Package io.github.kirstenali.deepj.training

Interface Trainable

All Known Subinterfaces:
CausalLM, Layer, NormLayer, Projection
All Known Implementing Classes:
Activation, Autoencoder, BatchNorm, BiasFreeLinear, Conv1D, Conv2D, DecoderOnlyModel, DeepJLoop, DeepJNova, DeepJNovaTransformerBlock, DeepJOrbit, DeepJOrbitTransformerBlock, DeepJOrigin, DeepJOriginTransformerBlock, DeepJPrism, DeepJPrismTransformerBlock, Dropout, Embedding, EncoderAttention, Flatten, FNN, GlobalAveragePool, GroupedQueryAttention, GRU, GRUCell, LayerNorm1D, Linear, LoopedTransformer, LSTM, LSTMCell, Module, MultiHeadLatentAttention, MultiHeadSelfAttention, Pool1D, Pool2D, PositionalEmbedding, Recurrent, RecurrentCell, Reshape, RMSNorm1D, RNN, RNNCell, RoPEMultiHeadSelfAttention, Sequential, SwiGLULayer, TiedEmbeddingProjection, TransformerEncoder, TransformerEncoderLayer, TransformerStack
public interface Trainable

  • Method Summary

    Modifier and Type
    Method
    Description
    List<Parameter>
    parameters()
     
    default void
    zeroGrad()
     

  • Method Details

    • parameters

      List<Parameter> parameters()

    • zeroGrad

      default void zeroGrad()