I have 2 dataframes as such:
dfA=pd.DataFrame([[1,2,3],[4,5,6]])
dfB=pd.DataFrame([[True,False,False],[False,True,False]])
How do i create a new dataframe where based on the same index,
2 ** (dfA+1) only if dfB is true
2 ** (dfA) only if dfB is false.
What I like to achieve is a dataframe as such:
df_output = pd.DataFrame([[4,4,8],[16,64,64]])
Thanks!
You can use DataFrame.mask:
(2 ** dfA).mask(dfB, 2 ** (dfA+1))
Out:
0 1 2
0 4 4 8
1 16 64 64
This will check dfB, if True, take the values from the original DataFrame ((2 ** dfA)) and if false, take the values from the other (2 ** (dfA+1)).
This is similar to numpy's where function:
np.where(dfB, 2 ** (dfA+1), 2 ** (dfA))
Out:
array([[ 4, 4, 8],
[16, 64, 64]])
However, it loses the index and columns so you may need to wrap the result in a DataFrame constructor:
pd.DataFrame(np.where(dfB, 2 ** (dfA+1), 2 ** (dfA)), dfA.index, dfA.columns)
Out:
0 1 2
0 4 4 8
1 16 64 64