Options
No examples after joining
Leonie0430
Member Posts: 2 
Newbie
in Help
Hi I am working on an assignment but I have a problem.
There are 2 datasets, the first has some flight information and the second has weather information. I want to predict whether a flight will depart with a delay of 15 minutes or more (using a decision tree). I have done the following:
- Import flight data -- select attributes -- split data -- parse numbers
I selected the following attributes: day, dep_delay_15, dep_time_planned, dest_airport, dest_city, month, origin_airport, origin_city, origin_id and year.
After this I used split to split the dep_time_planned attribute into 2 parts (the first with the hour, the second with the minute) and I've parsed this.
- Import weather data -- select attributes
Here I selected all of the attributes.
After this I have joined these datasets, but after the join all of the examples are gone (the attributes are still there). I have tried everything: left, right, inner and outer join. The key attribute couples I have made are:
- origin_id, airport_id
- day, day
- month, month
- dep_time_planned_1, hour
- year, year
Can somebody help me out?
- Import flight data -- select attributes -- split data -- parse numbers
I selected the following attributes: day, dep_delay_15, dep_time_planned, dest_airport, dest_city, month, origin_airport, origin_city, origin_id and year.
After this I used split to split the dep_time_planned attribute into 2 parts (the first with the hour, the second with the minute) and I've parsed this.
- Import weather data -- select attributes
Here I selected all of the attributes.
After this I have joined these datasets, but after the join all of the examples are gone (the attributes are still there). I have tried everything: left, right, inner and outer join. The key attribute couples I have made are:
- origin_id, airport_id
- day, day
- month, month
- dep_time_planned_1, hour
- year, year
Can somebody help me out?
Tagged:
0
Answers
Have you tried to put a Generate ID operator (to each of your 2 datasets) before joining by using use id attribute as key ?
In attached file, a process to help you to understand how to obtain a working process.
Regards,
Lionel
Thank you for your help, I have fixed the problem. The weird thing is when I hovered over the output of the join, it said my dataset was empty. I had put a filter examples operator after the join operator and filtered too much out. Because my dataset was empty after the join I thought it was because of the join, but it turned out the be because of the filter operator.
Glad you have solved your issue.
Regards,
Lionel