Run multiple seeds and summarize results.
Steps:
- •Implement a sweep script (scripts/sweep.py or scripts/run_sweep.sh) that:
- •accepts seeds list, parallelism, and config path
- •writes each seed to results/<exp_name>/seed_<k>/
- •Implement scripts/aggregate.py that:
- •loads all seeds
- •aligns x-axis exactly as paper (steps, episodes, wall-clock)
- •computes mean + CI/SEM as specified
- •applies the same smoothing/binning rules as paper (documented)
- •writes results/<exp_name>/aggregate/curve.csv and summary.json
- •Ensure failures are visible:
- •if a seed crashes, aggregator reports it and continues
- •Update docs/repro_spec.md with the exact aggregation definition.