CODEWITHSUNDEEP
pythonpandasdictionary
cookie1986
cookie1986

Reputation: 895

convert pandas DF into a formatted dictionary

Say I have the following dataframe:

myDF = pd.DataFrame([['A','red'],['B','green'],
       ['C','orange'],['D','purple']], columns = ['letter','color'])

How can I reformat this into a nested dictionary, such that the key of the outer dictionary is taken from the letter column, the key of the inner dictionary is taken from the color column-name, and that value of the inner dictionary is taken from the color column...

    key     |     value
-----------------------------
     A        {color: red}

     B        {color: green}

     C        {color: orange}

     D        {color: purple}

I have tried various takes on the df.to_dict without success. I have managed to get it in the right format if I export as csv, and then import using DictReader, but this is obviously long-winded.

myFile = csv.DictReader(open('file.csv'))
output = {}
for row in myFile:
    key = row.pop('letter')
    if key in output:
        pass
    output[key] = row

Upvotes: 0

Views: 55

Answers (5)

iampritamraj
iampritamraj

Reputation: 316

You should try this


import pandas as pd
 
myDF = pd.DataFrame([['A','red'],['B','green'],
       ['C','orange'],['D','purple']], columns = ['letter','color'])

mydict = {}
for i in range(0,len(myDF)):
    mydict[myDF['letter'][i]] = myDF['color'][i]
print(mydict)

Answer:-{'A': 'red', 'B': 'green', 'C': 'orange', 'D': 'purple'}

Upvotes: 0

shikbupt
shikbupt

Reputation: 36

Try this:

myDF.set_index('letter').to_dict('index')

Output:

{'A': {'color': 'red'}, 'B': {'color': 'green'}, 'C': {'color': 'orange'}, 'D': {'color': 'purple'}}

Upvotes: 2

Adrian Stępniak
Adrian Stępniak

Reputation: 150

That will do the job:

output_dict  = dict()
for (letter, color) in df.values:
    output_dict[letter] = {'color': color}

Upvotes: 0

Tom Ron
Tom Ron

Reputation: 6181

You can achieve it by setting letter as index and setting the orient property -

myDF.set_index("letter").to_dict(orient="index")

Upvotes: 1

Vishnudev Krishnadas
Vishnudev Krishnadas

Reputation: 10960

Use dict comprehension

{row.letter: {'color': row.color} for row in myDF.itertuples()}

Output

{'A': {'color': 'red'}, 'B': {'color': 'green'}, 'C': {'color': 'orange'}, 'D': {'color': 'purple'}}

Upvotes: 1

Related Questions