MLFunction Class Reference

A base class for machine learning functions, inheriting from Velox's VectorFunction. More...

#include <BaseFunction.h>

Inheritance diagram for MLFunction:

Public Member Functions
virtual	~MLFunction ()=default
	Virtual destructor.
virtual float *	getTensor () const =0
	Returns the tensor associated with this function.
virtual std::vector< int >	getDims ()
	Returns the dimensions of the function.
virtual std::string	getFuncName ()
	Returns the name of the function.
virtual int	getNumDims ()
	Returns the number of dimensions of the function.
virtual CostEstimate	getCost (std::vector< int > inputDims)
	Estimates the computational cost of applying the function.

Protected Member Functions
double	getWeightedCost (std::string name, float cost)
	Calculates the weighted cost of the function.
std::vector< double >	getCoefficientVector (std::string name)
	Retrieves the cost coefficients for the function.

Protected Attributes
std::vector< int >	dims
	Dimensions of the function.

Detailed Description

A base class for machine learning functions, inheriting from Velox's VectorFunction.

This class provides a common interface for machine learning functions, including methods for retrieving tensors, dimensions, and cost estimates. It also includes utility methods for calculating weighted costs and retrieving cost coefficients.

Member Function Documentation

getCoefficientVector()

std::vector< double > MLFunction::getCoefficientVector ( std::string name )

inlineprotected

Retrieves the cost coefficients for the function.

Parameters

name	The name of the function.

Returns

A vector of cost coefficients.

getCost()

virtual CostEstimate MLFunction::getCost ( std::vector< int > inputDims )

inlinevirtual

Estimates the computational cost of applying the function.

Parameters

inputDims

A vector containing the dimensions of the input.

Returns

A CostEstimate object representing the estimated cost.

Reimplemented in Argmax, BatchNorm1D, ChatGPT, ChatGPTRecommender, Concat, CosineSimilarity, DateToTimestamp, DotProduct, GetAge, GetBinaryClass, GetCustomerFeatures, GetTransactionFeatures, HuggingFaceServerless, HuggingFaceTokenizer, IntEncoder, IsWeekday, MatrixAddition, MatrixMultiply, MatrixMultiply_h, MinMaxScaler, ml::TreePrediction, ml::VeloxTreePrediction, MultiHotNormalizedEncoder, PositionEncoding, RAG, Relu, SequencePooling, Sigmoid, Softmax, StringEncoder, StringVariadicEncoder, SVD, TimeDiffInDays, TorchDNN2Level, TorchDNN, TorchDNNV2, and TorchDNNV2CUDA.

getDims()

virtual std::vector< int > MLFunction::getDims ( )

inlinevirtual

Returns the dimensions of the function.

Returns

A vector containing the dimensions.

getFuncName()

virtual std::string MLFunction::getFuncName ( )

inlinevirtual

Returns the name of the function.

Returns

The name of the function as a string.

Reimplemented in Argmax, Convolute, MatrixAddition, MatrixMultiply, MatrixMultiply_b, MatrixMultiply_Block, MatrixMultiply_h, MatrixVectorAddition, MaxPool, MinMaxScaler, ml::TreePrediction, ml::VeloxTreePrediction, RAG, Relu, Sigmoid, Softmax, SVD, TorchCNN, TorchConvolute, TorchDNN2Level, TorchDNN, TorchDNNKernel, TorchDNNV2, TorchDNNV2CUDA, and VectorScalarAddition.

getNumDims()

virtual int MLFunction::getNumDims ( )

inlinevirtual

Returns the number of dimensions of the function.

Returns

The number of dimensions.

getTensor()

virtual float * MLFunction::getTensor ( ) const

pure virtual

Returns the tensor associated with this function.

Returns

A pointer to the tensor data.

Implemented in Argmax, BatchNorm1D, ChangeRating, ChatGPT, ChatGPTRecommender, Concat, ConvertDoubleArrayToFloatArray, ConvertDoubleToFloatArray, ConvertToFloatArray, ConvertToIntArray, Convolute, CosineSimilarity, DateToTimestamp, DotProduct, Dropout, Embedding, FullyConnectWithBatchNormAndRelu, GetAge, GetBinaryClass, GetCustomerFeatures, GetTransactionFeatures, HuggingFaceServerless, HuggingFaceTokenizer, IntEncoder, IsWeekday, MatrixAddition, MatrixMultiply, MatrixMultiply_b, MatrixMultiply_Block, MatrixMultiply_h, MatrixVectorAddition, MatrixVectorAddition, MaxPool, MinMaxScaler, ml::TreePrediction, ml::VeloxTreePrediction, ml::XGBoostPrediction, MultiHotNormalizedEncoder, PositionEncoding, RAG, Relu, SequencePooling, Sigmoid, Softmax, StringEncoder, StringVariadicEncoder, SVD, TimeDiffInDays, TorchCNN, TorchConvolute, TorchDNN2Level, TorchDNN, TorchDNN_Multi, TorchDNNKernel, TorchDNNV2, TorchDNNV2CUDA, and VectorScalarAddition.

getWeightedCost()

double MLFunction::getWeightedCost ( std::string name, float cost )

inlineprotected

Calculates the weighted cost of the function.

Parameters

name	The name of the function.
cost	The base cost of the function.

Returns

The weighted cost as a double.

---

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

• BaseFunction.h