__init__.py 56.3 KB
Newer Older
1
2
# Event source for MAGIC calibrated data files.
# Requires uproot package (https://github.com/scikit-hep/uproot).
Moritz Huetten's avatar
Moritz Huetten committed
3
import logging
4
5
6
7

import glob
import re

8
import scipy
9
10
11
12
13
14
import numpy as np
import scipy.interpolate

from astropy import units as u
from astropy.time import Time
from ctapipe.io.eventsource import EventSource
15
from ctapipe.io.containers import DataContainer, EventAndMonDataContainer, TelescopePointingContainer, MonitoringCameraContainer, PedestalContainer, PixelStatusContainer
16
17
18
19
from ctapipe.instrument import TelescopeDescription, SubarrayDescription, OpticsDescription, CameraGeometry

__all__ = ['MAGICEventSource']

Moritz Huetten's avatar
Moritz Huetten committed
20
logger = logging.getLogger(__name__)
21

Ievgen Vovk's avatar
Ievgen Vovk committed
22

23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
class MAGICEventSource(EventSource):
    """
    EventSource for MAGIC calibrated data.

    This class operates with the MAGIC data run-wise. This means that the files
    corresponding to the same data run are loaded and processed together.
    """
    _count = 0

    def __init__(self, config=None, tool=None, **kwargs):
        """
        Constructor

        Parameters
        ----------
        config: traitlets.loader.Config
            Configuration specified by config file or cmdline arguments.
            Used to set traitlet values.
            Set to None if no configuration to pass.
        tool: ctapipe.core.Tool
            Tool executable that is calling this component.
            Passes the correct logger to the component.
            Set to None if no Tool to pass.
        kwargs: dict
            Additional parameters to be passed.
            NOTE: The file mask of the data to read can be passed with
            the 'input_url' parameter.
        """

        try:
            import uproot
        except ImportError:
            msg = "The `uproot` python module is required to access the MAGIC data"
            self.log.error(msg)
            raise

59
        self.file_list = glob.glob(kwargs['input_url'])
60
61
        if len(self.file_list) == 0:
            raise ValueError("Unreadable or wrong wildcard file path given.")
62
        self.file_list.sort()
63
64
65
66
67

        # EventSource can not handle file wild cards as input_url
        # To overcome this we substitute the input_url with first file matching
        # the specified file mask.
        del kwargs['input_url']
68
        super().__init__(input_url=self.file_list[0], **kwargs)
69
70

        # Retrieving the list of run numbers corresponding to the data files
Moritz Huetten's avatar
Moritz Huetten committed
71
72
        run_info = list(map(self._get_run_info_from_name, self.file_list))
        run_numbers = [i[0] for i in run_info]
73
        is_mc_runs = [i[1] for i in run_info]
Moritz Huetten's avatar
Moritz Huetten committed
74
75

        self.run_numbers, indices = np.unique(run_numbers, return_index=True)
76
77
78
79
80
81
        is_mc_runs = [is_mc_runs[i] for i in indices]
        is_mc_runs = np.unique(is_mc_runs)
        # Checking if runt type (data/MC) is consistent:
        if len(is_mc_runs) > 1:
            raise ValueError("Loaded files contain data and MC runs. Please load only data OR Monte Carlos.")
        self.is_mc = is_mc_runs[0]
82
83
84
85
86
87
88
89
90
91
92
93
94

        # # Setting up the current run with the first run present in the data
        # self.current_run = self._set_active_run(run_number=0)
        self.current_run = None

        # MAGIC telescope positions in m wrt. to the center of CTA simulations
        self.magic_tel_positions = {
            1: [-27.24, -146.66, 50.00] * u.m,
            2: [-96.44, -96.77, 51.00] * u.m
        }
        # MAGIC telescope description
        optics = OpticsDescription.from_name('MAGIC')
        geom = CameraGeometry.from_name('MAGICCam')
95
        self.magic_tel_description = TelescopeDescription(name='MAGIC', tel_type='MAGIC', optics=optics, camera=geom)
96
        self.magic_tel_descriptions = {1: self.magic_tel_description, 2: self.magic_tel_description}
97
        self._subarray_info = SubarrayDescription('MAGIC', self.magic_tel_positions, self.magic_tel_descriptions)
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135

    @staticmethod
    def is_compatible(file_mask):
        """
        This method checks if the specified file mask corresponds
        to MAGIC data files. The result will be True only if all
        the files are of ROOT format and contain an 'Events' tree.

        Parameters
        ----------
        file_mask: str
            A file mask to check

        Returns
        -------
        bool:
            True if the masked files are MAGIC data runs, False otherwise.

        """

        is_magic_root_file = True

        file_list = glob.glob(file_mask)

        for file_path in file_list:
            try:
                import uproot

                try:
                    with uproot.open(file_path) as input_data:
                        if 'Events' not in input_data:
                            is_magic_root_file = False
                except ValueError:
                    # uproot raises ValueError if the file is not a ROOT file
                    is_magic_root_file = False
                    pass

            except ImportError:
Ievgen Vovk's avatar
Ievgen Vovk committed
136
                if re.match(r'.+_m\d_.+root', file_path.lower()) is None:
137
138
139
140
141
                    is_magic_root_file = False

        return is_magic_root_file

    @staticmethod
Moritz Huetten's avatar
Moritz Huetten committed
142
    def _get_run_info_from_name(file_name):
143
        """
Moritz Huetten's avatar
Moritz Huetten committed
144
145
        This internal method extracts the run number and 
        type (data/MC) from the specified file name.
146
147
148
149
150
151
152
153
154
155
156
157

        Parameters
        ----------
        file_name: str
            A file name to process.

        Returns
        -------
        int:
            A run number of the file.
        """

Moritz Huetten's avatar
Moritz Huetten committed
158
        mask_data = r".*\d+_M\d+_(\d+)\.\d+_Y_.*"
Moritz Huetten's avatar
Moritz Huetten committed
159
160
161
162
163
164
165
166
167
168
169
        mask_mc = r".*_M\d_za\d+to\d+_\d_(\d+)_Y_.*"
        mask_mc_alt = r".*_M\d_\d_(\d+)_.*"
        if len(re.findall(mask_data, file_name)) > 0:
            parsed_info = re.findall(mask_data, file_name)
            is_mc = False
        elif len(re.findall(mask_mc, file_name)) > 0:
            parsed_info = re.findall(mask_mc, file_name)
            is_mc = True
        else:
            parsed_info = re.findall(mask_mc_alt, file_name)
            is_mc = True
170
171
172
173

        try:
            run_number = int(parsed_info[0])
        except IndexError:
Moritz Huetten's avatar
Moritz Huetten committed
174
            raise IndexError('Can not identify the run number and type (data/MC) of the file {:s}'.format(file_name))
175

Moritz Huetten's avatar
Moritz Huetten committed
176
        return run_number, is_mc
177

Moritz Huetten's avatar
Moritz Huetten committed
178
    def _set_active_run(self, run_number):
179
180
181
182
183
184
185
186
187
188
        """
        This internal method sets the run that will be used for data loading.

        Parameters
        ----------
        run_number: int
            The run number to use.

        Returns
        -------
Moritz Huetten's avatar
Moritz Huetten committed
189
        MarsRun:
Moritz Huetten's avatar
Moritz Huetten committed
190
            The run to use
191
192
193
194
195
196
197
198
        """

        input_path = '/'.join(self.input_url.split('/')[:-1])
        this_run_mask = input_path + '/*{:d}*root'.format(run_number)

        run = dict()
        run['number'] = run_number
        run['read_events'] = 0
Moritz Huetten's avatar
Moritz Huetten committed
199
        run['data'] = MarsRun(run_file_mask=this_run_mask, filter_list=self.file_list)
200
201
202

        return run

203
204
205
206
    @property
    def subarray(self):
        return self._subarray_info

207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
    def _generator(self):
        """
        The default event generator. Return the stereo event
        generator instance.

        Returns
        -------

        """

        return self._stereo_event_generator()

    def _stereo_event_generator(self):
        """
        Stereo event generator. Yields DataContainer instances, filled
        with the read event data.

        Returns
        -------

        """

        counter = 0

        # Data container - is initialized once, and data is replaced within it after each yield
232
233
234
235
        if not self.is_mc:
            data = EventAndMonDataContainer()
        else:
            data = DataContainer()
236
237
238
239
240
241

        # Telescopes with data:
        tels_in_file = ["m1", "m2"]
        tels_with_data = {1, 2}

        # Loop over the available data runs
242
        for run_number in self.run_numbers:
243
244
245
246
247
248

            # Removing the previously read data run from memory
            if self.current_run is not None:
                if 'data' in self.current_run:
                    del self.current_run['data']

Moritz Huetten's avatar
Moritz Huetten committed
249
            # Setting the new active run (class MarsRun object)
Moritz Huetten's avatar
Moritz Huetten committed
250
            self.current_run = self._set_active_run(run_number)
251

252
253
254
255
256
257
258
259
            # Set monitoring data:
            if not self.is_mc:
                
                monitoring_data = self.current_run['data'].monitoring_data
                
                for tel_i, tel_id in enumerate(tels_in_file):
                    monitoring_camera = MonitoringCameraContainer()
                    pedestal_info = PedestalContainer()
260
261
                    badpixel_info = PixelStatusContainer()
                    
262
263
264
265
266
267
268
269
270
271
272
                    time_tmp = Time(monitoring_data['M{:d}'.format(tel_i + 1)]['PedestalMJD'], scale='utc', format='mjd')
                    pedestal_info.sample_time = Time(time_tmp, format='unix', scale='utc', precision=9)
                    pedestal_info.n_events = 500 # hardcoded number of pedestal events averaged over
                    pedestal_info.charge_mean = []
                    pedestal_info.charge_mean.append(monitoring_data['M{:d}'.format(tel_i + 1)]['PedestalFundamental']['Mean'])
                    pedestal_info.charge_mean.append(monitoring_data['M{:d}'.format(tel_i + 1)]['PedestalFromExtractor']['Mean'])
                    pedestal_info.charge_mean.append(monitoring_data['M{:d}'.format(tel_i + 1)]['PedestalFromExtractorRndm']['Mean'])
                    pedestal_info.charge_std = []
                    pedestal_info.charge_std.append(monitoring_data['M{:d}'.format(tel_i + 1)]['PedestalFundamental']['Rms'])
                    pedestal_info.charge_std.append(monitoring_data['M{:d}'.format(tel_i + 1)]['PedestalFromExtractor']['Rms'])
                    pedestal_info.charge_std.append(monitoring_data['M{:d}'.format(tel_i + 1)]['PedestalFromExtractorRndm']['Rms'])
273
274
275
                    
                    badpixel_info.hardware_failing_pixels = monitoring_data['M{:d}'.format(tel_i + 1)]['badpixelinfo']
                    
276
                    monitoring_camera.pedestal = pedestal_info
277
                    monitoring_camera.pixel_status = badpixel_info
278
279
                    data.mon.tel[tel_i + 1] = monitoring_camera

280
281
282
283
284
285
286
287
288
            # Loop over the events
            for event_i in range(self.current_run['data'].n_stereo_events):
                # Event and run ids
                event_order_number = self.current_run['data'].stereo_ids[event_i][0]
                event_id = self.current_run['data'].event_data['M1']['stereo_event_number'][event_order_number]
                obs_id = self.current_run['number']

                # Reading event data
                event_data = self.current_run['data'].get_stereo_event_data(event_i)
Moritz Huetten's avatar
Moritz Huetten committed
289
                
290
                data.meta = event_data['mars_meta']
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313

                # Event counter
                data.count = counter

                # Setting up the R0 container
                data.r0.obs_id = obs_id
                data.r0.event_id = event_id
                data.r0.tel.clear()

                # Setting up the R1 container
                data.r1.obs_id = obs_id
                data.r1.event_id = event_id
                data.r1.tel.clear()

                # Setting up the DL0 container
                data.dl0.obs_id = obs_id
                data.dl0.event_id = event_id
                data.dl0.tel.clear()

                # Filling the DL1 container with the event data
                for tel_i, tel_id in enumerate(tels_in_file):
                    # Creating the telescope pointing container
                    pointing = TelescopePointingContainer()
314
315
316
317
                    pointing.azimuth = np.deg2rad(event_data['{:s}_pointing_az'.format(tel_id)]) * u.rad
                    pointing.altitude = np.deg2rad(90 - event_data['{:s}_pointing_zd'.format(tel_id)]) * u.rad
                    pointing.ra = np.deg2rad(event_data['{:s}_pointing_ra'.format(tel_id)]) * u.rad
                    pointing.dec = np.deg2rad(event_data['{:s}_pointing_dec'.format(tel_id)]) * u.rad
318
319
320
321
322
323

                    # Adding the pointing container to the event data
                    data.pointing[tel_i + 1] = pointing

                    # Adding event charge and peak positions per pixel
                    data.dl1.tel[tel_i + 1].image = event_data['{:s}_image'.format(tel_id)]
Ievgen Vovk's avatar
Ievgen Vovk committed
324
                    data.dl1.tel[tel_i + 1].pulse_time = event_data['{:s}_pulse_time'.format(tel_id)]
Moritz Huetten's avatar
Moritz Huetten committed
325

326
                if self.is_mc == False:
Moritz Huetten's avatar
Moritz Huetten committed
327
328
329
330
331
                    # Adding the event arrival time
                    time_tmp = Time(event_data['mjd'], scale='utc', format='mjd')
                    data.trig.gps_time = Time(time_tmp, format='unix', scale='utc', precision=9)
                else:
                    data.mc.energy = event_data['true_energy'] * u.GeV
332
                    data.mc.alt = (np.pi/2 - event_data['true_zd']) * u.rad
333
                    data.mc.az = -1 * (event_data['true_az'] - np.deg2rad(180 - 7)) * u.rad # check meaning of 7deg transformation (I.Vovk)
Moritz Huetten's avatar
Moritz Huetten committed
334
335
336
337
                    data.mc.shower_primary_id = 1 - event_data['true_shower_primary_id']
                    data.mc.h_first_int = event_data['true_h_first_int'] * u.cm
                    data.mc.core_x = event_data['true_core_x'] * u.cm
                    data.mc.core_y = event_data['true_core_y'] * u.cm
338
339
340
341
342
343

                # Setting the telescopes with data
                data.r0.tels_with_data = tels_with_data
                data.r1.tels_with_data = tels_with_data
                data.dl0.tels_with_data = tels_with_data
                data.trig.tels_with_trigger = tels_with_data
344
                
345
346
347
348
349
350
351
352
353
354
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

                yield data
                counter += 1

        return

    def _mono_event_generator(self, telescope):
        """
        Mono event generator. Yields DataContainer instances, filled
        with the read event data.

        Parameters
        ----------
        telescope: str
            The telescope for which to return events. Can be either "M1" or "M2".

        Returns
        -------

        """

        counter = 0
        telescope = telescope.upper()

        # Data container - is initialized once, and data is replaced within it after each yield
        data = DataContainer()

        # Telescopes with data:
        tels_in_file = ["M1", "M2"]

        if telescope not in tels_in_file:
            raise ValueError("Specified telescope {:s} is not in the allowed list {}".format(telescope, tels_in_file))

        tel_i = tels_in_file.index(telescope)
        tels_with_data = {tel_i + 1, }

        # Loop over the available data runs
382
        for run_number in self.run_numbers:
383
384
385
386
387
388
389

            # Removing the previously read data run from memory
            if self.current_run is not None:
                if 'data' in self.current_run:
                    del self.current_run['data']

            # Setting the new active run
Moritz Huetten's avatar
Moritz Huetten committed
390
            self.current_run = self._set_active_run(run_number)
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405

            if telescope == 'M1':
                n_events = self.current_run['data'].n_mono_events_m1
            else:
                n_events = self.current_run['data'].n_mono_events_m2

            # Loop over the events
            for event_i in range(n_events):
                # Event and run ids
                event_order_number = self.current_run['data'].mono_ids[telescope][event_i]
                event_id = self.current_run['data'].event_data[telescope]['stereo_event_number'][event_order_number]
                obs_id = self.current_run['number']

                # Reading event data
                event_data = self.current_run['data'].get_mono_event_data(event_i, telescope=telescope)
Moritz Huetten's avatar
Moritz Huetten committed
406
                
