python数据分析实战-第3章-numpy库

Posted learnfromnow

tags:

篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了python数据分析实战-第3章-numpy库相关的知识，希望对你有一定的参考价值。

第3章　NumPy库　　32

3.1　NumPy简史　　32

3.2　NumPy安装　　32

3.3　ndarray：NumPy库的心脏　　33

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

1
 
type(a), a.dtype, a.ndim, a.size, a.shape, a.itemsize
 
(numpy.ndarray, dtype(‘int64‘), 1, 3, (3,), 8)

1
 
b = np.array([[1.3, 2.4],[0.3, 4.1]])
 
1
 
b
 
array([[1.3, 2.4],
       [0.3, 4.1]])

1
 
type(b), b.dtype, b.ndim, b.size, b.shape, b.itemsize
 
(numpy.ndarray, dtype(‘float64‘), 2, 4, (2, 2), 8)

1
 
b.data, type(b.data)
 
(<memory at 0x7f3df8064558>, memoryview)

3.3.1　创建数组　　34

1
 
c = np.array([[1, 2, 3],[4, 5, 6]])
 
1
 
c
 
array([[1, 2, 3],
       [4, 5, 6]])

1
 
d = np.array(((1, 2, 3),(4, 5, 6)))
 
1
 
d
 
array([[1, 2, 3],
       [4, 5, 6]])

1
 
e = np.array([(1, 2, 3), [4, 5, 6], (7, 8, 9)])
 
1
 
e
 
array([[1, 2, 3],
       [4, 5, 6],
       [7, 8, 9]])

3.3.2　数据类型　　34

1
 
g = np.array([[‘a‘, ‘b‘],[‘c‘, ‘d‘]])
 
1
 
g
 
array([[‘a‘, ‘b‘],
       [‘c‘, ‘d‘]], dtype=‘<U1‘)

1
 
g.dtype, g.dtype.name
 
(dtype(‘<U1‘), ‘str32‘)

3.3.3　dtype选项　　35

1
 
f = np.array([[1, 2, 3],[4, 5, 6]], dtype=complex)
 
1
 
f
 
array([[1.+0.j, 2.+0.j, 3.+0.j],
       [4.+0.j, 5.+0.j, 6.+0.j]])

3.3.4　自带的数组创建方法　　36

1
 
np.zeros((3, 3))
 
array([[0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.]])

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

1
 
np.arange(0, 10)
 
array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

1
 
np.arange(4, 10)
 
array([4, 5, 6, 7, 8, 9])

1
 
np.arange(0, 12, 3)
 
array([0, 3, 6, 9])

1
 
np.arange(0, 6, 0.6)
 
array([0. , 0.6, 1.2, 1.8, 2.4, 3. , 3.6, 4.2, 4.8, 5.4])

1
 
np.arange(0, 12).reshape(3, 4)
 
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11]])

1
 
np.linspace(0,10,5)
 
array([ 0. ,  2.5,  5. ,  7.5, 10. ])

1
 
np.random.random(3)
 
array([0.71423775, 0.62229946, 0.97150108])

1
 
np.random.random((3,3))
 
array([[0.95155686, 0.10092046, 0.35220847],
       [0.96635154, 0.33764827, 0.44879963],
       [0.40573132, 0.97260092, 0.30105595]])

3.4　基本操作　　37

3.4.1　算术运算符　　37

1
 
a = np.arange(4)
 
1
 
a
 
array([0, 1, 2, 3])

1
 
a+4
 
array([4, 5, 6, 7])

1
 
a*2
 
array([0, 2, 4, 6])

1
 
b = np.arange(4,8)
 
1
 
b
 
array([4, 5, 6, 7])

1
 
a + b
 
array([ 4,  6,  8, 10])

1
 
a - b
 
array([-4, -4, -4, -4])

1
 
a * b
 
array([ 0,  5, 12, 21])

1
 
a * np.sin(b)
 
array([-0.        , -0.95892427, -0.558831  ,  1.9709598 ])

1
 
a * np.sqrt(b)
 
array([0.        , 2.23606798, 4.89897949, 7.93725393])

1
 
A = np.arange(0, 9).reshape(3, 3)
 
1
 
A
 
array([[0, 1, 2],
       [3, 4, 5],
       [6, 7, 8]])

1
 
B = np.ones((3, 3))
 
1
 
B
 
array([[1., 1., 1.],
       [1., 1., 1.],
       [1., 1., 1.]])

