search

cheungdaven / DeepRec

An Open-source Toolkit for Deep Learning based Recommendation with Tensorflow.
GNU General Public License v3.0
1.14k stars 293 forks source link

About hit radio@50 in the evaluation of AttRec #31

Closed blcunlp closed 3 months ago

blcunlp commented 4 years ago

hi, In the evaluation process of AttRec, when calculating the recommendation item rank of each user, why use the negative items instead of all items? this code in: DeepRec/utils/evaluation/SeqRecMetrics.py--->def evaluate1(self) image

In my understanding, It should be to calculate the score of each user and all items, then rank them, and finally take the top 50 to calculate HT@50.

Calculate it the way the code says, it does not have any positive items in the prediction process.

cheungdaven commented 4 years ago

Usually, we only consider recommending those items that the user has never seen. If an item is liked by the user, we don't want to repeatedly recommend it to him.