evaluation.py 17.5 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
from __future__ import absolute_import, division, print_function

from pathlib import Path
import os
import numpy as np
import matplotlib.pyplot as plt

from multiprocessing import Pool, cpu_count
from functools import partial

from astropy import units as u
from astropy.io import ascii
from astropy.wcs import WCS
from astropy.utils.console import ProgressBar
from astropy.table import vstack

from scipy.optimize import curve_fit

from nikamap import NikaMap, Jackknife
from nikamap.utils import pos_uniform
from astropy.io import fits
from astropy.table import Table, MaskedColumn
import sys
from mpl_toolkits.axes_grid1 import make_axes_locatable
LUSTIG Peter's avatar
LUSTIG Peter committed
25
26
27
28
29
import dill as pickle
from matplotlib.ticker import FormatStrFormatter
from collections import OrderedDict


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
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
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
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
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
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260

import os
os.getcwd()
'''
%load_ext autoreload
%autoreload 2
%matplotlib tk
'''

plt.ion()



def add_axis(name, range, bins, unit=None, i_axe=3, log=False):
    """Define a dictionnary for additionnal wcs axes (linear or log)"""

    header = {'CTYPE{}'.format(i_axe): name,
              'CRPIX{}'.format(i_axe): 1,
              'CUNIT{}'.format(i_axe): unit}

    if log:
        # Log scale (edges definition)
        log_step = (np.log(range[1]) - np.log(range[0])) / bins
        center_start = np.exp(np.log(range[0]) + log_step / 2)
        header['CTYPE{}'.format(i_axe)] += '-LOG'
        header['CRVAL{}'.format(i_axe)] = center_start
        header['CDELT{}'.format(i_axe)] = log_step * center_start

        # Log scale (center definition)
        # log_step = (np.log(flux_range[1]) - np.log(flux_range[0])) / (bins-1)
        # center_start = range[0]

    else:
        # Linear scale (edges definition)
        step = (range[1] - range[0]) / bins
        header['CRVAL{}'.format(i_axe)] = range[0] + step / 2
        header['CDELT{}'.format(i_axe)] = step

        # Linear scale (center definition)
        # step = (range[1] - range[0]) / (bins-1)

    return header


def completness_purity_wcs(shape, wcs, bins=30,
                           flux_range=(0, 1), flux_bins=10, flux_log=False,
                           threshold_range=(0, 1), threshold_bins=10, threshold_log=False):
    """Build a wcs for the completness_purity function"""

    slice_step = np.ceil(np.asarray(shape) / bins).astype(int)
    celestial_slice = slice(0, shape[0], slice_step[0]), slice(0, shape[1], slice_step[1])

    # [WIP]: Shall we use a 4D WCS ? (ra/dec flux/threshold)
    # [WIP]: -TAB does not seems to be very easy to do with astropy
    # Basicaly Working... .
    header = wcs[celestial_slice[0], celestial_slice[1]].to_header()
    header['WCSAXES'] = 4

    header.update(add_axis('THRESHOLD', threshold_range, threshold_bins, i_axe=3))
    header.update(add_axis('FLUX', flux_range, flux_bins, i_axe=4, log=True))

    return (bins, bins, threshold_bins, flux_bins), WCS(header)



def completness_worker(shape, wcs, sources, fake_sources, min_threshold=2,
                       max_threshold=5):
    """Compute completness from the fake source catalog

    Parameters
    ----------
    shape : tuple
        the shape of the resulting image
    sources : :class:`astropy.table.Table`
        the detected sources
    fake_sources : :class:`astropy.table.Table`
        the fake sources table, with corresponding mask
    min_threshold : float
        the minimum SNR threshold requested
    max_threshold : float
        the maximum SNR threshold requested

    Returns
    -------
    _completness, _norm_comp
        corresponding 2D :class:`numpy.ndarray`
    """
    # If one wanted to used a histogramdd, one would need a threshold axis
    # covering ALL possible SNR, otherwise loose flux, or cap the thresholds...
    fake_snr = np.ma.array(sources[fake_sources['find_peak'].filled(0)]['SNR'],
                           mask=fake_sources['find_peak'].mask)

    # As we are interested by the cumulative numbers, keep all inside the
    # upper pixel
    fake_snr[fake_snr > max_threshold] = max_threshold

