Runs an analysis of the site. More...

Public Member Functions
	__construct (\core_analytics\local\analyser\base $analyser, bool $includetarget, core_analytics\local\analysis\result $result)
	Constructor. More...

	calculate (\core_analytics\local\time_splitting\base $timesplitting, array &$sampleids, array $ranges, core_analytics\local\target\base $target)
	Calculates indicators and targets. More...

	get_headers (\core_analytics\local\time_splitting\base $timesplitting, core_analytics\local\target\base $target)
	Returns the headers for the csv file based on the indicators and the target. More...

	process_analysable (\core_analytics\analysable $analysable)
	Processes an analysable. More...

	run (array $contexts=[])
	Runs the analysis. More...

Static Public Member Functions
static	clean_time_splitting_id ($timesplittingid)
	Remove all possibly problematic chars from the time splitting method id (id = its full class name). More...

static	fill_firstanalyses_cache (int $modelid, ?int $analysableid=null)
	Fills a cache containing the first time each analysable in the provided model was analysed. More...

Protected Member Functions
	add_context_metadata (\core_analytics\local\time_splitting\base $timesplitting, array &$dataset, core_analytics\local\target\base $target)
	Adds dataset context info. More...

	calculate_indicators (\core_analytics\local\time_splitting\base $timesplitting, array $sampleids, array $ranges)
	Calculates indicators. More...

	fill_dataset (\core_analytics\local\time_splitting\base $timesplitting, array &$dataset, ?array $calculatedtarget=null)
	Adds time range indicators and the target to each sample. More...

	filter_out_prediction_samples_and_ranges (array &$sampleids, array &$ranges, core_analytics\local\time_splitting\base $timesplitting)
	Filters out samples that have already been used for prediction. More...

	filter_out_train_samples (array &$sampleids, core_analytics\local\time_splitting\base $timesplitting)
	Filters out samples that have already been used for training. More...

	get_processed_analysables ()
	Get analysables that have been already processed. More...

	process_time_splitting (\core_analytics\local\time_splitting\base $timesplitting, core_analytics\analysable $analysable, core_analytics\local\target\base $target)
	Processes the analysable samples using the provided time splitting method. More...

	save_prediction_samples (array $sampleids, array $ranges, core_analytics\local\time_splitting\base $timesplitting, ?\stdClass $predictsamplesrecord=null)
	Saves samples that have just been used for prediction. More...

	save_train_samples (array $sampleids, core_analytics\local\time_splitting\base $timesplitting)
	Saves samples that have just been used for training. More...

	update_analysable_analysed_time (array $processedanalysables, int $analysableid)
	Updates the analysable analysis time. More...

Detailed Description

Runs an analysis of the site.

Copyright: 2019 David Monllao http://www.davidmonllao.com

License: http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later

Constructor & Destructor Documentation

◆ __construct()

core_analytics\analysis::__construct	(	\core_analytics\local\analyser\base	$analyser,
		bool	$includetarget,
		core_analytics\local\analysis\result	$result
	)

Constructor.

Parameters

core_analytics\local\analyser\base	$analyser
bool	$includetarget	Whether to calculate the target or not.
core_analytics\local\analysis\result	$result

Member Function Documentation

◆ add_context_metadata()

core_analytics\analysis::add_context_metadata	(	\core_analytics\local\time_splitting\base	$timesplitting,
		array &	$dataset,
		core_analytics\local\target\base	$target
	)

protected

Adds dataset context info.

The final dataset document will look like this:

metadata1,metadata2,metadata3,..... value1, value2, value3,.....

header1,header2,header3,header4,..... stud1value1,stud1value2,stud1value3,stud1value4,..... stud2value1,stud2value2,stud2value3,stud2value4,.....

.....

Parameters

core_analytics\local\time_splitting\base	$timesplitting
array	$dataset
core_analytics\local\target\base	$target

Return values

null

◆ calculate()

core_analytics\analysis::calculate	(	\core_analytics\local\time_splitting\base	$timesplitting,
		array &	$sampleids,
		array	$ranges,
		core_analytics\local\target\base	$target
	)

Calculates indicators and targets.

Parameters

core_analytics\local\time_splitting\base	$timesplitting
array	$sampleids
array	$ranges
core_analytics\local\target\base	$target

Return values

array|null

◆ calculate_indicators()

core_analytics\analysis::calculate_indicators	(	\core_analytics\local\time_splitting\base	$timesplitting,
		array	$sampleids,
		array	$ranges
	)

protected

Calculates indicators.

Parameters

core_analytics\local\time_splitting\base	$timesplitting
array	$sampleids
array	$ranges

Return values

array

◆ clean_time_splitting_id()

static core_analytics\analysis::clean_time_splitting_id ( $timesplittingid )

static

Remove all possibly problematic chars from the time splitting method id (id = its full class name).

