operators.h File Reference

#include "tensor_utilities.h"
#include "enum_map.h"
#include "forward_propagation.h"
#include "back_propagation.h"

Go to the source code of this file.

Classes
struct	opennn::Operator
	Base class for compute building blocks composed by layers (matmul, activation, dropout, etc.). More...
struct	opennn::AddOp
	Element-wise sum of several input tensors (used by residual connections). More...
struct	opennn::DropoutOp
	Inverted dropout: at training time zeros activations with probability rate and rescales survivors. More...
struct	opennn::ActivationOp
	Element-wise non-linear activation (Identity, Sigmoid, Tanh, ReLU, Softmax). More...
struct	opennn::CombinationOp
	Affine combination output = input * weights + bias (the dense matmul building block). More...
struct	opennn::CombinationReluOp
	Fused affine + ReLU activation (uses cuBLASLt epilogue on GPU when available). More...
struct	opennn::BatchNormOp
	Batch normalization with learnable scale/shift and running statistics for inference. More...
struct	opennn::ConvolutionOp
	2D convolution operator (NHWC layout) backed by Eigen on CPU and cuDNN on GPU. More...
struct	opennn::ConvolutionReluOp
	Fused 2D convolution + ReLU activation (uses cuDNN fused epilogue on GPU). More...
struct	opennn::LayerNormOp
	Layer normalization with learnable scale/shift, applied across the embedding dimension. More...
struct	opennn::MultiHeadProjectionOp
	Projects (input_features) into (heads * head_dim) and reshapes for multi-head attention. More...
struct	opennn::AttentionOp
	Scaled dot-product attention with optional causal mask and dropout. More...
struct	opennn::MergeOp
	Reshapes (batch, heads, seq, head_dim) tensors back into (batch, seq, embed); no parameters. More...
struct	opennn::PoolOp
	2D pooling operator supporting max and average reductions. More...
struct	opennn::Pool3dOp
	Sequence-wide 1D pooling over the embedding dimension (mean or max). More...
struct	opennn::EmbeddingLookupOp
	Token embedding lookup with optional scaling and additive positional encoding. More...
struct	opennn::FlatOp
	Flattens a multi-dimensional tensor into a 2D (batch, features) tensor. More...
struct	opennn::BoundOp
	Clamps each output channel to a configurable lower/upper interval. More...
struct	opennn::ScaleOp
	Scales inputs to a target range using per-feature minimum/maximum or mean/std statistics. More...
struct	opennn::UnscaleOp
	Inverse of ScaleOp: maps normalized outputs back to the original feature range. More...

Namespaces
namespace	opennn