Create pandas dataframe from a dictionary of series

Question

I have a dictionary of 68 keys whereby each key has a list with 50 values in it. For example, my dict is the following, whereby each series has 50 values in it, e.g. value1, value2....

key1 : Series1
key2 : Series2
 .   :  .
key50:  Series50

I now want to make the following dataframe out of the dictionary:

key1          key2
value1      value1
 .            .
 .            .
value50     value 50

I looked at other threads and tried the following command:

df= pd.DataFrame([dict])

However, this yields:

key1          key2
Series1       Series2

How doI get the values in the dataframe instead of the Series. In the end, I should get dataframe sized 50*68.

score 6 · Accepted Answer · answered Sep 17 '17 at 11:53

6

Just pass dict_ directly:

df = pd.DataFrame(dict_)

Also, don't use dict as a variable name, it's bad form, and it shadows the builtin class with the same name.

answered Sep 17 '17 at 11:53

cs95

379,657
97
704
746

Sometimes I think too much. Unpacking a single element list is nothing but passing it directly ;). – Bharath M Shetty Sep 17 '17 at 11:54
@Bharathshetty I thought to correct you, but I was itching to write an answer.. :-) – cs95 Sep 17 '17 at 11:55
Nowadays very less pandas questions. – Bharath M Shetty Sep 17 '17 at 11:56
thank you guys. I realized I had different indices for the several series. I resetted the indices and then it worked like a charm. :) Thanks! – freddy888 Sep 17 '17 at 11:59
1

`it shadows the builtin` this will leads to 'dict' object is not callable. I faced it a lot of times. – Bharath M Shetty Sep 17 '17 at 11:59
@FriedrichFranz That would've been the next issue, had you solved this one... :p – cs95 Sep 17 '17 at 12:00
@cᴏʟᴅsᴘᴇᴇᴅ Is this a duplicated? https://stackoverflow.com/questions/46263897/how-to-reassign-dict-functionality-back-to-the-dict – Bharath M Shetty Sep 17 '17 at 12:08

score 1 · Answer 2 · answered Sep 17 '17 at 11:51

unpacking the list might help i.e

df = pd.DataFrame(*[dict])

If you have dict of series like

df = pd.DataFrame({'A':[1,2,3,4,5,6],'B':[2,5,1,1,5,1]})
data = {'A' : df['A'], 'B' : df['B']}

Then

ndf = pd.DataFrame(*[data])

Create pandas dataframe from a dictionary of series

2 Answers2