Reshaping Arrays In Python

Question

Why flatten() is slower than ravel() function in reshaping arrays in python?

x = np.arange(1000000)
x.shape = 100, 100, 100
%timeit x.flatten() 
%timeit x.ravel()

NumPy offers many functions and some of them can produce same results I think it depends on being Inplace and not.

There aren't many numpy functions or methods that work in-place. But there is an important distinction between a view and copy. Operations that return views are faster. ravel is one; it is essentially a reshape. But sometimes even those have to return copies. Don't skip the official numpy docs, especially the beginners and essentials parts. And then keep the reference section handy for continual reference. — hpaulj
– hpaulj, Commented Jan 4, 2023 at 4:11

user2357112 · Accepted Answer · 2023-01-03 18:11:14Z

1

flatten always makes a copy. ravel makes a view if it can. Copying takes time.

answered Jan 3, 2023 at 18:11

user2357112

286k32 gold badges490 silver badges571 bronze badges

Sign up to request clarification or add additional context in comments.

1 Answer 1