nom::repr::NeuralNetOperator Class Reference

Inheritance diagram for nom::repr::NeuralNetOperator:

Public Types
enum	NNKind { Undefined, GenericOperator, NNPhi, While }
	Discriminator for LLVM-style RTTI (isa<>)
enum	NNLayout { Undefined, NCHW, NHWC }
	An optional tensor-type specifier.
Public Types inherited from nom::repr::Instruction
enum	Opcode {   Generic, TerminatorStart, Branch, Return,   TerminatorEnd, Phi }
	All the different types of execution.
Public Types inherited from nom::repr::Value
enum	ValueKind { Value, Instruction, Data }

Public Member Functions
	NeuralNetOperator (NNKind K, Opcode I, NNLayout L)
	NeuralNetOperator (NNKind K, Opcode I)
	NeuralNetOperator (NNKind K, NNLayout L)
	NeuralNetOperator (NNKind K)
NNKind	getKind () const
void	setLayout (NNLayout L)
NNLayout	getLayout () const
void	setAnnotation (std::unique_ptr< Annotation > extraAnnotation)
const Annotation *	getAnnotation () const
Annotation *	getMutableAnnotation ()
const std::string	getName () const
bool	checkInputsAndOutputs (std::vector< const NeuralNetData * > inputs, std::vector< const NeuralNetData * > outputs)
	Validate the inputs and outputs to this operator. More...
	NeuralNetOperator (const NeuralNetOperator &)=delete
NeuralNetOperator &	operator= (NeuralNetOperator &)=delete
Public Member Functions inherited from nom::repr::Instruction
	Instruction (Opcode op)
Opcode	getOpcode () const
Public Member Functions inherited from nom::repr::Value
	Value (ValueKind K)
ValueKind	getKind () const

Additional Inherited Members
Static Public Member Functions inherited from nom::repr::Instruction
static bool	classof (const Value *V)

Detailed Description

Definition at line 60 of file NeuralNet.h.

Member Function Documentation

bool nom::repr::NeuralNetOperator::checkInputsAndOutputs ( std::vector< const NeuralNetData * >  inputs, std::vector< const NeuralNetData * >  outputs  )

inline

Validate the inputs and outputs to this operator.

inputs A vector of references to NeuralNetData types that represent the data being fed into the operator. outputs A vector of references to NeuralNetData types that represent the data being outputted by the operator.

Returns

true if the inputs and outputs are compatible with the operator.

Definition at line 118 of file NeuralNet.h.

---

The documentation for this class was generated from the following files:

• caffe2/core/nomnigraph/include/nomnigraph/Representations/NeuralNet.h
• caffe2/core/nomnigraph/Representations/NeuralNet.cc