How can I measure the performance of NumPy operations?

How can I measure the performance of NumPy operations?

You can use the time module to measure execution time or timeit for more accurate benchmarking.

What is the difference between using time and timeit?

The time module gives a simple execution time, while timeit runs the code multiple times for a more reliable average.

Can you show me an example of using timeit with NumPy?

Sure! Here's a simple example:

import numpy as np
import timeit

# Define a function to benchmark

def array_sum():
    return np.sum(np.random.rand(1000))

# Measure performance
execution_time = timeit.timeit(array_sum, number=1000)
print(f'Execution time: {execution_time}')

What other metrics can I use to evaluate NumPy performance?

You can also look at memory usage, speedup ratios, and compare with other libraries like CuPy or Dask.