canonical_comparison.png
eval_result.json
evaluation_summary.json
run.json