---

class sts_select.gensim.FastTextModel(model_path=None)

Parameters:

model_path – Path to a saved model. If None, a new model will be created.

encode(sentences, batch_size=1, show_progress_bar=False)

Prepares sentence bpairs for the model and returns the embeddings.

Parameters:

sentences – List of sentences to encode.

fit(train_objectives=None, epochs=None, warmup_steps=None)

Fits the model to the given sentence pair/similarity dataset.

Parameters:

• train_objectives – List of tuples. Each tuple contains a dataloader and a loss function.

• epochs – Number of epochs to train the model.

• warmup_steps – Number of warmup steps.

save(path)

Saves the model to the specified path.

Parameters:

path – Path to save the model to.

class sts_select.gensim.SkipgramModel(model_path=None)

Parameters:

model_path – Path to a saved model. If None, a new model will be created.

encode(sentences, batch_size=1, show_progress_bar=False)

Prepares sentence bpairs for the model and returns the embeddings.

Parameters:

sentences – List of sentences to encode.

fit(train_objectives=None, epochs=None, warmup_steps=None, worker_count=-1)

Fits the model to the given sentence pair/similarity dataset.

Parameters:

• train_objectives – List of tuples. Each tuple contains a dataloader and a loss function.

• epochs – Number of epochs to train the model.

• warmup_steps – Number of warmup steps.

save(path)

Saves the model to the specified path.

Parameters:

path – Path to save the model to.