I have a dataframe, with the following columns, in this order;
'2','4','9','A','1','B','C'
I want the first 3 columns to be ABC but the rest it doesn't matter.
Output:
'A','B','C','3','2','9'... and so on
Is this possible?
(there are 100's of columns, so i can't put them all in a list
You can try to reorder like this:
first_cols = ['A','B','C']
last_cols = [col for col in df.columns if col not in first_cols]
df = df[first_cols+last_cols]
cols = ['2','4','9','A','1','B','C']
df = pd.DataFrame(1, range(3), cols)
df
2 4 9 A 1 B C
0 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
2 1 1 1 1 1 1 1
sorted with keykey = lambda x: (x != 'A', x != 'B', x != 'C')
df[sorted(df, key=key)]
A B C 2 4 9 1
0 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
2 1 1 1 1 1 1 1
Better suited for longer length of first column members
first_cols = ['A', 'B', 'C']
key = lambda x: tuple(y != x for y in first_cols)
df[sorted(df, key=key)]
Using columns.difference:
first = ['A', 'B', 'C']
out = df[first+list(df.columns.difference(first, sort=False))]
Or, if there is a chance that a name from first is not present in df:
first = ['A', 'B', 'C']
out = df.reindex(columns=first+list(df.columns.difference(first, sort=False)))
A,B,Cor all string columns should come first?