<?php
// This file is part of Moodle - http://moodle.org/
//
// Moodle is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// Moodle is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with Moodle. If not, see <http://www.gnu.org/licenses/>.
/**
* Abstract base target.
*
* @package core_analytics
* @copyright 2016 David Monllao {@link http://www.davidmonllao.com}
* @license http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
*/
namespace core_analytics\local\target;
defined('MOODLE_INTERNAL') || die();
/**
* Abstract base target.
*
* @package core_analytics
* @copyright 2016 David Monllao {@link http://www.davidmonllao.com}
* @license http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
*/
abstract class base extends \core_analytics\calculable {
/**
* This target have linear or discrete values.
*
* @return bool
*/
abstract public function is_linear();
/**
* Returns the analyser class that should be used along with this target.
*
* @return string The full class name as a string
*/
abstract public function get_analyser_class();
/**
* Allows the target to verify that the analysable is a good candidate.
*
* This method can be used as a quick way to discard invalid analysables.
* e.g. Imagine that your analysable don't have students and you need them.
*
* @param \core_analytics\analysable $analysable
* @param bool $fortraining
* @return true|string
*/
abstract public function is_valid_analysable(\core_analytics\analysable $analysable, $fortraining = true);
/**
* Is this sample from the $analysable valid?
*
* @param int $sampleid
* @param \core_analytics\analysable $analysable
* @param bool $fortraining
* @return bool
*/
abstract public function is_valid_sample($sampleid, \core_analytics\analysable $analysable, $fortraining = true);
/**
* Calculates this target for the provided samples.
*
* In case there are no values to return or the provided sample is not applicable just return null.
*
* @param int $sampleid
* @param \core_analytics\analysable $analysable
* @param int|false $starttime Limit calculations to start time
* @param int|false $endtime Limit calculations to end time
* @return float|null
*/
abstract protected function calculate_sample($sampleid, \core_analytics\analysable $analysable, $starttime = false, $endtime = false);
/**
* Can the provided time-splitting method be used on this target?.
*
* Time-splitting methods not matching the target requirements will not be selectable by models based on this target.
*
* @param \core_analytics\local\time_splitting\base $timesplitting
* @return bool
*/
abstract public function can_use_timesplitting(\core_analytics\local\time_splitting\base $timesplitting): bool;
/**
* Is this target generating insights?
*
* Defaults to true.
*
* @return bool
*/
public static function uses_insights() {
return true;
}
/**
* Should the insights of this model be linked from reports?
*
* @return bool
*/
public function link_insights_report(): bool {
return true;
}
/**
* Based on facts (processed by machine learning backends) by default.
*
* @return bool
*/
public static function based_on_assumptions() {
return false;
}
/**
* Update the last analysis time on analysable processed or always.
*
* If you overwrite this method to return false the last analysis time
* will only be recorded in DB when the element successfully analysed. You can
* safely return false for lightweight targets.
*
* @return bool
*/
public function always_update_analysis_time(): bool {
return true;
}
/**
* Suggested actions for a user.
*
* @param \core_analytics\prediction $prediction
* @param bool $includedetailsaction
* @param bool $isinsightuser Force all the available actions to be returned as it the user who
* receives the insight is the one logged in.
* @return \core_analytics\prediction_action[]
*/
public function prediction_actions(\core_analytics\prediction $prediction, $includedetailsaction = false,
$isinsightuser = false) {
global $PAGE;
$predictionid = $prediction->get_prediction_data()->id;
$contextid = $prediction->get_prediction_data()->contextid;
$modelid = $prediction->get_prediction_data()->modelid;
$actions = array();
if ($this->link_insights_report() && $includedetailsaction) {
$predictionurl = new \moodle_url('/report/insights/prediction.php', array('id' => $predictionid));
$detailstext = $this->get_view_details_text();
$actions[] = new \core_analytics\prediction_action(\core_analytics\prediction::ACTION_PREDICTION_DETAILS, $prediction,
$predictionurl, new \pix_icon('t/preview', $detailstext),
$detailstext, false, [], \core_analytics\action::TYPE_NEUTRAL);
}
return $actions;
}
/**
* Suggested bulk actions for a user.
*
* @param \core_analytics\prediction[] $predictions List of predictions suitable for the bulk actions to use.
* @return \core_analytics\bulk_action[] The list of bulk actions.
*/
public function bulk_actions(array $predictions) {
$analyserclass = $this->get_analyser_class();
if ($analyserclass::one_sample_per_analysable()) {
// Default actions are useful / not useful.
$actions = [
\core_analytics\default_bulk_actions::useful(),
\core_analytics\default_bulk_actions::not_useful()
];
} else {
// Accept and not applicable.
$actions = [
\core_analytics\default_bulk_actions::accept(),
\core_analytics\default_bulk_actions::not_applicable()
];
if (!self::based_on_assumptions()) {
// We include incorrectly flagged.
$actions[] = \core_analytics\default_bulk_actions::incorrectly_flagged();
}
}
return $actions;
}
/**
* Adds the JS required to run the bulk actions.
*/
public function add_bulk_actions_js() {
global $PAGE;
$PAGE->requires->js_call_amd('report_insights/actions', 'initBulk', ['.insights-bulk-actions']);
}
/**
* Returns the view details link text.
* @return string
*/
private function get_view_details_text() {
if ($this->based_on_assumptions()) {
$analyserclass = $this->get_analyser_class();
if ($analyserclass::one_sample_per_analysable()) {
$detailstext = get_string('viewinsightdetails', 'analytics');
} else {
$detailstext = get_string('viewdetails', 'analytics');
}
} else {
$detailstext = get_string('viewprediction', 'analytics');
}
return $detailstext;
}
/**
* Callback to execute once a prediction has been returned from the predictions processor.
*
* Note that the analytics_predictions db record is not yet inserted.
*
* @param int $modelid
* @param int $sampleid
* @param int $rangeindex
* @param \context $samplecontext
* @param float|int $prediction
* @param float $predictionscore
* @return void
*/
public function prediction_callback($modelid, $sampleid, $rangeindex, \context $samplecontext, $prediction, $predictionscore) {
return;
}
/**
* Generates insights notifications
*
* @param int $modelid
* @param \context[] $samplecontexts
* @param \core_analytics\prediction[] $predictions
* @return void
*/
public function generate_insight_notifications($modelid, $samplecontexts, array $predictions = []) {
// Delegate the processing of insights to the insights_generator.
$insightsgenerator = new \core_analytics\insights_generator($modelid, $this);
$insightsgenerator->generate($samplecontexts, $predictions);
}
/**
* Returns the list of users that will receive insights notifications.
*
* Feel free to overwrite if you need to but keep in mind that moodle/analytics:listinsights
* or moodle/analytics:listowninsights capability is required to access the list of insights.
*
* @param \context $context
* @return array
*/
public function get_insights_users(\context $context) {
if ($context->contextlevel === CONTEXT_USER) {
if (!has_capability('moodle/analytics:listowninsights', $context, $context->instanceid)) {
$users = [];
} else {
$users = [$context->instanceid => \core_user::get_user($context->instanceid)];
}
} else if ($context->contextlevel >= CONTEXT_COURSE) {
// At course level or below only enrolled users although this is not ideal for
// teachers assigned at category level.
< $users = get_enrolled_users($context, 'moodle/analytics:listinsights');
> $users = get_enrolled_users($context, 'moodle/analytics:listinsights', 0, 'u.*', null, 0, 0, true);
} else {
$users = get_users_by_capability($context, 'moodle/analytics:listinsights');
}
return $users;
}
/**
* URL to the insight.
*
* @param int $modelid
* @param \context $context
* @return \moodle_url
*/
public function get_insight_context_url($modelid, $context) {
return new \moodle_url('/report/insights/insights.php?modelid=' . $modelid . '&contextid=' . $context->id);
}
/**
* The insight notification subject.
*
* This is just a default message, you should overwrite it for a custom insight message.
*
* @param int $modelid
* @param \context $context
* @return string
*/
public function get_insight_subject(int $modelid, \context $context) {
return get_string('insightmessagesubject', 'analytics', $context->get_context_name());
}
/**
* Returns the body message for an insight with multiple predictions.