407
                data.meta = event_data['mars_meta']
408
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

                # Event counter
                data.count = counter

                # Setting up the R0 container
                data.r0.obs_id = obs_id
                data.r0.event_id = event_id
                data.r0.tel.clear()

                # Setting up the R1 container
                data.r1.obs_id = obs_id
                data.r1.event_id = event_id
                data.r1.tel.clear()

                # Setting up the DL0 container
                data.dl0.obs_id = obs_id
                data.dl0.event_id = event_id
                data.dl0.tel.clear()

                # Creating the telescope pointing container
                pointing = TelescopePointingContainer()
                pointing.azimuth = np.deg2rad(event_data['pointing_az']) * u.rad
                pointing.altitude = np.deg2rad(90 - event_data['pointing_zd']) * u.rad
                pointing.ra = np.deg2rad(event_data['pointing_ra']) * u.rad
                pointing.dec = np.deg2rad(event_data['pointing_dec']) * u.rad

                # Adding the pointing container to the event data
                data.pointing[tel_i + 1] = pointing

                # Adding event charge and peak positions per pixel
                data.dl1.tel[tel_i + 1].image = event_data['image']
Ievgen Vovk's avatar
Ievgen Vovk committed
439
                data.dl1.tel[tel_i + 1].pulse_time = event_data['pulse_time']
440

441
                if self.is_mc == False:
Moritz Huetten's avatar
Moritz Huetten committed
442
443
444
445
446
                    # Adding the event arrival time
                    time_tmp = Time(event_data['mjd'], scale='utc', format='mjd')
                    data.trig.gps_time = Time(time_tmp, format='unix', scale='utc', precision=9)
                else:
                    data.mc.energy = event_data['true_energy'] * u.GeV
447
                    data.mc.alt = (np.pi/2 - event_data['true_zd']) * u.rad
448
                    data.mc.az = -1 * (event_data['true_az'] - np.deg2rad(180 - 7)) * u.rad # check meaning of 7deg transformation (I.Vovk)
Moritz Huetten's avatar
Moritz Huetten committed
449
450
451
452
                    data.mc.shower_primary_id = 1 - event_data['true_shower_primary_id']
                    data.mc.h_first_int = event_data['true_h_first_int'] * u.cm
                    data.mc.core_x = event_data['true_core_x'] * u.cm
                    data.mc.core_y = event_data['true_core_y'] * u.cm
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495

                # Setting the telescopes with data
                data.r0.tels_with_data = tels_with_data
                data.r1.tels_with_data = tels_with_data
                data.dl0.tels_with_data = tels_with_data
                data.trig.tels_with_trigger = tels_with_data

                yield data
                counter += 1

        return

    def _pedestal_event_generator(self, telescope):
        """
        Pedestal event generator. Yields DataContainer instances, filled
        with the read event data.

        Parameters
        ----------
        telescope: str
            The telescope for which to return events. Can be either "M1" or "M2".

        Returns
        -------

        """

        counter = 0
        telescope = telescope.upper()

        # Data container - is initialized once, and data is replaced within it after each yield
        data = DataContainer()

        # Telescopes with data:
        tels_in_file = ["M1", "M2"]

        if telescope not in tels_in_file:
            raise ValueError("Specified telescope {:s} is not in the allowed list {}".format(telescope, tels_in_file))

        tel_i = tels_in_file.index(telescope)
        tels_with_data = {tel_i + 1, }

        # Loop over the available data runs
496
        for run_number in self.run_numbers:
497
498
499
500
501
502
503

            # Removing the previously read data run from memory
            if self.current_run is not None:
                if 'data' in self.current_run:
                    del self.current_run['data']

            # Setting the new active run
Moritz Huetten's avatar
Moritz Huetten committed
504
            self.current_run = self._set_active_run(run_number)
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519

            if telescope == 'M1':
                n_events = self.current_run['data'].n_pedestal_events_m1
            else:
                n_events = self.current_run['data'].n_pedestal_events_m2

            # Loop over the events
            for event_i in range(n_events):
                # Event and run ids
                event_order_number = self.current_run['data'].pedestal_ids[telescope][event_i]
                event_id = self.current_run['data'].event_data[telescope]['stereo_event_number'][event_order_number]
                obs_id = self.current_run['number']

                # Reading event data
                event_data = self.current_run['data'].get_pedestal_event_data(event_i, telescope=telescope)
Moritz Huetten's avatar
Moritz Huetten committed
520
                
521
                data.meta = event_data['mars_meta']
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552

                # Event counter
                data.count = counter

                # Setting up the R0 container
                data.r0.obs_id = obs_id
                data.r0.event_id = event_id
                data.r0.tel.clear()

                # Setting up the R1 container
                data.r1.obs_id = obs_id
                data.r1.event_id = event_id
                data.r1.tel.clear()

                # Setting up the DL0 container
                data.dl0.obs_id = obs_id
                data.dl0.event_id = event_id
                data.dl0.tel.clear()

                # Creating the telescope pointing container
                pointing = TelescopePointingContainer()
                pointing.azimuth = np.deg2rad(event_data['pointing_az']) * u.rad
                pointing.altitude = np.deg2rad(90 - event_data['pointing_zd']) * u.rad
                pointing.ra = np.deg2rad(event_data['pointing_ra']) * u.rad
                pointing.dec = np.deg2rad(event_data['pointing_dec']) * u.rad

