当前位置：网站首页>numpy. Where() usage and np.argsort() usage

numpy. Where() usage and np.argsort() usage

2022-07-29 01:35:00 【51CTO】

numpy.where() There are two uses ：

1. np.where(condition, x, y)

Meet the conditions (condition), Output x, Not satisfied with output y.

If it's a one-dimensional array , amount to [xv if c else yv for (c,xv,yv) in zip(condition,x,y)]

       
       >>> 
       
       aa 
       
       = 
       
       np.
       
       arange(
       
       10)
       
       >>> 
       
       np.
       
       where(
       
       aa,
       
       1,
       
       -
       
       1)
       
       array([
       
       -
       
       1,  
       
       1,  
       
       1,  
       
       1,  
       
       1,  
       
       1,  
       
       1,  
       
       1,  
       
       1,  
       
       1])  
       
       # 0 by False, So the first output -1
       
       >>> 
       
       np.
       
       where(
       
       aa 
       
       > 
       
       5,
       
       1,
       
       -
       
       1)
       
       array([
       
       -
       
       1, 
       
       -
       
       1, 
       
       -
       
       1, 
       
       -
       
       1, 
       
       -
       
       1, 
       
       -
       
       1,  
       
       1,  
       
       1,  
       
       1,  
       
       1])
       
       >>> 
       
       np.
       
       where([[
       
       True,
       
       False], [
       
       True,
       
       True]],    
       
       #  Examples on the official website 
       
             [[
       
       1,
       
       2], [
       
       3,
       
       4]],
       
             [[
       
       9,
       
       8], [
       
       7,
       
       6]])
       
       array([[
       
       1, 
       
       8],
       
       [
       
       3, 
       
       4]])
      
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

The condition of the above example is [[True,False], [True,False]], Respectively corresponding to the four values of the final output result . The first value is from [1,9] Middle selection , Because the condition is True, So choose 1. The second value is from [2,8] Middle selection , Because the condition is False, So choose 8, And so on . You can take another example of a similar problem ：

       
       >>> 
       
       a 
       
       = 
       
       10
       
       >>> 
       
       np.
       
       where([[
       
       a 
       
       > 
       
       5,
       
       a 
       
       < 
       
       5], [
       
       a 
       
       == 
       
       10,
       
       a 
       
       == 
       
       7]],
       
             [[
       
       "chosen",
       
       "not chosen"], [
       
       "chosen",
       
       "not chosen"]],
       
             [[
       
       "not chosen",
       
       "chosen"], [
       
       "not chosen",
       
       "chosen"]])
       
       array([[
       
       'chosen', 
       
       'chosen'],
       
       [
       
       'chosen', 
       
       'chosen']], 
       
       dtype
       
       =
       
       '<U10')
      
1.
2.
3.
4.
5.
6.
7.

2. np.where(condition)

Only the conditions (condition), No, x and y, Then the output meets the condition ( It's not 0) The coordinates of the elements ( Equivalent to numpy.nonzero). The coordinates here are expressed in tuple Given in the form of , How many dimensions does the original array usually have , Output tuple It contains several arrays , Corresponding to the coordinates of each dimension of the qualified element .

       
       >>> 
       
       a 
       
       = 
       
       np.
       
       array([
       
       2,
       
       4,
       
       6,
       
       8,
       
       10])
       
       >>> 
       
       np.
       
       where(
       
       a 
       
       > 
       
       5)                
       
       #  Returns the index 
       
(
       
       array([
       
       2, 
       
       3, 
       
       4]),)   
       
       >>> 
       
       a[
       
       np.
       
       where(
       
       a 
       
       > 
       
       5)]            
       
       #  Equivalent to  a[a>5]
       
       array([ 
       
       6,  
       
       8, 
       
       10])
       
       >>> 
       
       np.
       
       where([[
       
       0, 
       
       1], [
       
       1, 
       
       0]])
       
(
       
       array([
       
       0, 
       
       1]), 
       
       array([
       
       1, 
       
       0]))
      
1.
2.
3.
4.
5.
6.
7.
8.

In the above example condition [[0,1],[1,0]] The true value of is two 1, Their first dimensional coordinates are [0,1], The second coordinate is [1,0] .

