作用
实现的功能是一致的(将多维数组降位一维),两者的区别在于返回拷贝(copy)还是返回视图(view),numpy.flatten()返回一份拷贝,对拷贝所做的修改不会影响(reflects)原始矩阵,而numpy.ravel()返回的是视图(view,也颇有几分C/C++引用reference的意味),会影响(reflects)原始矩阵。
import numpy as np
x = np.array([[1, 2], [3, 4]])
array([[1, 2],
[3, 4]])
x.flatten()
array([1, 2, 3, 4])
x.ravel()
array([1, 2, 3, 4])#两者默认均是行序优先
x.flatten('F')
array([1, 3, 2, 4])
x.ravel('F')
array([1, 3, 2, 4])
x.reshape(-1)
array([1, 2, 3, 4])
x.T.reshape(-1)
array([1, 3, 2, 4])
区别
x = np.array([[1, 2], [3, 4]])
x.flatten()[1] = 100
array([[1, 2],
[3, 4]]) # flatten:返回的是拷贝
x.ravel()[1] = 100
array([[ 1, 100],
[ 3, 4]])
参考文献:
1.https://stackoverflow.com/questions/28930465/what-is-the-difference-between-flatten-and-ravel-functions-in-numpy
2.https://docs.scipy.org/doc/numpy/reference/generated/numpy.ravel.html#numpy-ravel
3.http://www.starmcu.com/archives/624