np.reshape(x, (-1,1)) vs x[:, np.newaxis]

Question

I am reading source code of an open source project recently. When the programmer wanted to convert a row vector like array([0, 1, 2]) to a column vector like array([[0], [1], [2]]), np.reshape(x, (-1,1)) was used. In the comment, it says reshape is necessary to preserve the data contiguity against vs [:, np.newaxis] that does not.

I tried the two ways, it seems like they will return the same results. Then what does the data contiguity preservation mean here?

Some part of this story is getting lost in the retelling. (For example, the parts about a "row vector" seem incorrect; an actual 1-by-whatever row vector would respond differently to these operations.) — user2357112, Sep 21 '17 at 01:35

score 10 · Accepted Answer · answered Sep 21 '17 at 02:30

Both ways return views of the exact same data, therefore the 'data contiguity' is likely a non-issue as the data is not change, only the view is changed. See Numpy: use reshape or newaxis to add dimensions.

However there might be a practical advantage of using .reshape((-1,1)), as it will reshape the array into 2d array regardless of the original shape. For [:, np.newaxis], the result will depend on the original shape of the array, considering these:

In [3]: a1 = np.array([0, 1, 2])

In [4]: a2 = np.array([[0, 1, 2]])

In [5]: a1.reshape((-1, 1))
Out[5]: 
array([[0],
       [1],
       [2]])

In [6]: a2.reshape((-1, 1))
Out[6]: 
array([[0],
       [1],
       [2]])

In [7]: a1[:, np.newaxis]
Out[7]: 
array([[0],
       [1],
       [2]])

In [8]: a2[:, np.newaxis]
Out[8]: array([[[0, 1, 2]]])

np.reshape(x, (-1,1)) vs x[:, np.newaxis]

1 Answers1