数据分析：是把隐藏在一些看似杂乱无章的数据背后的信息提炼出来，总结出所研究对象的内在规律

数据分析三剑客：Numpy,Pandas,Matplotlib

NumPy(Numerical Python) 是 Python 语言的一个扩展程序库，支持大量的维度数组与矩阵运算，此外也针对数组运算提供大量的数学函数库。¶

一、创建ndarray

1. 使用np.array()创建

一维数据创建

import numpy as np

In [2]:

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

Out[2]:

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

二维数组创建

In [5]:

np.array([[1,2,3],['a','b',1.1]])

Out[5]:

array([['1', '2', '3'],
       ['a', 'b', '1.1']], dtype='<U11')

注意：

numpy默认ndarray的所有元素的类型是相同的
如果传进来的列表中包含不同的类型，则统一为同一类型，优先级：str>float>int

使用matplotlib.pyplot获取一个numpy数组，数据来源于一张图片

import matplotlib.pyplot as plt

In [8]:

img_arr = plt.imread('./bobo.jpg')

In [9]:

plt.imshow(img_arr)

Out[9]:

<matplotlib.image.AxesImage at 0x7635048>

In [11]:

plt.imshow(img_arr - 66)

Out[11]:

<matplotlib.image.AxesImage at 0x75bfd68>

操作该numpy数据，该操作会同步到图片中

2. 使用np的routines函数创建

5) np.linspace(start, stop, num=50, endpoint=True, retstep=False, dtype=None) 等差数列

In [12]:

np.linspace(1,100,num=50)

Out[12]:

array([  1.        ,   3.02040816,   5.04081633,   7.06122449,
         9.08163265,  11.10204082,  13.12244898,  15.14285714,
        17.16326531,  19.18367347,  21.20408163,  23.2244898 ,
        25.24489796,  27.26530612,  29.28571429,  31.30612245,
        33.32653061,  35.34693878,  37.36734694,  39.3877551 ,
        41.40816327,  43.42857143,  45.44897959,  47.46938776,
        49.48979592,  51.51020408,  53.53061224,  55.55102041,
        57.57142857,  59.59183673,  61.6122449 ,  63.63265306,
        65.65306122,  67.67346939,  69.69387755,  71.71428571,
        73.73469388,  75.75510204,  77.7755102 ,  79.79591837,
        81.81632653,  83.83673469,  85.85714286,  87.87755102,
        89.89795918,  91.91836735,  93.93877551,  95.95918367,
        97.97959184, 100.        ])

6) np.arange([start, ]stop, [step, ]dtype=None)

In [13]:

np.arange(1,100,2)

Out[13]:

array([ 1,  3,  5,  7,  9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33,
       35, 37, 39, 41, 43, 45, 47, 49, 51, 53, 55, 57, 59, 61, 63, 65, 67,
       69, 71, 73, 75, 77, 79, 81, 83, 85, 87, 89, 91, 93, 95, 97, 99])

7) np.random.randint(low, high=None, size=None, dtype='l')

In [28]:

np.random.seed(10)
arr = np.random.randint(0,100,size=(5,6))

9) np.random.random(size=None)

生成0到1的随机数，左闭右开 np.random.seed(3)

In [21]:

np.random.random(size=(3,3))

Out[21]:

array([[0.765334  , 0.68742254, 0.12771576],
       [0.34878082, 0.46292111, 0.75355298],
       [0.08188152, 0.53189213, 0.17514265]])

二、ndarray的属性

4个必记参数： ndim：维度 shape：形状（各维度的长度） size：总长度

dtype：元素类型

In [14]:

img_arr.shape

Out[14]:

(626, 413, 3)

In [15]:

img_arr.ndim

Out[15]:

In [16]:

img_arr.size

Out[16]:

In [17]:

img_arr.dtype

Out[17]:

dtype('uint8')

In [18]:

type(img_arr)

Out[18]:

numpy.ndarray

三、ndarray的基本操作

1. 索引

一维与列表完全一致多维时同理

In [29]:

arr

Out[29]:

array([[ 9, 15, 64, 28, 89, 93],
       [29,  8, 73,  0, 40, 36],
       [16, 11, 54, 88, 62, 33],
       [72, 78, 49, 51, 54, 77],
       [69, 13, 25, 13, 92, 86]])

