（自用）axis跟expand_dims理解

3.8alc

已于 2023-11-10 14:00:34 修改

阅读量87

点赞数

文章标签： numpy

于 2023-11-09 17:49:13 首次发布

本文链接：https://blog.csdn.net/l3_8alc/article/details/134317044

版权

1:shape(1,2,3)从左到右依次为0轴，1轴，2轴
2：维度跟括号的关系——最深层括号内为shape最右侧的维度，最外层括号内为最左侧维度(0维)

import numpy as np 
 
a = np.array([[1, 2, 3], [4, 5, 6]])
b = np.array([[[1, 2, 3], [4, 5, 6]]])
c = np.array([[[1, 2, 3]], [[4, 5, 6]]])
d = np.array([[[1], [2], [3]], [[4], [5], [6]]])
 
print(a.shape) # a.shape(2,3) 最里面的括号内有三个数，最后一个维度(1轴)个数为3，把最里面的括号看成整体，倒数第二个维度(0轴)个数为2
print(b.shape) # b.shape(1,2,3)
print(c.shape) # c.shape(2,1,3)
print(d.shape) # d.shape(2,3,1)

3:expeand_dims增加维度后,shape如何，括号加在哪里

import numpy as np
a=np.array([[1,2,3],[4,5,6]])

b=np.expand_dims(a, axis=0)

b=np.expand_dims(a, axis=1)

b=np.expand_dims(a, axis=2)

初始

当加在0轴,axis=0，最外层加括号

当加在1轴,axis=1,倒数第二深度的向量会加括号


rng = np.random.default_rng(seed=42)
arr = rng.integers(1, 100, (3, 4))
arr

expanded = np.expand_dims(arr, axis=(1, 3, 4))
print(expanded.shape)
expanded

在这里插入图片描述

参考https://blog.csdn.net/qq_37924224/article/details/119816771?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522169951981916800188582549%2522%252C%2522scm%2522%253A%252220140713.130102334…%2522%257D&request_id=169951981916800188582549&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2_alltop_positive~default-1-119816771-null-null.142^v96pc_search_result_base8&utm_term=np.expand_dims&spm=1018.2226.3001.4187

https://blog.csdn.net/weixin_40244676/article/details/114830074?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522169951981916800192249057%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fall.%2522%257D&request_id=169951981916800192249057&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2_allfirst_rank_ecpm_v1~rank_v31_ecpm-18-114830074-null-null.142^v96pc_search_result_base8&utm_term=np.expand_dims&spm=1018.2226.3001.4187