Is it possible to get a non zero score using the starter code?

So, I’ve trained the model provided in the starter notebook and submitted the result_dataset.pt file to the site. It didn’t give me an error, so can I rest assured that I generated the file correctly? I mean, in the correct shape and I only have to focus on fine-tuning or architectural improvement or hyperparameter tuning of the model to satisfy the other two criteria, i.e Check the Goodness of Fit and Slope for X0 and X1 and assess the diversity of X0 and X1?

Hello @nikhilsinghisthe
Good job!
Yes, that’s correct. If you see your result on the Leaderboard, then your submission file has been generated correctly and you can focus on your model improving.

Best regards,
Onward Team

1 Like