In [31]:

arr[[1,2]]

Out[31]:

array([[29,  8, 73,  0, 40, 36],
       [16, 11, 54, 88, 62, 33]])

根据索引修改数据

2. 切片

一维与列表完全一致多维时同理

In [32]:

arr

Out[32]:

array([[ 9, 15, 64, 28, 89, 93],
       [29,  8, 73,  0, 40, 36],
       [16, 11, 54, 88, 62, 33],
       [72, 78, 49, 51, 54, 77],
       [69, 13, 25, 13, 92, 86]])

In [33]:

#获取二维数组前两行
arr[0:2]

Out[33]:

array([[ 9, 15, 64, 28, 89, 93],
       [29,  8, 73,  0, 40, 36]])

In [34]:

#获取二维数组前两列
arr[:,0:2]

Out[34]:

array([[ 9, 15],
       [29,  8],
       [16, 11],
       [72, 78],
       [69, 13]])

In [10]:

#获取二维数组前两行和前两列数据

In [36]:

arr[0:2,0:2]

Out[36]:

array([[ 9, 15],
       [29,  8]])

将数据反转，例如[1,2,3]---->[3,2,1]

::进行切片

In [37]:

arr

Out[37]:

array([[ 9, 15, 64, 28, 89, 93],
       [29,  8, 73,  0, 40, 36],
       [16, 11, 54, 88, 62, 33],
       [72, 78, 49, 51, 54, 77],
       [69, 13, 25, 13, 92, 86]])

In [38]:

#将数组的行倒序
arr[::-1]

Out[38]:

array([[69, 13, 25, 13, 92, 86],
       [72, 78, 49, 51, 54, 77],
       [16, 11, 54, 88, 62, 33],
       [29,  8, 73,  0, 40, 36],
       [ 9, 15, 64, 28, 89, 93]])

In [39]:

#列倒序
arr[:,::-1]

Out[39]:

array([[93, 89, 28, 64, 15,  9],
       [36, 40,  0, 73,  8, 29],
       [33, 62, 88, 54, 11, 16],
       [77, 54, 51, 49, 78, 72],
       [86, 92, 13, 25, 13, 69]])

In [40]:

#全部倒序
arr[::-1,::-1]

Out[40]:

array([[86, 92, 13, 25, 13, 69],
       [77, 54, 51, 49, 78, 72],
       [33, 62, 88, 54, 11, 16],
       [36, 40,  0, 73,  8, 29],
       [93, 89, 28, 64, 15,  9]])

In [41]:

#将图片进行全倒置操作
plt.imshow(img_arr)

Out[41]:

<matplotlib.image.AxesImage at 0x7668400>

In [42]:

plt.imshow(img_arr[:,::-1,:])

Out[42]:

<matplotlib.image.AxesImage at 0x76f9e48>

In [44]:

plt.imshow(img_arr[::-1,::-1,::-1])

Out[44]:

<matplotlib.image.AxesImage at 0x786ec18>

3. 变形

使用arr.reshape()函数，注意参数是一个tuple！

基本使用

1.将一维数组变形成多维数组

In [51]:

arr_1.reshape((-1,15))

Out[51]:

array([[ 9, 15, 64, 28, 89, 93, 29,  8, 73,  0, 40, 36, 16, 11, 54],
       [88, 62, 33, 72, 78, 49, 51, 54, 77, 69, 13, 25, 13, 92, 86]])

2.将多维数组变形成一维数组

In [47]:

arr_1 = arr.reshape((30,))

4. 级联

np.concatenate()

1.一维，二维，多维数组的级联，实际操作中级联多为二维数组

In [52]:

arr

Out[52]:

array([[ 9, 15, 64, 28, 89, 93],
       [29,  8, 73,  0, 40, 36],
       [16, 11, 54, 88, 62, 33],
       [72, 78, 49, 51, 54, 77],
       [69, 13, 25, 13, 92, 86]])

In [56]:

np.concatenate((arr,arr),axis=1)