Parameters

string $timesplittingid

Return values

string

◆ fill_dataset()

core_analytics\analysis::fill_dataset	(	\core_analytics\local\time_splitting\base	$timesplitting,
		array &	$dataset,
		?array	$calculatedtarget = `null`
	)

protected

Adds time range indicators and the target to each sample.

This will identify the sample as belonging to a specific range.

Parameters

core_analytics\local\time_splitting\base	$timesplitting
array	$dataset
array \| null	$calculatedtarget

Return values

null

◆ fill_firstanalyses_cache()

static core_analytics\analysis::fill_firstanalyses_cache	(	int	$modelid,
		?int	$analysableid = `null`
	)

static

Fills a cache containing the first time each analysable in the provided model was analysed.

Parameters

int	$modelid
int \| null	$analysableid

Return values

null

◆ filter_out_prediction_samples_and_ranges()

core_analytics\analysis::filter_out_prediction_samples_and_ranges	(	array &	$sampleids,
		array &	$ranges,
		core_analytics\local\time_splitting\base	$timesplitting
	)

protected

Filters out samples that have already been used for prediction.

Parameters

int[]	$sampleids
array	$ranges
core_analytics\local\time_splitting\base	$timesplitting

Return values

stdClass|null The analytics_predict_samples record or null

◆ filter_out_train_samples()

core_analytics\analysis::filter_out_train_samples	(	array &	$sampleids,
		core_analytics\local\time_splitting\base	$timesplitting
	)

protected

Filters out samples that have already been used for training.

Parameters

int[]	$sampleids
core_analytics\local\time_splitting\base	$timesplitting

Return values

null

◆ get_headers()

core_analytics\analysis::get_headers	(	\core_analytics\local\time_splitting\base	$timesplitting,
		core_analytics\local\target\base	$target
	)

Returns the headers for the csv file based on the indicators and the target.

Parameters

core_analytics\local\time_splitting\base	$timesplitting
core_analytics\local\target\base	$target

Return values

string[]

◆ get_processed_analysables()

core_analytics\analysis::get_processed_analysables ( )

protected

Get analysables that have been already processed.

Return values

stdClass[]

◆ process_analysable()

core_analytics\analysis::process_analysable ( \core_analytics\analysable $analysable )

Processes an analysable.

This method returns the general analysable status, an array of files by time splitting method and an error message if there is any problem.

Parameters

core_analytics\analysable $analysable

Return values

stdClass[] Results objects by time splitting method

◆ process_time_splitting()

core_analytics\analysis::process_time_splitting	(	\core_analytics\local\time_splitting\base	$timesplitting,
		core_analytics\analysable	$analysable,
		core_analytics\local\target\base	$target
	)

protected

Processes the analysable samples using the provided time splitting method.

Parameters

core_analytics\local\time_splitting\base	$timesplitting
core_analytics\analysable	$analysable
core_analytics\local\target\base	$target

Return values

stdClass Results object.

◆ run()

core_analytics\analysis::run ( array $contexts = [] )

Runs the analysis.

Parameters

context[] $contexts Restrict the analysis to these contexts. No context restrictions if null.

Return values

null

◆ save_prediction_samples()

core_analytics\analysis::save_prediction_samples	(	array	$sampleids,
		array	$ranges,
		core_analytics\local\time_splitting\base	$timesplitting,
		?\stdClass	$predictsamplesrecord = `null`
	)

protected

Saves samples that have just been used for prediction.

Parameters

int[]	$sampleids
array	$ranges
core_analytics\local\time_splitting\base	$timesplitting
stdClass \| null	$predictsamplesrecord	The existing record or null if there is no record yet.

Return values

null

◆ save_train_samples()

core_analytics\analysis::save_train_samples	(	array	$sampleids,
		core_analytics\local\time_splitting\base	$timesplitting
	)

protected

Saves samples that have just been used for training.

Parameters

int[]	$sampleids
core_analytics\local\time_splitting\base	$timesplitting

Return values

null

◆ update_analysable_analysed_time()

core_analytics\analysis::update_analysable_analysed_time	(	array	$processedanalysables,
		int	$analysableid
	)

protected

Updates the analysable analysis time.

Parameters

array	$processedanalysables
int	$analysableid

Return values

null

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

analytics/classes/analysis.php

Public Member Functions

Static Public Member Functions

Protected Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ __construct()

Member Function Documentation

◆ add_context_metadata()

The final dataset document will look like this:

.....

◆ calculate()

◆ calculate_indicators()

◆ clean_time_splitting_id()

◆ fill_dataset()

◆ fill_firstanalyses_cache()

◆ filter_out_prediction_samples_and_ranges()

◆ filter_out_train_samples()

◆ get_headers()

◆ get_processed_analysables()

◆ process_analysable()

◆ process_time_splitting()

◆ run()

◆ save_prediction_samples()

◆ save_train_samples()

◆ update_analysable_analysed_time()