*
* This default method is executed when the analysable used by the model generates multiple insight
* for each analysable (one_sample_per_analysable === false)
*
* @param \context $context
* @param string $contextname
* @param \stdClass $user
* @param \moodle_url $insighturl
* @return string[] The plain text message and the HTML message
*/
public function get_insight_body(\context $context, string $contextname, \stdClass $user, \moodle_url $insighturl): array {
global $OUTPUT;
$fullmessage = get_string('insightinfomessageplain', 'analytics', $insighturl->out(false));
$fullmessagehtml = $OUTPUT->render_from_template('core_analytics/insight_info_message',
['url' => $insighturl->out(false), 'insightinfomessage' => get_string('insightinfomessagehtml', 'analytics')]
);
return [$fullmessage, $fullmessagehtml];
}
/**
* Returns the body message for an insight for a single prediction.
*
* This default method is executed when the analysable used by the model generates one insight
* for each analysable (one_sample_per_analysable === true)
*
* @param \context $context
* @param \stdClass $user
* @param \core_analytics\prediction $prediction
* @param \core_analytics\action[] $actions Passed by reference to remove duplicate links to actions.
* @return array Plain text msg, HTML message and the main URL for this
* insight (you can return null if you are happy with the
* default insight URL calculated in prediction_info())
*/
public function get_insight_body_for_prediction(\context $context, \stdClass $user, \core_analytics\prediction $prediction,
array &$actions) {
// No extra message by default.
return [FORMAT_PLAIN => '', FORMAT_HTML => '', 'url' => null];
}
/**
* Returns an instance of the child class.
*
* Useful to reset cached data.
*
* @return \core_analytics\base\target
*/
public static function instance() {
return new static();
}
/**
* Defines a boundary to ignore predictions below the specified prediction score.
*
* Value should go from 0 to 1.
*
* @return float
*/
protected function min_prediction_score() {
// The default minimum discards predictions with a low score.
return \core_analytics\model::PREDICTION_MIN_SCORE;
}
/**
* This method determines if a prediction is interesing for the model or not.
*
* @param mixed $predictedvalue
* @param float $predictionscore
* @return bool
*/
public function triggers_callback($predictedvalue, $predictionscore) {
$minscore = floatval($this->min_prediction_score());
if ($minscore < 0) {
debugging(get_class($this) . ' minimum prediction score is below 0, please update it to a value between 0 and 1.');
} else if ($minscore > 1) {
debugging(get_class($this) . ' minimum prediction score is above 1, please update it to a value between 0 and 1.');
}
// We need to consider that targets may not have a min score.
if (!empty($minscore) && floatval($predictionscore) < $minscore) {
return false;
}
return true;
}
/**
* Calculates the target.
*
* Returns an array of values which size matches $sampleids size.
*
* Rows with null values will be skipped as invalid by time splitting methods.
*
* @param array $sampleids
* @param \core_analytics\analysable $analysable
* @param int $starttime
* @param int $endtime
* @return array The format to follow is [userid] = scalar|null
*/
public function calculate($sampleids, \core_analytics\analysable $analysable, $starttime = false, $endtime = false) {
if (!PHPUNIT_TEST && CLI_SCRIPT) {
echo '.';
}
$calculations = [];
foreach ($sampleids as $sampleid => $unusedsampleid) {
// No time limits when calculating the target to train models.
$calculatedvalue = $this->calculate_sample($sampleid, $analysable, $starttime, $endtime);
if (!is_null($calculatedvalue)) {
if ($this->is_linear() &&
($calculatedvalue > static::get_max_value() || $calculatedvalue < static::get_min_value())) {
throw new \coding_exception('Calculated values should be higher than ' . static::get_min_value() .
' and lower than ' . static::get_max_value() . '. ' . $calculatedvalue . ' received');
} else if (!$this->is_linear() && static::is_a_class($calculatedvalue) === false) {
throw new \coding_exception('Calculated values should be one of the target classes (' .
json_encode(static::get_classes()) . '). ' . $calculatedvalue . ' received');
}
}
$calculations[$sampleid] = $calculatedvalue;
}
return $calculations;
}
/**
* Filters out invalid samples for training.
*
* @param int[] $sampleids
* @param \core_analytics\analysable $analysable
* @param bool $fortraining
* @return void
*/
public function filter_out_invalid_samples(&$sampleids, \core_analytics\analysable $analysable, $fortraining = true) {
foreach ($sampleids as $sampleid => $unusedsampleid) {
if (!$this->is_valid_sample($sampleid, $analysable, $fortraining)) {
// Skip it and remove the sample from the list of calculated samples.
unset($sampleids[$sampleid]);
}
}
}
}