BoxSquaredEL

class mowl.models.BoxSquaredEL(dataset, embed_dim=50, margin=0.02, reg_norm=1, learning_rate=0.001, epochs=1000, batch_size=32768, delta=2.5, reg_factor=0.2, num_negs=4, model_filepath=None, device='cpu')[source]

Bases: EmbeddingELModel

Implementation based on [jackermeier2023].

Methods Summary

`eval_method`(data)
`get_embeddings`()
`init_module`()
`load_best_model`()
`train`([epochs, validate_every])	Abstract method for training the model.

Methods Documentation

eval_method(data)[source]

get_embeddings()[source]

init_module()[source]

load_best_model()[source]

train(epochs=None, validate_every=1)[source]: Abstract method for training the model. This method must be implemented in children classes