def rapid_stage(locus_data):
locus_properties = str(locus_data.get_properties())
objid = locus_properties['alert_id']
ra = locus_properties['ra']
dec = locus_properties['dec']
redshift = 0.
mwebv = 0.2
alert_id, mjd, ras, decs, passband, mag, magerr, zeropoint, = locus_data.get_time_series(
'ra', 'dec', 'ztf_fid', 'ztf_magpsf', 'ztf_sigmapsf', 'ztf_magzpsci')
flux = 10.**(-0.4 * (mag - zeropoint))
fluxerr = np.abs(flux * magerr * (np.log(10.) / 2.5))
photflag = [0] * int(len(mjd) / 2 - 3) + [
6144
] + [4096] * int(len(mjd) / 2 + 2)
passband = np.where((passband == 1) | (passband == '1.0'), 'g', passband)
passband = np.where((passband == 2) | (passband == '2.0'), 'r', passband)
deleteindexes = np.where((passband == 3) | (passband == '3.0'))
mjd, passband, flux, fluxerr, zeropoint, photflag = delete_indexes(
deleteindexes, mjd, passband, flux, fluxerr, zeropoint, photflag)
light_curve_list = [(mjd, flux, fluxerr, passband, zeropoint, photflag, ra,
dec, objid, redshift, mwebv)]
classification = Classify(light_curve_list)
predictions = classification.get_predictions()
print(predictions)
print(
locus_data.get_time_series('ra', 'dec', 'ztf_fid', 'ztf_magpsf',
'ztf_sigmapsf', 'ztf_magzpsci',
'ztf_diffmaglim'))
def main(graph=None, model=None):
"""
Example code to run astrorapid.
Ignore the graph and model parameter inputs unless you wish to do your own multithreading.
(Note: astrorapid already performs its own parallelisation based on a keras and tensorflow backend).
"""
mjd = [57433.4816, 57436.4815, 57439.4817, 57451.4604, 57454.4397, 57459.3963, 57462.418, 57465.4385, 57468.3768,
57473.3606, 57487.3364, 57490.3341, 57493.3154, 57496.3352, 57505.3144, 57513.2542, 57532.2717, 57536.2531,
57543.2545, 57546.2703, 57551.2115, 57555.2669, 57558.2769, 57561.1899, 57573.2133, 57433.5019, 57436.4609,
57439.4587, 57444.4357, 57459.4189, 57468.3142, 57476.355, 57479.3568, 57487.3586, 57490.3562, 57493.3352,
57496.2949, 57505.3557, 57509.2932, 57513.2934, 57518.2735, 57521.2739, 57536.2321, 57539.2115, 57543.2301,
57551.1701, 57555.2107, 57558.191, 57573.1923, 57576.1749, 57586.1854]
flux = [2.0357230e+00, -2.0382695e+00, 1.0084588e+02, 5.5482742e+01, 1.4867026e+01, -6.5136810e+01, 1.6740545e+01,
-5.7269131e+01, 1.0649184e+02, 1.5505235e+02, 3.2445984e+02, 2.8735449e+02, 2.0898877e+02, 2.8958893e+02,
1.9793906e+02, -1.3370536e+01, -3.9001358e+01, 7.4040916e+01, -1.7343750e+00, 2.7844931e+01, 6.0861992e+01,
4.2057487e+01, 7.1565346e+01, -2.6085690e-01, -6.8435440e+01, 17.573107, 41.445435, -110.72664, 111.328964,
-63.48336, 352.44907, 199.59058, 429.83075, 338.5255, 409.94604, 389.71262, 195.63905, 267.13318, 123.92461,
200.3431, 106.994514, 142.96387, 56.491238, 55.17521, 97.556946, -29.263103, 142.57687, -20.85057,
-0.67210346, 63.353024, -40.02601]
fluxerr = [42.784702, 43.83665, 99.98704, 45.26248, 43.040398, 44.00679, 41.856007, 49.354336, 105.86439, 114.0044,
45.697918, 44.15781, 60.574158, 93.08788, 66.04482, 44.26264, 91.525085, 42.768955, 43.228336, 44.178196,
62.15593, 109.270035, 174.49638, 72.6023, 48.021034, 44.86118, 48.659588, 100.97703, 148.94061, 44.98218,
139.11194, 71.4585, 47.766987, 45.77923, 45.610615, 60.50458, 105.11658, 71.41217, 43.945534, 45.154167,
43.84058, 52.93122, 44.722775, 44.250145, 43.95989, 68.101326, 127.122025, 124.1893, 49.952255, 54.50728,