1
 
A * B
 
array([[0., 1., 2.],
       [3., 4., 5.],
       [6., 7., 8.]])

3.4.2　矩阵积　　38

1
 
np.dot(A,B)
 
array([[ 3.,  3.,  3.],
       [12., 12., 12.],
       [21., 21., 21.]])

1
 
A.dot(B)
 
array([[ 3.,  3.,  3.],
       [12., 12., 12.],
       [21., 21., 21.]])

1
 
np.dot(B,A)
 
array([[ 9., 12., 15.],
       [ 9., 12., 15.],
       [ 9., 12., 15.]])

3.4.3　自增和自减运算符　　39

1
 
a = np.arange(4)
 
1
 
a
 
array([0, 1, 2, 3])

1
 
a += 1
 
1
 
a
 
array([1, 2, 3, 4])

1
 
a -= 1
 
1
 
a
 
array([0, 1, 2, 3])

1
 
a += 4
 
1
 
a
 
array([4, 5, 6, 7])

1
 
a *= 2
 
1
 
a
 
array([ 8, 10, 12, 14])

3.4.4　通用函数　　40

1
 
a = np.arange(1, 5)
 
1
 
a
 
array([1, 2, 3, 4])

1
 
np.sqrt(a)
 
array([1.        , 1.41421356, 1.73205081, 2.        ])

1
 
np.log(a)
 
array([0.        , 0.69314718, 1.09861229, 1.38629436])

1
 
np.sin(a)
 
array([ 0.84147098,  0.90929743,  0.14112001, -0.7568025 ])

3.4.5　聚合函数　　40

1
 
a = np.array([3.3, 4.5, 1.2, 5.7, 0.3])
 
1
 
a
 
array([3.3, 4.5, 1.2, 5.7, 0.3])

1
 
a.sum()
 
15.0

1
 
a.min()
 
0.3

1
 
a.max()
 
5.7

1
 
a.mean()
 
3.0

1
 
a.std()
 
2.0079840636817816

3.5　索引机制、切片和迭代方法　　41

3.5.1　索引机制　　41

1
 
a = np.arange(10, 16)
 
1
 
a
 
array([10, 11, 12, 13, 14, 15])

1
 
a[4], a[-1], a[-6]
 
(14, 15, 10)

1
 
a[[1, 3, 4]]
 
array([11, 13, 14])

1
 
A = np.arange(10, 19).reshape((3, 3))
 
1
 
A
 
array([[10, 11, 12],
       [13, 14, 15],
       [16, 17, 18]])

1
 
A[1, 2]
 
15

3.5.2　切片操作　　42

1
 
a = np.arange(10, 16)
 
1
 
a
 
array([10, 11, 12, 13, 14, 15])

1
 
a[1:5]
 
array([11, 12, 13, 14])

1
 
a[1:5:2]
 
array([11, 13])

1
 
a[::2]
 
array([10, 12, 14])

1
 
a[:5:2]
 
array([10, 12, 14])

1
 
a[:5:]
 
array([10, 11, 12, 13, 14])

1
 
A = np.arange(10, 19).reshape((3, 3))
 
1
 
A
 
array([[10, 11, 12],
       [13, 14, 15],
       [16, 17, 18]])

1
 
A[0,:]
 
array([10, 11, 12])

1
 
A[:,0]
 
array([10, 13, 16])

1
 
A[0:2, 0:2]
 
array([[10, 11],
       [13, 14]])

1
 
A[[0,2], 0:2]
 
array([[10, 11],
       [16, 17]])

3.5.3　数组迭代　　43

1
2
 
for i in a:
    print(i)
 
10
11
12
13
14
15

1
2
 
for row in A:
    print(row)
 
[10 11 12]
[13 14 15]
[16 17 18]

1
2
 
for item in A.flat:
    print(item)
 
10
11
12
13
14
15
16
17
18

1
 
np.apply_along_axis(np.mean, axis=0, arr=A)
 
array([13., 14., 15.])

