CrossEncoder class - cross_encoder library

Constructors

CrossEncoder({required int vocabSize, required int maxSeqLen, required int embedDim, int numLayers = 4, int numHeads = 4, double dropoutP = 0.0, PoolingMode pooling = PoolingMode.cls, int numLabels = 1, Device device = Device.CPU, int seed = 0})
CrossEncoder.wrap(TokenEncoder backbone, {PoolingMode pooling = PoolingMode.cls, int numLabels = 1, Device device = Device.CPU, int seed = 424242}): Wrap a pretrained backbone. Adds a fresh classification head.

backbone → TokenEncoder: final
embedDim → int: no setter
hashCode → int: The hash code for this object.
no setterinherited
head → Linear: final
numLabels → int: final
pooling → PoolingMode: final
runtimeType → Type: A representation of the runtime type of the object.
no setterinherited
training ↔ bool: Whether this module is in training mode. Layers that behave differently between training and inference (e.g. Dropout) read this flag in their call method. Defaults to training mode.
getter/setter pairinherited

call(Tensor pairTokens) → Tensor: Forward pass on a single tokenized pair. Returns [1, numLabels] raw logits.
eval() → void: Put this module (and any registered submodules) into evaluation mode.
inherited
noSuchMethod(Invocation invocation) → dynamic: Invoked when a nonexistent method or property is accessed.
inherited
parameters() → List<Tensor>: Trainable tensors owned by this module (and its submodules).
override
score(Tensor pairTokens) → double: Convenience: run under noGrad, return the first-label logit squashed through a sigmoid. Useful when numLabels == 1.
submodules() → List<Module>: Submodules owned by this module. Subclasses that compose other modules should override this so train() / eval() propagate. Default: empty.
override
toString() → String: A string representation of this object.
inherited
train() → void: Put this module (and any registered submodules) into training mode.
inherited
zeroGrad() → void: Zero every parameter's gradient. Safe to call before each backward.
inherited