ad8c7bc4e164b08b9b5
2024-03-26 17:26:14 +08:00
hiyouga
04423b916f release v0.6.0 (real)
Former-commit-id: 34e06bf408ccd21e674f896703f1c7b62e97e1ca
2024-03-25 23:37:48 +08:00
hiyouga
daab85e3e6 release v0.6.0
Former-commit-id: 51910d5803eb718e4976da0b3bfcdc5eeeea48eb
2024-03-25 22:38:56 +08:00
hiyouga
04e0fe9147 tiny fix
Former-commit-id: c39cf3439a3025f703d50ac414c10ef3c8486a1f
2024-03-25 21:18:08 +08:00
marko1616
7f99cb1817 pass ruff check
Former-commit-id: 8534b069a05121eb041371a6becccf0a1a23f268
2024-03-24 16:12:10 +08:00
marko1616
c555b2cce3 fix Llama lora merge crash
Former-commit-id: 46f7d8e6b85f73fb0c51c8b08bd9955c3b171d93
2024-03-24 03:06:11 +08:00
marko1616
2eba1c6851 fix Llama lora merge crash
Former-commit-id: a8bd8e9149ff79a2707fec9c6d006761cfdd0dee
2024-03-24 02:55:23 +08:00
marko1616
edeed55664 fix Llama lora merge crash
Former-commit-id: c29a2893f58cf7a916ff05b2725fadf1ad2c4c9a
2024-03-24 02:44:35 +08:00
hiyouga
c7af26a9e3 fix #2777 #2895
Former-commit-id: 54d5f62d29456a8d9d0c0dd3d0bbfffe48935803
2024-03-20 17:59:45 +08:00
hiyouga
ffdacaa618 fix packages
Former-commit-id: 2f9f334a123d43267bfb3dd26aaa1ad285ffe7a5
2024-03-17 22:32:03 +08:00
hiyouga
ed020579dc fix export
Former-commit-id: 4e996f194406d7eb27b2bde290a12c82c41219d0
2024-03-15 15:06:30 +08:00
hiyouga
623ee1bd88 tiny fix
Former-commit-id: bf8123669be334338b4268d0a8f7703ff2cf6255
2024-03-14 21:19:06 +08:00