重复的日期时间索引熊猫

Question

I have a dataframe with multiple datetime index like

I would like to split data_value_index into three columns with the same index. So far I have only separated the dataframe and merge it like

df1= df[df['data_value_index']==5]
df2= df[df['data_value_index']==9]
df3 = df[df['data_value_index']==14]
df=pd.merge(df1,df2,left_index=True,right_index=True)

Is there any simpler way to do it?

Answer 1

似乎您需要新密钥的cumcount ，然后再cumcount unstack

df.set_index(df.groupby(df.index).cumcount(),append=True)['data_value_index'].unstack()