Projection based on embedding's weight. For example, embedding map vocab_size to embed_size, than projection map embed_size back to vocab_siz

Add a new hypothesis to the list.

Write a piece of data into dataset. Args: data (Any): Serialization will be done using pickle. Example: >>>

Encode a string into ids.

Open the dataset in write mode and returns a writer. Args: path (str): Path to dataset. dbname (str): The name of the file to whi

Decode ids into a string.

If there are enough hypotheses and that none of the hypotheses being generated can become better than the worst one in the heap, then

Tokenize a string.

Model arguments

Training arguments.