feat: add experiment resume support and improve robustness#90
Open
YihanJIANG-lab wants to merge 1 commit intoSakanaAI:mainfrom
Open
feat: add experiment resume support and improve robustness#90YihanJIANG-lab wants to merge 1 commit intoSakanaAI:mainfrom
YihanJIANG-lab wants to merge 1 commit intoSakanaAI:mainfrom
Conversation
Major features: - Add --resume flag to launch_scientist_bfts.py for resuming from a previous experiment's checkpoint - Add AgentManager.from_checkpoint() classmethod for state restoration - Add perform_experiments_bfts resume_dir parameter with checkpoint discovery Robustness improvements: - launch_scientist_bfts.py: robust PDF discovery (find_pdf_path_for_review), dynamic log dir discovery (find_run_log_dir), guard against missing experiment_results dir, psutil import guard - semantic_scholar.py: add max_tries=8 to backoff decorators, fix duplicate __init__ assignments - Add literature_search.py: multi-source search (OpenAlex, Crossref, arXiv, S2) with cascade fallback - Add github_models_config.yaml: example config for GitHub Models API - perform_ideation_temp_free.py: use MultiSourceLiteratureSearchTool for more reliable literature search
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds the ability to resume experiments from a checkpoint, plus several robustness improvements:
Resume support:
--resumeflag inlaunch_scientist_bfts.pyto resume from a previous experiment directoryAgentManager.from_checkpoint()classmethod for state restorationperform_experiments_bftsRobustness improvements:
find_pdf_path_for_review) — guards against None returnfind_run_log_dir) instead of hardcodedlogs/0-run/psutilimport guard for environments without itmax_tries=8on Semantic Scholar backoff decoratorsmkdir(parents=True, exist_ok=True)for log directories