The pandas to_json function for some reason is converting the index of the dataframe into strings. Is there a way to avoid this?
>>> import pandas as pd
>>> df = pd.DataFrame({"a" : [1,2,3], "b" : [2,3,4]})
>>> df.to_json(orient = "index")
'{"0":{"a":1,"b":2},"1":{"a":2,"b":3},"2":{"a":3,"b":4}}'
>>> import json
>>> json.loads(df.to_json(orient="index"))
{'2': {'a': 3, 'b': 4}, '1': {'a': 2, 'b': 3}, '0': {'a': 1, 'b': 2}}
>>>
One way is to use orient='records' instead.
df.to_json(orient='records')
Then
json.loads(df.to_json(orient="records"))
will give output as
[{'a': 1, 'b': 2}, {'a': 2, 'b': 3}, {'a': 3, 'b': 4}]
Keys in json cannot be int. Better it is explained here.
There is one possible solution - use parameter split in to_json if need store index values as int:
df = pd.DataFrame({"a" : [1,2,3], "b" : [2,3,4]}, index=[100,200,300])
print (df)
a b
100 1 2
200 2 3
300 3 4
a = df.to_json(orient = "split")
print (a)
{"columns":["a","b"],"index":[100,200,300],"data":[[1,2],[2,3],[3,4]]}
print (pd.read_json(a, orient='split'))
a b
100 1 2
200 2 3
300 3 4
ValueError: Length of values (10) does not match length of index (2)