Convert list of strings to python dictinoary-CodePudding

I have a datastructure like this:

lst = ['name, age, sex, height, weight',
'underweight,overweight,normal',
'David, 22, M, 185, -,-,78',
'Lily, 18, F, 165,-,75,-',
..............................]

The weight is categorized as three more columns (the second row in the list). How can I write it to a pandas dataframe.

What I have done is writing the list as dataframe using :

pd.DataFrame(lst)

But this is not the whole solution, it has more complicated logic.

Help me out please

CodePudding user response：

The output you expect is not fully clear, but you can preprocess your data with a list comprehension:

lst2 = [list(map(str.strip, e.split(','))) for e in lst] # split on commas
pd.DataFrame(lst2[2:], columns=lst2[0][:-1] lst2[1])     # use first 2 item to build header
                                                         # rest is data

output:

    name age sex height underweight overweight normal
0  David  22   M    185           -          -     78
1   Lily  18   F    165           -         75      -

MultiIndex

Although feasible, I don't recommend this, it be be much harder to work with:

lst2 = [list(map(str.strip, e.split(','))) for e in lst]
cols = pd.MultiIndex.from_arrays([lst2[0][:-1] [lst2[0][-1]]*3,
                                  ['']*4 lst2[1]])
pd.DataFrame(lst2[2:], columns=cols)

output:

    name age sex height      weight                  
                        underweight overweight normal
0  David  22   M    185           -          -     78
1   Lily  18   F    165           -         75      -