I splited a dataframe into two parts and changed their column names seperately. Here's what I got:
df1 = df[df['colname'==0]] df2 = df[df['colname'==1]] df1.columns = [ 'a'+ x for x in df1.columns] df2.columns = [ 'b'+ x for x in df2.columns] And it turned out df2 has the columns start with 'ba' rather than 'b'. What happened?
I cannot simulate your problem, for me working nice.
Alternative solution should be add_prefix instead list comprehension:
df = pd.DataFrame({'colname':[0,1,0,0,0,1], 'C':[7,8,9,4,2,3], 'D':[1,3,5,7,1,0], 'E':[5,3,6,9,2,4], 'F':list('aaabbb')}) print (df) C D E F colname 0 7 1 5 a 0 1 8 3 3 a 1 2 9 5 6 a 0 3 4 7 9 b 0 4 2 1 2 b 0 5 3 0 4 b 1 df1 = df[df['colname']==0].add_prefix('a') df2 = df[df['colname']==1].add_prefix('b') print (df1) aC aD aE aF acolname 0 7 1 5 a 0 2 9 5 6 a 0 3 4 7 9 b 0 4 2 1 2 b 0 print (df2) bC bD bE bF bcolname 1 8 3 3 a 1 5 3 0 4 b 1 
df1 = df[df['colname']==0]anddf2 = df[df['colname']==1]? For me your solution working nice.