Skip to content

PSI Engine Benchmark

This benchmark compares scalar calculate_psi against calculate_psi_batch(engine='python') and calculate_psi_batch(engine='rust').

Environment

  • Generated at: 2026-04-06T03:53:15.075948Z
  • Python: 3.10.19
  • Platform: macOS-26.2-arm64-arm-64bit
  • CPU count: 10

Scenarios

rows features groups missing_rate scalar_ms batch_python_ms batch_rust_ms rust_speedup_vs_scalar python_speedup_vs_scalar py_batch_max_abs_diff py_batch_mean_abs_diff rust_batch_max_abs_diff rust_batch_mean_abs_diff avg_scalar_ms_per_feat_group avg_rust_ms_per_feat_group peak_rss_mb
100000 20 6 0.00 1443.598 744.934 163.198 8.85x 1.94x 0.000e+00 0.000e+00 6.939e-18 4.770e-19 12.029984 1.359984 319.91
100000 20 6 0.05 1345.450 617.339 193.774 6.94x 2.18x 0.000e+00 0.000e+00 1.388e-17 6.939e-19 11.212087 1.614786 425.50
100000 20 6 0.20 1179.149 534.981 150.942 7.81x 2.20x 0.000e+00 0.000e+00 2.776e-17 2.082e-18 9.826244 1.257850 431.47

Reproduce

newt-benchmark-psi

Use CLI options (--rows, --features, --groups, --missing-rates) to customize scenario grid.