软件应用丨Pandas入门系列（一）：深入理解Series和DataFrame

版权声明：本文为CSDN博主「向前走别回头」的原创文章，遵循 CC 4.0 BY-SA 版权协议，在文中已附上链接。

Series的创建

# 使用列表创建

>>> import numpy as np
>>> import pandas as pd
>>> s1 = pd.Series([1,2,3,4])
>>> s1
0    1
1    2
2    3
3    4
dtype: int64

# 查看s1的值和索引
>>> s1.values
array([1, 2, 3, 4], dtype=int64)
>>> s1.index
RangeIndex(start=0, stop=4, step=1) # 默认索引

左右滑动查看更多

# 使用数组创建

>>> s2 = pd.Series(np.arange(10))
>>> s2
0    0
1    1
2    2
3    3
4    4
5    5
6    6
7    7
8    8
9    9
dtype: int32

左右滑动查看更多

# 使用字典创建

>>> s3 = pd.Series({'1':1, '2':2, '3':3})
>>> s3
1    1
2    2
3    3
dtype: int64
>>> s3.values
array([1, 2, 3], dtype=int64)
>>> s3.index
Index(['1', '2', '3'], dtype='object')

左右滑动查看更多

Series的访问

>>> s4 =  pd.Series([1,2,3,4], index = ['a','b','c','d'])
>>> s4
a    1
b    2
c    3
d    4
dtype: int64
>>> s4.values
array([1, 2, 3, 4], dtype=int64)
>>> s4.index
Index(['a', 'b', 'c', 'd'], dtype='object')
>>> s4['a'] # 访问索引为a的值
1
>>> s4[s4>2] #访问s4中值大于2的Series
c    3
d    4
dtype: int64

左右滑动查看更多

# Series与字典的转换

>>> s4.to_dict()  # s4转换为字典
{'a': 1, 'b': 2, 'c': 3, 'd': 4}


>>> s5 = pd.Series(s4.to_dict())  # 字典转换为Series
>>> s5
a    1
b    2
c    3
d    4
dtype: int64

左右滑动查看更多

# e索引无值补充为NaN

>>> index_1 = ['a','b','c','d','e']
>>> s6 = pd.Series(s5, index = index_1)
>>> s6
a    1.0
b    2.0
c    3.0
d    4.0
e    NaN  # s5此处无值
dtype: float64

左右滑动查看更多

# NaN判断

>>> pd.isnull(s6)
a    False
b    False
c    False
d    False
e     True
dtype: bool
>>> pd.notnull(s6)
a     True
b     True
c     True
d     True
e    False
dtype: bool

左右滑动查看更多

# 命名修改

>>> s6.name = 'demo'   # s6的名字修改
>>> s6
a    1.0
b    2.0
c    3.0
d    4.0
e    NaN
Name: demo, dtype: float64

>>> s6.index.name = 'demo_index'  # s6的索引的名字的修改
>>> s6.index
Index(['a', 'b', 'c', 'd', 'e'], dtype='object', name='demo_index')

左右滑动查看更多

官网：

http://pandas.pydata.org/pandas-docs/version/0.14.1/

原文链接：

https://blog.csdn.net/weixin_39778570/article/details/80550179

DataFrame数据框

# 巧用复制黏贴

>>> import numpy as np
>>> import pandas as pd
>>> from pandas import Series, DataFrame
>>> import webbrowser
>>> link = 'http://www.tiobe.com/tiobe-index/'
>>> webbrowser.open(link)  # 打开一个网页 然后把要生成的数据复制到剪切板，以下选择了编程语言前10行
True

>>> df = pd.read_clipboard()  # 执行这段代码自动生存DataFrame对象
>>> df
   May 2018 May 2017  Change Programming Language  Ratings Change.1
0         1        1     NaN                 Java  16.380%   +1.74%
1         2        2     NaN                    C  14.000%   +7.00%
2         3        3     NaN                  C++   7.668%   +2.92%
3         4        4     NaN               Python   5.192%   +1.64%
4         5        5     NaN                   C#   4.402%   +0.95%
5         6        6     NaN    Visual Basic .NET   4.124%   +0.73%
6         7        9  change                  PHP   3.321%   +0.63%
7         8        7  change           JavaScript   2.923%   -0.15%
8         9        -  change                  SQL   1.987%   +1.99%
9        10       11  change                 Ruby   1.182%   -1.25%
>>> type(df)  # 查看类型
<class 'pandas.core.frame.DataFrame'>

左右滑动查看更多

# 查看DataFrame的内容

>>> df.columns # 查看所有列
Index(['May 2018', 'May 2017', 'Change', 'Programming Language', 'Ratings',
       'Change.1'],
      dtype='object')
>>> df.Ratings  # 查看Ratings这一列
0    16.380%
1    14.000%
2     7.668%
3     5.192%
4     4.402%
5     4.124%
6     3.321%
7     2.923%
8     1.987%
9     1.182%
Name: Ratings, dtype: object

