熊猫DataFrame到JSON格式文件

Question

I am trying to generate a text file with the following format:

{"0": ["n01440764", "tench"], "1": ["n01443537", "goldfish"], "2": ["n01484850", "great_white_shark"]}

I have input data in a DataFrame:

data = [('n02124075', 'egyptian_cat'),
 ('n04067472', 'reel'),
 ('n04540053', 'volleyball')]
df = pd.DataFrame(data)
df.head(n=2)

    0   1
0   n02124075   egyptian_cat
1   n04067472   reel

I have tried:

df.to_json(PATH/'my_file.json', orient='index')

json_f = json.dumps(df.to_dict(orient='list'))
with open(PATH/'my_file.json', 'w') as outfile:
    json.dump(json_tiny, outfile)

and various variations of these but cant work out how to generate output file with correct format. Ant ideas?

Answer 1

You'll need a specific orient called "list" that to_json does not offer, but to_dict does.

import json

with open('my_file.json', 'w') as f:
    json.dump(df.T.to_dict(orient='list'), f)

my_file.json

{
    "0": ["n02124075", "egyptian_cat"],
    "1": ["n04067472", "reel"],
    "2": ["n04540053", "volleyball"]
}