pandas(三) Series和DataFrame

Series

简单介绍

1.Series对象是一种带有标签的一维数组对象
2.能够保存任何数据类型
3.Series对象包含数据和数据索引
4.数据部分是numpy的数组（ndarray）类型

创建series

创建Series对象，在之前的一篇博文里面提到过三种方式。

pd.Series(data=, index=, dtype= )

import pandas as pd
aa = pd.Series([1,2,34,4],index=["a","b","c","d"])
aa

Out:

访问series

在学习访问之前，我们需要知道标签和位置的区别。

上面的图片中，从上到下的a b c d就是标签，它相对应的0 1 2 3就是位置。

通过标签下标访问

aa["c"]

Out:

34

通过位置下标访问

aa[2]

Out:

34

通过标签切片访问

通过标签切片访问数据的时候，我们需要注意切片包括结束标签数据。

aa["c":"d"]

Out:

通过位置切片访问

aa[2:3]

Out:

通过布尔数组访问

注意：参数是列表

aa[[False,False,True,True]]

Out:

通过花式下标访问

使用标签数组（或列表）或者位置数组（或列表），作为下标，称为“花式下标”

aa[[0,2]]
aa[["a","c"]]

Out:

DataFrame

简单介绍

1.由多个Series结构构成二维表格对象
2.每一列可以是不同的数据类型
3.行和列是带有标签的轴
4.数据部分是numpy的数组（ndarray）类型

创建DataFrame对象

pandas.DataFrame(data, index, columns, dtype)

使用列表创建

L = [["Henan","2000","Nanyang"],
     ["Hubei","2020","Wuhan"],
     ["Hunan","2030","Changsha"],
     ["Guangzhou","2040","Zhuhai"],
     ["Sichuan","2050","Yaan"]]
frame = pd.DataFrame(L,
                     columns=["province","year","city"])
frame

Out:

使用字典创建

data = {"province":["Henan","Hubei","Hunan","Guangdong","Sichuan"],
        "year":[2000,2020,2030,2040,2050],
        "city":["Nanyang","Wuhan","Changsha","Zhuhai","Yaan"]
       }
frame = pd.DataFrame(data)
frame

Out:

使用列表嵌套字典创建

data = [{"province":"Henan","year":2000,"city":"Nanyang"},
        {"province":"Hubei","year":2020,"city":"Wuhan"},
        {"province":"Hunan","year":2030,"city":"Changsha"}]
frame = pd.DataFrame(data)
frame

Out:

使用字典嵌套Series创建

data = {"province":pd.Series(["Henan","Hubei","Hunan","Guangdong","Sichuan"]),
        "year":pd.Series([2000,2020,2030,2040,2050]),
        "city":pd.Series(["Nanyang","Wuhan","Changsha","Zhuhai","Yaan"])
       }
frame = pd.DataFrame(data)
frame

Out:

data = {"province":pd.Series(["Henan","Hubei","Hunan","Guangdong","Sichuan"]),
        "year":pd.Series([2000,2020,2030,2040,2050]),
        "city":pd.Series(["Nanyang","Wuhan","Changsha","Zhuhai","Yaan"])
       }
frame = pd.DataFrame(data, index=["first", "second", "third", "forth", "fifth"])
frame

Out:

在使用前面几种方式创建DataFrame的时候，我们可以通过添加index参数为数据框添加行标签。
但是在使用字典嵌套Series的时候会出现问题。
我们可以用下面的代码来解决：

data = {"province":pd.Series(["Henan","Hubei","Hunan","Guangdong","Sichuan"]),
        "year":pd.Series([2000,2020,2030,2040,2050]),
        "city":pd.Series(["Nanyang","Wuhan","Changsha","Zhuhai","Yaan"])
       }
frame = pd.DataFrame(data)
frame.index=["first", "second", "third", "forth", "fifth"]
frame

Out: