Reads the CSV produced by the memo_reqenerator SQL query and plots the distribution of initialization times

plot_initialization_times.py

import matplotlib.pyplot
import numpy
import pandas

colnames=[
    'Image', 'Pixels', 'Series', 'Pixel type',
    'SizeX', 'SizeY', 'SizeZ', 'SizeC', 'SizeT',
    'Format', 'Time', 'Rank'] 
df = pandas.read_csv(
    "image_list_20250404.csv",
    names=colnames,
    parse_dates=["Time"],
    date_format='%M:%S.%f')
times = pandas.to_timedelta(df['Time']) / numpy.timedelta64(1, "s")
ax = times.hist(bins=20, color='darkorange')
ax.set_yscale('log')
ax.grid(False)
matplotlib.pyplot.xlabel('Initialization time (s)')
matplotlib.pyplot.ylabel('Number of filesets')
matplotlib.pyplot.show()
matplotlib.pyplot.tight_layout()