Function main

python/benchmarks/benchmark_data_processing.py:259–302

()

def main() -> None:
    parser = argparse.ArgumentParser(
        description="Benchmark openquant.data with per-function throughput and latency metrics."
    )
    parser.add_argument("--rows-per-symbol", type=int, default=200_000)
    parser.add_argument("--symbols", type=int, default=4)
    parser.add_argument("--iterations", type=int, default=7)
    parser.add_argument("--out", type=Path, default=None)
    parser.add_argument("--baseline", type=Path, default=None)
    args = parser.parse_args()

    result = run_benchmarks(
        rows_per_symbol=args.rows_per_symbol,
        symbols=args.symbols,
        iterations=args.iterations,
    )

    print("Core Data Functions")
    print(result["tables"]["core"])
    print()
    print("I/O + End-to-End Load")
    print(result["tables"]["io"])
    print()

    if args.baseline is not None:
        diff_rows = _compare_against_baseline(
            [BenchStats(**row) for row in result["functions"]],
            args.baseline,
        )
        result["baseline_comparison"] = diff_rows
        if diff_rows:
            print("Baseline Comparison (mean_ms regression; negative is faster)")
            for row in diff_rows:
                print(
                    f"{row['name']}: baseline={row['baseline_mean_ms']:.2f}ms "
                    f"current={row['current_mean_ms']:.2f}ms "
                    f"regression_pct={row['regression_pct']:+.2f}%"
                )
            print()

    if args.out is not None:
        args.out.parent.mkdir(parents=True, exist_ok=True)
        args.out.write_text(json.dumps(result, indent=2), encoding="utf-8")
        print(f"wrote benchmark metrics to {args.out}")


if __name__ == "__main__":