◆ ordered_logistic_glm()

__kernel void stan::math::opencl_kernels::ordered_logistic_glm	(	__global double *	location_sum,
		__global double *	logp_global,
		__global double *	location_derivative,
		__global double *	cuts_derivative,
		const __global int *	y_global,
		const __global double *	x,
		const __global double *	beta,
		const __global double *	cuts,
		const int	N_instances,
		const int	N_attributes,
		const int	N_classes,
		const int	is_y_vector,
		const int	need_location_derivative,
		const int	need_cuts_derivative
	)

GPU implementation of ordinal regression Generalized Linear Model (GLM).

Must be run with at least N_instances threads and local size equal to LOCAL_SIZE_.

Parameters

[out]	location_sum	partially summed location (1 value per work group)
[out]	logp_global	partially summed log probability (1 value per work group)
[out]	location_derivative	derivative wrt location
[out]	cuts_derivative	partially summed derivative wrt cuts (1 column per work group)
[in]	y_global	a scalar or vector of classes.
[in]	x	design matrix
[in]	beta	weight vector
[in]	cuts	cutpoints vector
	N_instances	number of cases
	N_attributes	number of attributes
	N_classes	number of classes
	is_y_vector	0 or 1 - whether y is a vector (alternatively it is a scalar)
	need_location_derivative	interpreted as boolean - whether location_derivative needs to be computed
	need_cuts_derivative	interpreted as boolean - whether cuts_derivative needs to be computed