使用数组的reshape方法，可以创建一个改变了尺寸的新数组，原数组的shape保持不变；
1 >>> a = np.array([1, 2, 3, 4]);b = np.array((5, 6, 7, 8));c = np.array([[1, 2, 3, 4],[4, 5, 6, 7], [7, 8, 9, 10]]) 2 >>> b 3 array([5, 6, 7, 8]) 4 >>> c 5 array([[ 1, 2, 3, 4], 6 [ 4, 5, 6, 7], 7 [ 7, 8, 9, 10]]) 8 >>> c.dtype 9 dtype('int32') 10 >>> d = a.reshape((2,2)) 11 >>> d 12 array([[1, 2], 13 [3, 4]]) 14 >>> d = a.reshape((1,2)) 15 Traceback (most recent call last): 16 File "<pyshell#27>", line 1, in <module> 17 d = a.reshape((1,2)) 18 ValueError: total size of new array must be unchanged 19 >>> d = a.reshape((1,-1)) 20 >>> d 21 array([[1, 2, 3, 4]])
>>> d = a.reshape((-1,1))
>>> d
array([[1],
[2],
[3],
[4]])
注意:a.reshape((1,-1))和a.reshape((1,2))和a.reshape((-1,1))