How to create one DataFrame using Pandas?

Question

Recently I am struggling to learn how to create DataFrames using Pandas being guided by Pandas' documentation. The main problems that I was having were related to the parameters of the DataFrame function. The attributes columns and index. I was using one object containing the data and was trying to create one DataFrame using that data.

The code that I am working on and having errors is this:

data = [["Orange", "Watermelon", "Banana", "Apple"], [10, 30, 23, 56]]
df = pd.DataFrame(data=data[1], columns=data[0])
print(df)

It does not work for me, and if I remove the columns parameter it works. What can be done to solve it?

You probably meant pd.DataFrame(data=[data[1]], columns=data[0]). data= is looking for a 2d list containing all the rows. In your case you have 1 row, so the outer list has 1 element with your 4 cells of data. — ggorlen
– ggorlen, Commented Jan 17, 2021 at 1:18

Mayank Porwal · Accepted Answer · 2021-01-17 07:43:00Z

1

You can simply do this:

In [1426]: df = pd.DataFrame(data=[data[1]], columns=data[0])

In [1427]: df
Out[1427]: 
   Orange  Watermelon  Banana  Apple
0      10          30      23     56

answered Jan 17, 2021 at 7:43

Mayank Porwal

34.2k9 gold badges45 silver badges65 bronze badges

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

Comments

Jancer Lima · Accepted Answer · 2021-01-18 12:58:09Z

For someone who has experience with DataFrame, the problem is clear and easy to find out, but for who is starting it can be very frustrating to have to deal with problems like that.

The problem with the DataFrame is that pandas will not be able to assign 4 data values as 4 columns if you do not guide pandas to do so. What I was trying to do was to create one DataFrame of 4 lines and 1 column (4,1) using 4 data values and 4 column names (4,4). For the code work, you can use the columns values as index and then transpose your DataFrame. What transpose() does is simply transform the lines of the DataFrame into columns.

data = [["Orange", "Watermelon", "Banana", "Apple"], [10, 30, 23, 56]]
df = pd.DataFrame(data=data[1], index=data[0]).transpose()
print(df)

   Orange  Watermelon  Banana  Apple
0      10          30      23     56

The first way will work just if you have enough data to create one 4x4 DataFrame, for example:

data = [["Orange", "Watermelon", "Banana", "Apple"], [
    [10, 30, 23, 56], [18, 44, 12, 73], [2, 24, 88, 10], [5, 71, 35, 62]]]
df = pd.DataFrame(data=data[1], columns=data[0])
print(df)

   Orange  Watermelon  Banana  Apple
0      10          30      23     56
1      18          44      12     73
2       2          24      88     10
3       5          71      35     62

EDIT

As @Mayank said, it can be simplified by:

df = pd.DataFrame(data=[data[1]], columns=data[0]) 

   Orange  Watermelon  Banana  Apple
0      10          30      23     56

Ian Pang · Accepted Answer · 2025-05-06 05:18:38Z

0

Your code is throwing the following exception:

ValueError: Shape of passed values is (4, 1), indices imply (4, 4)

This means there's a mismatch in the data shape: you're passing a column vector of shape (4, 1), but pandas expects a row of shape (1, 4) for the given column headers.

You can fix this by reshaping your data as follows:

df = pd.DataFrame(data=[data[1]], columns=data[0])

answered May 6 at 5:18

Ian Pang

1

1 Comment

julaine May 6 at 12:02

what does this add that the other two answers don't already have?

Collectives™ on Stack Overflow

How to create one DataFrame using Pandas?

3 Answers 3

Comments

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related