nalp.models.discriminators

Pre-defined discriminator architectures.

A package for already-implemented discriminator models.

class nalp.models.discriminators.ConvDiscriminator(n_samplings: Optional[int] = 3, alpha: Optional[float] = 0.3, dropout_rate: Optional[float] = 0.3)

Bases: nalp.core.Discriminator

A ConvDiscriminator class stands for the convolutional discriminative part of a Generative Adversarial Network.

__init__(self, n_samplings: Optional[int] = 3, alpha: Optional[float] = 0.3, dropout_rate: Optional[float] = 0.3)

Initialization method.

Parameters

  • n_samplings – Number of downsamplings to perform.

  • alpha – LeakyReLU activation threshold.

  • dropout_rate – Dropout activation rate.

property alpha(self)

LeakyReLU activation threshold.

call(self, x: tensorflow.Tensor, training: Optional[bool] = True)

Method that holds vital information whenever this class is called.

Parameters

  • x – A tensorflow’s tensor holding input data.

  • training – Whether architecture is under training or not.

Returns

The same tensor after passing through each defined layer.

Return type

(tf.Tensor)

class nalp.models.discriminators.EmbeddedTextDiscriminator(vocab_size: Optional[int] = 1, max_length: Optional[int] = 1, embedding_size: Optional[int] = 32, n_filters: Optional[Tuple[int, Ellipsis]] = 64, filters_size: Optional[Tuple[int, Ellipsis]] = 1, dropout_rate: Optional[float] = 0.25)

Bases: nalp.core.Discriminator

A EmbeddedTextDiscriminator class stands for the text-discriminative part of a Generative Adversarial Network.

__init__(self, vocab_size: Optional[int] = 1, max_length: Optional[int] = 1, embedding_size: Optional[int] = 32, n_filters: Optional[Tuple[int, Ellipsis]] = 64, filters_size: Optional[Tuple[int, Ellipsis]] = 1, dropout_rate: Optional[float] = 0.25)

Initialization method.

Parameters

  • vocab_size – The size of the vocabulary.

  • max_length – Maximum length of the sequences.

  • embedding_size – The size of the embedding layer.

  • n_filters – Number of filters to be applied.

  • filters_size – Size of filters to be applied.

  • dropout_rate – Dropout activation rate.

call(self, x: tensorflow.Tensor, training: Optional[bool] = True)

Method that holds vital information whenever this class is called.

Parameters

  • x – A tensorflow’s tensor holding input data.

  • training – Whether architecture is under training or not.

Returns

The same tensor after passing through each defined layer.

Return type

(tf.Tensor)

class nalp.models.discriminators.LSTMDiscriminator(embedding_size: Optional[int] = 32, hidden_size: Optional[int] = 64)

Bases: nalp.core.Discriminator

A LSTMDiscriminator class is the one in charge of a discriminative Long Short-Term Memory implementation.

References

  1. Hochreiter, Jürgen Schmidhuber. Long short-term memory. Neural computation 9.8 (1997).

__init__(self, embedding_size: Optional[int] = 32, hidden_size: Optional[int] = 64)

Initialization method.

Parameters

  • embedding_size – The size of the embedding layer.

  • hidden_size – The amount of hidden neurons.

call(self, x: tensorflow.Tensor)

Method that holds vital information whenever this class is called.

Parameters

x – A tensorflow’s tensor holding input data.

Returns

The same tensor after passing through each defined layer.

Return type

(tf.Tensor)

class nalp.models.discriminators.LinearDiscriminator(n_samplings: Optional[int] = 3, alpha: Optional[float] = 0.01)

Bases: nalp.core.Discriminator

A LinearDiscriminator class stands for the linear discriminative part of a Generative Adversarial Network.

__init__(self, n_samplings: Optional[int] = 3, alpha: Optional[float] = 0.01)

Initialization method.

Parameters

  • n_samplings – Number of downsamplings to perform.

  • alpha – LeakyReLU activation threshold.

property alpha(self)

LeakyReLU activation threshold.

call(self, x: tensorflow.Tensor, training: Optional[bool] = True)

Method that holds vital information whenever this class is called.

Parameters

  • x – A tensorflow’s tensor holding input data.

  • training – Whether architecture is under training or not.

Returns

The same tensor after passing through each defined layer.

Return type

(tf.Tensor)

class nalp.models.discriminators.TextDiscriminator(max_length: Optional[int] = 1, embedding_size: Optional[int] = 32, n_filters: Optional[Tuple[int, Ellipsis]] = 64, filters_size: Optional[Tuple[int, Ellipsis]] = 1, dropout_rate: Optional[float] = 0.25)

Bases: nalp.core.Discriminator

A TextDiscriminator class stands for the text-discriminative part of a Generative Adversarial Network.

__init__(self, max_length: Optional[int] = 1, embedding_size: Optional[int] = 32, n_filters: Optional[Tuple[int, Ellipsis]] = 64, filters_size: Optional[Tuple[int, Ellipsis]] = 1, dropout_rate: Optional[float] = 0.25)

Initialization method.

Parameters

  • max_length – Maximum length of the sequences.

  • embedding_size – The size of the embedding layer.

  • n_filters – Number of filters to be applied.

  • filters_size – Size of filters to be applied.

  • dropout_rate – Dropout activation rate.

call(self, x: tensorflow.Tensor, training: Optional[bool] = True)

Method that holds vital information whenever this class is called.

Parameters

  • x – A tensorflow’s tensor holding input data.

  • training – Whether architecture is under training or not.

Returns

The same tensor after passing through each defined layer.

Return type

(tf.Tensor)