I\'m trying to aggregate several csv files into one. After the aggregation I notice that pandas tries to do a float estimation even if i state that the datatype should be a