                # Adding the pointing container to the event data
                data.pointing[tel_i + 1] = pointing

                # Adding event charge and peak positions per pixel
                data.dl1.tel[tel_i + 1].image = event_data['image']
Ievgen Vovk's avatar
Ievgen Vovk committed
553
                data.dl1.tel[tel_i + 1].pulse_time = event_data['pulse_time']
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570

                # Adding the event arrival time
                time_tmp = Time(event_data['mjd'], scale='utc', format='mjd')
                data.trig.gps_time = Time(time_tmp, format='unix', scale='utc', precision=9)

                # Setting the telescopes with data
                data.r0.tels_with_data = tels_with_data
                data.r1.tels_with_data = tels_with_data
                data.dl0.tels_with_data = tels_with_data
                data.trig.tels_with_trigger = tels_with_data

                yield data
                counter += 1

        return


Moritz Huetten's avatar
Moritz Huetten committed
571
class MarsRun:
572
573
574
575
    """
    This class implements reading of the event data from a single MAGIC data run.
    """

Moritz Huetten's avatar
Moritz Huetten committed
576
    def __init__(self, run_file_mask, filter_list=None):
577
578
579
580
581
582
583
584
585
        """
        Constructor of the class. Defines the run to use and the camera pixel arrangement.

        Parameters
        ----------
        run_file_mask: str
            A path mask for files belonging to the run. Must correspond to a single run
            or an exception will be raised. Must correspond to calibrated ("sorcerer"-level)
            data.
586
587
588
        filter_list: list, optional
            A list of files, to which the run_file_mask should be applied. If None, all the
            files satisfying run_file_mask will be used. Defaults to None.
589
590
        """

591
592
        self.n_camera_pixels = 1039

593
594
595
596
        self.run_file_mask = run_file_mask

        # Preparing the lists of M1/2 data files
        file_list = glob.glob(run_file_mask)
597
598
599
600
601

        # Filtering out extra files if necessary
        if filter_list is not None:
            file_list = list(set(file_list) & set(filter_list))

602
603
        self.m1_file_list = list(filter(lambda name: '_M1_' in name, file_list))
        self.m2_file_list = list(filter(lambda name: '_M2_' in name, file_list))
604
        self.m1_file_list.sort()
605
606
607
        self.m2_file_list.sort()

        # Retrieving the list of run numbers corresponding to the data files
Moritz Huetten's avatar
Moritz Huetten committed
608
609
        run_info = list(map(MAGICEventSource._get_run_info_from_name, file_list))
        run_numbers = [i[0] for i in run_info]
610
611
612
613
614
615
616
617
618
        is_mc_runs   = [i[1] for i in run_info]

        run_numbers = np.unique(run_numbers)
        is_mc_runs = np.unique(is_mc_runs)
        # Checking if runt type (data/MC) is consistent:
        if len(is_mc_runs) > 1:
            raise ValueError("Run type is not consistently data or MC: {}".format(is_mc))
        
        self.is_mc = is_mc_runs[0]
619
620
621
622
623

        # Checking if a single run is going to be read
        if len(run_numbers) > 1:
            raise ValueError("Run mask corresponds to more than one run: {}".format(run_numbers))

624
625
626
627
628
        # Reading the data
        m1_data = self.load_events(self.m1_file_list, self.is_mc, self.n_camera_pixels)
        m2_data = self.load_events(self.m2_file_list, self.is_mc, self.n_camera_pixels)
        
        # Getting the event data
629
        self.event_data = dict()
630
631
632
633
634
635
636
        self.event_data['M1'] = m1_data[0]
        self.event_data['M2'] = m2_data[0]

        # Getting the monitoring data
        self.monitoring_data = dict()
        self.monitoring_data['M1'] = m1_data[1]
        self.monitoring_data['M2'] = m2_data[1]
637
638
639
640
641
642
643
644

        # Detecting pedestal events
        self.pedestal_ids = self._find_pedestal_events()
        # Detecting stereo events
        self.stereo_ids = self._find_stereo_events()
        # Detecting mono events
        self.mono_ids = self._find_mono_events()

645

646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
    @property
    def n_events_m1(self):
        return len(self.event_data['M1']['MJD'])

    @property
    def n_events_m2(self):
        return len(self.event_data['M2']['MJD'])

    @property
    def n_stereo_events(self):
        return len(self.stereo_ids)

    @property
    def n_mono_events_m1(self):
        return len(self.mono_ids['M1'])

    @property
    def n_mono_events_m2(self):
        return len(self.mono_ids['M2'])

    @property
    def n_pedestal_events_m1(self):
        return len(self.pedestal_ids['M1'])

    @property
    def n_pedestal_events_m2(self):
        return len(self.pedestal_ids['M2'])

    @staticmethod
675
    def load_events(file_list, is_mc, n_camera_pixels):
676
        """
677
        This method loads events and monitoring data from the pre-defiled file and returns them as a dictionary.
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699

        Parameters
        ----------
        file_name: str
            Name of the MAGIC calibrated file to use.

