Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Adds VLM Training support to SFTTrainer + VSFT script #1518

Merged
merged 23 commits into from
Apr 11, 2024
Merged

Adds VLM Training support to SFTTrainer + VSFT script #1518

Changes from 1 commit
Commits
Show all changes
23 commits
Select commit Hold shift + click to select a range
43b78e2
adds option to skip dataset preparation in SFTTrainer
edbeeching Apr 10, 2024
d98b66d
before changing the template
edbeeching Apr 10, 2024
142727a
adds support for new schema
edbeeching Apr 10, 2024
c683d89
a few fixes to data collator to support new schema
edbeeching Apr 10, 2024
4c0457d
updates args
edbeeching Apr 10, 2024
2bec041
precommit
edbeeching Apr 10, 2024
57d1ad8
adds sys prompt to chat template and other fixes
edbeeching Apr 11, 2024
26f84e8
updates template, fixes collator for multiple images
edbeeching Apr 11, 2024
2e7ad90
precommit
edbeeching Apr 11, 2024
8a55580
rename vsft to vstf_llava
edbeeching Apr 11, 2024
9d4d732
adding integration tests
edbeeching Apr 11, 2024
c6a7c6f
adds integration test for vsft
edbeeching Apr 11, 2024
798799a
precommit
edbeeching Apr 11, 2024
c65c43d
adds back chat template
edbeeching Apr 11, 2024
dad4eda
docs
edbeeching Apr 11, 2024
89e268e
typo
edbeeching Apr 11, 2024
1d81cc8
adds eval, precommit
edbeeching Apr 11, 2024
35dd6a3
adds peft launch args
edbeeching Apr 11, 2024
46acce9
Merge branch 'main' into vlm-sft-support
edbeeching Apr 11, 2024
09fe59c
Merge branch 'main' into vlm-sft-support
edbeeching Apr 11, 2024
da06efa
formatting
edbeeching Apr 11, 2024
36f942a
fixes no deps tests by checking if PIL lib exists
edbeeching Apr 11, 2024
9c1bf9a
Update __init__.py
younesbelkada Apr 11, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
typo
  • Loading branch information
@edbeeching
edbeeching committed Apr 11, 2024
commit 89e268ec9d0c8e82edcf57985b2595e028a101dd
2 changes: 1 addition & 1 deletion docs/source/sft_trainer.mdx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
Supervised fine-tuning (or SFT for short) is a crucial step in RLHF. In TRL we provide an easy-to-use API to create your SFT models and train them with few lines of code on your dataset.

Check out a complete flexible example at [`examples/scripts/sft.py`](https://github.com/huggingface/trl/tree/main/examples/scripts/sft.py).
Experimental support for Vision Language Models is also included int the example [`examples/scripts/vsft_llava.py`](https://github.com/huggingface/trl/tree/main/examples/scripts/vsft_llava.py).
Experimental support for Vision Language Models is also included in the example [`examples/scripts/vsft_llava.py`](https://github.com/huggingface/trl/tree/main/examples/scripts/vsft_llava.py).

## Quickstart

Expand Down