reshape(-1,1)什么意思 numpy.reshape

numpy.reshape(a, newshape, order='C')[source]，参数`newshape`是啥意思？

官方文档：https://docs.scipy.org/doc/numpy/reference/generated/numpy.reshape.html

newshape : int or tuple of ints
The new shape should be compatible with the original shape. If an integer, then the result will be a 1-D array of that length. One shape dimension can be -1. In this case, **the value is inferred from the length of the array and remaining dimensions**.

大意是说，数组新的shape属性应该要与原来的配套，如果等于-1的话，那么Numpy会根据剩下的维度计算出数组的另外一个shape属性值。

举几个例子或许就清楚了，有一个数组z，它的shape属性是(4, 4)

z = np.array([[1, 2, 3, 4],
          [5, 6, 7, 8],
          [9, 10, 11, 12],
          [13, 14, 15, 16]])
z.shape
(4, 4)

z.reshape(-1)

z.reshape(-1)
array([ 1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16])

z.reshape(-1, 1)

也就是说，先前我们不知道z的shape属性是多少，但是想让z变成只有一列，行数不知道多少，通过`z.reshape(-1,1)`，Numpy自动计算出有12行，新的数组shape属性为(16, 1)，与原来的(4, 4)配套。

z.reshape(-1,1)
 array([[ 1],
        [ 2],
        [ 3],
        [ 4],
        [ 5],
        [ 6],
        [ 7],
        [ 8],
        [ 9],
        [10],
        [11],
        [12],
        [13],
        [14],
        [15],
        [16]])
 

z.reshape(-1, 2)

newshape等于-1，列数等于2，行数未知，reshape后的shape等于(8, 2)

 z.reshape(-1, 2)
 array([[ 1,  2],
        [ 3,  4],
        [ 5,  6],
        [ 7,  8],
        [ 9, 10],
        [11, 12],
        [13, 14],
        [15, 16]])
 

同理，只给定行数，newshape等于-1，Numpy也可以自动计算出新数组的列数。

----------

参考

1.http://stackoverflow.com/questions/18691084/what-does-1-mean-in-numpy-reshape

转自知乎 https://www.zhihu.com/question/52684594