Package io.github.kirstenali.deepj.layers

Interface Layer

All Superinterfaces:

Trainable

All Known Subinterfaces:

NormLayer, Projection

All Known Implementing Classes:

DeepSeekTransformerBlock, FNN, GPTTransformerBlock, LayerNorm1D, Linear, LlamaTransformerBlock, MultiHeadLatentAttention, MultiHeadSelfAttention, RMSNorm1D, RoPEMultiHeadSelfAttention, SwiGLULayer, TransformerStack

public interface Layer
extends Trainable

Differentiable module mapping Tensor -> Tensor. Layers may expose parameters via Trainable (defaults to none).

Method Summary

Modifier and Type	Method	Description
Tensor	backward(Tensor gradOutput)
Tensor	forward(Tensor input)

Methods inherited from interface io.github.kirstenali.deepj.training.Trainable

parameters, zeroGrad

Method Details

forward

Tensor forward(Tensor input)

backward

Tensor backward(Tensor gradOutput)