neuralmonkey.decoders.sequence_regressor module

class neuralmonkey.decoders.sequence_regressor.SequenceRegressor(name: str, encoders: List[neuralmonkey.model.stateful.Stateful], data_id: str, layers: List[int] = None, activation_fn: Callable[[tensorflow.python.framework.ops.Tensor], tensorflow.python.framework.ops.Tensor] = <function relu>, dropout_keep_prob: float = 1.0, dimension: int = 1, save_checkpoint: str = None, load_checkpoint: str = None, initializers: List[Tuple[str, Callable]] = None) → None

Bases: neuralmonkey.model.model_part.ModelPart

A simple MLP regression over encoders.

The API pretends it is an RNN decoder which always generates a sequence of length exactly one.

__init__(name: str, encoders: List[neuralmonkey.model.stateful.Stateful], data_id: str, layers: List[int] = None, activation_fn: Callable[[tensorflow.python.framework.ops.Tensor], tensorflow.python.framework.ops.Tensor] = <function relu>, dropout_keep_prob: float = 1.0, dimension: int = 1, save_checkpoint: str = None, load_checkpoint: str = None, initializers: List[Tuple[str, Callable]] = None) → None

Initialize self. See help(type(self)) for accurate signature.

cost
decoded
feed_dict(dataset: neuralmonkey.dataset.dataset.Dataset, train: bool = False) → Dict[tensorflow.python.framework.ops.Tensor, Any]
predictions
runtime_loss
train_loss