Pandas pivoting string columns impossible?

Question

I am looking to pivot the feature column into column headers.

here is the df

for instance neither of the below work. I've tried multiple variations to no avail.

df.groupby(['feature', 'year'])['value'].unstack(fill_value=0)

df.pivot_table(index='year', columns='feature', values='value')

The end goal would look like this.

kindly post data, not images/pics. use this as a guide on how to post minimum viable reproducible example : [link](https://stackoverflow.com/questions/20109391/how-to-make-good-reproducible-pandas-examples) — sammywemmy, Mar 29 '20 at 02:03
can you provide your df as code as opposed to an excel screengrab? — Umar.H, Mar 29 '20 at 02:04
Use `pivot` instead of `pivot_table`. `pivot` is a simple reshaping of the data, which is what you want here. `pivot_table` is more when you want to aggregate (or reshape with MultiIndices). You'd need to use `aggfunc='first'` with `pivot_table` to deal with strings as the default assumption is you aggregate with `mean` with `pivot_table`. — ALollz, Mar 29 '20 at 02:18

score 2 · Answer 1 · answered Mar 29 '21 at 04:14

You are going from a long layout to wide layout. See the sample below.

import pandas as pd

df_long = pd.DataFrame({
        "student":
            ["Andy", "Bernie", "Cindy", "Deb",
             "Andy", "Bernie", "Cindy", "Deb",
             "Andy", "Bernie", "Cindy", "Deb"],
        "school":
            ["Z", "Y", "Z", "Y",
             "Z", "Y", "Z", "Y",
             "Z", "Y", "Z", "Y"],
        "class":
            ["english", "english", "english", "english",
             "math", "math", "math", "math",
             "physics", "physics", "physics", "physics"],
        "grade":
            [10, 100, 1000, 10000,
             20, 200, 2000, 20000,
             30, 300, 3000, 30000]
})
df_long

df_long.pivot_table(index=["student", "school"], 
                    columns='class', 
                    values='grade').reset_index()

score 1 · Answer 2 · answered Mar 29 '20 at 04:11

1

@ALollz tipped me off on this one. I should have been using pivot and not pivot_table.

df = pd.pivot(df, values='value', columns='feature', index='year')

answered Mar 29 '20 at 04:11

Joe Rivera

307
2
11

Pandas pivoting string columns impossible?

2 Answers2