numpy的属性

1
import numpy as np

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

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

1
print(array.ndim)#维度

1
2

1
print(array.shape)#形状

1
(3, 3)

1
print(array.size)#大小

1
9

1
print(array.dtype)#元素类型

1
int32

创建array

1
import numpy as np

1
2
a = np.array([1,2,3],dtype=np.int32)  #创建array的同时给定类型
print(a.dtype)

1
int32

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

1
float64

1
2
c = np.array([1,2,3])#一维数据
print(c)

1
[1 2 3]

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

1
2
[[1 2 3]
 [4 5 6]]

1
2
zero = np.zeros((2,3)) #生成2行3列全为0的矩阵  zeros矩阵
print(zero)

1
2
[[ 0.  0.  0.]
 [ 0.  0.  0.]]

1
2
one = np.ones((3,4)) #生成3行4列全为1的矩阵 ones矩阵
print(one)

1
2
3
[[ 1.  1.  1.  1.]
 [ 1.  1.  1.  1.]
 [ 1.  1.  1.  1.]]

1
2
empty = np.empty((3,2))#生成3行2列全都接近于0（不等于0）的矩阵
print(empty)

1
2
3
[[ 0.  0.]
 [ 0.  0.]
 [ 0.  0.]]

1
2
e = np.arange(10)   #arange的使用
print(e)

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

1
2
f = np.arange(4,12)
print(f)

1
[ 4  5  6  7  8  9 10 11]

1
2
g = np.arange(1,20,3)
print(g)

1
[ 1  4  7 10 13 16 19]

1
2
h = np.arange(8).reshape(4,2)#重新定义矩阵的形状
print(h)

1
2
3
4
[[0 1]
 [2 3]
 [4 5]
 [6 7]]

numpy的运算

1
import numpy as np

1
2
3
4
5
6
arr1 = np.array([[1,2,3],
                 [4,5,6]])
arr2 = np.array([[1,1,2],
                 [2,3,3]])
print(arr1)
print(arr2)

1
2
3
4
[[1 2 3]
 [4 5 6]]
[[1 1 2]
 [2 3 3]]

1
print(arr1 + arr2)   #数组的加法,按位加

1
2
[[2 3 5]
 [6 8 9]]

1
print(arr1 - arr2)  #数组减法,按位减

1
2
[[0 1 1]
 [2 2 3]]

1
print(arr1 * arr2)    #数组乘法,按位乘

1
2
[[ 1  2  6]
 [ 8 15 18]]

1
print(arr1 ** arr2)   #数组双乘号求幂

1
2
[[  1   2   9]
 [ 16 125 216]]

1
print(arr1 / arr2)    #数组除法,按位除

1
2
[[ 1.          2.          1.5       ]
 [ 2.          1.66666667  2.        ]]

1
print(arr1 % arr2)   #数组求余

1
2
[[0 0 1]
 [0 2 0]]

1
print(arr1 // arr2)

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

1
print(arr1+2) #数组所有的元素加2

1
2
[[3 4 5]
 [6 7 8]]

1
print(arr1*10)#数组所有的元素乘以10

1
2
[[10 20 30]
 [40 50 60]]

1
2
arr3 = arr1 > 3 #判断哪些元素大于3,返回的是布尔类型
print(arr3)

1
2
[[False False False]
 [ True  True  True]]

1
2
arr4 = np.ones((3,5))
print(arr4)

1
2
3
[[ 1.  1.  1.  1.  1.]
 [ 1.  1.  1.  1.  1.]
 [ 1.  1.  1.  1.  1.]]

1
print(arr1)

1
2
[[1 2 3]
 [4 5 6]]

1
np.dot(arr1,arr4)#矩阵乘法AB

1
2
array([[  6.,   6.,   6.,   6.,   6.],
       [ 15.,  15.,  15.,  15.,  15.]])

1
arr1.dot(arr4)#矩阵乘法

1
2
array([[  6.,   6.,   6.,   6.,   6.],
       [ 15.,  15.,  15.,  15.,  15.]])

1
2
3
print(arr1)
print(arr1.T)#矩阵转置
print(np.transpose(arr1))#矩阵转置

1
2
3
4
5
6
7
8
[[1 2 3]
 [4 5 6]]
[[1 4]
 [2 5]
 [3 6]]
[[1 4]
 [2 5]
 [3 6]]

随机数生成及矩阵的运算2

1
import numpy as np

1
2
sample1 = np.random.random((3,2))#生成3行2列从0到1的随机数,调用的是numpy中random的random函数
print(sample1)

1
2
3
[[ 0.42548654  0.60831272]
 [ 0.48034909  0.70289579]
 [ 0.96871932  0.33469266]]

1
2
sample2 = np.random.normal(size=(3,2))#生成3行2列符合标准正态分布的随机数,使用的是random中的normal函数(标准正态分布)
print(sample2)