114.91599]
passband = ['g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g',
'g', 'g', 'g', 'g', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r',
'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r']
zeropoint = [27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5,
27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5,
27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5]
photflag = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4096, 4096, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4096,
6144, 4096, 4096, 4096, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
objid = 'MSIP_01_NONIa-0001_10400862'
ra = 3.75464531293933
dec = 0.205076187109334
redshift = 0.233557
mwebv = 0.0228761
light_curve_list = [(mjd, flux, fluxerr, passband, zeropoint, photflag, ra, dec, objid, redshift, mwebv)]
classification = Classify(light_curve_list, known_redshift=True, graph=graph, model=model)
predictions, time_steps = classification.get_predictions(return_predictions_at_obstime=False)
print(predictions)
# classification.plot_light_curves_and_classifications(step=False)
# classification.plot_classification_animation()
# classification.plot_classification_animation_step()
predictions, time_steps = classification.get_predictions(return_predictions_at_obstime=True)
import matplotlib.pyplot as plt
for i, class_name in enumerate(classification.class_names):
plt.plot(time_steps[0], predictions[0][:, i], label=class_name)
plt.legend()
plt.show()
def rapid_stage(locus_data):
locus_properties = locus_data.get_properties()
objid = locus_properties['alert_id']
ra = locus_properties['ra']
dec = locus_properties['dec']
redshift = 0. # TODO: Get correct redshift
mwebv = 0.2 # TODO: Get correct extinction
alert_id, mjd, ras, decs, passband, mag, magerr, zeropoint, = locus_data.get_time_series(
'ra', 'dec', 'ztf_fid', 'ztf_magpsf', 'ztf_sigmapsf', 'ztf_magzpsci')
# Ignore mag Nonetype values
alertid, mjd, ras, decs, passband, mag, magerr, zeropoint = delete_indexes(
np.where(mag == None), alert_id, mjd, ras, decs, passband, mag, magerr,
zeropoint)
print(
locus_data.get_time_series('ra', 'dec', 'ztf_fid', 'ztf_magpsf',
'ztf_sigmapsf', 'ztf_magzpsci'))
if len(mjd) < 3:
print("less than 3 points")
return
zpt_median = np.median(zeropoint[zeropoint != None])
zeropoint[zeropoint == None] = zpt_median
zeropoint = np.asarray(zeropoint, dtype=np.float64)
mag = np.asarray(mag, dtype=np.float64)
flux = 10.**(-0.4 * (mag - zeropoint))
fluxerr = np.abs(flux * magerr * (np.log(10.) / 2.5))
passband = np.where((passband == 1) | (passband == '1.0'), 'g', passband)
passband = np.where((passband == 2) | (passband == '2.0'), 'r', passband)
# Set photflag detections when S/N > 5
photflag = np.zeros(len(flux))
photflag[flux / fluxerr > 5] = 4096
photflag[np.where(mjd == min(mjd[photflag == 4096]))] = 6144
deleteindexes = np.where((passband == 3) | (passband == '3.0')
| (np.isnan(mag)))
mjd, passband, flux, fluxerr, zeropoint, photflag = delete_indexes(
deleteindexes, mjd, passband, flux, fluxerr, zeropoint, photflag)
light_curve_list = [(mjd, flux, fluxerr, passband, zeropoint, photflag, ra,
dec, objid, redshift, mwebv)]
from keras import backend as K
K.clear_session()
classification = Classify(known_redshift=True)
predictions = classification.get_predictions(light_curve_list)