    # print(fake_snr)
    # TODO: Consider keeping all pixels information in fake_source and source...
    #       This would imply to do only a simple wcs_threshold here...
    xx, yy, zz = wcs.wcs_world2pix(fake_sources['ra'], fake_sources['dec'],
                                   fake_snr.filled(min_threshold), 0)

    # Number of fake sources recovered
    _completness, _ = np.histogramdd(np.asarray([xx, yy, zz]).T + 0.5,
                                     bins=np.asarray(shape),
                                     range=list(zip([0]*len(shape), shape)),
                                     weights=~fake_sources['find_peak'].mask)
    # Reverse cumulative sum to get all sources at the given threshold
    _completness = np.cumsum(_completness[..., ::-1], axis=2)[..., ::-1]

    # Number of fake sources (independant of threshold)
    _norm_comp, _, _ = np.histogram2d(xx + 0.5, yy + 0.5,
                                      bins=np.asarray(shape[0:2]),
                                      range=list(zip([0]*2, shape[0:2])))
    return _completness, _norm_comp


def purity_worker(shape, wcs, sources, max_threshold=2):
    """Compute completness from the fake source catalog

    Parameters
    ----------
    shape : tuple
        the shape of the resulting image
    sources : :class:`astropy.table.Table`
        the detected sources table, with corresponding match
    max_threshold : float
        the maximum threshold requested

    Returns
    -------
    _completness, _norm_comp
        corresponding 2D :class:`numpy.ndarray`
    """

    sources_snr = sources['SNR']
    # As we are interested by the cumulative numbers, keep all inside the
    # upper pixel
    sources_snr[sources_snr > max_threshold] = max_threshold
    xx, yy, zz = wcs.wcs_world2pix(sources['ra'], sources['dec'],
                                   sources_snr, 0)

    # Number of fake sources recovered
    _purity, _ = np.histogramdd(np.asarray([xx, yy, zz]).T + 0.5,
                                bins=np.asarray(shape),
                                range=list(zip([0]*len(shape), shape)),
                                weights=~sources['fake_sources'].mask)

    # Revese cumulative sum...
    _purity = np.cumsum(_purity[..., ::-1], axis=2)[..., ::-1]

    # Number of total detected sources at a given threshold
    _norm_pur, _ = np.histogramdd(np.asarray([xx, yy, zz]).T + 0.5,
                                  bins=np.asarray(shape),
                                  range=list(zip([0]*len(shape), shape)))
    _norm_pur = np.cumsum(_norm_pur[..., ::-1], axis=2)[..., ::-1]

    return _purity, _norm_pur


def completness_purity(sources, fake_sources, wcs=None,
                       shape=None):
    """Compute completness map for a given flux"""

    # print(flux)

    # wcs_celestial = wcs.celestial
    # Lower and upper edges ... take the center of the pixel for the upper edge
    min_threshold, max_threshold = wcs.sub([3]).all_pix2world([-0.5, shape[2]-1], 0)[0]

    completness = np.zeros(shape, dtype=np.float)
    norm_comp = np.zeros(shape[0:2], dtype=np.float)

    purity = np.zeros(shape, dtype=np.float)
    norm_pur = np.zeros(shape, dtype=np.float)


    # %load_ext snakeviz
    # %snakeviz the following line.... all is spend in the find_peaks /
    # fit_2d_gaussian
    # TODO: Change the find_peaks routine, or maybe just the
    # fit_2d_gaussian to be FAST ! (Maybe look into gcntrd.pro routine
    # or photutils.centroid.centroid_1dg maybe ?)

    _completness, _norm_comp = completness_worker(shape, wcs, sources,
                                                  fake_sources,
                                                  min_threshold,
                                                  max_threshold)

    # print(_completness)
    completness += _completness
    norm_comp += _norm_comp

    _purity, _norm_pur = purity_worker(shape, wcs, sources, max_threshold)

    purity += _purity
    norm_pur += _norm_pur

    # norm can be 0, so to avoid warning on invalid values...
    with np.errstate(divide='ignore', invalid='ignore'):
        completness /= norm_comp[..., np.newaxis]
        purity /= norm_pur