1
2
3
[[ 0.82645622 -0.63300866]
 [ 0.18604463 -0.30988056]
 [-1.50301955 -0.51466896]]

1
2
sample3 = np.random.randint(0,10,size=(3,2))#生成3行2列从0到10的随机整数  ,ranint(随机整数)
print(sample3)

1
2
3
[[2 4]
 [3 1]
 [0 3]]

1
np.sum(sample1)#求和sum

1
3.5204561139867017

1
np.min(sample1)#求最小值min

1
0.33469265548836047

1
np.max(sample1)#求最大值max

1
0.96871931960307933

1
np.sum(sample1,axis=0)#对列求和,关键字axis0就是列,1就是行

1
array([ 1.87455495,  1.64590117])

1
np.sum(sample1,axis=1)#对行求和

1
array([ 1.03379926,  1.18324488,  1.30341198])

1
print(sample1)

1
2
3
[[ 0.42548654  0.60831272]
 [ 0.48034909  0.70289579]
 [ 0.96871932  0.33469266]]

1
np.argmin(sample1)#求最小值的索引  argmin(逐行从0开始的,将整个数组看展开成一维的)

1
5

1
np.argmax(sample1)#求最大值的索引

1
4

1
2
print(np.mean(sample1))#求平均值
print(sample1.mean())#求平均值

1
2
0.586742685664
0.586742685664

1
np.median(sample1)#求中位数median

1
0.5443309058371042

1
np.sqrt(sample1)#开方

1
2
3
array([[ 0.65229329,  0.77994405],
       [ 0.69307221,  0.8383888 ],
       [ 0.9842354 ,  0.57852628]])

1
2
sample4 = np.random.randint(0,10,size=(1,10))
print(sample4)

1
[[9 2 3 0 2 8 1 3 2 8]]

1
np.sort(sample4)#排序

1
array([[0, 1, 2, 2, 2, 3, 3, 8, 8, 9]])

1
np.sort(sample1)

1
2
3
array([[ 0.42548654,  0.60831272],
       [ 0.48034909,  0.70289579],
       [ 0.33469266,  0.96871932]])

1
np.clip(sample4,2,7)#小于2就变成2，大于7就变为7  ,数据阈值改变

1
array([[7, 2, 3, 2, 2, 7, 2, 3, 2, 7]])

numpy的索引

1
import numpy as np

1
2
arr1 = np.arange(2,14)
print(arr1)

1
[ 2  3  4  5  6  7  8  9 10 11 12 13]

1
print(arr1[2])#第二个位置的数据,一维

1
4

1
print(arr1[1:4])#第一到第四个位置的数据

1
[3 4 5]

1
print(arr1[2:-1])#第二到倒数第一个位置的数据

1
[ 4  5  6  7  8  9 10 11 12]

1
print(arr1[:5])#前五个数据

1
[2 3 4 5 6]

1
print(arr1[-2:])#最后两个数据

1
[12 13]

1
2
arr2 = arr1.reshape(3,4)
print(arr2)

1
2
3
[[ 2  3  4  5]
 [ 6  7  8  9]
 [10 11 12 13]]

1
print(arr2[1])

1
[6 7 8 9]

1
print(arr2[1][1])

1
7

1
print(arr2[1,2])

1
8

1
print(arr2[:,2])

1
[ 4  8 12]

1
2
for i in arr2: #迭代行
    print(i)

1
2
3
[2 3 4 5]
[6 7 8 9]
[10 11 12 13]

1
2
for i in arr2.T:#迭代列,转置一下再进行迭代
    print(i)

1
2
3
4
[ 2  6 10]
[ 3  7 11]
[ 4  8 12]
[ 5  9 13]

1
2
for i in arr2.flat:#一个一个元素迭代flat
    print(i)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
2
3
4
5
6
7
8
9
10
11
12
13

array合并

1
import numpy as np

1
2
3
4
5
arr1 = np.array([1,2,3])
arr2 = np.array([4,5,6])
arr3 = np.vstack((arr1,arr2))#垂直合并vertical(垂直)stack(堆放)
print(arr3)
print(arr3.shape)

1
2
3
[[1 2 3]
 [4 5 6]]
(2, 3)

1
2
3
arr4 = np.hstack((arr1,arr2))#水平合并horizontal(水平)
print(arr4)
print(arr4.shape)

1
2
[1 2 3 4 5 6]
(6,)

1
2
arrv = np.vstack((arr1,arr2,arr3))
print(arrv)

1
2
3
4
[[1 2 3]
 [4 5 6]
 [1 2 3]
 [4 5 6]]

1
2
arrh = np.hstack((arr1,arr2,arr4))
print(arrh)

1
[1 2 3 4 5 6 1 2 3 4 5 6]

1
2
arr = np.concatenate((arr1,arr2,arr1))
print(arr)

1
[1 2 3 4 5 6 1 2 3]

1
2
arr = np.concatenate((arr3,arrv),axis=0)#合并的array维度要相同，array形状要匹配，axis=0纵向合并
print(arr)

