concatenate 2 vaex dataframes causing columns issue

Question

I am facing some issues concatenating 2 vaex data frames. When I concat both data frames, the column names are ignored.

First I read a CSV file using vaex

>>> import vaex as vx

>>> df = vx.read_csv("fl_name", header=None)

>>> df.column_names

['0', '1', '2', '3', '4', '5', '6']

Then I try to concat this data frame to an existing one

>>> df_original

['A', 'B', 'C', 'D', 'E', 'F', 'G']

To enable that, I matched the new data frame column names to existing ones

>>> df.column_names = df_original.column_names

>>> df_original.concat(df)

When I checked the resulting data frame columns, I got

['A', 'B', 'C', 'D', 'E', 'F', 'G', '0', '1', '2', '3', '4', '5', '6']

Is there any way to solve this issue and make vaex respect the column names?

Answer 1

After searching for some possible solutions, I found one in pandas.

The easiest way would be to read the file using the desired columns names

´´´df = vx.read_csv("fl_name", header=None, names=['A', 'B', 'C', 'D', 'E', 'F', 'G'])´´´

Doing that, the code will work properly.

Answer 2

I think the best way to do this is using the rename function and the vaex.concat function

import vaex

df1 = vaex.from_dict({
    "0":list(range(5)),
    "1":list(range(5)),
    "2":list(range(5)),
    "3":list(range(5)),
    "4":list(range(5))
})

df2 = vaex.from_arrays(
    A=list(range(5)),
    B=list(range(5)),
    C=list(range(5)),
    D=list(range(5)),
    E=list(range(5))
)

for old, new in zip(df1.get_column_names(), df2.get_column_names()):
    df1.rename(old, new)

vaex.concat([df2, df1])

Note that there have to be the same number of columns in df1 and df2