Python Numpy 位运算符

Python Numpy 位运算符和用于执行位运算的函数。它们是 `bitwise_and`、`&`、`bitwise_or`、`|`、`invert`（位非）、`left_shift`、`<<`、`right_shift` 和 `>>`。这些位运算符会比较两个值的二进制表示并返回结果。

您还可以将这些 Python Numpy 位运算符和函数用作比较运算符。我的意思是，将每个项与一个条件进行比较。

Python Numpy 位与运算符

Python Numpy 位与运算符 `bitwise_and` 函数在两个位值都返回 true 时返回 true，否则返回 false。在我们进行实际示例之前，让我通过下面的程序向您展示此位与运算背后的真值表。

import numpy as np
 
print('----and Example----')
print('True & True   = ', (True & True))
print('True & False  = ', (True & False))
print('False & True  = ', (False & True))
print('False & False = ', (False & False))
       
print('\n----bitwise_and Function Example----')
print('True bitwise_and True   = ', np.bitwise_and(True, True))
print('True bitwise_and False  = ', np.bitwise_and(True, False))
print('False bitwise_and True  = ', np.bitwise_and(False, True))
print('False bitwise_and False = ', np.bitwise_and(False, False))

Numpy 数组位与运算符输出

----and Example----
True & True   =  True
True & False  =  False
False & True  =  False
False & False =  False

----bitwise_and Function Example----
True bitwise_and True   =  True
True bitwise_and False  =  False
False bitwise_and True  =  False
False bitwise_and False =  False

在此示例中，我们声明了两个整数变量并对它们使用了位与运算。首先，它将它们转换为二进制格式，然后比较 a 的每个位与 b。这里，12 = 00001100，25 = 00011001，因此 00001100 & 00011001 返回 00001000，即 8。

import numpy as np
 
a = 12
b = 25
 
print('Binary Value of 12 = ', bin(a))
print('Binary Value of 25 = ', bin(b))
 
print('\nBinary Value of 12 = ', np.binary_repr(a))
print('Binary Value of 25 = ', np.binary_repr(b))
 
print('\nBitwise and Operator Result = ', a & b)
print('bitwise_and Function Result = ', np.bitwise_and(a, b))

位与运算输出

Binary Value of 12 =  0b1100
Binary Value of 25 =  0b11001

Binary Value of 12 =  1100
Binary Value of 25 =  11001

Bitwise and Operator Result =  8
bitwise_and Function Result =  8

在上面的 Python 示例中，我们将此 Numpy `bitwise_and` 用于单个值。让我尝试将此位与运算符和函数用于两个数组。这意味着 arr1 中的单个元素与 arr2 执行二进制与运算，并返回 `bitwise_and` 结果。

import numpy as np
 
arr1 = np.array([2, 12, 9, 12, 17, 11])
print(arr1)
arr2 = np.array([14, 13, 65, 25, 42, 65])
print(arr2)
 
print('\nNumpy Bitwise and Operator Result = ', (arr1 & arr2))
print('Numpy bitwise_and Function Result = ', np.bitwise_and(arr1, arr2))
print()
 
x = np.random.randint(1, 20, size = (3, 7))
print(x)
y = np.random.randint(50, 125, size = (3, 7))
print(y)
 
print('\n---Numpy Bitwise and Operator Result---\n', (x & y))
print('---Numpy bitwise_and Function Result---\n ', np.bitwise_and(x, y))

位与运算输出

[ 2 12  9 12 17 11]
[14 13 65 25 42 65]

Numpy Bitwise and Operator Result =  [ 2 12  1  8  0  1]
Numpy bitwise_and Function Result =  [ 2 12  1  8  0  1]

[[18  1 11 10  5 14  6]
 [13  6  3 14  3  7  6]
 [10 14 18 12  4 19  6]]
[[121  86  73  98  73  96  82]
 [ 72 119 113 103 110  65 120]
 [ 55  77  75  72  59  64 113]]

---Numpy Bitwise and Operator Result---
 [[16  0  9  2  1  0  2]
 [ 8  6  1  6  2  1  0]
 [ 2 12  2  8  0  0  0]]
---Numpy bitwise_and Function Result---
  [[16  0  9  2  1  0  2]
 [ 8  6  1  6  2  1  0]
 [ 2 12  2  8  0  0  0]]