print(predictions)
for i, name in enumerate(classification.class_names):
locus_data.set_property('rapid_class_probability_{}'.format(name),
predictions[0][-1][i])
def classify_lasair_light_curves(object_names=('ZTF18acsovsw', ), plot=True):
light_curve_list = []
for object_name in object_names:
try:
mjd, passband, mag, magerr, photflag, zeropoint, ra, dec, objid, redshift, mwebv = read_lasair_json(
object_name)
except Exception as e:
print(e)
continue
flux = 10.**(-0.4 * (mag - zeropoint))
fluxerr = np.abs(flux * magerr * (np.log(10.) / 2.5))
passband = np.where((passband == 1) | (passband == '1'), 'g', passband)
passband = np.where((passband == 2) | (passband == '2'), 'r', passband)
# Set photflag detections when S/N > 5
photflag2 = np.zeros(len(flux))
photflag2[flux / fluxerr > 5] = 4096
photflag2[np.where(mjd == min(mjd[photflag2 == 4096]))] = 6144
mjd_first_detection = min(mjd[photflag == 4096])
photflag[np.where(mjd == mjd_first_detection)] = 6144
deleteindexes = np.where(((passband == 3) | (passband == '3'))
| (mjd > mjd_first_detection)
& (photflag == 0))
print("Deleting indexes {} at mjd {} and passband {}".format(
deleteindexes, mjd[deleteindexes], passband[deleteindexes]))
mjd, passband, flux, fluxerr, zeropoint, photflag = delete_indexes(
deleteindexes, mjd, passband, flux, fluxerr, zeropoint, photflag)
light_curve_list += [(mjd, flux, fluxerr, passband, zeropoint,
photflag, ra, dec, objid, redshift, mwebv)]
classification = Classify(light_curve_list,
known_redshift=True,
bcut=False,
zcut=None)
predictions, time_steps = classification.get_predictions(
return_predictions_at_obstime=False)
print(predictions)
if plot:
classification.plot_light_curves_and_classifications(
step=True, use_interp_flux=False)
# classification.plot_light_curves_and_classifications(step=False, use_interp_flux=True)
# classification.plot_light_curves_and_classifications(step=False, use_interp_flux=False)
# classification.plot_classification_animation_step()
# classification.plot_classification_animation()
return classification.orig_lc, classification.timesX, classification.y_predict
def __init__(self, model: str = None):
"""
Creates a new instance of RAPID
Parameters
----------
model: str, optional
The filepath of the model to be loaded, if any.
"""
if model is not None:
self.classifier = Classify(known_redshift=True,
model_filepath=model)
else:
self.classifier = Classify(known_redshift=True)
def classify(self, lc):
mjd, flux, fluxerr, passband, ra, dec, objid, redshift, mwebv = lc[0]
photflag = [4096] * len(flux)
photflag[np.argmax(flux)] = 6144
photflag = np.array(photflag)
light_curve_info1 = (mjd, flux, fluxerr, passband, photflag, ra, dec,
objid, redshift, mwebv)
light_curve_list = [
light_curve_info1,
]
classification = Classify(known_redshift=True,
model_filepath='keras_model.hdf5')
predictions = classification.get_predictions(light_curve_list)
y_predict = predictions[0][0]
time_steps = predictions[1][0]
classified_lightcurve = {
'time': datetime.datetime.utcnow(),
'obj_id': objid,
'mjd': mjd,
'flux': flux,
'fluxerr': fluxerr,
'passband': passband,
'ra': ra,
'dec': dec,
'redshift': redshift,
'mwebv': mwebv,
'photflag': photflag.tolist(),
'predicted_y': y_predict.tolist(),
'timesteps': time_steps.tolist(),
}
print(classified_lightcurve['timesteps'])
client = MongoClient('ampelml-mongo')
