Convert certain column on a dataframe to a dict-CodePudding

i want to have DF below from 3 columns to 1 column and convert the remaining to a dict

Library TaskName    Number
Group 1 A   1
Group 2 B   2
Group 3 C   3
Group 4 D   4
Group 5 A   5
Group 6 B   6
Group 7 C   7


to this kind:
Library Dict
Group 1 {"TaskName":"A","Number":"1"}
Group 2 {"TaskName":"B","Number":"2"}
Group 3 {"TaskName":"C","Number":"3"}
Group 4 {"TaskName":"D","Number":"4"}
Group 5 {"TaskName":"A","Number":"5"}
Group 6 {"TaskName":"B","Number":"6"}
Group 7 {"TaskName":"C","Number":"7"}

CodePudding user response：

Let us do

df['dict_col'] = df[['TaskName','Number']].apply(lambda x : x.to_dict(),axis=1)
df
Out[462]: 
  Library TaskName  Number                        dict_col
0  Group1        A       1  {'TaskName': 'A', 'Number': 1}
1  Group2        B       2  {'TaskName': 'B', 'Number': 2}
2  Group3        C       3  {'TaskName': 'C', 'Number': 3}
3  Group4        D       4  {'TaskName': 'D', 'Number': 4}
4  Group5        A       5  {'TaskName': 'A', 'Number': 5}
5  Group6        B       6  {'TaskName': 'B', 'Number': 6}
6  Group7        C       7  {'TaskName': 'C', 'Number': 7}

CodePudding user response：

You can do:

import json
df['Dict'] = json.loads(df[['TaskName', 'Number']].to_json(orient='records'))
df = df.filter(['Library', 'Dict'])

output:

   Library                            Dict
0  Group 1  {'TaskName': 'A', 'Number': 1}
1  Group 2  {'TaskName': 'B', 'Number': 2}
2  Group 3  {'TaskName': 'C', 'Number': 3}
3  Group 4  {'TaskName': 'D', 'Number': 4}
4  Group 5  {'TaskName': 'A', 'Number': 5}
5  Group 6  {'TaskName': 'B', 'Number': 6}
6  Group 7  {'TaskName': 'C', 'Number': 7}

CodePudding user response：

You could select the columns you want, and then transform to dict in a record orient:

df['dict_col'] = df[['TaskName', 'Number']].to_dict('records')
df
 
   Library TaskName  Number                        dict_col
0  Group 1        A       1  {'TaskName': 'A', 'Number': 1}
1  Group 2        B       2  {'TaskName': 'B', 'Number': 2}
2  Group 3        C       3  {'TaskName': 'C', 'Number': 3}
3  Group 4        D       4  {'TaskName': 'D', 'Number': 4}
4  Group 5        A       5  {'TaskName': 'A', 'Number': 5}
5  Group 6        B       6  {'TaskName': 'B', 'Number': 6}
6  Group 7        C       7  {'TaskName': 'C', 'Number': 7}