feature(wzn): add LoRA training demo for Geo3K#41
Merged
puyuan1996 merged 8 commits intoopendilab:mainfrom Mar 16, 2026
Merged
Conversation
- Implement LoRA-aware model saving in FSDPV2Strategy, supporting HF/PEFT `save_pretrained` for adapters. - Add LoRA merging/unmerging logic in BroadcastManager to ensure inference engines receive effective weights during synchronization. - Optimize checkpointing in PPOTerVL to prioritize HF adapter saving for LoRA runs. - Add `run_grpo_geo3k_lora_qwen2.5_vl_7b.sh` as a reference LoRA training script. - Improve weight mapping for SGLang to handle PEFT-wrapped module names and base layer stripping.
- Added rotation logic for HF/LoRA adapters in PPO/SPMD trainers to honor the `max_ckpt_num` parameter. - Synced the cleanup mechanism with the `save_ckpt` implementation in `FSDPV2Strategy`.
zunian-wan
changed the title
zunian-wan
changed the title
puyuan1996
changed the title
puyuan1996
requested changes
Feb 10, 2026
…ture and improve LoRA training documentation in runing script
PaParaZz1
requested changes
Feb 22, 2026
- simplify Geo3K LoRA demo script configs and naming - adjust save paths for final/critic ckpts in training entry - unify checkpoint rotation via rotate_ckpt_dirs - improve FSDP/engine weight broadcast handling and errors
PaParaZz1
approved these changes
Mar 5, 2026
- Add test_geo3k_lora.py to seamlessly evaluate LoRA-finetuned vision-language models (e.g., Qwen2.5-VL) on the Geo3K dataset.
… geo3k lora evaluation script
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
📋 Summary
Purpose:
Add a demo for Geo3k training using LoRA with FSDP and SGLang
Type of Change:
🔗 Related Issues
Fixes #(issue number)
Related to #(issue number)
📝 Changes
What changed:
Why these changes:
Key implementation details:
🧪 Testing
BEFORE SUBMITTING, PLEASE READ: