Numpy shape is same, mean returns different shapes

Question

I have the following snippet.

values = [[0.1, 0.7, 0.5], [0.6, 0.3, 0.2], [0.2, 0.8, 0.77]]
A = np.array(values).reshape(3,3)
print A.shape
print np.mean(A, axis=1)

B = np.mat(np.random.rand(3, 3));
print B.shape
print np.mean(B, axis=1)

Output of print statements:

(3, 3)
[ 0.43333333  0.36666667  0.59      ]

(3, 3)
[[ 0.47252016]
 [ 0.44380355]
 [ 0.51070646]]

I have two same shaped numpy array's with different values as an input, one is generated with the rand function the other is a python list created with the array function and calling reshape on it.

However, the shape of the mean return's a different shape for both, even though the input shapes are the same. Any Ideas what can cause this?

That absolutely does, just figured out going over it... thanks, that explains it! — user1767754, Jan 20 '20 at 06:38
Many of the functions/methods that take an `axis` parameter also take a `keepdims` parameter. If `True`, it can make the array mean behave more like the matrix one. — hpaulj, Jan 20 '20 at 08:08

score 3 · Accepted Answer · answered Jan 20 '20 at 06:37

As explained in the documentation for numpy.matrix,

A matrix is a specialized 2-D array that retains its 2-D nature through operations

Additionally,

It is no longer recommended to use this class, even for linear algebra. Instead use regular arrays. The class may be removed in the future.

Numpy shape is same, mean returns different shapes

1 Answers1