-
1958463f02
Reformat
main
Kevin Black
2023-11-16 22:36:46 +00:00
-
378dd18298
Merge pull request #16 from sayakpaul/patch-1
Kevin Black
2023-10-06 16:46:20 -07:00
-
-
bfcba5e28e
Update README.md to include a note about the
trl integration
Sayak Paul
2023-09-30 15:07:48 +02:00
-
-
b590ec0a7c
Fix accelerate version
Kevin Black
2023-09-15 22:54:51 -07:00
-
500edd2b53
Update README.md
Kevin Black
2023-09-11 16:10:03 -07:00
-
e17ecd265d
Update README.md
Kevin Black
2023-09-11 16:01:38 -07:00
-
5955244f37
Fix gradient sync for lora
Kevin Black
2023-08-22 16:18:49 -07:00
-
d7a63516cb
Merge pull request #9 from desaixie/main
Kevin Black
2023-08-22 11:54:52 -07:00
-
-
3130ddfaff
Only log rewards from process 0
Desai Xie
2023-08-21 15:10:45 -07:00
-
-
173b2bb6e0
Update README.md (add reward curves)
Kevin Black
2023-07-13 12:37:22 -07:00
-
c67c2adfee
Enforce python version
Kevin Black
2023-07-06 10:28:54 -07:00
-
64a20bc01d
Update README.md
Kevin Black
2023-07-04 01:29:50 -07:00
-
8c45353cce
Update README.md
Kevin Black
2023-07-04 01:28:40 -07:00
-
1f067b16c8
Add teaser image
Kevin Black
2023-07-04 01:22:08 -07:00
-
b14022ea92
Update README.md
Kevin Black
2023-07-04 01:21:46 -07:00
-
26177ccf40
Create LICENSE
Kevin Black
2023-07-04 01:19:47 -07:00
-
c65dd3a39c
Update README
Kevin Black
2023-07-04 01:15:16 -07:00
-
953d59eb70
Fix pydantic issue in setup
Kevin Black
2023-07-04 00:40:42 -07:00
-
10fbec322a
Add activities asset
Kevin Black
2023-07-04 00:27:04 -07:00
-
beb8c2f86d
Update configs
Kevin Black
2023-07-04 00:25:37 -07:00
-
ec499edf84
Fix aesthetic score (again), add llava reward
Kevin Black
2023-07-04 00:23:33 -07:00
-
c0bc708549
Commenting pass
Kevin Black
2023-06-29 00:51:38 -07:00
-
8779f62a1c
Adding checkpointing and resuming
Kevin Black
2023-06-28 17:58:25 -07:00
-
ad28862b48
Add reward to image caption
Kevin Black
2023-06-28 10:42:47 -07:00
-
fe9ed8a25f
Fix aesthetic scorer
Kevin Black
2023-06-28 10:42:30 -07:00
-
28d2d8c40e
Minor changes; add train_timestep_fraction
Kevin Black
2023-06-27 22:17:32 -07:00
-
bae3f43f5f
Add aesthetic scorer reward function
Kevin Black
2023-06-27 10:40:36 -07:00
-
8cab96dea4
Minor changes, add assets
Kevin Black
2023-06-27 10:20:03 -07:00
-
4c5322ca85
Device specific seed
Kevin Black
2023-06-26 22:35:24 -07:00
-
1ce0994c8a
Fix stat tracking bug
Kevin Black
2023-06-26 22:25:43 -07:00
-
5c16a90ceb
Move config out of module
Kevin Black
2023-06-25 21:02:27 -07:00
-
269615a35e
Working non-lora training; other changes
Kevin Black
2023-06-25 11:28:42 -07:00
-
c680890d5c
Working on DGX
Kevin Black
2023-06-24 00:07:55 -07:00
-
92fc030123
Continue implementation
Kevin Black
2023-06-23 21:08:32 -07:00
-
6d848c3cdc
Remove pycache
Kevin Black
2023-06-23 21:08:19 -07:00
-
2fda3d4e78
Initial commit
Kevin Black
2023-06-23 19:25:54 -07:00