首先声明两者所要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。
1. 两者的功能
>>> x = np.array([[1, 2], [3, 4]])>>> xarray([[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])2. 两者的区别>>> x = np.array([[1, 2], [3, 4]])>>> x.flatten()[1] = 100>>> xarray([[1, 2], [3, 4]]) # flatten:返回的是拷贝>>> x.ravel()[1] = 100>>> xarray([[ 1, 100], [ 3, 4]])References[1] What is the difference between flatten and ravel functions in numpy?--------------------- https://blog.csdn.net/lanchunhui/article/details/50354978