上篇文章中的reshape(-1,2),有的时候不明白为何会有参数-1,能够经过查找文档中的reshape()去理解这个问题html
根据Numpy文档(https://docs.scipy.org/doc/numpy/reference/generated/numpy.reshape.html#numpy-reshape)的解释: python
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**.
数组新的shape属性应该要与原来的配套,若是等于-1的话,那么Numpy会根据剩下的维度计算出数组的另一个shape属性值。 数组
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)this
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,先前咱们不知道z的shape属性是多少,可是想让z变成只有1列,行数不知道多少,经过`z.reshape(-1,1)`,Numpy自动计算出有16行,新的数组shape属性为(16, 1),与原来的(4, 4)配套。spa
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)code
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也能够自动计算出新数组的列数。htm