KanLayer

class KanLayer<T : DType, V>(val config: KanConfig, val baseActivation: (Tensor<T, V>) -> Tensor<T, V>, initMixingWeights: Tensor<T, V>, initBasis: Tensor<T, V>, initBias: Tensor<T, V>?, val name: String) : Module<T, V> , ModuleParameters<T, V> (source)

Stub implementation for a Kolmogorov–Arnold Network layer.

The forward path intentionally throws until the spline basis and mixing logic are implemented. The class still wires into the DSL so models can be composed before kernels land.

Constructors

Link copied to clipboard
constructor(config: KanConfig, baseActivation: (Tensor<T, V>) -> Tensor<T, V>, initMixingWeights: Tensor<T, V>, initBasis: Tensor<T, V>, initBias: Tensor<T, V>?, name: String)

Properties

Link copied to clipboard
val baseActivation: (Tensor<T, V>) -> Tensor<T, V>
Link copied to clipboard
Link copied to clipboard
open override val modules: List<Module<T, V>>
Link copied to clipboard
open override val name: String
Link copied to clipboard
open override val params: List<ModuleParameter<T, V>>

Functions

Link copied to clipboard
open override fun forward(input: Tensor<T, V>, ctx: ExecutionContext): Tensor<T, V>