predict_kernel.cl File Reference

Defines the functions used for prediction for the C-SVM using the OpenCL backend. More...

Functions
__kernel void	device_kernel_w_linear (__global real_type *w_d, __global real_type *data_d, __global real_type *data_last_d, __global real_type *alpha_d, const kernel_index_type num_data_points, const kernel_index_type num_features)
	Calculate the w vector to speed up the prediction of the labels for data points using the linear kernel function. More...
__kernel void	device_kernel_predict_polynomial (__global real_type *out_d, __global const real_type *data_d, __global const real_type *data_last_d, __global const real_type *alpha_d, const kernel_index_type num_data_points, __global const real_type *points, const kernel_index_type num_predict_points, const kernel_index_type num_features, const int degree, const real_type gamma, const real_type coef0)
	Predicts the labels for data points using the polynomial kernel function. More...
__kernel void	device_kernel_predict_rbf (__global real_type *out_d, __global const real_type *data_d, __global const real_type *data_last_d, __global const real_type *alpha_d, const kernel_index_type num_data_points, __global const real_type *points, const kernel_index_type num_predict_points, const kernel_index_type num_features, const real_type gamma)
	Predicts the labels for data points using the radial basis functions kernel function. More...

Detailed Description

Defines the functions used for prediction for the C-SVM using the OpenCL backend.

Author

Alexander Van Craen

Marcel Breyer

Copyright

2018-today The PLSSVM project - All Rights Reserved

License

This file is part of the PLSSVM project which is released under the MIT license. See the LICENSE.md file in the project root for full license information.

Function Documentation

device_kernel_w_linear()

__kernel void device_kernel_w_linear	(	__global real_type *	w_d,
		__global real_type *	data_d,
		__global real_type *	data_last_d,
		__global real_type *	alpha_d,
		const kernel_index_type	num_data_points,
		const kernel_index_type	num_features
	)

Calculate the w vector to speed up the prediction of the labels for data points using the linear kernel function.

Supports multi-GPU execution.

Template Parameters

real_type

the type of the data

Parameters

[out]	w_d	the w vector to assemble
[in]	data_d	the one-dimension support vector matrix
[in]	data_last_d	the last row of the support vector matrix
[in]	alpha_d	the previously calculated weight for each data point
[in]	num_data_points	the total number of support vectors
[in]	num_features	the number of features per support vector

device_kernel_predict_polynomial()

__kernel void device_kernel_predict_polynomial	(	__global real_type *	out_d,
		__global const real_type *	data_d,
		__global const real_type *	data_last_d,
		__global const real_type *	alpha_d,
		const kernel_index_type	num_data_points,
		__global const real_type *	points,
		const kernel_index_type	num_predict_points,
		const kernel_index_type	num_features,
		const int	degree,
		const real_type	gamma,
		const real_type	coef0
	)

Predicts the labels for data points using the polynomial kernel function.

Currently only single GPU execution is supported.

Template Parameters

real_type

the type of the data

Parameters

[out]	out_d	the calculated predictions
[in]	data_d	the one-dimension support vector matrix
[in]	data_last_d	the last row of the support vector matrix
[in]	alpha_d	the previously calculated weight for each data point
[in]	num_data_points	the total number of support vectors
[in]	points	the data points to predict
[in]	num_predict_points	the total number of data points to predict
[in]	num_features	the number of features per support vector and point to predict
[in]	degree	the degree parameter used in the polynomial kernel function
[in]	gamma	the gamma parameter used in the polynomial kernel function
[in]	coef0	the coef0 parameter used in the polynomial kernel function

device_kernel_predict_rbf()

__kernel void device_kernel_predict_rbf	(	__global real_type *	out_d,
		__global const real_type *	data_d,
		__global const real_type *	data_last_d,
		__global const real_type *	alpha_d,
		const kernel_index_type	num_data_points,
		__global const real_type *	points,
		const kernel_index_type	num_predict_points,
		const kernel_index_type	num_features,
		const real_type	gamma
	)

Predicts the labels for data points using the radial basis functions kernel function.

Currently only single GPU execution is supported.

Template Parameters

real_type

the type of the data

Parameters

[out]	out_d	the calculated predictions
[in]	data_d	the one-dimension support vector matrix
[in]	data_last_d	the last row of the support vector matrix
[in]	alpha_d	the previously calculated weight for each data point
[in]	num_data_points	the total number of support vectors
[in]	points	the data points to predict
[in]	num_predict_points	the total number of data points to predict
[in]	num_features	the number of features per support vector and point to predict
[in]	gamma	the gamma parameter used in the rbf kernel function