ci: add real training jobs to nightly workflow#1313
Draft
garrett4wade wants to merge 6 commits into
Draft
Conversation
Replace dummy test with actual gsm8k GRPO training inside Docker containers (dev-sglang/dev-vllm). Add workflow_dispatch input to run arbitrary branches. Key changes: - Pull latest runtime images, run via docker exec in persistent containers - Install AReaL from source (uv pip install -e . --no-deps) - Round-robin training backend (fsdp/megatron/archon) by day - Add 'ref' input for branch/tag/SHA override on manual dispatch
Contributor
|
Note Gemini is unable to generate a review for this pull request due to the file types involved not being currently supported. |
Docker containers run as root, creating .pyc files owned by root:root on the bind-mounted workspace. The next matrix job fails when actions/checkout tries to git clean these files. Key changes: - Add sudo rm cleanup before checkout - Clean root-owned files in teardown step
Key changes: - Run sglang/vllm sequentially via shared run_variant function - Pre-download model to persistent /opt/hf_cache volume - Enable wandb online logging with nightly environment secret - Trial name includes backend+variant shortcodes (e.g., m.s-2026-05-09)
Runner user lacks passwordless sudo. The cache dir is pre-created on the instance.
Pass model ID to actor.path and let from_pretrained resolve from the HF cache. Pre-download just populates the cache.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
docker exec(matching test-areal.yml pattern)refinput to workflow_dispatch for running arbitrary branches