#导入numpy
# import指令，打开门让pandas进来
# as 给numpy取个昵称，不然每次调用的时候都要输入 numpy.函数，现在就是 np.函数
import numpy as np

#定义一个一维数组
a = np.array([1,2,3])
print('a的数据类型是{}'.format(type(a)))

a的数据类型是<class 'numpy.ndarray'>

#定义一个二维数组
b = np.array([[1,1],[2,2]])
b

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

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

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

#查看数组的维度
c.ndim

2

#查看数组形状
c.shape

(3, 3)

#查看数组中的数据类型
c.dtype

dtype('int32')

#创建一个零数组
np.zeros(10)

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

np.zeros((3,4,3))#或者 np.empty((3,4,3))

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

       [[0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]],

       [[0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]]])

np.empty((3,4,3))

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

       [[0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]],

       [[0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]]])

#创建全1矩阵
np.ones(10)

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

np.ones((3,4,3))

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

       [[1., 1., 1.],
        [1., 1., 1.],
        [1., 1., 1.],
        [1., 1., 1.]],

       [[1., 1., 1.],
        [1., 1., 1.],
        [1., 1., 1.],
        [1., 1., 1.]]])

#用随机数创造一个二维数组——np.random.randn(行数，列数)
data = np.random.randn(2,3)
data

array([[-0.14248726,  1.25007783, -0.46832963],
       [-0.06727693,  0.4056693 ,  0.92559399]])

#加法
data + data

array([[-0.28497451,  2.50015567, -0.93665925],
       [-0.13455386,  0.81133859,  1.85118798]])

#减法
data - data

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

#乘法
data * 10

array([[-1.42487256, 12.50077835, -4.68329626],
       [-0.67276931,  4.05669295,  9.25593989]])

#除法
1/data

array([[ -7.01817152,   0.79995019,  -2.13524822],
       [-14.86393597,   2.46506209,   1.08038731]])

#幂
data ** 0.5

<ipython-input-16-47d2e170e061>:2: RuntimeWarning: invalid value encountered in sqrt
  data ** 0.5

array([[       nan, 1.1180688 ,        nan],
       [       nan, 0.63692173, 0.96207795]])

import pandas as pd
my_series=pd.Series([1,2,3,4],index=['a','b','c','d'])
print('my_series的数据类型是{}'.format(type(my_series)))
print(my_series)

my_series的数据类型是<class 'pandas.core.series.Series'>
a    1
b    2
c    3
d    4
dtype: int64

my_series=pd.Series(['a','b','c','d'])
print(my_series)

0    a
1    b
2    c
3    d
dtype: object

my_list=['a','b','c','d']
my_series_2=pd.Series(my_list)
my_series_2

0    a
1    b
2    c
3    d
dtype: object

my_series=pd.Series(['a','b','c','d','e'],index=[0,1,2,3,'a'])
my_series

0    a
1    b
2    c
3    d
a    e
dtype: object

my_series['a']=6        #我们可以索引后修改series的值
print(my_series)

0    a
1    b
2    c
3    d
a    6
dtype: object

print(my_series['a'])   #类似于字典的操作方式

6

print(my_series[1])     #类似于数组的索引方式
print(my_series[1:3])   #切片

b
1    b
2    c
dtype: object

s1=pd.Series([1,2,3])
s2=pd.Series([4,5,6])
s1.append(s2)

0    1
1    2
2    3
0    4
1    5
2    6
dtype: int64

my_df=pd.DataFrame([[1,2,3,4],[5,6,7,8]],columns=['a','b','c','d'],index=['first','second'])
my_df

my_dict={'a':[1,2],'b':[3,4],'c':[5,6]}
my_df_2=pd.DataFrame(my_dict)
my_df_2

data=pd.read_csv('http://www.cs.nott.ac.uk/~pszgss/teaching/ML/Prac3/adult.data.small',header=None)

data.head()#head默认为5，查看前5行

len(data.columns)#查看列索引数量

15

index=['age', 'workclass', 'fnlwgt', 'education',
       'education-num', 'marital-status', 'occupation', 'relationship', 'race',
       'sex', 'capital-gain', 'capital-loss', 'hours-per-week',
       'native-country', 'salary']
len(index)

