代码与论文不一致 Code does not match the paper

[UesugiErii]

代码与论文不一致 Code does not match the paper

[UesugiErii]

And in tf.nn.sampled_softmax_loss, use num_sampled=self.neg_num * self.batch_size is incorrect, you can refer to THUDM/ComiRec#3

[UesugiErii]

In your paper, you write we treat all the items that the user has not interacted with as negative items
But in your code, you treat the result of index.search(user_embs, topN) as negative items, but index.search(user_embs, topN) may contain items that the user has interacted with

[Fred199683]

确实有很多和论文不一致的地方。

[chungewoqu2]

损失函数上，Lc在代码里被换成了另外一个，也跟论文对不上。