How can I measure the performance of NumPy operations in Python?

How can I measure the performance of NumPy operations in Python?

You can use the time module or timeit library to measure the execution time of NumPy operations.

Can you show me an example using timeit?

Sure! Here's a simple example:

import numpy as np
import timeit

# Define a function to measure
def numpy_operation():
    return np.random.rand(1000, 1000).dot(np.random.rand(1000, 1000))

# Measure performance
execution_time = timeit.timeit(numpy_operation, number=10)
print(f'Execution time: {execution_time}')

What does the number parameter do in timeit?

The number parameter specifies how many times to execute the function. It helps to get a more accurate measurement by averaging the time over multiple runs.

Are there any other tools for performance measurement?

Yes, you can also use profiling tools like cProfile or libraries like line_profiler for more detailed insights.