search

garethjns / Kaggle-EEG

Seizure prediction from EEG data using machine learning. 3rd place solution for Kaggle/Uni Melbourne seizure prediction competition.
101 stars 29 forks source link

Solution File #16

Open YasminAMassoud opened 4 years ago

YasminAMassoud commented 4 years ago

I was able to run the code and obtain the results, however i need to make a new solution file for the data i entered, can you guide me which part i need to work on in the code? @garethjns

garethjns commented 4 years ago

Hi @YasminAMassoud,

If I remember correctly, predict.m runs the predictions from both models then create the solution file using saveSub. It should dump a .csv file to the current working directory.

YasminAMassoud commented 3 years ago

@garethjns I was able to do produce thanks, however can you guide me how can i produce AUC for testing without needing to submit solution file ?

garethjns commented 3 years ago

@YasminAMassoud It's possible to calculate the AUC for the training data (see https://github.com/garethjns/Kaggle-EEG/issues/17), however you'll need test set labels to calculate it for the test set. I don't know if these are available or not - if they aren't, it might be worth creating a smaller holdout set from the training data and using that instead.

YasminAMassoud commented 3 years ago

Thanks

YasminAMassoud commented 3 years ago

-I have got the labels now, can you give any guidelines how to do that ? -I cant trace where is the part of comparing labels for the training set to get the AUC , can you point where that part is ?