到目前为止，我们一直在使用 Numpy 位与运算执行位运算。但是，您也可以使用它们执行比较运算。让我使用此位与运算来检查该项是否大于 1 且小于 9。如果条件为 True，则返回 True，否则返回 False。

import numpy as np
 
arr = np.array([2, 5, 0, 7, 8, 4, 10, 9, 0, 6, 1, 11])
print(arr)
 
print('--- Numpy Bitwise and Operator ---')
print((arr > 1) & (arr < 9))
 
print('--- Numpy bitwise_and Function ---')
print(np.bitwise_and((arr > 1), (arr < 9)))

在这里，我们使用 `bitwise_and` 函数和位与运算符对多维数组执行比较。

import numpy as np
 
arr = np.random.randn(3, 5)
print(arr)
 
print('--- Two Dimensional Numpy Bitwise and Operator ---')
print((arr > 0) & (arr < 0.5))
 
print('--- Two Dimensional Numpy bitwise_and Function ---')
print(np.bitwise_and((arr > 0), (arr < 0.5)))
 
arr2 = np.random.randint(1, 25, size = (2, 2, 4))
print('--- Three Dimensional Random Array ---')
print(arr2)
 
print('--- Three Dimensional Numpy Bitwise and Operator ---')
print((arr2 > 5) & (arr2 < 20))
 
print('--- Three Dimensional Numpy bitwise_and Function ---')
print(np.bitwise_and((arr2 > 5), (arr2 < 20)))

位与运算输出

[[-0.49370172  0.36784164 -0.5634175  -1.52734428  1.97063201]
 [ 0.33895941 -0.07471463  2.26536834 -1.80136643 -0.06620597]
 [-0.20766583  0.38012599 -0.85149217  0.01983919 -1.45677273]]
--- Two Dimensional Numpy Bitwise and Operator ---
[[False  True False False False]
 [ True False False False False]
 [False  True False  True False]]
--- Two Dimensional Numpy bitwise_and Function ---
[[False  True False False False]
 [ True False False False False]
 [False  True False  True False]]
--- Three Dimensional Random Array ---
[[[24 21 24 23]
  [10 11  7 13]]

 [[17 19 13 20]
  [ 5 24  4 16]]]
--- Three Dimensional Numpy Bitwise and Operator ---
[[[False False False False]
  [ True  True  True  True]]

 [[ True  True  True False]
  [False False False  True]]]
--- Three Dimensional Numpy bitwise_and Function ---
[[[False False False False]
  [ True  True  True  True]]

 [[ True  True  True False]
  [False False False  True]]]

Python Numpy 位或运算符

Python Numpy 位或运算符和 `bitwise_or` 函数在两个位值都返回 False 时返回 False，否则返回 True。让我通过下面的程序向您展示位或运算的真值表。

import numpy as np
 
print('----Bitwise or operator Example----')
print('True | True   = ', (True | True))
print('True | False  = ', (True | False))
print('False | True  = ', (False | True))
print('False | False = ', (False | False))
       
print('\n----bitwise_or Function Example----')
print('True bitwise_or True   = ', np.bitwise_or(True, True))
print('True bitwise_or False  = ', np.bitwise_or(True, False))
print('False bitwise_or True  = ', np.bitwise_or(False, True))
print('False bitwise_or False = ', np.bitwise_or(False, False))

位或运算输出

----Bitwise or operator Example----
True | True   =  True
True | False  =  True
False | True  =  True
False | False =  False

----bitwise_or Function Example----
True bitwise_or True   =  True
True bitwise_or False  =  True
False bitwise_or True  =  True
False bitwise_or False =  False

在这里，我们声明了两个整数变量，并对它们使用了位或运算符和 `bitwise_or`。正如我们已经知道的，12 = 00001100，25 = 00011001，所以 00001100 | 00011001 返回 00011101，即 29。

import numpy as np
 
a = 12
b = 25
 
print('Binary Value of 12 = ', np.binary_repr(a))
print('Binary Value of 25 = ', np.binary_repr(b))
 
print('Bitwise or Operator Result = ', a | b)
print('bitwise_or Function Result = ', np.bitwise_or(a, b))

Numpy 数组位或运算输出

Binary Value of 12 =  1100
Binary Value of 25 =  11001
Bitwise or Operator Result =  29
bitwise_or Function Result =  29

