Numpy的学习笔记

#numpy

自用内容

简写np，矩阵的运算，

下面是输出矩阵：

import numpy as np
array = np.array([[1,2,3],[2,3,4]])
print(array)
#维度
print("number of dim: ",array.ndim)
#形状 ？X ？的数组
print("shape:",array.shape)
#多少元素
print("size:",array.size)

[[1 2 3]
 [2 3 4]]
number of dim:  2
shape: (2, 3)
size: 6

创建array，定义type

a = np.array([2,23,4],dtype = np.int64)

生成全部为0 和1的数组三行四列

zeroArray = np.zeros((3,4))
print(zeroArray)
oneArray = np.ones((3,4))
print(oneArray)

[[0. 0. 0. 0.]
 [0. 0. 0. 0.]
 [0. 0. 0. 0.]]
[[1. 1. 1. 1.]
 [1. 1. 1. 1.]
 [1. 1. 1. 1.]]

1.生成有步长的数组：
2.生成0—11的数列

arangeArray = np.arange(10,20,2)
print(arangeArray)
numArray = np.arange(12).reshape((3,4))
print(numArray)

[10 12 14 16 18]
[[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]]

生成线段，生成5点4段数列

linArray = np.linspace(1,10,6).reshape((2,3))
print(linArray)

[[ 1.   2.8  4.6]
 [ 6.4  8.2 10. ]]

numpy基础运算

1.加减法
2.平方
3.三角函数的运算

aArray = np.array([10,20,30,40])
bArray = np.arange(4)
c= aArray - bArray
print(c)
d = c **2
print(d)
sinA = np.sin(10)
print(sinA)

[10 19 28 37]
[ 100  361  784 1369]
-0.5440211108893698

数组的逻辑运算

1.筛选出数组中比3小的数
2.等于3的数
3.数组中逐个相乘

print(bArray<3)
print(bArray == 3)
aShape = aArray.reshape(2,2)
bShape = bArray.reshape(2,2)
print(aShape)
print(bShape)
mutArray = aShape * bShape #逐个相乘
mut_dotArray = np.dot(aShape,bShape) #矩阵相乘
#或者
mut_Array = aShape.dot(bShape)

print(mutArray)
print(mut_dotArray)

[ True  True  True False]
[False False False  True]
[[10 20]
 [30 40]]
[[0 1]
 [2 3]]
[[  0  20]
 [ 60 120]]
[[ 40  70]
 [ 80 150]]

随机生成一个0-1的矩阵

求和np.sum()
最小值np.min()
最大值np.max()
axis = 1 行
axis = 0 列

randomArray = np.random.random((2,4))
print(randomArray)
print(np.sum(randomArray,axis = 1))

[[0.54661074 0.27343044 0.42980382 0.04123646]
 [0.10769581 0.77549246 0.28787614 0.27292254]]
[1.29108145 1.44398695]

#最小值的索引
print(np.argmin(randomArray))
#max index
print(np.argmax(randomArray))
#mean
print(np.mean(randomArray))
print(randomArray.mean())
#median
print(np.median(randomArray))
#累加
print(np.cumsum(randomArray))
#累差
print(np.diff(randomArray))
#排序sort,逐行排序
print(np.sort(randomArray))
#矩阵反向：行列对换,转置
print(np.transpose(randomArray))
print(randomArray.T)

3
5
0.34188355032710077
0.34188355032710077
0.2806532854593555
[0.54661074 0.82004118 1.249845   1.29108145 1.39877726 2.17426972
 2.46214586 2.7350684 ]
[[-0.27318031  0.15637338 -0.38856736]
 [ 0.66779665 -0.48761633 -0.01495359]]
[[0.04123646 0.27343044 0.42980382 0.54661074]
 [0.10769581 0.27292254 0.28787614 0.77549246]]
[[0.54661074 0.10769581]
 [0.27343044 0.77549246]
 [0.42980382 0.28787614]
 [0.04123646 0.27292254]]
[[0.54661074 0.10769581]
 [0.27343044 0.77549246]
 [0.42980382 0.28787614]
 [0.04123646 0.27292254]]

clip

让小于或大于设定的某个数，都近似于设定的数字

setA = np.arange(2,14).reshape(3,4)
print(setA)
#求列中的平均值
print(np.mean(setA,axis = 0))
print(np.clip(setA,5,9))

[[ 2  3  4  5]
 [ 6  7  8  9]
 [10 11 12 13]]
[6. 7. 8. 9.]
[[5 5 5 5]
 [6 7 8 9]
 [9 9 9 9]]

index

print(setA)
#第3行的所有数
print(setA[2])
print(setA[2,:])
#第2行第三列
print(setA[1,2])
print(setA[2][1])
#第2列：
print(setA[:,1])
#取【7，8】,取左不取右
print(setA[1,1:3])

[[ 2  3  4  5]
 [ 6  7  8  9]
 [10 11 12 13]]
[10 11 12 13]
[10 11 12 13]
8
11
[ 3  7 11]
[7 8]

for循环输出每行，row和column都只是参数名

for row in setA:
    print(row)

[2 3 4 5]
[6 7 8 9]
[10 11 12 13]

for循环输出每列

for column in setA.T:
    print(column)

[ 2  6 10]
[ 3  7 11]
[ 4  8 12]
[ 5  9 13]

迭代出每一个数

for item in setA.flat:
    print(item)

#第二行的每一个数

print(setA)
    
for itemR2 in setA[1].flat:
    print(itemR2)

2
3
4
5
6
7
8
9
10
11
12
13
[[ 2  3  4  5]
 [ 6  7  8  9]
 [10 11 12 13]]
6
7
8
9

array的数据处理

setD = np.array([1,1,1])
setC = np.array([2,2,2])
setE = np.array([3,3])
print(np.vstack((setD,setC)))#vertical stack 变成二维上下合并
print(np.hstack((setD,setC)))#左右合并

[[1 1 1]
 [2 2 2]]
[1 1 1 2 2 2]

一维数组变二维数组

#将一维数组增加了一个维度，变成二维数组（1X3）
print(setD[np.newaxis,:])
#变成（3x1）
print(setD[:,np.newaxis])

[[1 1 1]]
[[1]
 [1]
 [1]]

多个array的合并，任意维度

#左右合并 axis = 1
setRow = np.concatenate((setC[np.newaxis,:],setD[np.newaxis,:]),axis = 1)
print(setRow)
#上下合并 axis = 0
setColumn = np.concatenate((setC[np.newaxis,:],setD[np.newaxis,:]),axis = 0)
print(setColumn)

[[2 2 2 1 1 1]]
[[2 2 2]
 [1 1 1]]

array的分割：

myArr = np.arange(12).reshape(3,4)
print(myArr)
#左右分割
print(np.split(myArr,2,axis = 1))
print(np.hsplit(myArr,2))
#左右不等量分割,分出三个不等量的矩阵
print(np.array_split(myArr,3,axis = 1))
#上下分割
print(np.vsplit(myArr,3))

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

Numpy 的copy 和deep copy

在numpyArray中，改变了赋值的数，被赋值的数也会跟着改变

#copy,仅copy初始值，不copy后面可能会改变的
print(setC)

setCopy = setC.copy()
#1.更改数组中的某一个值,得是整数
setC[0] = 3
print(setC)
#2.同时改变几个数字：
setC[1:3] = [4,6]
print(setC)
print(setCopy)

[2 2 2]
[3 2 2]
[3 4 6]
[2 2 2]