sort values and create new column based on result-CodePudding

I have this example

import pandas as pd
import numpy as np

rng = np.random.default_rng()
df = pd.DataFrame(rng.integers(0, 5, size=(5, 3)), columns=list("ABC"))
print(df)

which produce this dataframe

i want to sort values from column A and B and C from higher to lower and put result column headers in new column D like this:

   A  B  C  D
0  4  4  7  C, A, B
1  1  9  6  B, C, A
2  2  9  1  B, A, C
3  0  5  8  C, B, A
4  3  5  5  B, C, A

I hope it's clear, thank you

CodePudding user response：

You can try:

df['D'] = df.apply(lambda x: ', '.join(x.sort_values(ascending=False).index), axis=1)


   A  B  C        D
0  4  4  7  C, B, A
1  1  9  6  B, C, A
2  2  9  1  B, A, C
3  0  5  8  C, B, A
4  3  5  5  C, B, A

CodePudding user response：

Another possible solution:

df['D'] = df.apply(lambda x: df.columns[np.argsort(-x)].str.cat(sep = ', '), axis=1)

Output:

    A   B   C   D
0   3   1   0   A, B, C
1   3   4   3   B, A, C
2   4   4   1   A, B, C
3   3   2   0   A, B, C
4   0   4   3   B, C, A