processor.php 14 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
<?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/>.

/**
 * Python predictions processor
 *
 * @package   mlbackend_python
 * @copyright 2016 David Monllao {@link http://www.davidmonllao.com}
 * @license   http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
 */

namespace mlbackend_python;

defined('MOODLE_INTERNAL') || die();

/**
 * Python predictions processor.
 *
 * @package   mlbackend_python
 * @copyright 2016 David Monllao {@link http://www.davidmonllao.com}
 * @license   http://www.gnu.org/copyleft/gpl.html GNU GPL v3 or later
 */
36
class processor implements  \core_analytics\classifier, \core_analytics\regressor, \core_analytics\packable {
37

38
39
40
    /**
     * The required version of the python package that performs all calculations.
     */
41
    const REQUIRED_PIP_PACKAGE_VERSION = '2.1.0';
42

43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
    /**
     * The path to the Python bin.
     *
     * @var string
     */
    protected $pathtopython;

    /**
     * The constructor.
     */
    public function __construct() {
        global $CFG;

        // Set the python location if there is a value.
        if (!empty($CFG->pathtopython)) {
            $this->pathtopython = $CFG->pathtopython;
        }
    }

62
63
64
    /**
     * Is the plugin ready to be used?.
     *
65
     * @return bool|string Returns true on success, a string detailing the error otherwise
66
     */
67
    public function is_ready() {
68
69
70
71
        if (empty($this->pathtopython)) {
            $settingurl = new \moodle_url('/admin/settings.php', array('section' => 'systempaths'));
            return get_string('pythonpathnotdefined', 'mlbackend_python', $settingurl->out());
        }
72

73
        // Check the installed pip package version.
74
        $cmd = "{$this->pathtopython} -m moodlemlbackend.version";
75
76
77
78
79
80

        $output = null;
        $exitcode = null;
        // Execute it sending the standard error to $output.
        $result = exec($cmd . ' 2>&1', $output, $exitcode);

81
82
83
        $vercheck = self::check_pip_package_version($result);

        if ($vercheck === 0) {
84
85
86
87
88
89
90
91
            return true;
        }

        if ($exitcode != 0) {
            return get_string('pythonpackagenotinstalled', 'mlbackend_python', $cmd);
        }

        if ($result) {
92
93
94
95
96
97
98
99
100
101
102
            $a = [
                'installed' => $result,
                'required' => self::REQUIRED_PIP_PACKAGE_VERSION,
            ];

            if ($vercheck < 0) {
                return get_string('packageinstalledshouldbe', 'mlbackend_python', $a);

            } else if ($vercheck > 0) {
                return get_string('packageinstalledtoohigh', 'mlbackend_python', $a);
            }
103
104
105
106
107
        }

        return get_string('pythonpackagenotinstalled', 'mlbackend_python', $cmd);
    }

108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
    /**
     * Delete the model version output directory.
     *
     * @param string $uniqueid
     * @param string $modelversionoutputdir
     * @return null
     */
    public function clear_model($uniqueid, $modelversionoutputdir) {
        remove_dir($modelversionoutputdir);
    }

    /**
     * Delete the model output directory.
     *
     * @param string $modeloutputdir
     * @return null
     */
    public function delete_output_dir($modeloutputdir) {
        remove_dir($modeloutputdir);
    }

129
130
131
132
133
134
135
136
    /**
     * Trains a machine learning algorithm with the provided dataset.
     *
     * @param string $uniqueid
     * @param \stored_file $dataset
     * @param string $outputdir
     * @return \stdClass
     */
137
    public function train_classification($uniqueid, \stored_file $dataset, $outputdir) {
138

139
        // Obtain the physical route to the file.
140
141
        $datasetpath = $this->get_file_path($dataset);

142
        $cmd = "{$this->pathtopython} -m moodlemlbackend.training " .
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
            escapeshellarg($uniqueid) . ' ' .
            escapeshellarg($outputdir) . ' ' .
            escapeshellarg($datasetpath);

        if (!PHPUNIT_TEST && CLI_SCRIPT) {
            debugging($cmd, DEBUG_DEVELOPER);
        }