    # TODO: One should probably return completness AND norm if one want to
    # combine several fluxes
    return completness, purity


def Evaluate(flux_ds_fs_list):
    # _flux = u.Quantity(pr.header['flux{}'.format(isimu)])
    _flux = flux_ds_fs_list[0]
    sources = flux_ds_fs_list[1]
    fake_sources = flux_ds_fs_list[2]
    fluxval = _flux.to_value(u.mJy)
    '''
    _flux.to_value(u.mJy)
    sources = Table.read(hdul['DETECTED_SOURCES{}'.format(_flux)])
    fake_sources = Table.read(hdul['FAKE_SOURCES{}'.format(_flux)])
    '''
    print('{} data loaded'.format(_flux))

    # sources = df['DETECTED_SOURCES{}'.format(flux)]
    # print(fake_sources)
    # sys.exit()
    completness, purity = completness_purity(sources, fake_sources,
                                             wcs=wcs_4D.sub([1, 2, 3]),
                                             shape=shape_4D[0:3])

    return fluxval, completness, purity


LUSTIG Peter's avatar
LUSTIG Peter committed
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
def Evaluate2(_flux, hdul):
    print(fits.HDUList(hdul).info())
    # _flux = u.Quantity(pr.header['flux{}'.format(isimu)])
    fluxval = _flux.to_value(u.mJy)
    sources = Table.read(hdul['DETECTED_SOURCES{}'.format(_flux)])
    fake_sources = Table.read(hdul['FAKE_SOURCES{}'.format(_flux)])

    print('{} data loaded'.format(_flux))

    completness, purity = completness_purity(sources, fake_sources,
                                             wcs=wcs_4D.sub([1, 2, 3]),
                                             shape=shape_4D[0:3])

    print(fluxval, completness, purity)
    return fluxval, completness, purity


278
279
280
281
282
283
284
285
286
287
288
289
plt.close('all')

# _data = next(Jackknife(filenames, n=None))
# # TODO: Should in principle be the same, but is not... check....
# _ = plt.hist((data.data - _data.data)[~data.mask & ~_data.mask],
# bins=1000, range=[-0.1, 0.1], log=True)


# Create the flux and threshold axes...

DATA_DIR = "/home/peter/Dokumente/Uni/Paris/Stage/data/v_1"
data = NikaMap.read(Path(DATA_DIR) / '..' / 'map.fits')
LUSTIG Peter's avatar
LUSTIG Peter committed
290
bins = 19
291
292
293
294

flux_bins = 2
flux_range = [0.1, 10]

LUSTIG Peter's avatar
LUSTIG Peter committed
295
threshold_bins = 5
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
threshold_range = [2, 5]
threshold_range = [3, 7.5]


# Does not really make sense... better define edges
fluxes = np.logspace(np.log10(flux_range[0]), np.log10(flux_range[1]),
                     flux_bins)*u.mJy
fluxes_edges = np.logspace(np.log10(flux_range[0]), np.log10(flux_range[1]),
                           flux_bins + 1)*u.mJy

# Does not really make sense... better define edges
threshold = np.linspace(threshold_range[0], threshold_range[1], threshold_bins)
threshold_edges = np.linspace(threshold_range[0], threshold_range[1],
                              threshold_bins+1)

shape_4D, wcs_4D = completness_purity_wcs(data.shape, data.wcs, bins=bins,
                                          flux_range=flux_range,
                                          flux_bins=flux_bins, flux_log=True,
                                          threshold_range=threshold_range,
                                          threshold_bins=threshold_bins)


# Testing the lower edges
wcs_threshold = wcs_4D.sub([3])
assert np.all(np.abs(wcs_threshold.all_pix2world(np.arange(threshold_bins+1)-0.5, 0) - threshold_edges) < 1e-15)
wcs_flux = wcs_4D.sub([4])
assert np.all(np.abs(wcs_flux.all_pix2world(np.arange(flux_bins+1)-0.5, 0) - fluxes_edges.value) < 1e-13)


# DEBUG :
# flux, nsources, within, wcs, shape, nsim, jk_filenames = 10*u.mJy, 8**2,
# (0, 1), wcs_4D.sub([1, 2, 3]), shape_4D[0:3], np.multiply(*shape_4D[0:2])
# * 100, filenames

# This is a single run check for a single flux

LUSTIG Peter's avatar
LUSTIG Peter committed
332
333
334
hdul = fits.HDUList(fits.open('/home/peter/Dokumente/Uni/Paris/Stage/FirstSteps/'
                 'Completness/combined_tables_long.fits'))
# sys.exit()
335
336
337
338
339
nfluxes = hdul[0].header['NFLUXES']
print('{} different fluxes found'.format(nfluxes))

