massquantity

Thanks for implementing this! I've seen the paper you mentioned roughly, and I'll assume you're using TensorFlow1. It seems that the main difference is they add a weight(w_l) parameter in...

What do you mean by various loss variables? Training loss or model's variables?

For example, If I want to get the `self.bpr_loss`, run `sess.run` in the [training function](https://github.com/massquantity/LibRecommender/blob/master/libreco/training/tf_trainer.py#L356), which can get the variable and its shape. ```python for data in data_generator(shuffle, self.batch_size): self.sess.run(...

bpr loss belongs to pairwise loss, which is complicated to process with users providing both negative and positive samples. So it only supports negative sampling, see *Caution* in [Loss](https://librecommender.readthedocs.io/en/latest/user_guide/model_train.html#loss).

Hi, thanks for your suggestion. If I want to implement this, I may pass the expired items as an argument into the `recommend_user` function and filter out them. Another way...

First of all, there are two kinds of ids in this library, i.e. original id and inner id. I will assume the expired items you've got are represented as original...

> This can be a problem if you want to measure a model's degradation over time. Let's say you have an updated test set with new items every few weeks....

The save/load API is used for inference, so u should pass `inference=True` in the `ytb_retrieval.save` method. See [Save/Load](https://librecommender.readthedocs.io/en/latest/user_guide/evaluation_save_load.html#save-load-model)

Hi, you can inherit from the [LightGCN](https://github.com/massquantity/LibRecommender/blob/master/libreco/algorithms/lightgcn.py) class. Internally it uses a pytorch module [self.torch_model](https://github.com/massquantity/LibRecommender/blob/master/libreco/algorithms/lightgcn.py#L118) to perform the forward propagation. So you can add layers after this module.

You can set `inference_only` to False, then `rebuild_model`. Setting `inference_only` to True will only save the embeddings and drop the entire torch model.