case class BertLossInput(input: BertPretrainInput, maskedLanguageModelTarget: STen, wholeSentenceTarget: STen)
Input to BertLoss module
- input: feature data, see documentation of BertPretrainInput
- maskedLanguageModelTarget: long tensor of (batch size, masked positions (variable)). Values are the true tokens masked out at the positions in input.positions
- wholeSentenceTarget: float tensor of size (batch size). Values are truth targets for the whole sentence loss which is a BCEWithLogitLoss. Values are floats in [0,1].
- Companion:
- object
trait Serializable
trait Product
trait Equals
class Object
trait Matchable
class Any