You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello! First of all, Thank you so much for all the efforts in creating such a beautiful repo.
I am currently working on LLM Based Recommendation and have been trying to work on something similar to TALLRec. However, in CTR prediction papers like TALLRec or ReLLa, I found the metrics of comparision with the baselines like SaSRec, BERT4Rec etc. to be the AUC score. I am a little uncertain as of how to get AUC scores from SASRec type of models which inherently are designed for ranking predictions metrics like NDCG@k, HitRate@k etc. Is it possible to implement the same? Could you please guide me how to work on it?
Thanks and Regards
Millennium Bismay
The text was updated successfully, but these errors were encountered:
Hello! First of all, Thank you so much for all the efforts in creating such a beautiful repo.
I am currently working on LLM Based Recommendation and have been trying to work on something similar to TALLRec. However, in CTR prediction papers like TALLRec or ReLLa, I found the metrics of comparision with the baselines like SaSRec, BERT4Rec etc. to be the AUC score. I am a little uncertain as of how to get AUC scores from SASRec type of models which inherently are designed for ranking predictions metrics like NDCG@k, HitRate@k etc. Is it possible to implement the same? Could you please guide me how to work on it?
Thanks and Regards
Millennium Bismay
The text was updated successfully, but these errors were encountered: