gen_gc_summary.py

# script to generate GC summary from statistics gathered with
# https://github.com/python/cpython/pull/100958

import sys
import pandas as pd
import numpy as np
import argparse


def get_gc_summary(df):
    return df.groupby('generation_number').aggregate(
        num_collections=('generation_number', 'count'),
        mean_collected_cycles=('collected_cycles', 'mean'),
        mean_time=('collection_time', 'mean'),
        mean_total_objects=('total_objects', 'mean'),
        total_time=('collection_time', np.sum),
        )


def main():
    parser = argparse.ArgumentParser()
    parser.add_argument('files', nargs='+')
    args = parser.parse_args()

    summary = pd.DataFrame()
    for fn in args.files:
        df = pd.read_csv(fn)
        sdf = get_gc_summary(df)
        sdf['filename'] = fn
        summary = pd.concat([summary, sdf])

    summary['mean_time'] *= 1e9 # convert to ns
    summary['time_per_total'] = summary['mean_time'] / summary['mean_total_objects']
    summary['time_per_collected'] = summary['mean_time'] / summary['mean_collected_cycles']
    summary['ratio_collected'] = summary['mean_collected_cycles'] / summary['mean_total_objects']
    summary.to_csv(sys.stdout)

if __name__ == "__main__":
    main()