在此示例中，我们在两个数组上使用 Numpy `bitwise_or` 函数和位或运算符。

import numpy as np
 
arr1 = np.array([2, 5, 0, 12, 8, 10])
print(arr1)
 
arr2 = np.array([12, 30, 6, 25, 12, 65])
print(arr2)
 
print('\nNumpy Bitwise or Operator Result = ', (arr1 | arr2))
print('Numpy bitwise_or Function Result = ', np.bitwise_or(arr1, arr2))
print()
 
x = np.random.randint(1, 15, size = (3, 5))
print(x)
y = np.random.randint(10, 25, size = (3, 5))
print(y)
 
print('\n---Numpy Bitwise or Operator Result---\n', (x | y))
print('---Numpy bitwise_or Function Result---\n ', np.bitwise_or(x, y))

Numpy 数组位或运算输出

[ 2  5  0 12  8 10]
[12 30  6 25 12 65]

Numpy Bitwise or Operator Result =  [14 31  6 29 12 75]
Numpy bitwise_or Function Result =  [14 31  6 29 12 75]

[[11  7 12  7 10]
 [ 4  2  5 12 12]
 [ 9 12  3 11 13]]
[[16 17 11 15 10]
 [18 19 19 22 19]
 [22 17 14 19 23]]

---Numpy Bitwise or Operator Result---
 [[27 23 15 15 10]
 [22 19 23 30 31]
 [31 29 15 27 31]]
---Numpy bitwise_or Function Result---
  [[27 23 15 15 10]
 [22 19 23 30 31]
 [31 29 15 27 31]]

与位与运算类似，您也可以使用位或运算和 `bitwise_or` 函数来执行逻辑比较。让我使用此位或运算符来检查数组中的项是否小于 3 或大于 8。如果任一条件为 True，则返回 True，否则返回 False。

import numpy as np
 
arr = np.array([2, 5, 0, 7, 8, 4, 10, 9, 0, 6, 1, 11])
print(arr)
 
print('--- Numpy Bitwise or Operator ---')
print((arr < 3) | (arr > 8))
 
print('--- Numpy bitwise_or Function ---')
print(np.bitwise_or((arr < 3), (arr > 8)))

位或运算输出

[ 2  5  0  7  8  4 10  9  0  6  1 11]
--- Numpy Bitwise or Operator ---
[ True False  True False False False  True  True  True False  True  True]
--- Numpy bitwise_or Function ---
[ True False  True False False False  True  True  True False  True  True]

在这里，我们使用 `bitwise_or` 和位或运算符对多维数组执行比较。

import numpy as np
 
arr = np.random.randn(3, 5)
print(arr)
 
print('--- Two Dimensional Numpy Bitwise or Operator ---')
print((arr < 0) | (arr > 1))
 
print('--- Two Dimensional Numpy bitwise_or Function ---')
print(np.bitwise_or((arr < 0), (arr > 1)))
 
arr2 = np.random.randint(1, 25, size = (2, 2, 4))
print('--- Three Dimensional Random Array ---')
print(arr2)
 
print('--- Three Dimensional Numpy Bitwise or Operator ---')
print((arr2 < 8) | (arr2 > 17))
 
print('--- Three Dimensional Numpy bitwise_or Function ---')
print(np.bitwise_or((arr2 < 8), (arr2 > 17)))

输出

[[ 1.04390004  0.66190755  0.27436456 -1.80262006 -0.263908  ]
 [ 0.4266228  -0.08927048  0.48680432 -3.34951675  0.36628762]
 [-1.41136647 -0.42538889 -1.88675276 -1.11568904  0.47588306]]
--- Two Dimensional Numpy Bitwise or Operator ---
[[ True False False  True  True]
 [False  True False  True False]
 [ True  True  True  True False]]
--- Two Dimensional Numpy bitwise_or Function ---
[[ True False False  True  True]
 [False  True False  True False]
 [ True  True  True  True False]]
--- Three Dimensional Random Array ---
[[[ 8 17 10  5]
  [ 9 16 10 23]]

 [[11 11  6 10]
  [ 9  4  6 15]]]
--- Three Dimensional Numpy Bitwise or Operator ---
[[[False False False  True]
  [False False False  True]]

