Skip to main content

Crate zyx_nn

Crate zyx_nn 

Source
Expand description

nn modules for zyx ML library

Structs§

BatchNorm
Batch norm
CausalSelfAttention
Causal self attention
Conv2d
Applies a 2D convolution over an input signal composed of several input planes.
Embedding
Embedding layer
GRUCell
GRU cell (PyTorch-style)
GroupNorm
Group normalization
LSTMCell
A single LSTM (Long Short-Term Memory) cell.
LayerNorm
A Layer Normalization layer.
Linear
Linear layer
MultiheadAttention
Implements multi-head attention as described in “Attention Is All You Need”.
PositionalEncoding
Sinusoidal positional encoding module for transformers.
RMSNorm
RMS norm layer
RNNCell
An Elman RNN cell with optional nonlinearity.
TransformerDecoderLayer
A single layer of a Transformer decoder.
TransformerEncoderLayer
A single Transformer Encoder layer, analogous to torch.nn.TransformerEncoderLayer.

Derive Macros§

Module
Procedural macro Module