# 某一列的类型为时间序列
>>> type(df['May 2018'])
<class 'pandas.core.series.Series'>

左右滑动查看更多

# 从df中提取指定的列

>>> df_new = DataFrame(df, columns = ['May 2018', 'Change.1']) # 从df中提取两列生成新的DataFrame
>>> df_new
   May 2018 Change.1
0         1   +1.74%
1         2   +7.00%
2         3   +2.92%
3         4   +1.64%
4         5   +0.95%
5         6   +0.73%
6         7   +0.63%
7         8   -0.15%
8         9   +1.99%
9        10   -1.25%

左右滑动查看更多

# 如果列里面多加了不存在的  会自动赋值为空

>>> df_new = DataFrame(df, columns=['May 2018', 'Change', 'Sep 2019'])
>>> df_new
   May 2018  Change  Sep 2019 # Sep 2019 这一列是不存在的
0         1     NaN       NaN
1         2     NaN       NaN
2         3     NaN       NaN
3         4     NaN       NaN
4         5     NaN       NaN
5         6     NaN       NaN
6         7  change       NaN
7         8  change       NaN
8         9  change       NaN
9        10  change       NaN

左右滑动查看更多

# 列的赋值

# 对序列进行赋值，使用range函数

>>> df_new['Sep 2019'] = range(0,10) # 赋值 0-9 这10个数给 Sep 2019 这一列
>>> df_new
   May 2018  Change  Sep 2019
0         1     NaN         0
1         2     NaN         1
2         3     NaN         2
3         4     NaN         3
4         5     NaN         4
5         6     NaN         5
6         7  change         6
7         8  change         7
8         9  change         8
9        10  change         9

左右滑动查看更多

# 使用np下的arange（数组）函数

>>> df_new['Sep 2019'] = np.arange(1,11)
>>> df_new
   May 2018  Change  Sep 2019
0         1     NaN         1
1         2     NaN         2
2         3     NaN         3
3         4     NaN         4
4         5     NaN         5
5         6     NaN         6
6         7  change         7
7         8  change         8
8         9  change         9
9        10  change        10

左右滑动查看更多

# 使用序列修改

>>> df_new['Sep 2019'] = pd.Series(np.arange(2,12))
>>> df_new
   May 2018  Change  Sep 2019
0         1     NaN         2
1         2     NaN         3
2         3     NaN         4
3         4     NaN         5
4         5     NaN         6
5         6     NaN         7
6         7  change         8
7         8  change         9
8         9  change        10
9        10  change        11

左右滑动查看更多

# 对单数某一列下的某些行进行赋值

>>> df_new['Sep 2019'] = pd.Series([100,200], index=[1,2])
>>> df_new
   May 2018  Change  Sep 2019
0         1     NaN       NaN
1         2     NaN     100.0
2         3     NaN     200.0
3         4     NaN       NaN
4         5     NaN       NaN
5         6     NaN       NaN
6         7  change       NaN
7         8  change       NaN
8         9  change       NaN

左右滑动查看更多

官网：

http://pandas.pydata.org/pandas-docs/version/0.14.1/

原文链接：

https://blog.csdn.net/weixin_39778570/article/details/80550480

DataFrame和Series

# 导入需要的模块

>>> import pandas as pd
>>> import numpy as np
>>> from pandas import Series, DataFrame

左右滑动查看更多

# 首先创建一个字典

>>> data = {'Student':['XiaoMing','XiaoHong','XiaoWang'],'Grade':[100,90,20],'Class':['RG1','RG2','RG3']}

左右滑动查看更多

# 创建一个Series对象

>>> s1 = pd.Series(data['Student'])
>>> s1
0    XiaoMing
1    XiaoHong
2    XiaoWang
dtype: object

左右滑动查看更多

# 查看一下这个对象的属性

>>> s1.values
array(['XiaoMing', 'XiaoHong', 'XiaoWang'], dtype=object)

左右滑动查看更多

# 当没有给索引赋值的时候默认为

>>> s1.index
RangeIndex(start=0, stop=3, step=1)

左右滑动查看更多

# 创建并修改默认索引

>>> s1 = pd.Series(data['Student'], index = ['first','second','three'])
>>> s1
first     XiaoMing
second    XiaoHong
three     XiaoWang
dtype: object

左右滑动查看更多

# 给它起个列名‘GDUF’

>>> s1 = pd.Series(data['Student'], index = ['first','second','three'], name = 'GDUF')
>>> s1
first     XiaoMing
second    XiaoHong
three     XiaoWang
Name: GDUF, dtype: object
# 注意：索引和列名是可以修改的，方式如下:
>>> s1.name = 'haha'
>>> s1.index = ['first','second','three']

