Accumulator for loss statistics. Currently calculates: * accuracy * perplexity * elapsed time

Accumulator for loss statistics. Currently calculates: * accuracy * perplexity * elapsed time

Accumulator for loss statistics. Currently calculates: * accuracy * perplexity * elapsed time

Accumulator for loss statistics. Currently calculates: * accuracy * perplexity * elapsed time

Accumulator for loss statistics. Currently calculates: * accuracy * perplexity * elapsed time

Ordered Iterator Class

Ordered Iterator Class

Ordered Iterator Class

Ordered Iterator Class

Ordered Iterator Class

A two-layer Feed-Forward-Network. Args: d_model (int): the size of input for the first-layer of the FFN. d_ff (int): the h

A two-layer Feed-Forward-Network. Args: d_model (int): the size of input for the first-layer of the FFN. d_ff (int): the h

A two-layer Feed-Forward-Network. Args: d_model (int): the size of input for the first-layer of the FFN. d_ff (int): the h

A two-layer Feed-Forward-Network with residual layer norm. Args: d_model (int): the size of input for the first-layer of the FFN

A two-layer Feed-Forward-Network. Args: d_model (int): the size of input for the first-layer of the FFN. d_ff (int): the h

Beam Search

Beam Search

Beam Search

Beam Search

Beam Search

An Ordered Dataset Iterator, supporting multiple datasets, and lazy loading. Args: datsets (list): a list of datasets, which are lazil

An Ordered Dataset Iterator, supporting multiple datasets, and lazy loading. Args: datsets (list): a list of datasets, which are lazil

An Ordered Dataset Iterator, supporting multiple datasets, and lazy loading. Args: datsets (list): a list of datasets, which are lazil

An Ordered Dataset Iterator, supporting multiple datasets, and lazy loading. Args: datsets (list): a list of datasets, which are lazil

An Ordered Dataset Iterator, supporting multiple datasets, and lazy loading. Args: datsets (list): a list of datasets, which are lazil

A class that listens for exceptions in children processes and propagates the tracebacks to the parent process.

A class that listens for exceptions in children processes and propagates the tracebacks to the parent process.

A class that listens for exceptions in children processes and propagates the tracebacks to the parent process.

A class that listens for exceptions in children processes and propagates the tracebacks to the parent process.

A class that listens for exceptions in children processes and propagates the tracebacks to the parent process.

With label smoothing, KL-divergence between q_{smoothed ground truth prob.}(w) and p_{prob. computed by model}(w) is minimized.

With label smoothing, KL-divergence between q_{smoothed ground truth prob.}(w) and p_{prob. computed by model}(w) is minimized.

With label smoothing, KL-divergence between q_{smoothed ground truth prob.}(w) and p_{prob. computed by model}(w) is minimized.

With label smoothing, KL-divergence between q_{smoothed ground truth prob.}(w) and p_{prob. computed by model}(w) is minimized.

With label smoothing, KL-divergence between q_{smoothed ground truth prob.}(w) and p_{prob. computed by model}(w) is minimized.

Base class for model saving operations Inherited classes must implement private methods: * `_save` * `_rm_che

Base class for model saving operations Inherited classes must implement private methods: * `_save` * `_rm_che

Base class for model saving operations Inherited classes must implement private methods: * `_save` * `_rm_che

Base class for model saving operations Inherited classes must implement private methods: * `_save` * `_rm_che

Base class for model saving operations Inherited classes must implement private methods: * `_save` * `_rm_che

Implement multiple optimizers needed for sparse adam

Implement multiple optimizers needed for sparse adam

Implement multiple optimizers needed for sparse adam

Implement multiple optimizers needed for sparse adam

Implement multiple optimizers needed for sparse adam

Standard NMT Loss Computation.

Standard NMT Loss Computation.

Standard NMT Loss Computation.

Standard NMT Loss Computation.

Standard NMT Loss Computation.

Controller class for optimization. Mostly a thin wrapper for `optim`, but also useful for implementing rate scheduling beyond what is cur

Controller class for optimization. Mostly a thin wrapper for `optim`, but also useful for implementing rate scheduling beyond what is cur

Controller class for optimization. Mostly a thin wrapper for `optim`, but also useful for implementing rate scheduling beyond what is cur

Controller class for optimization. Mostly a thin wrapper for `optim`, but also useful for implementing rate scheduling beyond what is cur

Controller class for optimization. Mostly a thin wrapper for `optim`, but also useful for implementing rate scheduling beyond what is cur

Class that controls the training process. Args: model(:py:class:`onmt.models.model.NMTModel`): translation model to train

Class that controls the training process. Args: model(:py:class:`onmt.models.model.NMTModel`): translation model to train

Class that controls the training process. Args: model(:py:class:`onmt.models.model.NMTModel`): translation model to train

Class that controls the training process. Args: model(:py:class:`onmt.models.model.NMTModel`): translation model to train

Class that controls the training process. Args: model(:py:class:`onmt.models.model.NMTModel`): translation model to train