[FIXED] Iterate over rows in pandas DataFrame and create a dict

Issue

I am trying to create a dictionary using pandas DataFrame.

import pandas as pd

df = pd.DataFrame({'name': ["p1","p1","p2","p2","p2"],
                   'val': [0, 1, np.nan, 1, 0] })

I'd like to create a dictionary with name column as key and val column as values. Values will be a list. I am using a for loop as the DataFrame is only 100 rows.

mydict = dict()
valList = []

for index, row in df.iterrows():

    # create list of values
    ......

    mydict[row['name']] = valList

Expected output:

{
 'p1': [0, 1],
 'p2': [nan, 1, 0]
}

Open to other approaches, but I have a slight preference for using for loop and iterrows().

Solution

You could use groupby and a dictionary comprehension:

d = {k:list(v) for k,v in df.groupby('name')['val']}

output:

{'p1': [0.0, 1.0], 'p2': [nan, 1.0, 0.0]}

using iterrows (not my favorite option)

NB. this will be quite slower on large dataframes

from collections import defaultdict

d = defaultdict(list)

for _, row in df.iterrows():
    d[row['name']].append(row['val'])
    
dict(d)

Answered By - mozway