I currently have this code running

def single_iteration(iter:int):
   some_parameters =100
   do_somthing_get data1, data2, data3, data4
   result = []
   for i in range 100:
       data = {'Col1':data1, 'Col2':data1, 'Col3':data3, 'Col4':data4}
       results.append(data)
   df =  pd.DataFrame(result)
   return df

if __name__ == "__main__":
   run_stop = 100
   number_of_cores = int(os.environ['SLURM_CPUS_PER_TASK'])
   with multiprocessing.Pool(number_of_cores) as pool:
      results = pool.map(single_iteration, range(run_stop))
   df = pd.concat(results, ignore_index=True)
   df.to_csv(path, file_name)

However, now I want two different csvs , like with df1 = pd.DaFrame[{'Col1':data1, 'Col2':data2}] and df2 = pd.DaFrame[{'Col3':data3, 'Col4':data4}] and return them in each run and then concat them separately and save them.