Pandas: Create dataframe from data and column order

Question

what i'm asking must be something very easy, but i honestly can't see it.... :(

I have an array, lets say

data = [[1, 2, 3], 
        [4, 5, 6],
        [7, 8, 9],
        [10,11,12]]

and i want to put it in a dataframe.
I do df = pd.Dataframe(data, columns={'col1', 'col2', 'col3'})

aiming for:

col1 col2 col3
1    2    3
4    5    6
7    8    9
10   11   12

but i am getting:

col3 col1 col2
1    2    3
4    5    6
7    8    9
10   11   12

(notice the discrepancy between column names and data)

I know i can re-arrange the column names order in the dataframe creation, but i'm trying to understand how it works.

Am i doing something wrong, or it's normal behaviour? (why though?)

Answer 1

You have to pass a tuple or list as value for columns property.

In your example you're using a set of columns which is an unordered collection.

df = pd.DataFrame(data, columns=['col1', 'col2', 'col3'])

Answer 2

You are are using a {set} of columns, which is NOT an ordered collection (neither are dictionaries). Try with a (tuple) , o simply a [list]

df = pd.Dataframe(data, columns=['col1', 'col2', 'col3'])