The definition of variables does not match the data in the train and test data. Variables like ATA(actual time of arrival), ATD(actual time of departure) etc do not appear in the train and test data
Yes, it was originally going to be a multi-class classification problem and they defined the variables in that way, but then they turned it into a regression problem in the last minute. So some variables that are defined here won't match the variables in the data. But for your mentionned variables, you can calculate that by adding the delay to the estimated time of arrival to get the actual time of arrival, and same for departure.
Yes, it was originally going to be a multi-class classification problem and they defined the variables in that way, but then they turned it into a regression problem in the last minute. So some variables that are defined here won't match the variables in the data. But for your mentionned variables, you can calculate that by adding the delay to the estimated time of arrival to get the actual time of arrival, and same for departure.
Thanks but which one is the delay?
should be a variable called 'TARGET', and the last one ,in order , in the train file.
Is the 'TARGET' in seconds or minutes?