Functions
void	buffer_size (int n, int batch_size, int frequency, int start_leveltrend_len, int start_season_len, int components_len, int error_len, int leveltrend_coef_shift, int season_coef_shift)

void	fit (const raft::handle_t &handle, int n, int batch_size, int frequency, int start_periods, ML::SeasonalType seasonal, float epsilon, float data, float level_d, float trend_d, float season_d, float *error_d)

void	fit (const raft::handle_t &handle, int n, int batch_size, int frequency, int start_periods, ML::SeasonalType seasonal, double epsilon, double data, double level_d, double trend_d, double season_d, double *error_d)

void	forecast (const raft::handle_t &handle, int n, int batch_size, int frequency, int h, ML::SeasonalType seasonal, float level_d, float trend_d, float season_d, float forecast_d)

void	forecast (const raft::handle_t &handle, int n, int batch_size, int frequency, int h, ML::SeasonalType seasonal, double level_d, double trend_d, double season_d, double forecast_d)

Function Documentation

◆ buffer_size()

void ML::HoltWinters::buffer_size	(	int	n,
		int	batch_size,
		int	frequency,
		int *	start_leveltrend_len,
		int *	start_season_len,
		int *	components_len,
		int *	error_len,
		int *	leveltrend_coef_shift,
		int *	season_coef_shift
	)

Provides buffer sizes for HoltWinters algorithm

Parameters

[in]	n	n_samples in time-series
[in]	batch_size	number of time-series in X
[in]	frequency	number of periods in a season of the time-series
[out]	start_leveltrend_len	pointer which will hold the length of the level/trend array buffers
[out]	start_season_len	pointer which will hold the length of the seasonal array buffer
[out]	components_len	pointer which will hold the length of all three components
[out]	error_len	pointer which will hold the length of the SSE Error
[out]	leveltrend_coef_shift	pointer which will hold the offset to level/trend arrays
[out]	season_coef_shift	pointer which will hold the offset to season array

Fits a HoltWinters model

Parameters

[in]	handle	cuml handle to use across the algorithm
[in]	n	n_samples in time-series
[in]	batch_size	number of time-series in X
[in]	frequency	number of periods in a season of the time-series
[in]	start_periods	number of seasons to be used for seasonal seed values
[in]	seasonal	type of seasonal component (ADDITIVE or MULTIPLICATIVE)
[in]	epsilon	the error tolerance value for optimization
[in]	data	device pointer to the data to fit on
[out]	level_d	device pointer to array which will hold level components
[out]	trend_d	device pointer to array which will hold trend components
[out]	season_d	device pointer to array which will hold season components
[out]	error_d	device pointer to array which will hold training SSE error

void ML::HoltWinters::forecast	(	const raft::handle_t &	handle,
		int	n,
		int	batch_size,
		int	frequency,
		int	h,
		ML::SeasonalType	seasonal,
		double *	level_d,
		double *	trend_d,
		double *	season_d,
		double *	forecast_d
	)

void ML::HoltWinters::forecast	(	const raft::handle_t &	handle,
		int	n,
		int	batch_size,
		int	frequency,
		int	h,
		ML::SeasonalType	seasonal,
		float *	level_d,
		float *	trend_d,
		float *	season_d,
		float *	forecast_d
	)

Forecasts future points from fitted HoltWinters model

Parameters

[in]	handle	cuml handle to use across the algorithm
[in]	n	n_samples in time-series
[in]	batch_size	number of time-series in X
[in]	frequency	number of periods in a season of the time-series
[in]	h	number of future points to predict in the time-series
[in]	seasonal	type of seasonal component (ADDITIVE or MULTIPLICATIVE)
[out]	level_d	device pointer to array which holds level components
[out]	trend_d	device pointer to array which holds trend components
[out]	season_d	device pointer to array which holds season components
[out]	forecast_d	device pointer to array which will hold the forecast points