Delegate squeeze to avoid modifying :func:`onmt.translate.translator.Translator.translate_batch()`

Assert all arguments have the same value

Initialize for decoding. Repeat src objects `beam_size` times.

Tiles x on dimension dim count times.

Args: opt: the option in current environment. text_field(TextMultiField): word and feats field. for_encoder(bool): build

compute accuracy

Select next tokens randomly from the top k possible next tokens. Args: log_probs (FloatTensor): Shaped ``(batch_size, vocab_size)

Log translation.

compute cross entropy

Builds the optimizer from options. Args: cls: The ``Optimizer`` class to instantiate. model: The model to optimize.

Finalize scores and predictions.

Adds a Test to TestModel according to settings Args: param_setting: list of tuples of (param, setting) methodname: name of t

Detect old-style vocabs (``List[Tuple[str, torchtext.data.Vocab]]``). Args: vocab: some object loaded from a *.vocab.pt file Returns

Alternate constructor. Args: model (onmt.modules.NMTModel): See :func:`__init__()`. fields (dict[str, torchtext.data.

compute perplexity

Sets the random seed.

yield a `list` containing `shard_size` line of `path`, or repeatly generate `default` if `path` is None.

Update the model parameters based on current gradients. Optionally, will employ gradient modification or update learning rate.

Move the model to CPU and clear CUDA cache.

Move the model to GPU.

Line-by-line read a file as bytes.

This returns user-defined train/validate data iterator for the trainer to iterate over. We implement simple ordered iterator strategy here,

Various decoder dispatcher function. Args: opt: the option in current environment. embeddings (Embeddings): vocab embeddings

path: location of a corpus file with whitespace-delimited tokens and ￨-delimited features within the token returns: the n

Initialize decoder state with last state of the encoder.

Read data into dicts. Args: vecs (str or Iterable[str]): Sequence of feature vector paths or path to file containi

rnn factory, Use pytorch version when available.

Creates a boolean mask from sequence lengths.

Read the config file and pre-/load the models.

Tokenize a single sequence. Args: sequence (str): The sequence to tokenize. Returns: tok (str): The tokenize

Adds a Test to TestData according to settings Args: param_setting: list of tuples of (param, setting) methodname: name of th

Detect old-style nesting (``dict[str, List[Tuple[str, Field]]]``).

Raise missing dep exception with standard error message.

Yield elements from data in chunks of batch_size, where each chunk size is a multiple of batch_size_multiple. This is an extended version of

Various encoder dispatcher function. Args: opt: the option in current environment. embeddings (Embeddings): vocab embeddings

Return True if check pass; if check fails and abort is True, raise an Exception, othereise return False.

utility for retrieving polyak averaged params

Initialize for decoding.

Returns the current learning rate.

Update a legacy vocab/field format. Args: vocab: a list of (field name, torchtext.vocab.Vocab) pairs. This is the format form

The vocab/fields need updated.

Pad a batch of examples to the length of the longest example. Args: minibatch (List[torch.FloatTensor]): A list of audio data,

Prune statistics dict for efficiency of max() The frequency of a symbol pair never increases, so pruning is generally safe (until we the most

Translate `inputs` using this model Args: inputs (List[dict[str, str]]): [{"src": "..."},{"src": ...}] Returns:

The main training loop by iterating over `train_iter` and possibly running validation on `valid_iter`. Args: tra

Manually unload a model. It will free the memory and cancel the timer

Update statistics by suming values with another `Statistics` object Args: stat: another statistic object upd

Creates a boolean if gpu used

Zero the gradients of optimized parameters.

Initializes the controller. Args: optimizer: A ``torch.optim.Optimizer`` instance. learning_rate: The initial learning rate.