Python基础——numpy.ndarray一维数组与多维数组

shape为(3, 1)（表多维数组）与shape为(3, )（表一维数组）用法上有很大的不同

>>> x = np.random.randn(3)
>>> x
array([ 0.53150381, -0.45205789, -0.6655509 ])
>>> x.shape
(3,)

>>> y = np.random.randn(3, 1)
>>> y
array([[ 0.69918993],
       [ 1.17769064],
       [-0.53024913]])
>>> y.shape
(3, 1)

不同之一：与list进行转换

>>> list(x)
[0.53150381185761253, -0.45205789280422431, -0.66555089846812798]
>>> list(y)
[array([ 0.69918993]), array([ 1.17769064]), array([-0.53024913])]