Convert Pandas column containing NaNs to dtype `int` May 9, 2022 by Tarik The lack of NaN rep in integer columns is a pandas “gotcha”. The usual workaround is to simply use floats.