numpy常见操作汇总

numpy常见操作汇总

• 本质是多维数组

import numpy as np

# numpy版本
np.__version__

'1.21.2'

python list

list1 = list(range(10))
list1

[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]

list1[5]='hello list'
list1

[0, 1, 2, 3, 4, 'hello list', 6, 7, 8, 9]

array

import array

arr1 = array.array('i',list(range(10)))
arr1

array('i', [0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

arr1[5]='hello array'
arr1

---------------------------------------------------------------------------

TypeError                                 Traceback (most recent call last)

 in 
----> 1 arr1[5]='hello array'
      2 arr1


TypeError: an integer is required (got type str)

arr1[5] = 78
arr1

array('i', [0, 1, 2, 3, 4, 78, 6, 7, 8, 9])

numpy.ndarray

arr = np.array(list(range(10)))
arr

array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

type(arr)

numpy.ndarray

arr.dtype

dtype('int32')

arr[3] = 33
arr

array([ 0,  1,  2, 33,  4,  5,  6,  7,  8,  9])

arr[2]

2

arr[2] = 'hello ndarray'
arr[2]

---------------------------------------------------------------------------

ValueError                                Traceback (most recent call last)

 in 
----> 1 arr[2] = 'hello ndarray'
      2 arr[2]


ValueError: invalid literal for int() with base 10: 'hello ndarray'

arr[3] = 13.4
arr

array([ 0,  1,  2, 13,  4,  5,  6,  7,  8,  9])

arr1 = np.array([1,2,3.0])
arr1

array([1., 2., 3.])

arr1.dtype

dtype('float64')

arr2 = np.array([1,2,3],dtype=float)

arr2.dtype

dtype('float64')

def p_test(n):
    a = [i**2 for i in range(n)]
    b = [i**3 for i in range(n)]
    c = []
    for i in range(n):
        c.append(a[i]+b[i])
    return c
p_test(10)

[0, 2, 12, 36, 80, 150, 252, 392, 576, 810]

# 用numpy实现上述函数功能
def p_test1(n):
    a = np.arange(n) ** 2
    b = np.arange(n) ** 3
    c = a + b
    return c
p_test1(10)

array([  0,   2,  12,  36,  80, 150, 252, 392, 576, 810], dtype=int32)

比较用时，numpy更快

%time res = p_test(10000000)

Wall time: 16.7 s

%time res = p_test1(10000000)

Wall time: 363 ms

矩阵和随机数的产生

np.array([1,2,3])

array([1, 2, 3])

np.array(range(10))

array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

np.arange(10)

array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

np.arange(2,20,2)

array([ 2,  4,  6,  8, 10, 12, 14, 16, 18])

np.arange(2,20,0.3)

array([ 2. ,  2.3,  2.6,  2.9,  3.2,  3.5,  3.8,  4.1,  4.4,  4.7,  5. ,
        5.3,  5.6,  5.9,  6.2,  6.5,  6.8,  7.1,  7.4,  7.7,  8. ,  8.3,
        8.6,  8.9,  9.2,  9.5,  9.8, 10.1, 10.4, 10.7, 11. , 11.3, 11.6,
       11.9, 12.2, 12.5, 12.8, 13.1, 13.4, 13.7, 14. , 14.3, 14.6, 14.9,
       15.2, 15.5, 15.8, 16.1, 16.4, 16.7, 17. , 17.3, 17.6, 17.9, 18.2,
       18.5, 18.8, 19.1, 19.4, 19.7])

np.zeros(10,dtype=int)

array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0])

np.zeros(10)

array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])

np.zeros(shape=(3,5),dtype=int)

array([[0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0]])

np.ones(10)

array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1.])

np.ones(10,dtype=int)

array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1])

np.ones(shape=(4,4),dtype=int)

array([[1, 1, 1, 1],
       [1, 1, 1, 1],
       [1, 1, 1, 1],
       [1, 1, 1, 1]])

np.full(10,99)

array([99, 99, 99, 99, 99, 99, 99, 99, 99, 99])

np.full((3,5),99)

array([[99, 99, 99, 99, 99],
       [99, 99, 99, 99, 99],
       [99, 99, 99, 99, 99]])

