CODEWITHSUNDEEP
pythonpandascsvdictionary
Greedo
Greedo

Reputation: 5533

Panda array from list of dictionaries

I have a list of dictionaries like this

[{"Key":[val1,val2,...]}, {"Key2":[val1,val2,...]}, ...]

I would like to convert this to a csv format where the keys are column headers, and he values form the column values

To do this I had intended to use a pandas dataframe (which exports to csv readily), but I can't work out how to get my values into the data frame.

I've tried myPanda = pd.DataFrame.from_records(data) but alas, to no avail.

Upvotes: 1

Views: 240

Answers (1)

cs95
cs95

Reputation: 402423

I'm assuming your keys are unique, otherwise this wouldn't make much sense.

Option 1

  1. merge your dictionaries to a single dict
  2. pass the dict to pd.DataFrame
  3. save as CSV using df.to_csv
dct = [{...}, {...}, ...]

r = {}
for d in dct:
    r.update(d)

pd.DataFrame(r).to_csv('file.csv')

Option 2
Using cytoolz, there's another possibility for dictionary merging - 

from cytoolz.dicttoolz import merge
df = pd.DataFrame(merge(*dct))

Option 3
Create separate dataframes and concatenate them -

df = pd.concat(list(map(pd.DataFrame, dct)), axis=1)

This option works even if your lists are uneven.


Credit to piRSquared for the last two solutions!.

Upvotes: 4

Related Questions