        $output = null;
        $exitcode = null;
        $result = exec($cmd, $output, $exitcode);

        if (!$result) {
            throw new \moodle_exception('errornopredictresults', 'analytics');
        }

        if (!$resultobj = json_decode($result)) {
            throw new \moodle_exception('errorpredictwrongformat', 'analytics', '', json_last_error_msg());
        }

        if ($exitcode != 0) {
164
165
166
167
168
169
170
171
172
173
174
            if (!empty($resultobj->errors)) {
                $errors = $resultobj->errors;
                if (is_array($errors)) {
                    $errors = implode(', ', $errors);
                }
            } else if (!empty($resultobj->info)) {
                // Show info if no errors are returned.
                $errors = $resultobj->info;
                if (is_array($errors)) {
                    $errors = implode(', ', $errors);
                }
175
176
            }
            $resultobj->info = array(get_string('errorpredictionsprocessor', 'analytics', $errors));
177
178
179
180
181
        }

        return $resultobj;
    }

182
    /**
183
     * Classifies the provided dataset samples.
184
185
186
187
188
189
     *
     * @param string $uniqueid
     * @param \stored_file $dataset
     * @param string $outputdir
     * @return \stdClass
     */
190
    public function classify($uniqueid, \stored_file $dataset, $outputdir) {
191

192
        // Obtain the physical route to the file.
193
194
        $datasetpath = $this->get_file_path($dataset);

195
        $cmd = "{$this->pathtopython} -m moodlemlbackend.prediction " .
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
            escapeshellarg($uniqueid) . ' ' .
            escapeshellarg($outputdir) . ' ' .
            escapeshellarg($datasetpath);

        if (!PHPUNIT_TEST && CLI_SCRIPT) {
            debugging($cmd, DEBUG_DEVELOPER);
        }

        $output = null;
        $exitcode = null;
        $result = exec($cmd, $output, $exitcode);

        if (!$result) {
            throw new \moodle_exception('errornopredictresults', 'analytics');
        }

        if (!$resultobj = json_decode($result)) {
            throw new \moodle_exception('errorpredictwrongformat', 'analytics', '', json_last_error_msg());
        }

        if ($exitcode != 0) {
217
218
219
220
221
222
223
224
225
226
227
            if (!empty($resultobj->errors)) {
                $errors = $resultobj->errors;
                if (is_array($errors)) {
                    $errors = implode(', ', $errors);
                }
            } else if (!empty($resultobj->info)) {
                // Show info if no errors are returned.
                $errors = $resultobj->info;
                if (is_array($errors)) {
                    $errors = implode(', ', $errors);
                }
228
229
            }
            $resultobj->info = array(get_string('errorpredictionsprocessor', 'analytics', $errors));
230
231
232
233
234
        }

        return $resultobj;
    }

235
    /**
236
     * Evaluates this processor classification model using the provided supervised learning dataset.
237
238
239
240
241
242
     *
     * @param string $uniqueid
     * @param float $maxdeviation
     * @param int $niterations
     * @param \stored_file $dataset
     * @param string $outputdir
243
     * @param  string $trainedmodeldir
244
245
     * @return \stdClass
     */
246
247
    public function evaluate_classification($uniqueid, $maxdeviation, $niterations, \stored_file $dataset,
            $outputdir, $trainedmodeldir) {
248

249
        // Obtain the physical route to the file.
250
251
        $datasetpath = $this->get_file_path($dataset);

252
        $cmd = "{$this->pathtopython} -m moodlemlbackend.evaluation " .
253
254
255
256
257
258
259
            escapeshellarg($uniqueid) . ' ' .
            escapeshellarg($outputdir) . ' ' .
            escapeshellarg($datasetpath) . ' ' .
            escapeshellarg(\core_analytics\model::MIN_SCORE) . ' ' .
            escapeshellarg($maxdeviation) . ' ' .
            escapeshellarg($niterations);

260
261
262
263
        if ($trainedmodeldir) {
            $cmd .= ' ' . escapeshellarg($trainedmodeldir);
        }

264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
        if (!PHPUNIT_TEST && CLI_SCRIPT) {
            debugging($cmd, DEBUG_DEVELOPER);
        }

        $output = null;
        $exitcode = null;
        $result = exec($cmd, $output, $exitcode);

        if (!$result) {
            throw new \moodle_exception('errornopredictresults', 'analytics');
        }

        if (!$resultobj = json_decode($result)) {
            throw new \moodle_exception('errorpredictwrongformat', 'analytics', '', json_last_error_msg());
        }

        return $resultobj;
    }

283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
    /**
     * Exports the machine learning model.
     *
     * @throws \moodle_exception
     * @param  string $uniqueid  The model unique id
     * @param  string $modeldir  The directory that contains the trained model.
     * @return string            The path to the directory that contains the exported model.
     */
    public function export(string $uniqueid, string $modeldir) : string {

        // We include an exporttmpdir as we want to be sure that the file is not deleted after the
        // python process finishes.
        $exporttmpdir = make_request_directory('mlbackend_python_export');

