Convert JSON keys to set in different columns with preset headers in pandas - json

Question

Want to implement dataframe expansion from JSON data with preset values. Collating all the key values within JSON data to be set as column headers
id |Name | MonthWise
0 |ABC |{'102022':{'val':100, 'count':1}}
1 |XYZ |{'102022':{'val':20,'count':5},'092022':{'val':20,'count':2}}
2 |DEF |{}
3 |PQR |{'082022':{'val':50,'count':3}}
Here df containing MonthWise column which has JSON objects, which needs to transposed into 12 different columns like 'MMYYYY' (Yearly data)
Something like:
id |Name |042022.val | 042022.count |....|102022.val | 102022.count|....| 032023.val| 032023.count
0 |ABC |nan|nan|....|100|1|....|nan|nan
1 |XYZ |nan|nan|....|20|5|....|nan|nan
2 |DEF |nan|nan|....|nan|nan|....|nan|nan
3 |PQR |nan|nan|....|nan|nan|....|nan|nan
I have tried with df['MonthWise'].apply(pd.json_normalize(x, max_level=1)) but no success.

Answer 1

There is no need for apply in this case. You may use pd.json_normalize as follows
import pandas as pd
# sample data
df = pd.DataFrame({
'id': [0, 1, 2, 3],
'Name': ['ABC', 'XYZ', 'DEF', 'PQR'],
'MonthWise': [{'102022':{'val':100, 'count':1}}, {'102022':{'val':20,'count':5},'092022':{'val':20,'count':2}}, {}, {'082022':{'val':50,'count':3}}]
})
# normalize
result = pd.concat([df[['id','Name']], pd.json_normalize(df['MonthWise'])], axis=1)
This returns
id Name 102022.val 102022.count 092022.val 092022.count 082022.val 082022.count
0 0 ABC 100.0 1.0 NaN NaN NaN NaN
1 1 XYZ 20.0 5.0 20.0 2.0 NaN NaN
2 2 DEF NaN NaN NaN NaN NaN NaN
3 3 PQR NaN NaN NaN NaN 50.0 3.0
(I believe that the expected result in your original post is inconsistent with the input dataframe)