diffusers
[Research Projects] ORPO diffusion for alignment
#7423
Merged

[Research Projects] ORPO diffusion for alignment #7423

sayakpaul merged 24 commits into main from orpo-lora
sayakpaul
sayakpaul barebones orpo
e6b734e8
sayakpaul remove reference model.
2c041bd0
sayakpaul full implementation
58f441cd
sayakpaul change default of beta_orpo
cc0cff21
sayakpaul add a training command.
6aa8fc8e
sayakpaul fix: dataloading issues.
83af3add
sayakpaul interpreting the formulation.
fe7c9236
sayakpaul sayakpaul requested a review from kashif kashif 1 year ago
sayakpaul revert styling
2f801b96
HuggingFaceDocBuilderDev
sayakpaul add: wds full blown version
07872301
sayakpaul Merge branch 'main' into orpo-lora
67767858
sayakpaul fix: per_gpu_batch_siz
98fe76c7
sayakpaul start debuggin
a7ed27c1
sayakpaul debugging
675f7bb4
sayakpaul remove print
4ca3ec69
sayakpaul fix
34681cb9
sayakpaul remove filter keys.
2d4d84f2
sayakpaul turn on non-blocking calls.
4b43d793
sayakpaul device_placement
e695c3dd
sayakpaul let's see.
48a75f24
sayakpaul add bigger training run command
6b43a5d8
sayakpaul Merge branch 'main' into orpo-lora
1fa16c15
sayakpaul
reinitialize generator for fair repro
7b05bf82
sayakpaul Merge branch 'main' into orpo-lora
698903a3
sayakpaul add: detailed readme and requirements
4240024c
sayakpaul
sayakpaul sayakpaul merged e29f16cf into main 1 year ago
sayakpaul sayakpaul deleted the orpo-lora branch 1 year ago
feffy380
feffy380 commented on 2024-05-15

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone