This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong. ====== Performance Comparison of libdsepio_read and read_trk Functions ====== This script benchmarks the performance of two functions, libdsepio_read and read_trk, used for reading tracking data from file paths. It measures the execution time for reading the same track file multiple times using both Python loops and leveraging a library possibly written in C++. Additionally, it employs the pandas library to organize the timing results into a dataframe and outputs these results into a CSV file. Finally, it prints the execution timings for each function to the console, providing insight into which method is more efficient for reading track data files. <code python> from pysep.io.trk.read import libdsepio_read from pysep.io.trk import read_trk from datetime import datetime import numpy as np import pandas as pd path = 'example.track' # Total = np.linspace(1, 100, 25) # # print("========= Python Loop Test ==========") # PythonTimes = list() # for t in Total: # start = datetime.now() # for i in range(int(t)): # n = libdsepio_read([path], 60) # end = datetime.now() # PythonTimes.append((end-start).total_seconds()) # # print("========= C++ Loop Test ===========") # Ctimes = list() # for t in Total: # start = datetime.now() # n = libdsepio_read([path]*int(t), 60) # end = datetime.now() # Ctimes.append((end-start).total_seconds()) # # ActualTotal = list() # for t in Total: # ActualTotal.append(int(t)) # # payload = {'files':ActualTotal, 'PyTime':PythonTimes, 'CTime':Ctimes} # df = pd.DataFrame(payload) # df.to_csv('res.csv', index=False) print('=========== libdsepio timeing ==============') start = datetime.now() n = libdsepio_read([path]*50, 60) end = datetime.now() print(f"Took: {(end-start).total_seconds()} s") print('=========== read_trk timeing ==============') start = datetime.now() n,m = read_trk([path]*50) end = datetime.now() print(f"Took: {(end-start).total_seconds()} s") # print(type(n)) </code>