        Returns
        -------
        dict:
            A dictionary with the even properties: charge / arrival time data, trigger, direction etc.
        """

        try:
            import uproot
        except ImportError:
            msg = "The `uproot` python module is required to access the MAGIC data"
            raise ImportError(msg)

        event_data = dict()

        event_data['charge'] = []
        event_data['arrival_time'] = []
700
701
702
703
704
705
706
        event_data['trigger_pattern'] = scipy.array([])
        event_data['stereo_event_number'] = scipy.array([])
        event_data['pointing_zd'] = scipy.array([])
        event_data['pointing_az'] = scipy.array([])
        event_data['pointing_ra'] = scipy.array([])
        event_data['pointing_dec'] = scipy.array([])
        event_data['MJD'] = scipy.array([])
Moritz Huetten's avatar
Moritz Huetten committed
707
        event_data['mars_meta'] = []
708
709

        # run-wise meta information (same for all events)
Moritz Huetten's avatar
Moritz Huetten committed
710
        mars_meta = dict()
711
712
713

        # monitoring information (updated from time to time)
        monitoring_data = dict()
714
        
715
716
717
718
719
720
721
722
723
724
725
726
        monitoring_data['badpixelinfo'] = [] 
        monitoring_data['PedestalMJD'] = scipy.array([])
        monitoring_data['PedestalFundamental'] = dict()
        monitoring_data['PedestalFundamental']['Mean'] = []
        monitoring_data['PedestalFundamental']['Rms'] = []
        monitoring_data['PedestalFromExtractor'] = dict()
        monitoring_data['PedestalFromExtractor']['Mean'] = []
        monitoring_data['PedestalFromExtractor']['Rms'] = []
        monitoring_data['PedestalFromExtractorRndm'] = dict()
        monitoring_data['PedestalFromExtractorRndm']['Mean'] = []
        monitoring_data['PedestalFromExtractorRndm']['Rms'] = []

727
728
        event_data['file_edges'] = [0]

729
730
731
732
        degrees_per_hour = 15.0
        seconds_per_day = 86400.0
        seconds_per_hour = 3600.

Moritz Huetten's avatar
Moritz Huetten committed
733
734
735
736
737
738
739
        evt_common_list = [
            'MCerPhotEvt.fPixels.fPhot', 
            'MArrivalTime.fData',
            'MTriggerPattern.fPrescaled',
            'MRawEvtHeader.fStereoEvtNumber', 
            'MRawEvtHeader.fDAQEvtNumber',
            ]
740
        
Moritz Huetten's avatar
Moritz Huetten committed
741
742
743
744
745
746
747
748
749
750
751
752
        # Separately, because only used with pre-processed MARS data 
        # to create MPointingPos container
        pointing_array_list = [
            'MPointingPos.fZd', 
            'MPointingPos.fAz', 
            'MPointingPos.fRa', 
            'MPointingPos.fDec', 
            'MPointingPos.fDevZd',
            'MPointingPos.fDevAz', 
            'MPointingPos.fDevHa', 
            'MPointingPos.fDevDec',
            ]
753
        
Moritz Huetten's avatar
Moritz Huetten committed
754
755
756
757
758
759
        # Info only applicable for data:
        time_array_list = [
            'MTime.fMjd',
            'MTime.fTime.fMilliSec',
            'MTime.fNanoSec', 
            ]
Moritz Huetten's avatar
Moritz Huetten committed
760
        
Moritz Huetten's avatar
Moritz Huetten committed
761
762
763
764
765
766
767
768
        drive_array_list = [
            'MReportDrive.fMjd',
            'MReportDrive.fCurrentZd',
            'MReportDrive.fCurrentAz',
            'MReportDrive.fRa',
            'MReportDrive.fDec'
            ]

769
770
771
772
773
774
775
776
777
778
779
780
        pedestal_array_list = [
            'MTimePedestals.fMjd',
            'MTimePedestals.fTime.fMilliSec',
            'MTimePedestals.fNanoSec',
            'MPedPhotFundamental.fArray.fMean',
            'MPedPhotFundamental.fArray.fRms',
            'MPedPhotFromExtractor.fArray.fMean',
            'MPedPhotFromExtractor.fArray.fRms',
            'MPedPhotFromExtractorRndm.fArray.fMean',
            'MPedPhotFromExtractorRndm.fArray.fRms'
            ]

Moritz Huetten's avatar
Moritz Huetten committed
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
        # Info only applicable for MC:
        mc_list = [
            'MMcEvt.fEnergy',
            'MMcEvt.fTheta',
            'MMcEvt.fPhi',
            'MMcEvt.fPartId',
            'MMcEvt.fZFirstInteraction',
            'MMcEvt.fCoreX',
            'MMcEvt.fCoreY', 
            ]

        # Metadata, currently not strictly required
        metainfo_array_list = [
            'MRawRunHeader.fRunNumber',
            'MRawRunHeader.fRunType',
            'MRawRunHeader.fSubRunIndex',
            'MRawRunHeader.fSourceRA',
            'MRawRunHeader.fSourceDEC',
            'MRawRunHeader.fTelescopeNumber']
800

801
802
803
804
        for file_name in file_list:

            input_file = uproot.open(file_name)

Moritz Huetten's avatar
Moritz Huetten committed
805
            events = input_file['Events'].arrays(evt_common_list)
806

807
            # Reading the info common to MC and real data
808
809
810
811
            charge = events[b'MCerPhotEvt.fPixels.fPhot']
            arrival_time = events[b'MArrivalTime.fData']
            trigger_pattern = events[b'MTriggerPattern.fPrescaled']
            stereo_event_number = events[b'MRawEvtHeader.fStereoEvtNumber']
812

Moritz Huetten's avatar
Moritz Huetten committed
813
            # Reading meta information:
Moritz Huetten's avatar
Moritz Huetten committed
814
            mars_meta['is_simulation'] = is_mc
815
816
            
