How to transform table with JSON documents columns to several panda data frames [closed]

Question

. 。 This question needs details or clarity . It is not currently accepting answers.

---

Add details and clarify the problem by editing this post .

Closed 7 months ago .

Improve this question

I have a dataframe in Python pandas with rows of NoSQL documents and json strings column and I want to transform it into flat table with specific keys as column name and values as data in rows. Following is the example table I begin with:

	diagnoses
1	a
2	b

Where 'a' & 'b' looks something like:

[{"icd10":"I611","icd10Name":"Intracerebral haemorrhage in hemisphere, cortical","diagType":"1","diagTypeName":"Principal Diagnosis"},{"icd10":"I10","icd10Name":"Essential (primary) hypertension","diagType":"2","diagTypeName":"Comorbidity (โรคอื่นที่เป็นร่วมด้วย)"},{"icd10":"E789","icd10Name":"Disorder of lipoprotein metabolism,unspecified","diagType":"2","diagTypeName":"Comorbidity (โรคอื่นที่เป็นร่วมด้วย)"}]

I want to extract this table to another table like:

	diagnoses	Principal Diagnoses	Comorbidity
1	a	I611	I10, E789
2	b	...	...

Please feel free to refer to the image here:

Thank you so much in advance. Also, please note that "Principal Diagnoses" doesn't always come first.

Answer 1

try working this into a lambda function. loop through dictionaries to separate your values. Assumes principal diagnosis is always single value, but can have 1 or more secondary.

data_dict = [{"icd10":"I611","icd10Name":"Intracerebral haemorrhage in hemisphere, cortical","diagType":"1","diagTypeName":"Principal Diagnosis"},{"icd10":"I10","icd10Name":"Essential (primary) hypertension","diagType":"2","diagTypeName":"Comorbidity (โรคอื่นที่เป็นร่วมด้วย)"},{"icd10":"E789","icd10Name":"Disorder of lipoprotein metabolism,unspecified","diagType":"2","diagTypeName":"Comorbidity (โรคอื่นที่เป็นร่วมด้วย)"}]


co_diag = []
for d in data_dict:
    if 'Principal Diagnosis' in d.values():
        pd = d['icd10']
    else:
        co_diag.append(d['icd10'])
        co_diag_str = ', '.join(co_diag)

within the steps you should be able to write columns you need.

Output if the loop:

In [19]: co_diag_str
Out[19]: 'I10, E789'

In [20]: pd
Out[20]: 'I611'