等差数列

np.linspace(0,10,10)

array([ 0.        ,  1.11111111,  2.22222222,  3.33333333,  4.44444444,
        5.55555556,  6.66666667,  7.77777778,  8.88888889, 10.        ])

np.linspace(0,10,10,dtype=int)

array([ 0,  1,  2,  3,  4,  5,  6,  7,  8, 10])

随机数

np.random.randint(0,20)

9

np.random.randint(0,10,size=5)

array([0, 8, 0, 5, 1])

np.random.randint(0,10,size=(3,5))

array([[9, 4, 9, 1, 6],
       [9, 8, 7, 8, 3],
       [6, 0, 3, 6, 4]])

# 控制随机数不变
np.random.seed(2)
np.random.randint(1,20,size=(4,5))

array([[ 9, 16, 14,  9, 12],
       [19, 12,  9,  8,  3],
       [18, 12, 16,  6,  8],
       [ 4,  7,  5, 11, 12]])

np.random.random()

0.7197542323569591

np.random.random((3,5))

array([[0.25849809, 0.54620732, 0.40730783, 0.17698462, 0.96963241],
       [0.29701836, 0.28786882, 0.11619332, 0.18172704, 0.49428977],
       [0.56576513, 0.22183517, 0.76749117, 0.57730807, 0.16782331]])

ndarray基础操作

a = np.ones((3,5))

a

array([[1., 1., 1., 1., 1.],
       [1., 1., 1., 1., 1.],
       [1., 1., 1., 1., 1.]])

a.ndim

2

a.shape

(3, 5)

a.size

15

a.reshape(5,3)

array([[1., 1., 1.],
       [1., 1., 1.],
       [1., 1., 1.],
       [1., 1., 1.],
       [1., 1., 1.]])

取值

x = np.arange(15).reshape(3,5)
x

array([[ 0,  1,  2,  3,  4],
       [ 5,  6,  7,  8,  9],
       [10, 11, 12, 13, 14]])

x[0]

array([0, 1, 2, 3, 4])

x[-1]

array([10, 11, 12, 13, 14])

x[0][1]

1

x[(0,1)]

1

x[0,1]

1

x[(0,1),(2,3)]

array([2, 8])

切片

y = np.arange(10)
y

array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

y[0:5]

array([0, 1, 2, 3, 4])

y[:5]

array([0, 1, 2, 3, 4])

y[5:]

array([5, 6, 7, 8, 9])

y[0:8:2]

array([0, 2, 4, 6])

y[::2]

array([0, 2, 4, 6, 8])

y[::-1]

array([9, 8, 7, 6, 5, 4, 3, 2, 1, 0])

a = np.arange(16).reshape(4,4)
a

array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11],
       [12, 13, 14, 15]])

# 取前两行，前三列
a[0:2,0:3]

array([[0, 1, 2],
       [4, 5, 6]])

a[:2,:3]

array([[0, 1, 2],
       [4, 5, 6]])

a[:2,::2]

array([[0, 2],
       [4, 6]])

a[::-1,::-1]

array([[15, 14, 13, 12],
       [11, 10,  9,  8],
       [ 7,  6,  5,  4],
       [ 3,  2,  1,  0]])

a.T

array([[ 0,  4,  8, 12],
       [ 1,  5,  9, 13],
       [ 2,  6, 10, 14],
       [ 3,  7, 11, 15]])

矩阵合并

x1 = np.array([[20,178],[23,180]])
x1

array([[ 20, 178],
       [ 23, 180]])

x2 = np.array([[1],[0]])
x2

array([[1],
       [0]])

x = np.concatenate([x1,x2],axis=1)
x

array([[ 20, 178,   1],
       [ 23, 180,   0]])

聚合操作

m = np.arange(16).reshape(4,4)
m

array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11],
       [12, 13, 14, 15]])

np.sum(m)

120

np.max(m)

15

np.min(m)

0

np.mean(m)

7.5