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

根据Numpy文档的解释：

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 2	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.stackoverflow: What does -1 mean in numpy reshape?