Huifeng Guo et al. “DeepFM: A Factorization-Machine based Neural Network for CTR Prediction.” in IJCAI 2017.
DeepFM is a DNN enhanced FM which both use a DNN and a FM to calculate feature interaction. Also DeepFM can be seen as a combination of FNN and FM.
Calculate the training loss for a batch data.
interaction (Interaction) – Interaction class of the batch.
Training loss, shape: 
- Return type
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.