Let's take a look at a complicated example ：

       
       >>> 
       
       a 
       
       = 
       
       np.
       
       arange(
       
       27).
       
       reshape(
       
       3,
       
       3,
       
       3)
       
       >>> 
       
       a
       
       array([[[ 
       
       0,  
       
       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]]])
       
       >>> 
       
       np.
       
       where(
       
       a 
       
       > 
       
       5)
       
(
       
       array([
       
       0, 
       
       0, 
       
       0, 
       
       1, 
       
       1, 
       
       1, 
       
       1, 
       
       1, 
       
       1, 
       
       1, 
       
       1, 
       
       1, 
       
       2, 
       
       2, 
       
       2, 
       
       2, 
       
       2, 
       
       2, 
       
       2, 
       
       2, 
       
       2]),
       
       array([
       
       2, 
       
       2, 
       
       2, 
       
       0, 
       
       0, 
       
       0, 
       
       1, 
       
       1, 
       
       1, 
       
       2, 
       
       2, 
       
       2, 
       
       0, 
       
       0, 
       
       0, 
       
       1, 
       
       1, 
       
       1, 
       
       2, 
       
       2, 
       
       2]),
       
       array([
       
       0, 
       
       1, 
       
       2, 
       
       0, 
       
       1, 
       
       2, 
       
       0, 
       
       1, 
       
       2, 
       
       0, 
       
       1, 
       
       2, 
       
       0, 
       
       1, 
       
       2, 
       
       0, 
       
       1, 
       
       2, 
       
       0, 
       
       1, 
       
       2]))
       
       #  The eligible elements are 
       
       [ 
       
       6,  
       
       7,  
       
       8]
       
       ],
       
      [[ 
       
       9, 
       
       10, 
       
       11],
       
       [
       
       12, 
       
       13, 
       
       14],
       
       [
       
       15, 
       
       16, 
       
       17]],
       
      [[
       
       18, 
       
       19, 
       
       20],
       
       [
       
       21, 
       
       22, 
       
       23],
       
       [
       
       24, 
       
       25, 
       
       26]]
       
       ]
      
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.

therefore np.where The corresponding coordinates of each element will be output , Because the original array has three dimensions , therefore tuple There are three arrays in .

One thing to note is that , The input cannot be directly list, Need to change to array Or for array Talent . such as range(10) and np.arange(10) The latter returns an array , Use np.where In order to achieve the effect .

np.argsort() Usage of

       
       numpy.
       
       argsort(
       
       a, 
       
       axis
       
       =
       
       -
       
       1, 
       
       kind
       
       =
       
       ’quicksort’, 
       
       order
       
       =
       
       None)
      
1.

argsort(a)# obtain a An array arranged from small to large

argsort(-a)# obtain a Array arranged from large to small

argmin(a)# obtain a Minimum subscript

argmax(a)# obtain a The maximum subscript

function : The matrix a according to axis Sort , And return the sorted subscript
Parameters : a: Input matrix , axis: Dimensions to be sorted
Return value : Output the sorted subscript

( One dimensional array )

You can find ,argsort() Yes, it will X After the elements in are sorted from small to large , Extract the corresponding index index, Then output to y
Such as x[3]=-1 Minimum ,x[5]=9 Maximum

So take the array x The minimum value of can be written as :

Or use argmin() function

  
       x[
       
       x.
       
       argmin()]
      
1.
2.

Array x The maximum of , It's written in ：

       
       x[
       
       x.
       
       argsort()[
       
       -
       
       1]]  # 
       
       -
       
       1
       
        Represents the index from back to front 
      
1.

Or use argmax() function , No more details

Output sorted array

       
       x[x.argsort()]
       
#  or 
       
x[np.argsort(x)]
      
1.
2.
3.

( Two dimensional array )

       
       x = np.array([[1,5,4],[-1,6,9]])
       
# [[ 1  5  4]
       
# [-1  6  9]]
      
1.
2.
3.

       
        Down the line ( Each column ) The elements are sorted 
      
1.

       
       np.argsort(x,axis=0)
       
# array([[1, 0, 0],
       
#        [0, 1, 1]], dtype=int64)
      
1.
2.
3.

Go right along the column ( Each row ) The elements are sorted

       
       np.argsort(x,axis=1)
       
# array([[0, 2, 1],
       
#        [0, 1, 2]], dtype=int64)
      
1.
2.
3.

author ： Your Rego

The copyright of this article belongs to the author , Welcome to reprint , But without the author's consent, the original link must be given on the article page , Otherwise, the right to pursue legal responsibility is reserved .

原网站

版权声明
本文为[51CTO]所创，转载请带上原文链接，感谢
https://yzsam.com/2022/210/202207290032357235.html

当前位置：网站首页>numpy. Where() usage and np.argsort() usage

numpy. Where() usage and np.argsort() usage

1. np.where(condition, x, y)

2. np.where(condition)

边栏推荐

猜你喜欢

随机推荐