1
 
np.apply_along_axis(np.mean, axis=1, arr=A)
 
array([11., 14., 17.])

1
2
 
def foo(x):
    return x/2
 
1
 
np.apply_along_axis(foo, axis=1, arr=A)
 
array([[5. , 5.5, 6. ],
       [6.5, 7. , 7.5],
       [8. , 8.5, 9. ]])

1
 
np.apply_along_axis(foo, axis=0, arr=A)
 
array([[5. , 5.5, 6. ],
       [6.5, 7. , 7.5],
       [8. , 8.5, 9. ]])

3.6　条件和布尔数组　　45

1
 
A = np.random.random((4, 4))
 
1
 
A
 
array([[0.92122811, 0.55810739, 0.16604965, 0.98756188],
       [0.89990698, 0.45441885, 0.85233406, 0.32187075],
       [0.58032546, 0.01878004, 0.91573127, 0.63144651],
       [0.52282585, 0.57666091, 0.97865543, 0.54473589]])

1
 
A < 0.5
 
array([[False, False,  True, False],
       [False,  True, False,  True],
       [False,  True, False, False],
       [False, False, False, False]])

1
 
A[A < 0.5]
 
array([0.04869294, 0.07669828, 0.26766679, 0.04023488, 0.10839496,
       0.45508628, 0.48141253])

3.7　形状变换　　45

1
 
a = np.random.random(12)
 
1
 
a
 
array([0.93015885, 0.01067508, 0.43200938, 0.24035759, 0.01564892,
       0.09507133, 0.70143444, 0.46887118, 0.85551184, 0.61722919,
       0.94688342, 0.92520351])

1
 
A = a.reshape(3, 4)
 
1
 
A
 
array([[0.93015885, 0.01067508, 0.43200938, 0.24035759],
       [0.01564892, 0.09507133, 0.70143444, 0.46887118],
       [0.85551184, 0.61722919, 0.94688342, 0.92520351]])

1
 
a.shape = (3, 4)
 
1
 
a
 
array([[0.93015885, 0.01067508, 0.43200938, 0.24035759],
       [0.01564892, 0.09507133, 0.70143444, 0.46887118],
       [0.85551184, 0.61722919, 0.94688342, 0.92520351]])

1
 
a.ravel()
 
array([0.93015885, 0.01067508, 0.43200938, 0.24035759, 0.01564892,
       0.09507133, 0.70143444, 0.46887118, 0.85551184, 0.61722919,
       0.94688342, 0.92520351])

1
 
a.shape = (12)
 
1
 
a
 
array([0.93015885, 0.01067508, 0.43200938, 0.24035759, 0.01564892,
       0.09507133, 0.70143444, 0.46887118, 0.85551184, 0.61722919,
       0.94688342, 0.92520351])

1
 
A.transpose()
 
array([[0.93015885, 0.01564892, 0.85551184],
       [0.01067508, 0.09507133, 0.61722919],
       [0.43200938, 0.70143444, 0.94688342],
       [0.24035759, 0.46887118, 0.92520351]])

3.8　数组操作　　46

3.8.1　连接数组　　46

1
2
 
A = np.ones((3, 3))
B = np.zeros((3, 3))
 
1
 
np.vstack((A, B))
 
array([[1., 1., 1.],
       [1., 1., 1.],
       [1., 1., 1.],
       [0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.]])

1
 
np.hstack((A,B))
 
array([[1., 1., 1., 0., 0., 0.],
       [1., 1., 1., 0., 0., 0.],
       [1., 1., 1., 0., 0., 0.]])

