numpy修改数组维度

numpy学习专题

十二、numpy修改数组维度

修改数组维度

1.broadcast() 用于模仿广播对象，它返回一个对象，对象中封装了将一个数组广播到另一个数组的结果

import numpy as np
x = np.array([[1],[2],[4]])
y = np.array([4,5,6])
print(x)
print(y)

b = np.broadcast(x,y)
#广播后返回的是一个对象
print(b)

r,c = b.iters
# r c 均为一个迭代器对象
print(r)
print(c)

# 广播的对象包含一个三行三列的数组
# 所以迭代9次后，无法再次迭代
print(next(r),next(c))
print(next(r),next(c))
print(next(r),next(c))
print(next(r),next(c))
print(next(r),next(c))
print(next(r),next(c))
print(next(r),next(c))
print(next(r),next(c))
print(next(r),next(c))

b = np.broadcast(x,y)
print(b.shape)
c = np.empty(b.shape)
print("\n",c)
print("\n",x+y,"\n")

[[1]
 [2]
 [4]]
[4 5 6]
<numpy.broadcast object at 0x000001859F9C4EE0>
<numpy.flatiter object at 0x000001859B715F20>
<numpy.flatiter object at 0x000001859FAE0E10>
1 4
1 5
1 6
2 4
2 5
2 6
4 4
4 5
4 6
(3, 3)

 [[6.23042070e-307 4.67296746e-307 1.69121096e-306]
 [1.11258192e-307 1.89146896e-307 7.56571288e-307]
 [3.11525958e-307 1.24610723e-306 2.56765117e-312]]

 [[ 5  6  7]
 [ 6  7  8]
 [ 8  9 10]] 

broadcast_to()
原型：numpy.broadcast_to(array,shape,subok = False)
作用：将数组广播到新形状。它在原始数组上返回只读视图。它通常不连续。如果新形状不符合numpy的广播规则，则该函数可能抛出ValueError
参数           说明
array          待修改的形状
shape          修改后的形状

import numpy as np
arr = np.arange(1,5).reshape(1,4)
arr2=np.broadcast_to(arr,(5,4))
print(arr2)

[[1 2 3 4]
 [1 2 3 4]
 [1 2 3 4]
 [1 2 3 4]
 [1 2 3 4]]

expand_dims
原型：numpy.squeeze(arr,axis = None)
作用：通过在指定位置插入新的轴来扩展数组
参数             说明
arr             输入数组
axis            新轴插入的位置

import numpy as np
arr = np.arange(1,5).reshape(2,2)
arr2 = np.expand_dims(arr,0)
print("arr:\n",arr,"\n")
print("\n",arr.shape,"\n")
print("arr:\n",arr2,"\n")
print("\n",arr2.shape,"\n")

arr:
 [[1 2]
 [3 4]] 


 (2, 2) 

arr:
 [[[1 2]
  [3 4]]] 


 (1, 2, 2) 

squeeze()
原型:numpy.squeeze(arr,axis)
作用:从给定数组的形状中删除一维的条目
注意：只有轴为1才能删除，但凡大于1都不行
参数           说明
arr           输入数组
axis         删除轴的位置

import numpy as np
arr = np.arange(1,13).reshape(1,1,3,4)
print(arr,"\n")
print(arr.shape,"\n")
arr2 = np.squeeze(arr)
print(arr2,"\n")
print(arr2.shape,"\n")

arr3 = np.squeeze(arr,1)
print(arr3,"\n")
print(arr3.shape,"\n")

arr4 = np.arange(1,61).reshape(3,1,4,5)
arr5 = np.squeeze(arr4)
print(arr4,"\n")
print(arr4.shape,"\n")
print(arr5,"\n")
print(arr5.shape,"\n")

[[[[ 1  2  3  4]
   [ 5  6  7  8]
   [ 9 10 11 12]]]] 

(1, 1, 3, 4) 

[[ 1  2  3  4]
 [ 5  6  7  8]
 [ 9 10 11 12]] 

(3, 4) 

[[[ 1  2  3  4]
  [ 5  6  7  8]
  [ 9 10 11 12]]] 

(1, 3, 4) 

[[[[ 1  2  3  4  5]
   [ 6  7  8  9 10]
   [11 12 13 14 15]
   [16 17 18 19 20]]]


 [[[21 22 23 24 25]
   [26 27 28 29 30]
   [31 32 33 34 35]
   [36 37 38 39 40]]]


 [[[41 42 43 44 45]
   [46 47 48 49 50]
   [51 52 53 54 55]
   [56 57 58 59 60]]]] 

(3, 1, 4, 5) 

[[[ 1  2  3  4  5]
  [ 6  7  8  9 10]
  [11 12 13 14 15]
  [16 17 18 19 20]]

 [[21 22 23 24 25]
  [26 27 28 29 30]
  [31 32 33 34 35]
  [36 37 38 39 40]]

 [[41 42 43 44 45]
  [46 47 48 49 50]
  [51 52 53 54 55]
  [56 57 58 59 60]]] 

(3, 4, 5) 

一定要支持作者哦