pandas DataFrame常用总结

以pandas读取csv文件为例：

1. DataFrame数据取用

import pandas

df = pandas.read_csv('suites.csv',sep=',') 读取csv文件，以,分隔
print(df)
    Status  ...                            Description
0   passed  ...  \n            用例描述：期权强平市价委托\n        
1   passed  ...  \n            用例描述：期权强平市价委托\n        
2   passed  ...  \n            用例描述：期权强平市价委托\n        
3   passed  ...  \n            用例描述：期权强平市价委托\n        
4   passed  ...  \n            用例描述：期权强平市价委托\n        
5   passed  ...  \n            用例描述：期权强平市价委托\n        
6   passed  ...  \n            用例描述：期权强平市价委托\n        
7   passed  ...                                    NaN
8   passed  ...  \n            用例描述：期权强平市价委托\n        
9   passed  ...                                    NaN
10  passed  ...  \n            用例描述：期权强平市价委托\n        
11  passed  ...  \n            用例描述：期权强平市价委托\n        
12  passed  ...  \n            用例描述：期权强平市价委托\n        
13  passed  ...  \n            用例描述：期权强平市价委托\n        
14  passed  ...  \n            用例描述：期权强平市价委托\n        
15  passed  ...  \n            用例描述：期权强平市价委托\n        
16  passed  ...  \n            用例描述：期权强平市价委托\n        
17  passed  ...  \n            用例描述：期权强平市价委托\n        
18  passed  ...  \n            用例描述：期权强平市价委托\n        
19  passed  ...                                    NaN
20  passed  ...                                    NaN
21  passed  ...  \n            用例描述：期权强平市价委托\n        

[22 rows x 11 columns]
print(df.head())  #打印前五行
   Status  ...                            Description
0  passed  ...  \n            用例描述：期权强平市价委托\n        
1  passed  ...  \n            用例描述：期权强平市价委托\n        
2  passed  ...  \n            用例描述：期权强平市价委托\n        
3  passed  ...  \n            用例描述：期权强平市价委托\n        
4  passed  ...  \n            用例描述：期权强平市价委托\n        

[5 rows x 11 columns]
print(df[:3])   #打印前三行
print(len(df.index))  #获取文件行数
print(df.columns)   #获取文件表头
Index(['Status', 'Start Time', 'Stop Time', 'Duration in ms', 'Parent Suite',
       'Suite', 'Sub Suite', 'Test Class', 'Test Method', 'Name',
       'Description'],
      dtype='object')
print(df.loc[0,'Status'])  #获取Status列第0行
print(df['Stop Time'])   #打印Stop Time列

0     Mon Jul 01 20:22:29 CST 2019
1     Mon Jul 01 20:22:25 CST 2019
2     Mon Jul 01 20:22:25 CST 2019
3     Mon Jul 01 20:22:31 CST 2019
4     Mon Jul 01 20:22:25 CST 2019
5     Mon Jul 01 20:22:25 CST 2019
6     Mon Jul 01 20:22:31 CST 2019
7     Mon Jul 01 20:22:33 CST 2019
8     Mon Jul 01 20:22:31 CST 2019
9     Mon Jul 01 20:22:31 CST 2019
10    Mon Jul 01 20:22:25 CST 2019
11    Mon Jul 01 20:22:31 CST 2019
12    Mon Jul 01 20:22:25 CST 2019
13    Mon Jul 01 20:22:25 CST 2019
14    Mon Jul 01 20:22:25 CST 2019
15    Mon Jul 01 20:22:31 CST 2019
16    Mon Jul 01 20:22:31 CST 2019
17    Mon Jul 01 20:22:31 CST 2019
18    Mon Jul 01 20:22:25 CST 2019
19    Mon Jul 01 20:22:25 CST 2019
20    Mon Jul 01 20:22:25 CST 2019
21    Mon Jul 01 20:22:31 CST 2019
Name: Stop Time, dtype: object


df_compare = df.compare(df2, align_axis=1)   #对比两个dataframe

2. 数据取交集、差集

df_insertaction = df.merge(df2)  #取df和df2的交集

df_app = df.append(df_insertaction)
df_app.drop_dupicates(keep=False)   #取df与df_insertaction的差集

3. dataframe列

'A' in df.columns    #'A'是否属于df的列

4. 合并多个sheet内容

df = pd.ExcelFile('test.xlsx')
df.sheet_names  # 查看所有sheet名字
df.parse('sheet1')  #读取“sheet1"工作薄中的内容
df_concat= pd.concat([ pd.read_excel(df, sheet) for sheet in df.sheet_names])
#将所有sheet中数据合并到一个df中