db = client.ampel_ml
classified_lightcurves = db.classified_lightcurves
class_lc_id = classified_lightcurves.insert_one(
classified_lightcurve).inserted_id
last_class_lc = classified_lightcurves.find().sort([('time', -1)
]).limit(1)[0]
return dumps(last_class_lc)
def main():
mjd = [
57433.4816, 57436.4815, 57439.4817, 57451.4604, 57454.4397, 57459.3963,
57462.418, 57465.4385, 57468.3768, 57473.3606, 57487.3364, 57490.3341,
57493.3154, 57496.3352, 57505.3144, 57513.2542, 57532.2717, 57536.2531,
57543.2545, 57546.2703, 57551.2115, 57555.2669, 57558.2769, 57561.1899,
57573.2133, 57433.5019, 57436.4609, 57439.4587, 57444.4357, 57459.4189,
57468.3142, 57476.355, 57479.3568, 57487.3586, 57490.3562, 57493.3352,
57496.2949, 57505.3557, 57509.2932, 57513.2934, 57518.2735, 57521.2739,
57536.2321, 57539.2115, 57543.2301, 57551.1701, 57555.2107, 57558.191,
57573.1923, 57576.1749, 57586.1854
]
flux = [
2.0357230e+00, -2.0382695e+00, 1.0084588e+02, 5.5482742e+01,
1.4867026e+01, -6.5136810e+01, 1.6740545e+01, -5.7269131e+01,
1.0649184e+02, 1.5505235e+02, 3.2445984e+02, 2.8735449e+02,
2.0898877e+02, 2.8958893e+02, 1.9793906e+02, -1.3370536e+01,
-3.9001358e+01, 7.4040916e+01, -1.7343750e+00, 2.7844931e+01,
6.0861992e+01, 4.2057487e+01, 7.1565346e+01, -2.6085690e-01,
-6.8435440e+01, 17.573107, 41.445435, -110.72664, 111.328964,
-63.48336, 352.44907, 199.59058, 429.83075, 338.5255, 409.94604,
389.71262, 195.63905, 267.13318, 123.92461, 200.3431, 106.994514,
142.96387, 56.491238, 55.17521, 97.556946, -29.263103, 142.57687,
-20.85057, -0.67210346, 63.353024, -40.02601
]
fluxerr = [
42.784702, 43.83665, 99.98704, 45.26248, 43.040398, 44.00679,
41.856007, 49.354336, 105.86439, 114.0044, 45.697918, 44.15781,
60.574158, 93.08788, 66.04482, 44.26264, 91.525085, 42.768955,
43.228336, 44.178196, 62.15593, 109.270035, 174.49638, 72.6023,
48.021034, 44.86118, 48.659588, 100.97703, 148.94061, 44.98218,
139.11194, 71.4585, 47.766987, 45.77923, 45.610615, 60.50458,
105.11658, 71.41217, 43.945534, 45.154167, 43.84058, 52.93122,
44.722775, 44.250145, 43.95989, 68.101326, 127.122025, 124.1893,
49.952255, 54.50728, 114.91599
]
passband = [
'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g',
'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'r', 'r', 'r',
'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r',
'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r'
]
zeropoint = [
27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5,
27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5,
27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5,
27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5, 27.5,
27.5, 27.5, 27.5
]
photflag = [
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4096, 4096, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 4096, 6144, 4096, 4096, 4096, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
]
objid = 'MSIP_01_NONIa-0001_10400862'
ra = 3.75464531293933
dec = 0.205076187109334
redshift = 0.233557
mwebv = 0.0228761
light_curve_list = [(mjd, flux, fluxerr, passband, zeropoint, photflag, ra,
dec, objid, redshift, mwebv)]
classification = Classify(light_curve_list)
predictions = classification.get_predictions()
print(predictions)
classification.plot_light_curves_and_classifications()
classification.plot_classification_animation()