            # try to read RunHeaders tree (soft fail if not present)
Moritz Huetten's avatar
Moritz Huetten committed
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
            try:
                meta_info = input_file['RunHeaders'].arrays(metainfo_array_list)
                
                mars_meta['origin'] = "MAGIC"
                mars_meta['input_url'] = file_name
    
                mars_meta['number'] = int(meta_info[b'MRawRunHeader.fRunNumber'][0])
                #mars_meta['number_subrun'] = int(meta_info[b'MRawRunHeader.fSubRunIndex'][0])
                mars_meta['source_ra'] = meta_info[b'MRawRunHeader.fSourceRA'][0] / seconds_per_hour * degrees_per_hour * u.deg
                mars_meta['source_dec'] = meta_info[b'MRawRunHeader.fSourceDEC'][0] / seconds_per_hour * u.deg
    
                is_simulation = int(meta_info[b'MRawRunHeader.fRunType'][0])
                if is_simulation == 0:
                    is_simulation = False
                elif is_simulation == 256:
                    is_simulation = True
                else:
                    msg = "Run type (Data or MC) of MAGIC data file not recognised."
Moritz Huetten's avatar
Moritz Huetten committed
835
836
                    logger.error(msg)
                    raise ValueError(msg)
Moritz Huetten's avatar
Moritz Huetten committed
837
838
                if is_simulation != is_mc:
                    msg = "Inconsistent run type (data or MC) between file name and runheader content."
Moritz Huetten's avatar
Moritz Huetten committed
839
840
                    logger.error(msg)
                    raise ValueError(msg)
Moritz Huetten's avatar
Moritz Huetten committed
841
                
Moritz Huetten's avatar
Moritz Huetten committed
842
                # Reading the info only contained in real data
843
                if not is_simulation:
Moritz Huetten's avatar
Moritz Huetten committed
844
845
846
847
848
849
850
                    badpixelinfo = input_file['RunHeaders']['MBadPixelsCam.fArray.fInfo'].array(uproot.asjagged(uproot.asdtype(np.int32))).flatten().reshape((4, 1183), order='F')
                    # now we have 3 axes:
                    # 1st axis: Unsuitable pixels
                    # 2nd axis: Uncalibrated pixels (says why pixel is unsuitable)
                    # 3rd axis: Bad hardware pixels (says why pixel is unsuitable)
                    # Each axis cointains a 32bit integer encoding more information about the specific problem, see MARS software, MBADPixelsPix.h
                    # Here, we however discard this additional information and only grep the "unsuitable" axis.
851
852
                    monitoring_data['badpixelinfo'].append(badpixelinfo[1].astype(bool)[:n_camera_pixels])

Moritz Huetten's avatar
Moritz Huetten committed
853
            except KeyError:
854
                logger.warning("RunHeaders tree not present in file. Cannot read meta information - will assume it is a real data run.")
Moritz Huetten's avatar
Moritz Huetten committed
855
                is_simulation = False
856

857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
            # try to read Pedestals tree (soft fail if not present)
            if not is_simulation:
                try:
                    pedestal_info = input_file['Pedestals'].arrays(pedestal_array_list)
                    
                    pedestal_mjd = pedestal_info[b'MTimePedestals.fMjd']
                    pedestal_millisec = pedestal_info[b'MTimePedestals.fTime.fMilliSec']
                    pedestal_nanosec = pedestal_info[b'MTimePedestals.fNanoSec']
                    n_pedestals = len(pedestal_mjd)
                    pedestal_mjd = pedestal_mjd + (pedestal_millisec / 1e3 + pedestal_nanosec / 1e9) / seconds_per_day
                    monitoring_data['PedestalMJD'] = scipy.concatenate((monitoring_data['PedestalMJD'], pedestal_mjd))
                    for quantity in ['Mean', 'Rms']:
                        for i_pedestal in range(n_pedestals):
                            monitoring_data['PedestalFundamental'][quantity].append(pedestal_info['MPedPhotFundamental.fArray.f{:s}'.format(quantity).encode()][i_pedestal][:n_camera_pixels])
                            monitoring_data['PedestalFromExtractor'][quantity].append(pedestal_info['MPedPhotFromExtractor.fArray.f{:s}'.format(quantity).encode()][i_pedestal][:n_camera_pixels])
                            monitoring_data['PedestalFromExtractorRndm'][quantity].append(pedestal_info['MPedPhotFromExtractorRndm.fArray.f{:s}'.format(quantity).encode()][i_pedestal][:n_camera_pixels])
                
                except KeyError:
                    logger.warning("Pedestals tree not present in file.")

            # Reading event timing information:
            if not is_simulation:
Moritz Huetten's avatar
Moritz Huetten committed
879
880
                event_times = input_file['Events'].arrays(time_array_list)
                # Computing the event arrival time
881
                
882
883
884
                event_mjd = event_times[b'MTime.fMjd']
                event_millisec = event_times[b'MTime.fTime.fMilliSec']
                event_nanosec = event_times[b'MTime.fNanoSec']
Moritz Huetten's avatar
Moritz Huetten committed
885
    
886
887
                event_mjd = event_mjd + (event_millisec / 1e3 + event_nanosec / 1e9) / seconds_per_day
                event_data['MJD'] = scipy.concatenate((event_data['MJD'], event_mjd))
888
889

            # Reading pointing information (in units of degrees):
890
891
            if 'MPointingPos.' in input_file['Events']:
                # Retrieving the telescope pointing direction
892
893
                pointing = input_file['Events'].arrays(pointing_array_list)