# Get fluxlist:

LUSTIG Peter's avatar
LUSTIG Peter committed
340
indata = []
341
342
343
344
345
346
347
348
for isimu in range(nfluxes):
    _FLUX = u.Quantity(hdul[0].header['flux{}'.format(isimu)])

    _SOURCES = Table.read(hdul['DETECTED_SOURCES{}'
                               .format(_FLUX)])
    _FAKE_SOURCES = Table.read(hdul['FAKE_SOURCES{}'
                                    .format(_FLUX)])
    indata.append([_FLUX, _SOURCES, _FAKE_SOURCES])
LUSTIG Peter's avatar
LUSTIG Peter committed
349
350
351
352
353
'''
flux = []
for isimu in range(nfluxes):
    flux.append(u.Quantity(hdul[0].header['flux{}'.format(isimu)]))
'''
354
355
# helpfunc = partial(Evaluate, **{'hdul': hdul})
p = Pool(cpu_count())
LUSTIG Peter's avatar
LUSTIG Peter committed
356

357
358
359
360
361
362
363
364
365
366
367
368
res = p.map(Evaluate, indata)
res = list(zip(*res))
FLUX = np.array(res[0])
COMPLETNESS = np.array(res[1])
PURITY = np.array(res[2])

idxsort = np.argsort(FLUX)
FLUX = FLUX[idxsort]
COMPLETNESS = COMPLETNESS[idxsort]
PURITY = PURITY[idxsort]

midbin = int(bins/2)
LUSTIG Peter's avatar
LUSTIG Peter committed
369

370
371
372
373
374
print(midbin)
# sys.exit()
# %% PlotFigure


LUSTIG Peter's avatar
LUSTIG Peter committed
375
376
377
378
379
380
381
382
def find_nearest(array, values):
    x, y = np.meshgrid(array, values)
    ev = np.abs(x - y)
    return np.argmin(ev, axis=1)


def PlotEvaluation(data, title='', flux=np.array([]), thresh=[],
                   nfluxlabels=None, nthreshlabels=None, **kwargs):
383
384
    tickfs = 20
    labelfs = 25
LUSTIG Peter's avatar
LUSTIG Peter committed
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410

    if nfluxlabels is not None:
        _label_flux = np.geomspace(flux[0], flux[-1], nfluxlabels)
        _f_idx = find_nearest(flux, _label_flux)
        flblpos, _flbl = _f_idx, flux[_f_idx]
    else:
        flblpos, _flbl = np.arange(len(flux)), flux

    if nthreshlabels is not None:
        _label_thresh = np.linspace(thresh[0], thresh[-1], nthreshlabels)
        _t_idx = find_nearest(thresh, _label_thresh)
        print(_t_idx)
        tlblpos, _tlbl = _t_idx, thresh[_t_idx]
    else:
        tlblpos, _tlbl = np.arange(len(thresh)), thresh

    flbl = []
    for i in range(len(_flbl)):
        flbl.append('{:.1f}'.format(_flbl[i]))

    tlbl = []
    for i in range(len(_tlbl)):
        tlbl.append('{:.1f}'.format(_tlbl[i]))
        # print(i, tlbl[i])

    # print(np.array([tlblpos, tlbl]).T)
411
412
413
414
    plt.figure()
    plt.title(title, fontsize=30)
    plt.xlabel('Detection Threshold [SNR]', fontsize=labelfs)
    plt.ylabel('Flux [mJy]', fontsize=labelfs)
LUSTIG Peter's avatar
LUSTIG Peter committed
415
416
417
418
419
420

    plt.xticks(tlblpos, tlbl, fontsize=tickfs)
    # ax = plt.gca()
    plt.yticks(flblpos, flbl, fontsize=tickfs)
    # ax.yaxis.set_major_formatter(FormatStrFormatter('%.2f'))

421
422
423
424
425
426
    plt.imshow(data, origin='lower', **kwargs)
    cbar = plt.colorbar()
    cbar.ax.tick_params(labelsize=tickfs)


PlotEvaluation(COMPLETNESS[:, midbin, midbin, :], title='Completness',
LUSTIG Peter's avatar
LUSTIG Peter committed
427
428
429
               flux=np.array(FLUX), thresh=threshold, cmap='bone',
               nfluxlabels=10, nthreshlabels=5, aspect='auto')