def do(self,debug=False):
# run this under Admin
user = User.objects.get(username='******')
# get New, Watch, FollowupRequested, Following
transients_to_classify = \
Transient.objects.filter(Q(status__name = 'New') |
Q(status__name = 'Watch') |
Q(status__name = 'FollowupRequested') |
Q(status__name = 'Following'))
light_curve_list_z,light_curve_list_noz,transient_list_z,transient_list_noz = [],[],[],[]
for t in transients_to_classify: #.filter(Q(name = '2019np') | Q(name = '2019gf')):
ra, dec, objid, redshift = t.ra, t.dec, t.name,t.z_or_hostz()
if not t.mw_ebv is None:
mwebv = t.mw_ebv
else:
mwebv = 0.0
photdata = get_all_phot_for_transient(user, t.id)
if not photdata: continue
gobs = photdata.filter(band__name = 'g-ZTF')
robs = photdata.filter(band__name = 'r-ZTF')
if not len(gobs) and not len(robs): continue
mjd, passband, flux, fluxerr, mag, magerr, zeropoint, photflag = \
np.array([]),np.array([]),[],[],[],[],[],[]
if redshift: transient_list_z += [t]
else: transient_list_noz += [t]
first_detection_set = False
for obs,filt in zip([gobs.order_by('obs_date'),robs.order_by('obs_date')],['g','r']):
for p in obs:
if p.data_quality: continue
if len(np.where((filt == passband) & (np.abs(mjd - date_to_mjd(p.obs_date)) < 0.001))[0]): continue
mag += [p.mag]
if p.mag_err:
magerr += [p.mag_err]
fluxerr_obs = 0.4*np.log(10)*p.mag_err
else:
magerr += [0.001]
fluxerr_obs = 0.4*np.log(10)*0.001
flux_obs = 10**(-0.4*(p.mag-27.5))
mjd = np.append(mjd,[date_to_mjd(p.obs_date)])
flux += [flux_obs]
fluxerr += [fluxerr_obs]
zeropoint += [27.5]
passband = np.append(passband,[filt])
if flux_obs/fluxerr_obs > 5 and not first_detection_set:
photflag += [6144]
first_detection_set = True
elif flux_obs/fluxerr_obs > 5:
photflag += [4096]
else: photflag += [0]
#except: import pdb; pdb.set_trace()
try:
if redshift:
light_curve_info = (mjd, flux, fluxerr, passband,
zeropoint, photflag, ra, dec, objid, redshift, mwebv)
light_curve_list_z += [light_curve_info,]
else:
light_curve_info = (mjd, flux, fluxerr, passband, zeropoint, photflag, ra, dec, objid, None, mwebv)
light_curve_list_noz += [light_curve_info,]
except:
import pdb; pdb.set_trace()
if len(light_curve_list_noz):
classification_noz = Classify(light_curve_list_noz, known_redshift=False, bcut=False, zcut=None)
predictions_noz = classification_noz.get_predictions()
if len(light_curve_list_z):
classification_z = Classify(light_curve_list_z, known_redshift=True, bcut=False, zcut=None)
predictions_z = classification_z.get_predictions()
if debug:
import matplotlib
matplotlib.use('MacOSX')
import matplotlib.pyplot as plt
plt.ion()
classification_z.plot_light_curves_and_classifications()
for tl in [transient_list_z,transient_list_noz]:
for t,i in zip(tl,range(len(tl))):
best_predictions = predictions_z[0][i][-1,:]
adjusted_best_predictions = np.zeros(10)
idx,outclassnames,PIa = 0,[],0
for j in range(len(classification_z.class_names)):
if classification_z.class_names[j] == 'Pre-explosion': continue
elif classification_z.class_names[j].startswith('SNIa'): PIa += best_predictions[j]
else:
outclassnames += [classification_z.class_names[j]]
adjusted_best_predictions[idx] = best_predictions[j]
idx += 1
outclassnames += ['SN Ia']
outclassnames = np.array(outclassnames)
adjusted_best_predictions[9] = PIa
