pandas long to wide: melt + transposing

Asked Mar 13 '20 at 14:50

Active Mar 13 '20 at 14:50

Viewed 9 times

I have this dataframe:

np.random.seed(0)
df = pd.DataFrame({"Status": np.random.choice(["Married", "Single", "Single", "Single", "Widow"], 10000),
               "Genre": np.random.choice(["Masculine", "Masculine", "Masculine", "Femenine"], 10000),
               "Count": 1}).groupby(["Status", "Genre"]).agg("sum").reset_index()
df


    Status  Genre       Count
0   Married Femenine    485
1   Married Masculine   1520
2   Single  Femenine    1493
3   Single  Masculine   4552
4   Widow   Femenine    520
5   Widow   Masculine   1430

I'm looking to change it where the columns are the Genre values, and an index with the Status values, like:

            Single  Married  Widow  
Masculine   4552    1520     1430
Femenine    1493    485      520

I have tried with a mix of melt() and .T, but still isn't working. I'm looking for a general approach, as I have 5 Genre values and 20 Status values.

asked Mar 13 '20 at 14:50

Chris

2,019
5
22
67

`df.pivot('Genre', 'Status', 'Count').rename_axis(columns=None, index=None)` – ansev Mar 13 '20 at 14:52
or `df.set_index(['Status','Genre']).unstack('Status')` – ansev Mar 13 '20 at 14:54

pandas long to wide: melt + transposing

0 Answers0