Keras 3 API documentation / KerasNLP / Models / Roberta / RobertaBackbone model

RobertaBackbone model

[source]

RobertaBackbone class

keras_nlp.models.RobertaBackbone(
    vocabulary_size,
    num_layers,
    num_heads,
    hidden_dim,
    intermediate_dim,
    dropout=0.1,
    max_sequence_length=512,
    dtype=None,
    **kwargs
)

A RoBERTa encoder network.

This network implements a bi-directional Transformer-based encoder as described in "RoBERTa: A Robustly Optimized BERT Pretraining Approach". It includes the embedding lookups and transformer layers, but does not include the masked language model head used during pretraining.

The default constructor gives a fully customizable, randomly initialized RoBERTa encoder with any number of layers, heads, and embedding dimensions. To load preset architectures and weights, use the from_preset() constructor.

Disclaimer: Pre-trained models are provided on an "as is" basis, without warranties or conditions of any kind. The underlying model is provided by a third party and subject to a separate license, available here.

Arguments

  • vocabulary_size: int. The size of the token vocabulary.
  • num_layers: int. The number of transformer layers.
  • num_heads: int. The number of attention heads for each transformer. The hidden size must be divisible by the number of attention heads.
  • hidden_dim: int. The size of the transformer encoding layer.
  • intermediate_dim: int. The output dimension of the first Dense layer in a two-layer feedforward network for each transformer.
  • dropout: float. Dropout probability for the Transformer encoder.
  • max_sequence_length: int. The maximum sequence length this encoder can consume. The sequence length of the input must be less than max_sequence_length default value. This determines the variable shape for positional embeddings.
  • dtype: string or keras.mixed_precision.DTypePolicy. The dtype to use for model computations and weights. Note that some computations, such as softmax and layer normalization, will always be done at float32 precision regardless of dtype.

Examples

input_data = {
    "token_ids": np.ones(shape=(1, 12), dtype="int32"),
    "padding_mask": np.array(
        [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0], shape=(1, 12)),
}

# Pretrained RoBERTa encoder
model = keras_nlp.models.RobertaBackbone.from_preset("roberta_base_en")
model(input_data)

# Randomly initialized RoBERTa model with custom config
model = keras_nlp.models.RobertaBackbone(
    vocabulary_size=50265,
    num_layers=4,
    num_heads=4,
    hidden_dim=256,
    intermediate_dim=512,
    max_sequence_length=128,
)
model(input_data)

[source]

from_preset method

RobertaBackbone.from_preset()

Instantiate RobertaBackbone model from preset architecture and weights.

Arguments

  • preset: string. Must be one of "roberta_base_en", "roberta_large_en".
  • load_weights: Whether to load pre-trained weights into model. Defaults to True.

Examples

# Load architecture and weights from preset
model = keras_nlp.models.RobertaBackbone.from_preset(
    "roberta_base_en"
)

# Load randomly initialized model from preset architecture
model = keras_nlp.models.RobertaBackbone.from_preset(
    "roberta_base_en",
    load_weights=False
)
Preset name Parameters Description
roberta_base_en 124.05M 12-layer RoBERTa model where case is maintained.Trained on English Wikipedia, BooksCorpus, CommonCraw, and OpenWebText.
roberta_large_en 354.31M 24-layer RoBERTa model where case is maintained.Trained on English Wikipedia, BooksCorpus, CommonCraw, and OpenWebText.

token_embedding property

keras_nlp.models.RobertaBackbone.token_embedding

A keras.layers.Embedding instance for embedding token ids.

This layer embeds integer token ids to the hidden dim of the model.