CpuMultiHeadAttentionOp.ixx File Reference

Implementation of the CPU-based attention operation for neural networks. More...

#include <string>
#include <memory>
#include <vector>
#include <stdexcept>
#include <cmath>
import Compute.CpuDevice;
import Compute.OperationType;
import Compute.OperationAttributes;
import Compute.OperationRegistry;
import Compute.UnaryOperation;
import Compute.OperationBase;
import Compute.DeviceContext;
import Dnn.ComponentConfig;
import Compute.DeviceType;
import Dnn.Tensor;
import Dnn.Modules.Attention;

Classes
class	Mila::Dnn::Compute::CpuMultiHeadAttentionOp
	CPU implementation of the Multi-Head Attention operation for neural networks. More...
class	Mila::Dnn::Compute::CpuMultiHeadAttentionOpRegistrar
	Class responsible for registering the CpuMultiHeadAttention operation. More...

Namespaces
namespace	Mila
namespace	Mila::Dnn
namespace	Mila::Dnn::Compute

Detailed Description

Implementation of the CPU-based attention operation for neural networks.