RoFormerForMultipleChoice¶
- class lucid.models.RoFormerForMultipleChoice(config: RoFormerConfig)¶
The RoFormerForMultipleChoice class scores candidate choices by flattening [batch, num_choices, seq] inputs, running RoFormer encoding, and reshaping logits back to [batch, num_choices].
Class Signature¶
class RoFormerForMultipleChoice(config: RoFormerConfig)
Parameters¶
config (RoFormerConfig): RoFormer configuration with pooling enabled.
Methods¶
- RoFormerForMultipleChoice.forward(input_ids: LongTensor | None = None, attention_mask: Tensor | None = None, token_type_ids: LongTensor | None = None, position_ids: LongTensor | None = None, inputs_embeds: FloatTensor | None = None) Tensor
- RoFormerForMultipleChoice.get_loss(labels: Tensor, input_ids: LongTensor | None = None, attention_mask: Tensor | None = None, token_type_ids: LongTensor | None = None, position_ids: LongTensor | None = None, inputs_embeds: FloatTensor | None = None, *, reduction: str | None = 'mean') Tensor
- RoFormerForMultipleChoice.predict_labels(input_ids: LongTensor | None = None, attention_mask: Tensor | None = None, token_type_ids: LongTensor | None = None, position_ids: LongTensor | None = None, inputs_embeds: FloatTensor | None = None) Tensor
Examples¶
>>> import lucid.models as models
>>> config = models.RoFormerConfig.base(vocab_size=50000)
>>> model = models.RoFormerForMultipleChoice(config)
>>> print(model)
RoFormerForMultipleChoice(...)