使用字典熊猫填充 NaN 值

Question

I have a dataframe dfp with columns Brand_ID and Brand_Name (and some more columns like Product_ID , Product_Name etc.)

Some Brand names are NaN because of multiple brand_IDs separated by comma (see picture)

I want to fill those NaNs with the actual brand names separated by comma. I have a reference dictionary that I can use for this

Answer 1

For rows with missing values use lambda function for split values, match in dictionary and join:

df = pd.DataFrame({'Brand_ID':['11,12,15','10','15,11'],
                   'Brand_Name':[np.nan, 'aaa', np.nan]})


x = {'11':'ww', '12':'oup','15':'ret'}
m = df['Brand_Name'].isna()
f = lambda y: ','.join(x[z] for z in y.split(',') if z in x)
df.loc[m, 'Brand_Name'] = df.loc[m, 'Brand_ID'].apply(f)

print (df)
   Brand_ID  Brand_Name
0  11,12,15  ww,oup,ret
1        10         aaa
2     15,11      ret,ww