GPUMLib
0.2.2
GPU Machine Learning Library

▼NGPUMLib  
CArgument  This class acts like a placeholder for an argument, composed of a parameter and a attribute/value 
CBackPropagation  Represents a feedforward network that can be trained using the CUDA implementation of the BackPropagation algorithm 
CBaseArray  Base class for HostArray and DeviceArray classes (Array base class) 
CBaseMatrix  Base class for HostMatrix and DeviceMatrix classes (Matrix base class) 
CCudaArray  Create an array of any type, that automatically manages the memory used to hold its elements (data will be stored both on the host and on the device) 
CCudaMatrix  Create a matrix of any type, that automatically manages the memory used to hold its elements (data will be stored both on the host and on the device) 
CCudaStream  Represents a CUDA stream 
CDBN  Represents a Deep Belief Network (Device  GPU) 
CDBNhost  Represents a Deep Belief Network (Host  CPU) 
CDeviceAccessibleVariable  Represents a variable residing in memory that is pagelocked and accessible to the device 
CDeviceArray  Create an array of any type, on the device, that automatically manages the memory used to hold its elements 
CDeviceMatrix  Create a matrix of any type, on the device, that automatically manages the memory used to hold its elements 
CHostArray  Create an array of any type, on the host, that automatically manages the memory used to hold its elements 
CHostMatrix  Create a matrix of any type, on the host, that automatically manages the memory used to hold its elements 
CKMeans  Represents a clustering algorithm using the KMeans technique, implemented in CUDA 
CMultipleBackPropagation  Represents a multiple feedforward network that can be trained using the CUDA implementation of the Multiple BackPropagation algorithm 
CNMF  Base class for all NonNegative Matrix Factorization classes 
CNMF_AdditiveDivergence  Represents a NonNegative Matrix Factorization (NMF) algorithm that uses additive update rules and the (KullbackLeibler) divergence metric 
CNMF_AdditiveEuclidian  Represents a NonNegative Matrix Factorization (NMF) algorithm that uses additive update rules and the Euclidean distance metric 
CNMF_MultiplicativeDivergence  Represents a NonNegative Matrix Factorization (NMF) algorithm that uses multiplicative update rules and the (KullbackLeibler) divergence metric 
CNMF_MultiplicativeEuclidianDistance  Represents a NonNegative Matrix Factorization (NMF) algorithm that uses multiplicative update rules and the Euclidean distance metric 
CRadialBasisFunction  Represents a radial basis function network that can be trained using the CUDA implementation of the Radial Basis Function algorithm 
CRandom  Class for generating random values on the device. Uses the CURAND library 
CRBM  Represents a Restricted Boltzman Machine (GPU) 
CRBMhost  Represents a Restricted Boltzman Machine (Host  CPU) 
CReduction  Provides reduction functions (Sum, Average, Max, Min, ...) 
CResourceAllocatingNetwork  Represents a resource allocating network with long term memory that can be trained using a CUDA implementation of the algorithm 
CSettings  Utility class to parse main()'s arguments, store them in a convenient list and access when needed 
CSVM  Represents an SVM which can be used to train and classify datasets on the GPU device 