1
2
3
 
a = np.array([0, 1, 2])
b = np.array([3, 4, 5])
c = np.array([6, 7, 8])
 
1
 
np.column_stack((a, b, c))
 
array([[0, 3, 6],
       [1, 4, 7],
       [2, 5, 8]])

1
 
np.row_stack((a, b, c))
 
array([[0, 1, 2],
       [3, 4, 5],
       [6, 7, 8]])

3.8.2　数组切分　　47

1
 
A = np.arange(16).reshape((4, 4))
 
1
 
A
 
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11],
       [12, 13, 14, 15]])

1
 
[B,C] = np.hsplit(A, 2)
 
1
 
B
 
array([[ 0,  1],
       [ 4,  5],
       [ 8,  9],
       [12, 13]])

1
 
C
 
array([[ 2,  3],
       [ 6,  7],
       [10, 11],
       [14, 15]])

1
 
[B,C] = np.vsplit(A, 2)
 
1
 
B
 
array([[0, 1, 2, 3],
       [4, 5, 6, 7]])

1
 
C
 
array([[ 8,  9, 10, 11],
       [12, 13, 14, 15]])

1
 
[A1,A2,A3] = np.split(A,[1,3],axis=1)
 
1
 
A1
 
array([[ 0],
       [ 4],
       [ 8],
       [12]])

1
 
A2
 
array([[ 1,  2],
       [ 5,  6],
       [ 9, 10],
       [13, 14]])

1
 
A3
 
array([[ 3],
       [ 7],
       [11],
       [15]])

1
 
[A1,A2,A3] = np.split(A,[1,3],axis=0)
 
1
 
A1
 
array([[0, 1, 2, 3]])

1
 
A2
 
array([[ 4,  5,  6,  7],
       [ 8,  9, 10, 11]])

1
 
A3
 
array([[12, 13, 14, 15]])

3.9　常用概念　　49

3.9.1　对象的副本或视图　　49

1
2
 
a = np.array([1, 2, 3, 4])
b = a
 
1
 
b
 
array([1, 2, 3, 4])

1
 
a[2] = 0
 
1
 
b
 
array([1, 2, 0, 4])

1
 
c = a[0:2]
 
1
 
c
 
array([1, 2])

1
 
a[0] = 0
 
1
 
c
 
array([0, 2])

1
 
a = np.array([1, 2, 3, 4])
 
1
 
c = a.copy()
 
1
 
c
 
array([1, 2, 3, 4])

1
 
a[0] = 0
 
1
 
c
 
array([1, 2, 3, 4])

3.9.2　向量化　　50

3.9.3　广播机制　　50

1
 
A = np.arange(16).reshape(4, 4)
 
1
 
b = np.arange(4)
 
1
 
A
 
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11],
       [12, 13, 14, 15]])

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

1
 
A + b
 
array([[ 0,  2,  4,  6],
       [ 4,  6,  8, 10],
       [ 8, 10, 12, 14],
       [12, 14, 16, 18]])

1
 
m = np.arange(6).reshape(3, 1, 2)
 
1
 
n = np.arange(6).reshape(3, 2, 1)
 
1
 
m
 
array([[[0, 1]],

       [[2, 3]],

       [[4, 5]]])

1
 
n
 
array([[[0],
        [1]],

       [[2],
        [3]],

       [[4],
        [5]]])

1
 
m + n
 
array([[[ 0,  1],
        [ 1,  2]],

       [[ 4,  5],
        [ 5,  6]],

       [[ 8,  9],
        [ 9, 10]]])

3.10　结构化数组　　52

1
 
structured = np.array([(1, ‘First‘, 0.5, 1+2j),(2, ‘Second‘, 1.3, 2-2j), (3, ‘Third‘, 0.8, 1+3j)],dtype=(‘i2, a6, f4, c8‘))
 
1
 
structured
 
array([(1, b‘First‘, 0.5, 1.+2.j), (2, b‘Second‘, 1.3, 2.-2.j),
       (3, b‘Third‘, 0.8, 1.+3.j)],
      dtype=[(‘f0‘, ‘<i2‘), (‘f1‘, ‘S6‘), (‘f2‘, ‘<f4‘), (‘f3‘, ‘<c8‘)])

1
 
structured = np.array([(1, ‘First‘, 0.5, 1+2j),(2, ‘Second‘, 1.3,2-2j),(3, ‘Third‘, 0.8, 1+3j)],dtype=(‘int16, a6, float32, complex64‘))
 
