Print the evaluation logs

Evaluate the model on test or valid datasets

Read triples and map them into ids.

Transform a PyTorch Dataloader into python iterator

Save the parameters of the model and the optimizer, as well as some other variables such as step and learning_rate

Get frequency of a partial triple like (head, relation) or (relation, tail) The frequency will be used for subsampling like word2vec

Build a dictionary of true triples that will be used to filter these true triples for negative sampling

Override model and data configuration

Write logs to checkpoint and console

A single train step. Apply back-propation and return the loss

Forward function that calculate the score of a batch of triples. In the 'single' mode, sample is a batch of triple. In the 'h