Loads checkpoints from inputs and returns a model with averaged weights. Args: inputs: An iterable of string paths of checkpoints to load f

Build the :class:`~fairseq.criterions.FairseqCriterion` instance for this task. Args: args (argparse.Namespace):

Build a new model instance.

Helper function to build shared embeddings for a set of languages after checking that all dicts corresponding to those languages are

BPE-encode a sentence (or multiple sentences). Every sequence begins with a beginning-of-sentence (`<s>`) symbol. Every sent

Get the number of parameters updates.

Helper to get index of mask symbol

Parse the configuration of lambda coefficient (for scheduling). x = "3" # lambda will be a constant equal to x x = "0:1,

Do forward pass in evaluation mode.

Weight-normalized Conv1d layer optimized for decoding

Cut a sentence based on the numbers of tokens to be cut from beginning and end Represent the sentence as dataset idx and return

Fetch a block of tokens based on its dataset idx

Constructs a NoisingDataset and the corresponding ``LanguagePairDataset(NoisingDataset(src), src)``. If *append_eos_to_tgt* i

Mask tokens for Masked Language Model training Samples mask_ratio tokens that will be predicted by LM. Note:This function ma

This function registers the callback all-reduction function for the NCCL backend. All gradients will be all reduced in one single ste

Generate a random integer which is not in skip_ids. Sample range is [0, total) TODO: ids in skip_ids should be consecutive, we can ex

In order to be efficient, the AdaptiveSoftMax does not compute the scores for all the word of the vocabulary for all the examples. It

increments counters for the sum of log probs of current word and next word (given context ending at current word). Since the next word mi

Asserts that the last token of each sentence in x is not EOS

left and right trim

Build the dictionary Args: filenames (list): list of filenames workers (int): number of concurrent workers

we expect output from a decoder is a tuple with the following constraint: - the first element is a torch.Tensor - the second element can be an

we expect encoder_output to be a dict with the following key/value pairs: - encoder_out: a Torch.Tensor - encoder_padding_mask: a binary T