python pandas dataframe在分组时导致两列

Question

I have a dataframe df :

PID    AID    Ethnicity
1      A      Asian
1      B      Asian
1      C      Arab
1      D      African
2      A      Asian
2      D      African
2      E      Caucasian 
2      F      African
2      B      Asian

I want to generate a frame that tells me for each PID how many AIDs it has, and how many Ethnic groups:

So for the above the resulting newdf would be:

PID    numAID    numEthnicities
1      4         3
2      5         3

I know how to find numAID:

newdf = df[['PID','AID']].groupby('PID',  
as_index=False).count().rename(columns={'AID':'numAID'})

I'm not sure how to add the third column to the dataframe .

Answer 1

This will work:

df.groupby('PID').agg({'AID':'count','Ethnicity':pd.Series.nunique}).add_prefix('num')

     numAID  numEthnicity
PID                
1      4          3
2      5          3

Answer 2

您可以添加第三列，如下所示：

newdf['numEthnicities'] = df[['PID, 'Ethnicity']].groupby('PID', as_index=False).count()

Answer 3

since you have found out newdf , you could try to use join function.)

df = df.set_index('PID')
newdf = newdf.set_index('PID')
result = df.join(newdf, lsuffix='df', rsuffix='newdf')