左右滑动查看更多

# 使用字典创建DataFrame对象

>>> df1 = pd.DataFrame(data)
>>> df1
    Student  Grade Class
0  XiaoMing    100   RG1
1  XiaoHong     90   RG2
2  XiaoWang     20   RG3

左右滑动查看更多

# 查看DataFrame的属性

# 某一列
>>> cou = df1['Student']
>>> cou
0    XiaoMing
1    XiaoHong
2    XiaoWang
Name: Student, dtype: object   # Series是有名字的

# 列的类型为Series
>>> type(cou)
<class 'pandas.core.series.Series'>
# 默认索引
>>> df1.index
RangeIndex(start=0, stop=3, step=1)

# 查看某一列某一行
>>> c1 = df1['Student'][0]
>>> c1
'XiaoMing'
>>> type(c1)
<class 'str'>

# 查看某一行的属性
>>> for row in df1.iterrows():
  print(row), print(type(row)),print(type(row[0])), print(type(row[1]))
  break

(0, Student    XiaoLi
Grade         100
Class         RG1
Name: 0, dtype: object) # 每一行默认的名字为0
<class 'tuple'>       # 每一行为一个tuple对象
<class 'int'>        # 默认索引为int类型
<class 'pandas.core.series.Series'>   # 值为Series类型
(None, None, None, None)

左右滑动查看更多

# 通过Series构建DataFrame对象

>>> data
{'Student': ['XiaoMing', 'XiaoHong', 'XiaoWang'], 'Grade': [100, 90, 20], 'Class': ['RG1', 'RG2', 'RG3']}
>>> s1 = pd.Series(data['Student'])
>>> s2 = pd.Series(data['Grade'])
>>> s3 = pd.Series(data['Class'])
>>> df_new = pd.DataFrame([s1,s2,s3]) # 当以列表的形式构建的时候会按行来放，有时候可以优先选择按行放 
>>> df_new
          0         1         2
0  XiaoMing  XiaoHong  XiaoWang
1       100        90        20
2       RG1       RG2       RG3
>>> df_new.T            # 进行行列转置
          0    1    2
0  XiaoMing  100  RG1
1  XiaoHong   90  RG2
2  XiaoWang   20  RG3

左右滑动查看更多

# 修改索引（通过字典直接构建，修改索引）

>>> df_new = DataFrame(data)
>>> df_new
    Student  Grade Class
0  XiaoMing    100   RG1
1  XiaoHong     90   RG2
2  XiaoWang     20   RG3
>>> df_new.index = ['first', 'second', 'thrid']
>>> df_new
         Student  Grade Class
first   XiaoMing    100   RG1
second  XiaoHong     90   RG2
thrid   XiaoWang     20   RG3

左右滑动查看更多

# 看一下这行代码

# 构建Series对象的时候，同时指定索引，名字

>>> s1 = pd.Series(data['Student'], name = 'Student', index = ['one','two','three'])

左右滑动查看更多

# 这样子DataFrame的结构就呼之欲出了，即列（colums），索引（index），值（values）

>>> df2 = pd.DataFrame(s1)
>>> df2
        Student
one    XiaoMing
two    XiaoHong
three  XiaoWang

>>> s1 = pd.Series(data['Student'], name = 'Student', index = ['one','two','three'])
>>> s2 = pd.Series(data['Grade'], name = 'Grade', index = ['one','two','three'])
>>> s3 = pd.Series(data['Class'], name = 'Class', index = ['one','two','three'])
>>> df2 = pd.DataFrame([s1,s2,s3])
>>> df2
>>> df2
              one       two     three
Student  XiaoMing  XiaoHong  XiaoWang
Grade         100        90        20
Class         RG1       RG2       RG3

# 用列表构建是按行放置的，需要装置一下

>>> df2.T
        Student Grade Class
one    XiaoMing   100   RG1
two    XiaoHong    90   RG2
three  XiaoWang    20   RG3

左右滑动查看更多

# 下面这种也是一种比较直观的构建方式。但是比较繁琐 （字典中的值为Series）

>>> d = {'one': Series([1., 2., 3.], index=['a', 'b', 'c']), 'two': Series([1., 2., 3., 4.], index=['a', 'b', 'c', 'd'])}
>>> df = DataFrame(d)
>>> df
   one  two
a  1.0  1.0
b  2.0  2.0
c  3.0  3.0
d  NaN  4.0

官网：

http://pandas.pydata.org/pandas-docs/version/0.14.1/

原文链接：

https://blog.csdn.net/weixin_39778570/article/details/80551032

·END·

点击搜索你感兴趣的内容吧

数据Seminar

这里是大数据、分析技术与学术研究的三叉路口

出处：CSDN作者：向前走别回头推荐：青酱排版编辑：青酱 


    欢迎扫描👇二维码添加关注