1
2
3
4
5
6
[[1 2 3]
 [4 5 6]
 [1 2 3]
 [4 5 6]
 [1 2 3]
 [4 5 6]]

1
2
arr = np.concatenate((arr3,arr3),axis=1)#合并的array维度要相同，array形状要匹配，axis=1横向合并
print(arr)

1
2
[[1 2 3 1 2 3]
 [4 5 6 4 5 6]]

1
2
arr1.T 
print(arr1.T) #一维的array不能转置

1
[1 2 3]

1
print(arr1.shape)

1
(3,)

1
2
3
arr1_1 = arr1[np.newaxis,:]  #添加一个维度到行
print(arr1_1)
print(arr1_1.shape)

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

1
print(arr1_1.T)

1
2
3
[[1]
 [2]
 [3]]

1
2
3
arr1_2 = arr1[:,np.newaxis]  #添加维度到列
print(arr1_2)
print(arr1_2.shape)

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

1
2
3
arr1_3 = np.atleast_2d(arr1)   #atleast_2d  将数据变成2维,维度比他高的不发生改变
print(arr1_3)
print(arr1_3.T)

1
2
3
4
[[1 2 3]]
[[1]
 [2]
 [3]]

array的分割

1
import numpy as np

1
2
arr1 = np.arange(12).reshape((3,4))
print(arr1)

1
2
3
[[ 0  1  2  3]
 [ 4  5  6  7]
 [ 8  9 10 11]]

1
2
3
arr2,arr3 = np.split(arr1,2,axis=1)#水平方向分割，分成2份split(分割)  将arr1切隔成2个arr2和arr3,在水平方向
print(arr2)
print(arr3)

1
2
3
4
5
6
[[0 1]
 [4 5]
 [8 9]]
[[ 2  3]
 [ 6  7]
 [10 11]]

1
2
3
4
arr4,arr5,arr6 = np.split(arr1,3,axis=0)#垂直方向分割，分成3份
print(arr4)
print(arr5)
print(arr6)

1
2
3
[[0 1 2 3]]
[[4 5 6 7]]
[[ 8  9 10 11]]

1
2
3
4
arr2,arr3,arr4 = np.split(arr1,3,axis=1)#水平方向分割，分成3份,没办法切割成相同大小的部分
print(arr2)
print(arr3)
print(arr4)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
---------------------------------------------------------------------------

TypeError                                 Traceback (most recent call last)

~/anaconda3/lib/python3.6/site-packages/numpy/lib/shape_base.py in split(ary, indices_or_sections, axis)
    552     try:
--> 553         len(indices_or_sections)
    554     except TypeError:


TypeError: object of type 'int' has no len()

​
​ During handling of the above exception, another exception occurred:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
ValueError                                Traceback (most recent call last)

<ipython-input-5-2961433b0366> in <module>()
----> 1 arr2,arr3,arr4 = np.split(arr1,3,axis=1)#水平方向分割，分成3份
      2 print(arr2)
      3 print(arr3)
      4 print(arr4)


~/anaconda3/lib/python3.6/site-packages/numpy/lib/shape_base.py in split(ary, indices_or_sections, axis)
    557         if N % sections:
    558             raise ValueError(
--> 559                 'array split does not result in an equal division')
    560     res = array_split(ary, indices_or_sections, axis)
    561     return res


ValueError: array split does not result in an equal division

1
2
3
4
arr7,arr8,arr9 = np.array_split(arr1,3,axis=1)#水平方向分割，分成3份，不等分割
print(arr7)
print(arr8)
print(arr9)

1
2
3
4
5
6
7
8
9
[[0 1]
 [4 5]
 [8 9]]
[[ 2]
 [ 6]
 [10]]
[[ 3]
 [ 7]
 [11]]

1
2
3
4
arrv1,arrv2,arrv3 = np.vsplit(arr1,3)#垂直分割
print(arrv1)
print(arrv2)
print(arrv3)

1
2
3
[[0 1 2 3]]
[[4 5 6 7]]
[[ 8  9 10 11]]

1
2
3
arrh1,arrh2 = np.hsplit(arr1,2)#水平分割
print(arrh1)
print(arrh2)

1
2
3
4
5
6
[[0 1]
 [4 5]
 [8 9]]
[[ 2  3]
 [ 6  7]
 [10 11]]

numpy的深拷贝,浅拷贝

1
import numpy as np

1
arr1 = np.array([1,2,3])

1
arr2 = arr1#arr1,arr2共享一块内存，浅拷贝

1
2
3
arr2[0] = 5   #浅拷贝,同步改变
print(arr1)
print(arr2)

1
2
[5 2 3]
[5 2 3]

1
arr3 = arr1.copy()#深拷贝这里就相当于普通变量的deepcopy

1
2
3
arr3[0] = 10
print(arr1)
print(arr3)

1
2
[5 2 3]
[10  2  3]