functions.h File Reference

Header file containing various machine learning functions and utilities. More...

#include <torch/torch.h>
#include <Eigen/Dense>
#include <chrono>
#include <filesystem>
#include "BaseFunction.h"
#include "BatchNorm.h"
#include "ChatGPT.h"
#include "ComplexLayer.h"
#include "Concat.h"
#include "CosineSimilarity.h"
#include "DecisionForest.h"
#include "DecisionTree.h"
#include "DotProduct.h"
#include "Dropout.h"
#include "Embedding.h"
#include "Encoder.h"
#include "HuggingFaceServerless.h"
#include "HuggingFaceTokenizer.h"
#include "PositionEncoding.h"
#include "RAG.h"
#include "SequencePooling.h"
#include "XGBoost.h"
#include "velox/vector/tests/utils/VectorMaker.h"

Go to the source code of this file.

Classes
class	MatrixMultiply
	Class for performing matrix multiplication. More...
class	MatrixMultiply_b
	Class for performing blocked matrix multiplication. More...
class	MatrixMultiply_h
	Class for performing matrix multiplication with a hierarchical approach. More...
class	MatrixMultiply_Block
	Class for performing blocked matrix multiplication. More...
class	MatrixAddition
	Class for performing matrix addition. More...
class	MatrixVectorAddition
	A class that performs matrix-vector addition, inheriting from MLFunction. More...
class	Sigmoid
	A class that implements the Sigmoid activation function, inheriting from MLFunction. More...
class	Relu
	A class that implements the Rectified Linear Unit (ReLU) activation function, inheriting from MLFunction. More...
class	Softmax
	A class that implements the Softmax activation function, inheriting from MLFunction. More...
class	Argmax
	A class that implements the Argmax function, inheriting from MLFunction. More...
class	MinMaxScaler
	A class that implements Min-Max scaling, inheriting from MLFunction. More...
class	TorchDNN2Level
	A class that implements a 2-level deep neural network using PyTorch, inheriting from MLFunction. More...
class	TorchDNN
	A class that implements a deep neural network using PyTorch, inheriting from MLFunction. More...
class	TorchDNNV2
	A class that implements a configurable deep neural network using PyTorch, inheriting from MLFunction. More...
class	TorchDNNV2CUDA
	A class that implements a configurable deep neural network using PyTorch with CUDA support, inheriting from MLFunction. More...
class	TorchDNNKernel
	A class that implements a single-layer neural network kernel using PyTorch, inheriting from MLFunction. More...
class	TorchDNN_Multi
	A class that implements a multi-layer deep neural network using PyTorch, inheriting from MLFunction. More...
class	Convolute
	A class for performing 2D convolution operations as part of a machine learning function. More...
class	TorchConvolute
	A class that implements a 2D convolution operation using PyTorch, inheriting from MLFunction. More...
class	TorchCNN
	A class that implements a convolutional neural network (CNN) using PyTorch, inheriting from MLFunction. More...
class	VectorScalarAddition
	A class that implements vector-scalar addition, inheriting from MLFunction. More...
class	MaxPool
	A class that implements max pooling, inheriting from MLFunction. More...

Namespaces
namespace	velox::dl
	Namespace for deep learning-related utilities and kernels.

Enumerations
enum class	velox::dl::KernelType {   velox::dl::MatMul , velox::dl::MatAdd , velox::dl::ReLU , velox::dl::Softmax ,   velox::dl::BatchNorm , velox::dl::Argmax , velox::dl::Sigmoid }
	Enumeration of kernel types used in deep learning operations. More...

Functions
std::string	velox::dl::kernelTypeToString (KernelType kernelType)
	Converts a KernelType enum value to its string representation.
std::ostream &	velox::dl::operator<< (std::ostream &os, KernelType kernelType)
	Overloads the << operator for KernelType.

Detailed Description

Header file containing various machine learning functions and utilities.

This file defines a collection of classes and functions for performing machine learning operations such as matrix multiplication, addition, activation functions, and more. It also includes utility functions for cost estimation and tensor manipulation.