15

data.columns=index
data.head()

import pandas as pd

# 问题1：路径不对，大家只写了路径，没写文件名字和后缀
# 问题2：文件名字里，不要有空格。如果非要区分，用下划线连接。比如：重庆_渝北。
# 问题3：路径里，不要有中文
# 问题4: 有的人的电脑，在路径里需要有两个斜线进行分割。

data=pd.read_csv('http://www.cs.nott.ac.uk/~pszgss/teaching/ML/Prac3/adult.data.small',header=None)
# 就是导入文件的操作

# 如果大家是要读取excel文件，可以使用read_excel方法，括号里同样是填写文件路径和文件名字
# data=pd.read_excel(路径和文件名)
# data=pd.read_csv(r"C:\Users\sh\Desktop\mydata.csv")

data.head()

data.head(2)
# 这就是我们的数据表，使用 head() 操作可以让我们直观地看到数据的前5行。
# 当然，我们也可以查看数据的前n行，试试data.head(n)看看有什么效果？

data.columns

Int64Index([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14], dtype='int64')

index=['age', 'workclass', 'fnlwgt', 'education',
       'education-num', 'marital-status', 'occupation', 'relationship', 'race',
       'sex', 'capital-gain', 'capital-loss', 'hours-per-week',
       'native-country', 'salary']

data.columns=index
data.head()

data.columns

Index(['age', 'workclass', 'fnlwgt', 'education', 'education-num',
       'marital-status', 'occupation', 'relationship', 'race', 'sex',
       'capital-gain', 'capital-loss', 'hours-per-week', 'native-country',
       'salary'],
      dtype='object')

print('我们的这组数据有{}个样本，每个样本有{}个特征\n'.format(len(data),len(data.columns)))
print('这些特征是: {}'.format(' '.join(data.columns)))

我们的这组数据有9000个样本，每个样本有15个特征

这些特征是: age workclass fnlwgt education education-num marital-status occupation relationship race sex capital-gain capital-loss hours-per-week native-country salary

data.describe()

data[['age']]

print("data['age']的数据类型是{}\n".format(type(data['age'])))
data['age'].head()

data['age']的数据类型是<class 'pandas.core.series.Series'>

0    39
1    50
2    38
3    53
4    28
Name: age, dtype: int64

data.age.head()

0    39
1    50
2    38
3    53
4    28
Name: age, dtype: int64

print("data['age']的数据类型是{}\n".format(type(data[['age','workclass']])))
data[['age','workclass']].head()

data['age']的数据类型是<class 'pandas.core.frame.DataFrame'>

data.head(1)

data.iloc[0]

age                           39
workclass              State-gov
fnlwgt                     77516
education              Bachelors
education-num                 13
marital-status     Never-married
occupation          Adm-clerical
relationship       Not-in-family
race                       White
sex                         Male
capital-gain                2174
capital-loss                   0
hours-per-week                40
native-country     United-States
salary                     <=50K
Name: 0, dtype: object

data.loc[data['age']<37].head()

data.loc[(data.age<37) & (data['education-num']==13)]
# 在这里，‘&’符号表示的是"与"。表示两个条件同时满足才为真。

(data.loc[(data.age<37) & (data['education-num']==13)])['workclass']