Out[56]:

array([[ 9, 15, 64, 28, 89, 93,  9, 15, 64, 28, 89, 93],
       [29,  8, 73,  0, 40, 36, 29,  8, 73,  0, 40, 36],
       [16, 11, 54, 88, 62, 33, 16, 11, 54, 88, 62, 33],
       [72, 78, 49, 51, 54, 77, 72, 78, 49, 51, 54, 77],
       [69, 13, 25, 13, 92, 86, 69, 13, 25, 13, 92, 86]])

In [57]:

arr1 = np.random.randint(0,100,size=(5,5))
arr1

Out[57]:

array([[30, 30, 89, 12, 65],
       [31, 57, 36, 27, 18],
       [93, 77, 22, 23, 94],
       [11, 28, 74, 88,  9],
       [15, 18, 80, 71, 88]])

In [60]:

np.concatenate((arr,arr1),axis=1)

Out[60]:

array([[ 9, 15, 64, 28, 89, 93, 30, 30, 89, 12, 65],
       [29,  8, 73,  0, 40, 36, 31, 57, 36, 27, 18],
       [16, 11, 54, 88, 62, 33, 93, 77, 22, 23, 94],
       [72, 78, 49, 51, 54, 77, 11, 28, 74, 88,  9],
       [69, 13, 25, 13, 92, 86, 15, 18, 80, 71, 88]])

2.合并两张照片

In [62]:

img_3 = np.concatenate((img_arr,img_arr,img_arr),axis=1)
imgs = np.concatenate((img_3,img_3,img_3),axis=0)
plt.imshow(imgs)

Out[62]:

<matplotlib.image.AxesImage at 0x79d4e80>

3.np.hstack与np.vstack

级联需要注意的点：

级联的参数是列表：一定要加中括号或小括号
维度必须相同
形状相符:在维度保持一致的前提下，如果进行横向（axis=1）级联，必须保证进行级联的数组行数保持一致。如果进行纵向（axis=0）级联，必须保证进行级联的数组列数保持一致。
可通过axis参数改变级联的方向

切割

In [64]:

plt.imshow(img_arr)

Out[64]:

<matplotlib.image.AxesImage at 0x7d05358>

In [67]:

plt.imshow(img_arr[50:390,85:340])

Out[67]:

<matplotlib.image.AxesImage at 0x7e01630>

In [65]:

plt.imshow(np.split(img_arr,[400],axis=0)[0])

Out[65]:

<matplotlib.image.AxesImage at 0x7d54ba8>

四、ndarray的聚合操作

1. 求和np.sum

In [70]:

arr.sum(axis=0)

Out[70]:

array([195, 125, 265, 180, 337, 325])

2. 最大最小值：np.max/ np.min

同理

3.平均值：np.mean()

3. 其他聚合操作

Function Name   NaN-safe Version    Description
np.sum  np.nansum   Compute sum of elements
np.prod np.nanprod  Compute product of elements
np.mean np.nanmean  Compute mean of elements
np.std  np.nanstd   Compute standard deviation
np.var  np.nanvar   Compute variance
np.min  np.nanmin   Find minimum value
np.max  np.nanmax   Find maximum value
np.argmin   np.nanargmin    Find index of minimum value
np.argmax   np.nanargmax    Find index of maximum value
np.median   np.nanmedian    Compute median of elements
np.percentile   np.nanpercentile    Compute rank-based statistics of elements
np.any  N/A Evaluate whether any elements are true
np.all  N/A Evaluate whether all elements are true
np.power 幂运算

六、ndarray的排序

1. 快速排序

np.sort()与ndarray.sort()都可以，但有区别：

np.sort()不改变输入
ndarray.sort()本地处理，不占用空间，但改变输入

In [72]:

np.sort(arr,axis=0)

Out[72]:

array([[ 9,  8, 25,  0, 40, 33],
       [16, 11, 49, 13, 54, 36],
       [29, 13, 54, 28, 62, 77],
       [69, 15, 64, 51, 89, 86],
       [72, 78, 73, 88, 92, 93]])

In [73]:

arr.sort(axis=0)

In [74]:

arr

Out[74]:

array([[ 9,  8, 25,  0, 40, 33],
       [16, 11, 49, 13, 54, 36],
       [29, 13, 54, 28, 62, 77],
       [69, 15, 64, 51, 89, 86],
       [72, 78, 73, 88, 92, 93]])

陪伴is最长情的告白

python数据分析——numpy基本知识点