Pandas 基于另一个 dataframe 将多个列和行值设置为 nan

Question

I have two large dataframes, df1 :

Col1    Col2    Val1    Val2    Val3
asd     ASYL    4.2             4.2
ppq     CONE            35      35
DA      HU      100     100

And df2 is

Col1    Col2    Val1    Val2  Val3
asd     ASYL    7       12    17
ppq     CONE    17      19    19
DA      HU      5       14    13

Both dataframes have same index columns Col1 and Col2 and same value columns Val1, Val2, Val3 .

I want to set the values in df2 to nan where it is Null in df1 to get the following:

Col1    Col2    Val1    Val2  Val3
asd     ASYL    7             17
ppq     CONE            19    19
DA      HU      5       14      

I tried the following:

idx = df1.isnull()
df2.loc[idx] = np.nan

But it is not working.

Answer 1

You can use df.where with df.notna

df2.where(df1.notna())

Col1    Col2    Val1   Val2  Val3
asd     ASYL    7      NaN    17
ppq     CONE   NaN      19    19
DA      HU      5       14    NaN   

Answer 2

Another way

df1.mask(df1.notna(),df2)

Alternatively

m=df1.notna()

pd.DataFrame(np.where(m, df2, np.nan))