4          Private
11       State-gov
12         Private
42         Private
60         Private
           ...
8922       Private
8923       Private
8928       Private
8950       Private
8964       Private
Name: workclass, Length: 713, dtype: object

data.loc[(data.age<37) & (data['education-num']==13),['workclass','age']]

data.loc[data.salary=='<=50K','salary']=1
data.loc[data.salary=='>50K','salary']=0
data.head(20)

data.age=20
data.head()

data['check']=False

data.head()

data.loc[0:10,'check']='true'
data.head()

data.head()

data.drop('capital-loss',axis=1)
# axis=1表示纵向，axis=0表示横向操作。

data.drop([0,1,2],axis=0)

data.head()

data=data.drop('capital-loss',axis=1)#从新赋值

#将处理后的文件导出
data.to_csv(r"C:\Users\sh\Desktop\2021_01_05.csv")

data.head()

data # 展示的是原汁原味

print(data)

age          workclass  fnlwgt      education  education-num  \
0      20          State-gov   77516      Bachelors             13
1      20   Self-emp-not-inc   83311      Bachelors             13
2      20            Private  215646        HS-grad              9
3      20            Private  234721           11th              7
4      20            Private  338409      Bachelors             13
...   ...                ...     ...            ...            ...
8995   20            Private  390817        5th-6th              3
8996   20                  ?  145964   Some-college             10
8997   20            Private   30424           11th              7
8998   20            Private  548361        HS-grad              9
8999   20            Private  189148        HS-grad              9

           marital-status          occupation    relationship    race  \
0           Never-married        Adm-clerical   Not-in-family   White
1      Married-civ-spouse     Exec-managerial         Husband   White
2                Divorced   Handlers-cleaners   Not-in-family   White
3      Married-civ-spouse   Handlers-cleaners         Husband   Black
4      Married-civ-spouse      Prof-specialty            Wife   Black
...                   ...                 ...             ...     ...
8995        Never-married   Handlers-cleaners   Not-in-family   White
8996        Never-married                   ?   Not-in-family   White
8997            Separated       Other-service       Unmarried   White
8998   Married-civ-spouse        Craft-repair         Husband   White
8999   Married-civ-spouse   Handlers-cleaners         Husband   White

          sex  capital-gain  hours-per-week  native-country salary  check
0        Male          2174              40   United-States      1   true
1        Male             0              13   United-States      1   true
2        Male             0              40   United-States      1   true
3        Male             0              40   United-States      1   true
4      Female             0              40            Cuba      1   true
...       ...           ...             ...             ...    ...    ...
8995     Male             0              25          Mexico      1  False
8996     Male             0              40   United-States      1  False
8997   Female             0              38   United-States      1  False
8998     Male             0              40   United-States      1  False
8999     Male             0              48   United-States      1  False

[9000 rows x 15 columns]

import matplotlib
import matplotlib.pyplot as plt
import numpy as np

x1=[i for i in range(100)]
y1=np.sin(x1)
x2=[i for i in range(100)]
y2=[3* i for i in x2]

plt.plot(x1,y1)
plt.xlabel('avsf')

Text(0.5, 0, 'avsf')

plt.figure(1)
ax1=plt.subplot(211)#创建两行一列绘图区，选择第一个绘图区
ax1.set_title('a')
plt.xlabel("a")
plt.plot(x1,y1)
ax2=plt.subplot(212)#创建两行一列绘图区，选择第二个绘图区
plt.plot(x2,y2)

[<matplotlib.lines.Line2D at 0x2393be53b00>]

fig,axes=plt.subplots(2,2)#2*2的绘图区
print(axes)
axes[0][0].plot(x1,y1)
plt.xlabel='a'
plt.ylabel='b'
axes[0][1].plot(x2,y2)

[[<matplotlib.axes._subplots.AxesSubplot object at 0x000002393BDC4DA0>
  <matplotlib.axes._subplots.AxesSubplot object at 0x000002393BED16A0>]
 [<matplotlib.axes._subplots.AxesSubplot object at 0x000002393BEFFC18>
  <matplotlib.axes._subplots.AxesSubplot object at 0x000002393BF3D1D0>]]

[<matplotlib.lines.Line2D at 0x2393be9c278>]