LUSTIG Peter's avatar
LUSTIG Peter committed
430
431
432
433
434
PlotEvaluation(PURITY[:, midbin, midbin, :], title='Purity',
               flux=np.array(FLUX), thresh=threshold, cmap='bone',
               nfluxlabels=10, nthreshlabels=5, aspect='auto')
plt.show(block=True)

LUSTIG Peter's avatar
LUSTIG Peter committed
435
436
437
438
439

# %% 2D plot


def PlotFixedThreshold(thresholds, bin, completness, allthresholds, flux,
LUSTIG Peter's avatar
LUSTIG Peter committed
440
                       nfluxlabels=None, hlines=None):
LUSTIG Peter's avatar
LUSTIG Peter committed
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466

    linestyles = ['-', '--', '-.', ':']
    real_thresholds = find_nearest(allthresholds, thresholds)
    for i in range(len(real_thresholds)):
        _x = flux
        _y = completness[:, bin[0], bin[1], real_thresholds[i]]
        plt.plot(_x, _y, linestyle=linestyles[i],
                 label='{:.1f}'.format(allthresholds[real_thresholds[i]]))

    if hlines is not None:
        for i, val in enumerate(hlines):
            plt.axhline(val, color='r')
    plt.title('Fixed Threshold', fontsize=30, y=1.02)
    plt.xlabel('Source Flux [mJy]', fontsize=25)
    plt.ylabel('Completness', fontsize=25)
    plt.yticks(fontsize=20)
    plt.xticks(fontsize=20)
    plt.subplots_adjust(left=0.12)
    ax = plt.gca()
    ax.set_xscale("log", nonposx='clip')
    # legend = plt.legend(fontsize=25, title='SNR', loc='lower right')
    legend = plt.legend(fontsize=25, title='SNR', loc='upper left',
                        framealpha=1)
    plt.setp(legend.get_title(), fontsize=25)
    plt.show(block=True)

467
# cmap bone hot
LUSTIG Peter's avatar
LUSTIG Peter committed
468
469
470
471
472
473
# print(np.array(FLUX))


PlotFixedThreshold(np.array((3, 5, 7)), (midbin, midbin), COMPLETNESS,
                   threshold, np.array(FLUX), nfluxlabels=None,
                   hlines=[.5, .8, .9])
LUSTIG Peter's avatar
LUSTIG Peter committed
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522

# %% Plot map
# Plot_CompPur(completness, purity, threshold, nsim=None, savename=None,
#                  flux=None):

plotidx = np.array([24, 30, 40, 49])
# plotthreshidx = np.array([0, 4, 8])

plotcomp = COMPLETNESS[plotidx]
plotpur = PURITY[plotidx]
print(plotpur.shape)
# plotthresh = threshold[plotthreshidx]
plotflux = np.array(FLUX[plotidx])


def Plot_CompPur(completness, purity, threshold, nsim=None, savename=None,
                 flux=None):
    threshold_bins = completness.shape[-1]
    print('subplots:', threshold_bins)
    fig, axes = plt.subplots(nrows=2, ncols=threshold_bins, sharex=True,
                             sharey=True)
    print('axes shape', axes.shape)
    print('completness shape', completness.shape)
    for i in range(threshold_bins):
        print(i)
        axes[0, i].imshow(completness[:, :, i], vmin=0, vmax=1)
        im = axes[1, i].imshow(purity[:, :, i], vmin=0, vmax=1)
        axes[1, i].set_xlabel("thresh={:.2f}".format(threshold[i]))
        if i == (threshold_bins-1):
            # print('-----------')
            divider = make_axes_locatable(axes[1, i])
            cax = divider.append_axes('right', size='5%', pad=0.0)
            fig = plt.gcf()
            fig.colorbar(im, cax=cax, orientation='vertical')
    if nsim is not None:
        axes[0, 0].set_title("{} simulations".format(nsim))
    if flux is not None:
        axes[0, 1].set_title("{}".format(flux))
    axes[0, 0].set_ylabel("completness")
    axes[1, 0].set_ylabel("purity")
    if savename is not None:
        plt.savefig(savename)


for i in range(len(plotidx)):
    _comp = plotcomp[i]
    _pur = plotpur[i]
    Plot_CompPur(_comp, _pur, threshold, flux=plotflux[i])
plt.show(block=True)