I have 3 columns col_1, col_2, col_3. The unique values of the columns are

col_1 => [1,2]

col_2 => [“yes”, “no”]

col_3 => [4, 5, 6]

I want to generate a pandas df that has all possible combinations as rows:

so, the number of rows are 2 x 2 x 4 = 12

Is there any way to generate such table using pandas?

CodePudding user response：

You can use pandas.MultiIndex.from_product:

col_1 = [1,2]
col_2 = ["yes", "no"]
col_3 = [4, 5, 6]

df = (pd.MultiIndex.from_product([col_1, col_2, col_3],
                                names=['col_1', 'col_2', 'col_3'])
        .to_frame().reset_index(drop=True)
     )

output:

    col_1 col_2  col_3
0       1   yes      4
1       1   yes      5
2       1   yes      6
3       1    no      4
4       1    no      5
5       1    no      6
6       2   yes      4
7       2   yes      5
8       2   yes      6
9       2    no      4
10      2    no      5
11      2    no      6

CodePudding user response：

from itertools import product
import pandas as pd

list1 = [1,2]
list2 =  ['yes', 'no']
list3 = [4, 5, 6]

df = pd.DataFrame(product(list1,list2, list3), columns =['Num1', 'Yes/No', 'Num2'])

print(df)


    Num1 Yes/No  Num2
0      1    yes     4
1      1    yes     5
2      1    yes     6
3      1     no     4
4      1     no     5
5      1     no     6
6      2    yes     4
7      2    yes     5
8      2    yes     6
9      2     no     4
10     2     no     5
11     2     no     6

CodePudding user response：

Here is your answer. Just copy and paste it.

import itertools
import pandas as pd
l1 = [1, 2]
l2 = ["yes", "no"]
l3 = [4, 5, 6]
all_list = [l1, l2, l3]
combs = list(itertools.product(*all_list))

final_df = []
for value in range(len(combs)):
    final_df.append(pd.DataFrame({"col_1": [combs[value][0]],
                                  "col_2": [combs[value][1]],
                                  "col_3": [combs[value][2]]}))
    
final_df = pd.concat(final_df, ignore_index=True)
final_df