deep_learning_Function_tensorflow_reshape()

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

根据Numpy文档(https://docs.scipy.org/doc/numpy/reference/generated/numpy.reshape.html#numpy-reshape)的解释：

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)

1. z.reshape(-1)
2. 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)

1.  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。

转自：https://blog.csdn.net/qq_38409301/article/details/88559889