Convert Pandas Series to Dictionary Without Index

Question

I need to convert Pandas Series to a Dictionary, without Index (like pandas.DataFrame.to_dict('r')) - code is below:

grouped_df = df.groupby(index_column)
for key, val in tqdm(grouped):
    json_dict[key] = val.apply(lambda x: x.to_dict(), axis=1).to_dict()

Currently, I get output like so:

{
   "15717":{
      "col1":1.61,
      "col2":1.53,
      "col3":1.0
   },
   "15718":{
      "col1":10.97,
      "col2":5.79,
      "col3":2.0
   },
   "15719":{
      "col1":15.38,
      "col2":12.81,
      "col3":1.0
   }
}

but i need output like:

[
   {
      "col1":1.61,
      "col2":1.53,
      "col3":1.0
   },
   {
      "col1":10.97,
      "col2":5.79,
      "col3":2.0
   },
   {
      "col1":15.38,
      "col2":12.81,
      "col3":1.0
   }
]

Thanks for your help!

Edit: Here is the original dataframe:

        col1   col2 col3
2751    5.46    1.0 1.11
2752    16.47   0.0 6.54
2753    26.51   0.0 18.25
2754    31.04   1.0 28.95
2755    36.45   0.0 32.91

Could you please add the original dataframe, so one could reproduce your scenario? — Roy2012, Jun 23 '20 at 15:28

score 2 · Accepted Answer · answered Jun 23 '20 at 15:32

Two ways of doing that:

[v for _, v in df.to_dict(orient="index").items()]

Another one:

df.to_dict(orient="records")

The output, either way, is:

[{'col1': 1.61, 'col2': 1.53, 'col3': 1.0},
 {'col1': 10.97, 'col2': 5.79, 'col3': 2.0},
 {'col1': 15.38, 'col2': 12.81, 'col3': 1.0}]

score 1 · Answer 2 · answered Jun 23 '20 at 15:29

1

You can try:

df.T.to_dict('r')

Output:

[{'col1': 1.61, 'col2': 1.53, 'col3': 1.0},
 {'col1': 10.97, 'col2': 5.79, 'col3': 2.0},
 {'col1': 15.38, 'col2': 12.81, 'col3': 1.0}]

answered Jun 23 '20 at 15:29

Quang Hoang

146,074
10
56
74

Convert Pandas Series to Dictionary Without Index

2 Answers2

Linked