Returns the batch size for a batch of examples

Returns the batch size of the curren batch based on the passed features.

Performs a single step of Beam Search Decoding. Args: time_: Beam search time step, should start at 0. At time 0 we assume that all beams

Internal while_loop body. Args: time: scalar int32 tensor. outputs_ta: structure of TensorArray. state: (structure of)

Expects a binary class matrix instead of a vector of scalar classes

Chooses the top-k beams as successors.

Computes the loss for this model. Loss = seq_loss + schema_copy_loss. seq_loss is the cross entropy loss for the output sequence. #

Computes the loss for this model. Loss = seq_loss + schema_copy_loss. seq_loss is the cross entropy loss for the output sequence.

Computes the loss for this model. Loss = seq_loss + word_copy_loss. seq_loss is the cross entropy loss for the output sequence.

Computes the loss for this model. Loss = seq_loss + schema_copy_loss + word_copy_loss. seq_loss is the cross entropy loss for the outp

Interleaves training and evaluation. The frequency of evaluation is controlled by the `train_steps_per_iteration` (via constructor). The mode

Creates a new Experiment instance. Args: output_dir: Output directory for model checkpoints and summaries.

Creates an instance of `BeamState` that can be used on the first call to `beam_step`. Args: config: A BeamSearchConfig Returns: An ins

Creates an input function that can be used with tf.learn estimators. Note that you must pass "factory funcitons" for both the data provider and

Creates a function that decays the learning rate. Args: decay_steps: How often to apply decay. decay_rate: A Python number. The decay rate.

Creates the metric op

Creates the model

Creates a temporary TFRecords file. Args: source: List of source sentences target: List of target sentences Returns: A tuple (sou

Creates a temporary TFRecords file. Args: source: List of source sentences target: List of target sentences Returns: A tuple (sou

Creates a temporary vocabulary file. Args: words: List of words in the vocabulary Returns: A temporary file object with one word per

Calculates the per-example cross-entropy loss for a sequence of logits and masks out all losses passed the sequence length. Args: logits: L

Returns a dictionary of default parameters for this model.