fit a ridge regression model (l2 regularized least squares)

Functions
void	ML::GLM::ridgeFit (const raft::handle_t &handle, float *input, size_t n_rows, size_t n_cols, float *labels, float *alpha, int n_alpha, float *coef, float *intercept, bool fit_intercept, bool normalize, int algo=0, float *sample_weight=nullptr)
void	ML::GLM::ridgeFit (const raft::handle_t &handle, double *input, size_t n_rows, size_t n_cols, double *labels, double *alpha, int n_alpha, double *coef, double *intercept, bool fit_intercept, bool normalize, int algo=0, double *sample_weight=nullptr)

Detailed Description

Parameters

input	device pointer to feature matrix n_rows x n_cols
n_rows	number of rows of the feature matrix
n_cols	number of columns of the feature matrix
labels	device pointer to label vector of length n_rows
alpha	host pointer to parameters of the l2 regularizer
n_alpha	number of regularization parameters
coef	device pointer to hold the solution for weights of size n_cols
intercept	host pointer to hold the solution for bias term of size 1
fit_intercept	if true, fit intercept
normalize	if true, normalize data to zero mean, unit variance
algo	specifies which solver to use (0: SVD, 1: Eigendecomposition)
sample_weight	device pointer to sample weight vector of length n_rows (nullptr for uniform weights) This vector is modified during the computation

Function Documentation

ridgeFit() [1/2]

void ML::GLM::ridgeFit	(	const raft::handle_t &	handle,
		double *	input,
		size_t	n_rows,
		size_t	n_cols,
		double *	labels,
		double *	alpha,
		int	n_alpha,
		double *	coef,
		double *	intercept,
		bool	fit_intercept,
		bool	normalize,
		int	algo = 0,
		double *	sample_weight = nullptr
	)

ridgeFit() [2/2]

void ML::GLM::ridgeFit	(	const raft::handle_t &	handle,
		float *	input,
		size_t	n_rows,
		size_t	n_cols,
		float *	labels,
		float *	alpha,
		int	n_alpha,
		float *	coef,
		float *	intercept,
		bool	fit_intercept,
		bool	normalize,
		int	algo = 0,
		float *	sample_weight = nullptr
	)