Adding row to pandas dataframe where column contains additional value that should be another row?

user_4107349 0
Viewed 535

Simplified df where some rows contain additional entries in another flavor that should be another row:

   values more values   flavor another flavor
0       6         foo  caramel      chocolate
1       4         baz  vanilla            NaN

df = pd.DataFrame({"values": [6, 4],"more values": ["foo",  "baz"],"flavor": ["caramel", "vanilla"],"another flavor": ["chocolate",  np.nan],})

We need to add another row containing values from other columns, populating flavor with values from another flavor. We'd then drop another flavor to get desired_df:

   values more values     flavor
0       6         foo    caramel
1       6         foo  chocolate
2       4         baz    vanilla

desired_df = pd.DataFrame({"values": [6, 6, 4],"more values": ["foo", "foo", "baz"],"flavor": ["caramel", "chocolate",  "vanilla"],})

What's a practical way to do this? Is there an expression for this I could search for as a keyword?

pandas dataframe python
1 Answers

Please use the loc accessor to slice suitable columns and then append. You can sort values if needed.

df.loc[:,:'flavor'].append(df.loc[:, df.columns != 'flavor'].rename(columns={'another flavor':'flavor'}),ignore_index=True).dropna().sort_index().
user_89869750
Related