[recipe, doc] feat: add Boogu-Image FlowGRPO v1 sync recipe - #30
Open
knlnguyen1802 wants to merge 1 commit into
Open
[recipe, doc] feat: add Boogu-Image FlowGRPO v1 sync recipe#30knlnguyen1802 wants to merge 1 commit into
knlnguyen1802 wants to merge 1 commit into
Conversation
Port the existing OCR LoRA recipe onto the v1 sync trainer so this verl-project#389 path can opt into TransferQueue + ReplayBuffer without changing adapters, losses, or the colocated 4-GPU layout. AI assistance (Cursor) was used for this change. Co-authored-by: Cursor <cursoragent@cursor.com> Signed-off-by: knlnguyen1802 <knlnguyen1802@gmail.com> Co-authored-by: Cursor <cursoragent@cursor.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
examples/flowgrpo_trainer/boogu_image/run_boogu_image_ocr_lora_v1.sh, a recipe-only v1 sync counterpart of the Boogu-Image OCR LoRA FlowGRPO script.main_diffusion_v1,trainer.use_v1=true,trainer.v1.trainer_mode=sync, andtrainer.experiment_name=boogu_image_ocr_lora_v1.fsdp_layer_prefixes, request-level packing, 512 resolution, colocated 4-GPU actor/rollout/reward). Does not rewrite adapters or losses.examples/flowgrpo_trainer/boogu_image/README.mdonly (docs/examples/boogu_image/flowgrpo_trainer_boogu_image.mdis a symlink to that README).Tracked by verl-project#389.
Why this is not duplicating an existing PR
verl-project/verl-omniandknlnguyen1802/verl-omni:gh pr list --search "389 in:body"gh pr list --search "boogu"/"boogu-image"/"v1"booguhits are Qwen-Image-Edit ([7/N][recipe] feat: add Qwen-Image-Edit FlowGRPO v1 sync recipe #25), not this recipe.Test plan
bash -n examples/flowgrpo_trainer/boogu_image/run_boogu_image_ocr_lora_v1.sh— passPR_TITLE='[recipe, doc] feat: add Boogu-Image FlowGRPO v1 sync recipe' python tests/special_sanity/check_pr_title.py— pass (modules: ['recipe', 'doc'], type: feat)run_boogu_image_ocr_lora.sh: header + shebang +main_diffusion_v1+experiment_namesuffix +trainer.use_v1=true+trainer.v1.trainer_mode=synconly; LoRA,fsdp_layer_prefixes, packing, resolution, and colocated 4-GPU reward layout unchanged.bash examples/flowgrpo_trainer/boogu_image/run_boogu_image_ocr_lora_v1.sh trainer.total_training_steps=1 trainer.logger='["console"]') — not run here (no GPU in this environment; recipe-only).examples/flowgrpo_trainer/README.mdanddocs/start/diffusion_v1.mdwere not edited.AI assistance
AI assistance (Cursor) was used to draft the recipe, local README update, commit, and this PR. A human submitter has reviewed every changed line and is responsible for defending the change end-to-end. This is not a pure code-agent PR.
Made with Cursor