Profiling NumPy operations - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the module used for profiling NumPy operations.

NumPy

import [1]

Drag options to blanks, or click blank then click option'

Amatplotlib

Bpandas

CcProfile

Dtimeit

Attempts:

3 left

💡 Hint

Common Mistakes

Importing unrelated modules like pandas or matplotlib.

Using timeit instead of cProfile for detailed profiling.

The cProfile module is used to profile Python code, including NumPy operations.

2fill in blank

medium

Complete the code to start profiling a function named compute_array.

NumPy

profiler = cProfile.Profile()
profiler.[1]('compute_array()')
profiler.print_stats()

Drag options to blanks, or click blank then click option'

Arun

Brunctx

Cenable

Ddisable

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'enable' or 'disable' instead of 'run'.

Using 'runctx' which requires extra context parameters.

The run method runs the function under the profiler.

3fill in blank

hard

Fix the error in the code to measure execution time of a NumPy operation using timeit.

NumPy

import numpy as np
import timeit

arr = np.arange(1000)
time = timeit.timeit(lambda: np.[1](arr), number=1000)
print(time)

Drag options to blanks, or click blank then click option'

Aappend

Bsort

Cadd

Dsum

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'add' which requires two arrays.

Using 'append' which is not a NumPy function.

Using 'sort' which is a method, not a function.

The sum function correctly computes the sum of array elements. add is a ufunc needing two arrays, append is not a NumPy function, and sort is a method, not a function.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that profiles functions with timeit and filters results greater than 0.001 seconds.

NumPy

functions = [np.sum, np.mean, np.std]
results = {func.__name__: timeit.timeit(lambda: func(np.arange(1000)), number=1000) for func in functions if [1] [2] 0.001}

Drag options to blanks, or click blank then click option'

Atimeit.timeit(lambda: func(np.arange(1000)), number=1000)

B>

C<

Dfunc.__name__

Attempts:

3 left

💡 Hint

Common Mistakes

Using function name instead of timing expression in the first blank.

Using '<' instead of '>' for filtering.

The comprehension measures time for each function and filters those with time greater than 0.001 seconds.

5fill in blank

hard

Fill all three blanks to create a profiling dictionary that stores function names, their execution times, and filters times less than 0.005 seconds.

NumPy

functions = [np.min, np.max, np.median]
profile = [1]: [2] for func in functions if [3] < 0.005

Drag options to blanks, or click blank then click option'

Afunc.__name__

Btimeit.timeit(lambda: func(np.arange(1000)), number=1000)

Dfunc

Attempts:

3 left

💡 Hint

Common Mistakes

Using function object instead of function name as key.

Using function object instead of timing expression for value or condition.

Mixing up the order of keys and values.

The dictionary comprehension keys are function names, values are execution times, and filtering is done on the execution time less than 0.005 seconds.