Search code examples
pythonpandas

ValueError: No axis named "type" for object type <class 'pandas.core.frame.DataFrame

This is my data frame df received after some preprocessing:

    name    type    range   time                number
0   XXXX    11111   1-120   [1-120, 261.05]     [1-120, 23229]
1   XXXX    11111   121-240 [121-240, 277.79]   [121-240, 6382]
2   XXXX    11111   241-360 [241-360, 269.64]   [241-360, 2769]
3   XXXX    11111   361-480 [361-480, 279.09]   [361-480, 1331]

I want to transform it into the following json string:

[{"name":"XXXX","type":11111,"time":[["1-120",261.05],["121-240",277.79],["241-360",269.64],["361-480",279.09]],"number":[["1-120",23229],["121-240",6382],["241-360",2769],["361-480",1331]]

If I do it this way, I get an error ValueError: No axis named "type" for object type <class 'pandas.core.frame.DataFrame'>:

jsonresult = df.groupby('name','type')['time','number'].agg(lambda x : x.tolist()).reset_index().to_json(orient='records')
jsonresult

It only works if I do df.groupby('name') or df.groupby('type').

Solution

  • You can use convert to tuple instead list:

    jsonresult = df.groupby(['name','type'])['time','number']
               .agg(lambda x: tuple(x)).reset_index().to_json(orient='records')

print (jsonresult)
[{"name":"XXXX",
  "type":11111,
  "time":[[-119,261.05],[-119,277.79],[-119,269.64],[-119,279.09]],
  "number":[[-119,23229],[-119,6382],[-119,2769],[-119,1331]]}]