 [[False False  True False]
  [False  True  True False]]]
--- Three Dimensional Numpy bitwise_or Function ---
[[[False False False  True]
  [False False False  True]]

 [[False False  True False]
  [False  True  True False]]]

Python Numpy 左移运算符

Python Numpy 位左移运算符将二进制数向左移动指定的位数。例如，a<<1 或 `left_shift(a, 1)` 将 12 转换为二进制值，然后向左移动一位。a = 00001100<<1 = 00011000 = 24。如果 a<<2，则为 00110000。

import numpy as np
 
a = 12
b = 25
 
print('Binary Value of 12 = ', np.binary_repr(a))
print('Binary Value of 25 = ', np.binary_repr(b))
 
print('\nLeft Shift Operator Result = ', a << 1)
print('left_shift Function Result = ', np.left_shift(a, 1))
 
print('\nLeft Shift Operator Result = ', b << 1)
print('left_shift Function Result = ', np.left_shift(b, 1))

左移输出

Binary Value of 12 =  1100
Binary Value of 25 =  11001

Left Shift Operator Result =  24
left_shift Function Result =  24

Left Shift Operator Result =  50
left_shift Function Result =  50

在此示例中，我们在数组上使用 Numpy `left_shift` 函数和左移运算符。第一个语句 (arr1 << 1) 将对数组中的每一项执行二进制左移一位。第二个语句 (arr1 << arr2) 根据 arr2 的项执行左移。我的意思是，2 << 1、12 << 2、9 << 3、12 << 4、17 << 5、11 << 6

import numpy as np
 
arr1 = np.array([2, 12, 9, 12, 17, 11])
print(arr1)
 
arr2 = np.array([1, 2, 3, 4, 5, 6])
print(arr2)
 
print('\nNumpy Left Shift Operator Result = ', (arr1 << 1))
print('Numpy left_shift Function Result = ', np.left_shift(arr1, 1))
 
print('\nNumpy Left Shift Operator Result = ', (arr1 << arr2))
print('Numpy left_shift Function Result = ', np.left_shift(arr1, arr2))
print()
 
x = np.random.randint(1, 20, size = (3, 7))
print(x)
y = np.random.randint(1, 5, size = (3, 7))
print(y)
 
print('\n---Numpy Left Shift Operator Result---\n', (x << y))
print('---Numpy left_shift Function Result---\n ', np.left_shift(x, y))

左移输出

[ 2 12  9 12 17 11]
[1 2 3 4 5 6]

Numpy Left Shift Operator Result =  [ 4 24 18 24 34 22]
Numpy left_shift Function Result =  [ 4 24 18 24 34 22]

Numpy Left Shift Operator Result =  [  4  48  72 192 544 704]
Numpy left_shift Function Result =  [  4  48  72 192 544 704]

[[ 3 13 18 14  5  4  2]
 [13 14  7  9  7  6  4]
 [ 8 13  4  3 18  4  8]]
[[2 3 1 3 4 2 1]
 [2 2 3 2 2 2 1]
 [4 2 2 2 4 3 4]]

---Numpy Left Shift Operator Result---
 [[ 12 104  36 112  80  16   4]
 [ 52  56  56  36  28  24   8]
 [128  52  16  12 288  32 128]]
---Numpy left_shift Function Result---
  [[ 12 104  36 112  80  16   4]
 [ 52  56  56  36  28  24   8]
 [128  52  16  12 288  32 128]]

Python Numpy 右移

Python Numpy 位右移运算符将二进制数向右移动给定的位数。例如 b>>1 或 `right_shift(b, 1)` 将 25 转换为二进制值，然后向右移动一位。b>>1 = 00011001>>1 = 00001100 = 12。

import numpy as np
 
a = 12
b = 25
 
print('Binary Value of 12 = ', np.binary_repr(a))
print('Binary Value of 25 = ', np.binary_repr(b))
 
print('\nRight Shift Operator Result = ', a >> 1)
print('right_shift Function Result = ', np.right_shift(a, 1))
 
print('\nRight Shift Operator Result = ', b >> 1)
print('right_shift Function Result = ', np.right_shift(b, 1))

