Adbot Ad Engagement Forecasting Challenge
'%3e%3cpath fill='%23001489' d='M0 0v6h9V0z'/%3e%3cpath fill='%23e03c31' d='M0 0v3h9V0z'/%3e%3cg stroke='%23fff' stroke-width='2'%3e%3cpath id='d' d='m0 0 4.5 3L0 6m4.5-3H9'/%3e%3cuse xlink:href='%23b' stroke='%23ffb81c' clip-path='url(%23c)'/%3e%3c/g%3e%3cuse xlink:href='%23d' fill='none' stroke='%23007749' stroke-width='1.2'/%3e%3c/g%3e%3c/svg%3e)
Helping South Africa
$500 USD
Completed (over 2 years ago)
Forecast
456 joined
112 active
Start
04 Apr 24
Close
19 May 24
Reveal
19 May 24
Few advise on how to achieve a high score (20.03251794 on Public LB)
Platform Ā· 12 Apr 2024, 08:50 Ā· 5
1. Keep your model as simple as possible. The key to achieving high score is the ability of your model to generalize well.
2. You may try ensembling models but I don't think it offers much value.
3. Spend some time to understand the dataset. An understanding of the dataset will help to choose the right model.
4. Don't be afraid to try as much models as possible.
good luck
Are you using a recurrent neural network or ARIMA model for the forecasting? Thanks for the tips.
In my opinion, RNN will be an over-kill. You may try it and let us know how it goes.
I tried it and the R2 score was good but once I uploaded the submission file the R2 score was way higher. But I think I'm probably not doing something right.
I will advise that you have a look at your submission file and do a sense check. The error could be in the submission file
Yeah, I'll check it out.