1
 
structured
 
array([(1, b‘First‘, 0.5, 1.+2.j), (2, b‘Second‘, 1.3, 2.-2.j),
       (3, b‘Third‘, 0.8, 1.+3.j)],
      dtype=[(‘f0‘, ‘<i2‘), (‘f1‘, ‘S6‘), (‘f2‘, ‘<f4‘), (‘f3‘, ‘<c8‘)])

1
 
structured[1]
 
(2, b‘Second‘, 1.3, 2.-2.j)

1
 
structured[‘f1‘]
 
array([b‘First‘, b‘Second‘, b‘Third‘], dtype=‘|S6‘)

1
 
structured = np.array([(1,‘First‘,0.5,1+2j),(2,‘Second‘,1.3,2-2j),(3,‘Third‘,0.8,1+3j)],dtype=[(‘id‘,‘i2‘),(‘position‘,‘a6‘),(‘value‘,‘f4‘),(‘complex‘,‘c8‘)])
 
1
 
structured
 
array([(1, b‘First‘, 0.5, 1.+2.j), (2, b‘Second‘, 1.3, 2.-2.j),
       (3, b‘Third‘, 0.8, 1.+3.j)],
      dtype=[(‘id‘, ‘<i2‘), (‘position‘, ‘S6‘), (‘value‘, ‘<f4‘), (‘complex‘, ‘<c8‘)])

1
 
structured.dtype.names = (‘id‘,‘order‘,‘value‘,‘complex‘)
 
1
 
structured[‘order‘]
 
array([b‘First‘, b‘Second‘, b‘Third‘], dtype=‘|S6‘)

3.11　数组数据文件的读写　　53

3.11.1　二进制文件的读写　　54

1
 
data = np.random.random(16).reshape(4,4)
 
1
 
data
 
array([[0.87899377, 0.11342252, 0.6682891 , 0.90832539],
       [0.33643193, 0.49913359, 0.46395367, 0.38503354],
       [0.30343769, 0.74571053, 0.16758969, 0.11507577],
       [0.00213888, 0.12661618, 0.71188585, 0.95622255]])

1
 
np.save(‘saved_data‘,data)
 
1
 
loaded_data = np.load(‘saved_data.npy‘)
 
1
 
loaded_data
 
array([[0.87899377, 0.11342252, 0.6682891 , 0.90832539],
       [0.33643193, 0.49913359, 0.46395367, 0.38503354],
       [0.30343769, 0.74571053, 0.16758969, 0.11507577],
       [0.00213888, 0.12661618, 0.71188585, 0.95622255]])

3.11.2　读取文件中的列表形式数据　　54

1
 
data = np.genfromtxt(‘data.csv‘, delimiter=‘,‘, names=True)
 
1
 
data
 
array([(1., 123., 1.4, 23.), (2., 110., 0.5, 18.), (3., 164., 2.1, 19.)],
      dtype=[(‘id‘, ‘<f8‘), (‘value1‘, ‘<f8‘), (‘value2‘, ‘<f8‘), (‘value3‘, ‘<f8‘)])

1
 
data2 = np.genfromtxt(‘data2.csv‘, delimiter=‘,‘, names=True)
 
1
 
data2
 
array([(1., 123., 1.4, 23.), (2., 110., nan, 18.), (3.,  nan, 2.1, 19.)],
      dtype=[(‘id‘, ‘<f8‘), (‘value1‘, ‘<f8‘), (‘value2‘, ‘<f8‘), (‘value3‘, ‘<f8‘)])

3.12　小结　　55

















































































































































































































































































































































































以上是关于python数据分析实战-第3章-numpy库的主要内容，如果未能解决你的问题，请参考以下文章

《Python数据分析实战》3 NumPy库

python数据分析实战-第10章-ipythonNotebook内嵌JavaScript库D3

python数据分析实战-第8章-用scikit-learn库实现机器学习

高端实战 Python数据分析与机器学习实战 Numpy/Pandas/Matplotlib等常用库

python数据分析实战-第7章-用matplotlib实现数据可视化

第3章 实战项目需求分析