如何使用python pandas进行以下分析

Question

我有一些村庄的产量数据，样本数据请参见附件。enter image description here

Village Yield(in Kg)    Date
Village1    0.22    01/06/18
Village1    0.23    02/06/18
Village1    0.55    01/06/18
Village1    0.2     02/06/18
Village2    0.88    31/05/18
Village2    0.89    30/05/18
Village2    0.63    30/05/18
Village2    0.55    30/05/18

现在，我想说明，与实验date.so，village1对应的产量数据有4个产量值。

请参阅附件。enter image description here

Village     Yield-1                Yield-2              Yield-3 
Village1    0.22    01/06/18        0.23    02/06/18    0.55    01/06/18    

任何帮助都是有帮助的。谢谢

Answer 1

尝试使用groupby，然后获取其值，然后将group转换为字典，然后在字典之外创建一个数据框，然后转置它，然后使用mean为平均值创建一个新列

import pandas as pd
df = pd.DataFrame({'Village': ['Village1', 'Village1',
                               'Village1', 'Village1', 'Village2',
                               'Village2', 'Village2', 'Village2'],
                   'Yield (in kg)': [0.22,0.23,0.55,0.2, 0.88, 0.89, 0.63, 0.55]})
group = df.groupby('Village')['Yield (in kg)'].apply(lambda x: x.values)
df = pd.DataFrame(group.to_dict()).T
df.columns = df.columns.astype(str)
df['Average'] = df.mean(axis=1)
print(df)

输出：

          0     1     2     3      Average
Village1  0.22  0.23  0.55  0.20   0.3000
Village2  0.88  0.89  0.63  0.55   0.7375

要重命名列，请执行以下操作：

df.columns = ['Yield (in kg)-'+i for i in df.columns if i != 'Average']

输出：

          Yield (in kg)-0  Yield (in kg)-1  Yield (in kg)-2  Yield (in kg)-3   /
Village1   0.22             0.23             0.55             0.20   
Village2   0.88             0.89             0.63             0.55   

                         Average  
Village1                 0.3000  
Village2                 0.7375

Answer 2

尝尝这个,

df.groupby(['Village']).apply(lambda x:pd.Series(zip(x['Yield(in Kg)'],x['Date']))).reset_index()


    Village                 0                 1                 2  \
0  Village1  (0.22, 01/06/18)  (0.23, 02/06/18)  (0.55, 01/06/18)   
1  Village2  (0.88, 31/05/18)  (0.89, 30/05/18)  (0.63, 30/05/18)   

                  3  
0   (0.2, 02/06/18)  
1  (0.55, 30/05/18)

要重命名列，请执行以下操作：

col1=df.filter(regex='\d+').columns.values
col2=['Yield - '+str(col+1) for col in col1]
df.rename(columns= dict(zip(col1,col2)),inplace=True)

    Village         Yield - 1         Yield - 2         Yield - 3  \
0  Village1  (0.22, 01/06/18)  (0.23, 02/06/18)  (0.55, 01/06/18)   
1  Village2  (0.88, 31/05/18)  (0.89, 30/05/18)  (0.63, 30/05/18)   

          Yield - 4  
0   (0.2, 02/06/18)  
1  (0.55, 30/05/18)