Xiangnan He et al. “Neural Collaborative Filtering.” in WWW 2017.
NeuMF is an neural network enhanced matrix factorization model. It replace the dot product to mlp for a more precise user-item interaction.
Our implementation only contains a rough pretraining function.
Calculate the training loss for a batch data.
interaction (Interaction) – Interaction class of the batch.
Training loss, shape: 
- Return type
A simple implementation of dumping model parameters for pretrain.
Defines the computation performed at every call.
Should be overridden by all subclasses.
Although the recipe for forward pass needs to be defined within this function, one should call the
Moduleinstance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.
A simple implementation of loading pretrained parameters.