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

Public Member Functions
	device_kernel_w_linear (real_type w_d, const real_type data_d, const real_type data_last_d, const real_type alpha_d, const kernel_index_type num_data_points, const kernel_index_type num_features)
	Construct a new device kernel generating the `w` vector used to speedup the prediction when using the linear kernel function. More...

void	operator() (::sycl::id< 1 > index) const
	Function call operator overload performing the actual calculation. More...

Detailed Description

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

T	the type of the data

Constructor & Destructor Documentation

template<typename T >

plssvm::sycl::detail::device_kernel_w_linear< T >::device_kernel_w_linear	(	real_type *	w_d,
		const real_type *	data_d,
		const real_type *	data_last_d,
		const real_type *	alpha_d,
		const kernel_index_type	num_data_points,
		const kernel_index_type	num_features
	)

inline

Construct a new device kernel generating the w vector used to speedup the prediction when using the linear kernel function.

Currently only single GPU execution is supported.

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

template<typename T >

void plssvm::sycl::detail::device_kernel_w_linear< T >::operator() ( ::sycl::id< 1 > index ) const

inline

Function call operator overload performing the actual calculation.

Parameters

[in] index the sycl::id identifying an instance of the functor executing at each point in a sycl::range

The documentation for this class was generated from the following file: