Putting dictionary values into data frame new column in python pandas

Question

I have a python dictionary with key and values and I wish to create a new pandas data frame object with a new column constructed from dictionary values only. Whats is the most optimised way to achieve this?

Can you add sample? Maybe works simple df = pd.DataFrame(d) — jezrael
– jezrael, Commented Dec 13, 2016 at 18:44
For example, I have a dictionary as name (key) and score(value) [tom:100, jerry:150, micky:400, donald:250] Now all I want is a data frame with column name as "scores" having all scores(values) from this dictionary. Say like df['scores'] = ... — Vijayendra
– Vijayendra, Commented Dec 13, 2016 at 18:54
No, put the example in the original question not in a comment. — juanpa.arrivillaga
– juanpa.arrivillaga, Commented Dec 13, 2016 at 19:07

jezrael · Accepted Answer · 2016-12-13 19:24:13Z

4

I think you need DataFrame.from_dict:

d = {'tom':100, 'jerry':150, 'micky':400, 'donald':250}

df = pd.DataFrame.from_dict(d, orient='index')
df.columns = ['scores']
print (df)
        scores
micky      400
donald     250
jerry      150
tom        100

Another solution is add dict with new column name to DataFrame constructor:

df = pd.DataFrame({'scores':d})

print (df)
        scores
donald     250
jerry      150
micky      400
tom        100

Last if need remove names from index add reset_index:

print (df.reset_index(drop=True))
   scores
0     400
1     250
2     150
3     100

Another solution is use only values of dict, but python dictionaries are always unordered, so get random order of values in df:

print (d.values())
dict_values([400, 250, 150, 100])

#python 3 add list
df = pd.DataFrame({'scores': list(d.values())})
print (df)
   scores
0     400
1     250
2     150
3     100

edited Dec 13, 2016 at 19:24

answered Dec 13, 2016 at 18:57

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Vijayendra Over a year ago

Thanks @jezrael for your quick reply. I wish to achieve outcome as shown by your last step - without names. I am looking for an optimised way to achieve this as my example was a small dataset. What if we have a huge dataset and yet we first load whole dictionary into data frame and then drop names. Can we directly load values into data frame?

jezrael Over a year ago

I add last solution, I believe it can be fastest.

Collectives™ on Stack Overflow

Putting dictionary values into data frame new column in python pandas

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related