        $cmd = "{$this->pathtopython} -m moodlemlbackend.export " .
            escapeshellarg($uniqueid) . ' ' .
            escapeshellarg($modeldir) . ' ' .
            escapeshellarg($exporttmpdir);

        if (!PHPUNIT_TEST && CLI_SCRIPT) {
            debugging($cmd, DEBUG_DEVELOPER);
        }

        $output = null;
        $exitcode = null;
        $exportdir = exec($cmd, $output, $exitcode);

        if ($exitcode != 0) {
            throw new \moodle_exception('errorexportmodelresult', 'analytics');
        }

        if (!$exportdir) {
            throw new \moodle_exception('errorexportmodelresult', 'analytics');
        }

        return $exportdir;
    }

    /**
     * Imports the provided machine learning model.
     *
     * @param  string $uniqueid The model unique id
     * @param  string $modeldir  The directory that will contain the trained model.
     * @param  string $importdir The directory that contains the files to import.
     * @return bool Success
     */
    public function import(string $uniqueid, string $modeldir, string $importdir) : bool {

        $cmd = "{$this->pathtopython} -m moodlemlbackend.import " .
            escapeshellarg($uniqueid) . ' ' .
            escapeshellarg($modeldir) . ' ' .
            escapeshellarg($importdir);

        if (!PHPUNIT_TEST && CLI_SCRIPT) {
            debugging($cmd, DEBUG_DEVELOPER);
        }

        $output = null;
        $exitcode = null;
        $success = exec($cmd, $output, $exitcode);

        if ($exitcode != 0) {
            throw new \moodle_exception('errorimportmodelresult', 'analytics');
        }

        if (!$success) {
            throw new \moodle_exception('errorimportmodelresult', 'analytics');
        }

        return $success;
    }

355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
    /**
     * Train this processor regression model using the provided supervised learning dataset.
     *
     * @throws new \coding_exception
     * @param string $uniqueid
     * @param \stored_file $dataset
     * @param string $outputdir
     * @return \stdClass
     */
    public function train_regression($uniqueid, \stored_file $dataset, $outputdir) {
        throw new \coding_exception('This predictor does not support regression yet.');
    }

    /**
     * Estimates linear values for the provided dataset samples.
     *
     * @throws new \coding_exception
     * @param string $uniqueid
     * @param \stored_file $dataset
     * @param mixed $outputdir
     * @return void
     */
    public function estimate($uniqueid, \stored_file $dataset, $outputdir) {
        throw new \coding_exception('This predictor does not support regression yet.');
    }

    /**
     * Evaluates this processor regression model using the provided supervised learning dataset.
     *
     * @throws new \coding_exception
     * @param string $uniqueid
     * @param float $maxdeviation
     * @param int $niterations
     * @param \stored_file $dataset
     * @param string $outputdir
390
     * @param  string $trainedmodeldir
391
392
     * @return \stdClass
     */
393
394
    public function evaluate_regression($uniqueid, $maxdeviation, $niterations, \stored_file $dataset,
            $outputdir, $trainedmodeldir) {
395
396
397
        throw new \coding_exception('This predictor does not support regression yet.');
    }

398
399
400
401
402
403
    /**
     * Returns the path to the dataset file.
     *
     * @param \stored_file $file
     * @return string
     */
404
405
406
407
408
    protected function get_file_path(\stored_file $file) {
        // From moodle filesystem to the local file system.
        // This is not ideal, but there is no read access to moodle filesystem files.
        return $file->copy_content_to_temp('core_analytics');
    }
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443

    /**
     * Check that the given package version can be used and return the error status.
     *
     * When evaluating the version, we assume the sematic versioning scheme as described at
     * https://semver.org/.
     *
     * @param string $actual The actual Python package version
     * @param string $required The required version of the package
     * @return int -1 = actual version is too low, 1 = actual version too high, 0 = actual version is ok
     */
    public static function check_pip_package_version($actual, $required = self::REQUIRED_PIP_PACKAGE_VERSION) {

        if (empty($actual)) {
            return -1;
        }

        if (version_compare($actual, $required, '<')) {
            return -1;
        }

        $parts = explode('.', $required);
        $requiredapiver = reset($parts);

        $parts = explode('.', $actual);
        $actualapiver = reset($parts);

        if ($requiredapiver > 0 || $actualapiver > 1) {
            if (version_compare($actual, $requiredapiver + 1, '>=')) {
                return 1;
            }
        }

        return 0;
    }
444
}