Numpy 数组右移输出

Binary Value of 12 =  1100
Binary Value of 25 =  11001

Right Shift Operator Result =  6
right_shift Function Result =  6

Right Shift Operator Result =  12
right_shift Function Result =  12

在这里，我们在数组上使用 Numpy `right_shift` 函数和右移运算符。第一个 Numpy 语句 (arr1 >> 1) 对数组中的每一项执行二进制右移一位。第二个语句 (arr1 >> arr2) 根据 arr2 的项执行右移。我的意思是，2 >> 1、12 >> 2、9 >> 3、12 >> 4、17 >> 5、11 >> 6

import numpy as np
 
arr1 = np.array([2, 12, 9, 12, 17, 11])
print(arr1)
 
arr2 = np.array([1, 2, 3, 4, 5, 6])
print(arr2)
 
print('\nNumpy right Shift Operator Result = ', (arr1 >> 1))
print('Numpy right_shift Function Result = ', np.right_shift(arr1, 1))
 
print('\nNumpy right Shift Operator Result = ', (arr1 >> arr2))
print('Numpy right_shift Function Result = ', np.right_shift(arr1, arr2))
print()
 
x = np.random.randint(1, 20, size = (3, 7))
print(x)
y = np.random.randint(1, 5, size = (3, 7))
print(y)
 
print('\n---Numpy right Shift Operator Result---\n', (x >> y))
print('---Numpy right_shift Function Result---\n ', np.right_shift(x, y))

Numpy 数组右移输出

[ 2 12  9 12 17 11]
[1 2 3 4 5 6]

Numpy right Shift Operator Result =  [1 6 4 6 8 5]
Numpy right_shift Function Result =  [1 6 4 6 8 5]

Numpy right Shift Operator Result =  [1 3 1 0 0 0]
Numpy right_shift Function Result =  [1 3 1 0 0 0]

[[11 12 12 12  2 12  4]
 [ 1 14 14 14 17 12  2]
 [12 11 10 12  2  9 11]]
[[3 3 1 1 3 3 3]
 [4 4 4 1 2 4 4]
 [1 3 1 2 4 3 1]]

---Numpy right Shift Operator Result---
 [[1 1 6 6 0 1 0]
 [0 0 0 7 4 0 0]
 [6 1 5 3 0 1 5]]
---Numpy right_shift Function Result---
  [[1 1 6 6 0 1 0]
 [0 0 0 7 4 0 0]
 [6 1 5 3 0 1 5]]

Python Numpy 翻转

Python Numpy `invert` 函数与位非运算符相同。我们通过下面显示的 Python 示例来展示这一点。

import numpy as np
 
arr1 = np.array([10], dtype = np.uint8)
print(arr1)
 
print('Invert Value of arr1 = ', np.invert(arr1))
 
print('Binary Representation of arr1 = ', np.binary_repr(10, 8))
print('Binary Representation of x    = ', np.binary_repr(245, 8))
 
arr2 = np.array([1, 2, 3, 4, 5, 6], dtype = np.uint8)
print(arr2)
print('Invert Value of arr1 = ', np.invert(arr2))
 
for x in arr2:
    print('Binary Value of arr2  =', np.binary_repr(x, 8))
     
print('Binary Representation of 254 = ', np.binary_repr(254, 8))
print('Binary Representation of 253 = ', np.binary_repr(253, 8))
print('Binary Representation of 252 = ', np.binary_repr(252, 8))
print('Binary Representation of 251 = ', np.binary_repr(251, 8))
print('Binary Representation of 250 = ', np.binary_repr(250, 8))
print('Binary Representation of 249 = ', np.binary_repr(249, 8))

Python Numpy 数组翻转函数

[10]
Invert Value of arr1 =  [245]
Binary Representation of arr1 =  00001010
Binary Representation of x    =  11110101
[1 2 3 4 5 6]
Invert Value of arr1 =  [254 253 252 251 250 249]
Binary Value of arr2  = 00000001
Binary Value of arr2  = 00000010
Binary Value of arr2  = 00000011
Binary Value of arr2  = 00000100
Binary Value of arr2  = 00000101
Binary Value of arr2  = 00000110
Binary Representation of 254 =  11111110
Binary Representation of 253 =  11111101
Binary Representation of 252 =  11111100
Binary Representation of 251 =  11111011
Binary Representation of 250 =  11111010
Binary Representation of 249 =  11111001