wiki-speedrun / src /benchmark /__init__.py
jwlutz's picture
Initial commit: Wikipedia Speedrun benchmark with Flask dashboard
0e61be5
Raw
History Blame Contribute Delete
240 Bytes
"""
Benchmark module.
Provides infrastructure for running and analyzing benchmarks:
- Problem: Defines a benchmark problem
- BenchmarkRunner: Runs agents on problem sets
- BenchmarkResult: Stores results
- Metrics: Computes statistics
"""