894
895
896
897
                pointing_zd = pointing[b'MPointingPos.fZd'] - pointing[b'MPointingPos.fDevZd']
                pointing_az = pointing[b'MPointingPos.fAz'] - pointing[b'MPointingPos.fDevAz']
                pointing_ra = (pointing[b'MPointingPos.fRa'] + pointing[b'MPointingPos.fDevHa']) * degrees_per_hour # N.B. the positive sign here, as HA = local sidereal time - ra
                pointing_dec = pointing[b'MPointingPos.fDec'] - pointing[b'MPointingPos.fDevDec']
898
899
            else:
                # Getting the telescope drive info
900
901
902
903
904
                drive = input_file['Drive'].arrays(drive_array_list)

                drive_mjd = drive[b'MReportDrive.fMjd']
                drive_zd = drive[b'MReportDrive.fCurrentZd']
                drive_az = drive[b'MReportDrive.fCurrentAz']
905
                drive_ra = drive[b'MReportDrive.fRa'] * degrees_per_hour
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
                drive_dec = drive[b'MReportDrive.fDec']

                # Finding only non-repeating drive entries
                # Repeating entries lead to failure in pointing interpolation
                non_repeating = scipy.diff(drive_mjd) > 0
                non_repeating = scipy.concatenate((non_repeating, [True]))

                # Filtering out the repeating ones
                drive_mjd = drive_mjd[non_repeating]
                drive_zd = drive_zd[non_repeating]
                drive_az = drive_az[non_repeating]
                drive_ra = drive_ra[non_repeating]
                drive_dec = drive_dec[non_repeating]

                if len(drive_zd) > 2:
                    # If there are enough drive data, creating azimuth and zenith angles interpolators
                    drive_zd_pointing_interpolator = scipy.interpolate.interp1d(drive_mjd, drive_zd, fill_value="extrapolate")
                    drive_az_pointing_interpolator = scipy.interpolate.interp1d(drive_mjd, drive_az, fill_value="extrapolate")

                    # Creating azimuth and zenith angles interpolators
                    drive_ra_pointing_interpolator = scipy.interpolate.interp1d(drive_mjd, drive_ra, fill_value="extrapolate")
                    drive_dec_pointing_interpolator = scipy.interpolate.interp1d(drive_mjd, drive_dec, fill_value="extrapolate")

                    # Interpolating the drive pointing to the event time stamps
930
931
932
933
                    pointing_zd = drive_zd_pointing_interpolator(event_mjd)
                    pointing_az = drive_az_pointing_interpolator(event_mjd)
                    pointing_ra = drive_ra_pointing_interpolator(event_mjd)
                    pointing_dec = drive_dec_pointing_interpolator(event_mjd)
934
935
936

                else:
                    # Not enough data to interpolate the pointing direction.
937
938
939
940
                    pointing_zd = scipy.repeat(-1, len(event_mjd))
                    pointing_az = scipy.repeat(-1, len(event_mjd))
                    pointing_ra = scipy.repeat(-1, len(event_mjd))
                    pointing_dec = scipy.repeat(-1, len(event_mjd))
941
942
943

            event_data['charge'].append(charge)
            event_data['arrival_time'].append(arrival_time)
Moritz Huetten's avatar
Moritz Huetten committed
944
            event_data['mars_meta'].append(mars_meta)
945
            event_data['trigger_pattern'] = scipy.concatenate((event_data['trigger_pattern'], trigger_pattern))
946
            event_data['stereo_event_number'] = scipy.concatenate((event_data['stereo_event_number'], stereo_event_number)).astype(dtype='int')
947
948
949
950
            event_data['pointing_zd'] = scipy.concatenate((event_data['pointing_zd'], pointing_zd))
            event_data['pointing_az'] = scipy.concatenate((event_data['pointing_az'], pointing_az))
            event_data['pointing_ra'] = scipy.concatenate((event_data['pointing_ra'], pointing_ra))
            event_data['pointing_dec'] = scipy.concatenate((event_data['pointing_dec'], pointing_dec))
951
952
953

            # Reading MC only information:
            if is_simulation:
Moritz Huetten's avatar
Moritz Huetten committed
954
955
956
957
958
959
960
961
962
                mc_info = input_file['Events'].arrays(mc_list)
                # N.B.: For MC, there is only one subrun
                event_data['true_energy'] = mc_info[b'MMcEvt.fEnergy']
                event_data['true_zd'] = mc_info[b'MMcEvt.fTheta']
                event_data['true_az'] = mc_info[b'MMcEvt.fPhi']
                event_data['true_shower_primary_id'] = mc_info[b'MMcEvt.fPartId']
                event_data['true_h_first_int'] = mc_info[b'MMcEvt.fZFirstInteraction']
                event_data['true_core_x'] = mc_info[b'MMcEvt.fCoreX']
                event_data['true_core_y'] = mc_info[b'MMcEvt.fCoreY']
963
964
965

            event_data['file_edges'].append(len(event_data['trigger_pattern']))

966
        return event_data, monitoring_data
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997

    def _find_pedestal_events(self):
        """
        This internal method identifies the IDs (order numbers) of the
        pedestal events in the run.

        Returns
        -------
        dict:
            A dictionary of pedestal event IDs in M1/2 separately.
        """

        pedestal_ids = dict()

        pedestal_trigger_pattern = 8

        for telescope in self.event_data:
            ped_triggers = np.where(self.event_data[telescope]['trigger_pattern'] == pedestal_trigger_pattern)
            pedestal_ids[telescope] = ped_triggers[0]

        return pedestal_ids

    def _find_stereo_events(self):
        """
        This internal methods identifies stereo events in the run.

        Returns
        -------
        list:
            A list of pairs (M1_id, M2_id) corresponding to stereo events in the run.
        """
998
999
1000

        stereo_ids = []