Reputation: 143
I am comparing two data frames with master_df and create a new column based on a new condition if available.
for example I have master_df and two region df as asia_df and europe_df. I want to check if company of master_df is available in any of the region data frames and create a new column as region as Europe and Asia
master_df
company product
ABC Apple
BCA Mango
DCA Apple
ERT Mango
NFT Oranges
europe_df
account sales
ABC 12
BCA 13
DCA 12
asia_df
account sales
DCA 15
ERT 34
My final output dataframe is expected to be
company product region
ABC Apple Europe
BCA Mango Europe
DCA Apple Europe
DCA Apple Asia
ERT Mango Asia
NFT Oranges Others
When I try to merge and compare, some datas are removed. I need help on how to fix this issues
final_df = europe_df.merge(master_df, left_on='company', right_on='account', how='left').drop_duplicates()
final1_df = asia_df.merge(master_df, left_on='company', right_on='account', how='left').drop_duplicates()
final['region'] = np.where(final_df['account'] == final_df['company'] ,'Europe','Others')
final['region'] = np.where(final1_df['account'] == final1_df['company'] ,'Asia','Others')
Upvotes: 1
Views: 178
Reputation: 71689
First using pd.concat
concat the dataframes asia_df
and europe_df
then use DataFrame.merge
to merge them with master_df
, finally use Series.fillna
to fill NaN
values in Region
with Others
:
r = pd.concat([europe_df.assign(Region='Europe'), asia_df.assign(Region='Asia')])\
.rename(columns={'account': 'company'})[['company', 'Region']]
df = master_df.merge(r, on='company', how='left')
df['Region'] = df['Region'].fillna('Others')
Result:
print(df)
company product Region
0 ABC Apple Europe
1 BCA Mango Europe
2 DCA Apple Europe
3 DCA Apple Asia
4 ERT Mango Asia
5 NFT Oranges Others
Upvotes: 1