|
Mila 0.13.48
Deep Neural Network Library
|
Helpers to standardize registration of unary/binary/paired ops. More...
#include <memory>#include <string>#include <type_traits>#include <stdexcept>import Compute.UnaryOperation;import Compute.BinaryOperation;import Compute.ExecutionContext;import Compute.DeviceType;import Compute.OperationRegistry;import Dnn.TensorDataType;import Compute.PairedOperation;import Compute.IExecutionContext;import Dnn.ComponentConfig;Classes | |
| struct | Mila::Dnn::Compute::always_false< T > |
Namespaces | |
| namespace | Mila |
| Mila main API namespace. | |
| namespace | Mila::Dnn |
| namespace | Mila::Dnn::Compute |
Functions | |
| template<typename OpType, DeviceType DT, typename ConfigT> | |
| std::shared_ptr< OpType > | Mila::Dnn::Compute::makeOpInstance (IExecutionContext *ctx, const ConfigT &cfg) |
| Attempt to construct an OpType instance from a raw IExecutionContext*. | |
| template<DeviceType TDataType, typename OpType, Dnn::TensorDataType TA, Dnn::TensorDataType TB, Dnn::TensorDataType TP = TA> | |
| void | Mila::Dnn::Compute::registerBinaryOpType (const std::string &opName) |
| Register a binary operation type with OperationRegistry using a common factory pattern. | |
| template<DeviceType TDataType, typename OpType, Dnn::TensorDataType TA, Dnn::TensorDataType TB = TA, Dnn::TensorDataType TP = TA> | |
| void | Mila::Dnn::Compute::registerPairedOpType (const std::string &opName) |
| Register a paired operation type with OperationRegistry using a common factory pattern. | |
| template<DeviceType TDataType, typename OpType, Dnn::TensorDataType TInput, Dnn::TensorDataType TPrecision = TInput> | |
| void | Mila::Dnn::Compute::registerUnaryOpType (std::string_view op_name) |
| Register a unary operation type with OperationRegistry using a common factory pattern. | |
Helpers to standardize registration of unary/binary/paired ops.
Small helpers that produce consistent factory lambdas for OperationRegistry.