print(t.name,outclassnames[adjusted_best_predictions == np.max(adjusted_best_predictions)][0])
transient_class = outclassnames[adjusted_best_predictions == np.max(adjusted_best_predictions)][0]
photo_class = TransientClass.objects.filter(name = classdict[transient_class])
if len(photo_class):
t.photo_class = photo_class[0]
t.save()
else:
print('class %s not in DB'%classdict[transient_class])
raise RuntimeError('class %s not in DB'%classdict[transient_class])
def classify_lasair_light_curves(
object_names=('ZTF18acsovsw', ), plot=True, figdir='.'):
light_curve_list = []
peakfluxes_g, peakfluxes_r = [], []
mjds, passbands, mags, magerrs, zeropoints, photflags = [], [], [], [], [], []
obj_names = []
ras, decs, objids, redshifts, mwebvs = [], [], [], [], []
peakmags_g, peakmags_r = [], []
for object_name in object_names:
try:
mjd, passband, mag, magerr, photflag, zeropoint, ra, dec, objid, redshift, mwebv = read_lasair_json(
object_name)
sortidx = np.argsort(mjd)
mjds.append(mjd[sortidx])
passbands.append(passband[sortidx])
mags.append(mag[sortidx])
magerrs.append(magerr[sortidx])
zeropoints.append(zeropoint[sortidx])
photflags.append(photflag[sortidx])
obj_names.append(object_name)
ras.append(ra)
decs.append(dec)
objids.append(objid)
redshifts.append(redshift)
mwebvs.append(mwebv)
peakmags_g.append(min(mag[passband == 1]))
peakmags_r.append(min(mag[passband == 2]))
except Exception as e:
print(e)
continue
flux = 10.**(-0.4 * (mag - zeropoint))
fluxerr = np.abs(flux * magerr * (np.log(10.) / 2.5))
passband = np.where((passband == 1) | (passband == '1'), 'g', passband)
passband = np.where((passband == 2) | (passband == '2'), 'r', passband)
# Set photflag detections when S/N > 5
photflag2 = np.zeros(len(flux))
photflag2[flux / fluxerr > 5] = 4096
photflag2[np.where(mjd == min(mjd[photflag2 == 4096]))] = 6144
mjd_first_detection = min(mjd[photflag == 4096])
photflag[np.where(mjd == mjd_first_detection)] = 6144
deleteindexes = np.where(((passband == 3) | (passband == '3'))
| (mjd > mjd_first_detection)
& (photflag == 0))
if deleteindexes[0].size > 0:
print("Deleting indexes {} at mjd {} and passband {}".format(
deleteindexes, mjd[deleteindexes], passband[deleteindexes]))
mjd, passband, flux, fluxerr, zeropoint, photflag = delete_indexes(
deleteindexes, mjd, passband, flux, fluxerr, zeropoint, photflag)
light_curve_list += [(mjd, flux, fluxerr, passband, photflag, ra, dec,
objid, redshift, mwebv)]
try:
dummy = max(flux[passband == 'g'])
dummy = max(flux[passband == 'r'])
except Exception as e:
print(e)
continue
peakfluxes_g.append(max(flux[passband == 'g']))
peakfluxes_r.append(max(flux[passband == 'r']))
# import sys
# import pickle
# with open('save_real_ZTF_unprocessed_data_snia_osc_12nov2019.npz', 'wb') as f:
# pickle.dump([mjds, passbands, mags, magerrs, photflags, zeropoints, ras, decs, objids, redshifts, mwebvs], f)
# # np.savez('save_real_ZTF_unprocessed_data_snia_osc_12nov2019.npz', mjds=mjds, passbands=passbands, mags=mags, magerrs=magerrs, photflags=photflags, zeropoints=zeropoints, ras=ras, decs=decs, objids=objids, redshifts=redshifts, mwebvs=mwebvs)# , peakflux_g=peakfluxes_g, peakflux_r=peakfluxes_r)
# print("finished")
# # # sys.exit(0)
# # with open('save_real_ZTF_unprocessed_data_snia_osc_12nov2019.npz', 'rb') as f:
# # a = pickle.load(f)
classification = Classify(known_redshift=True, bcut=False, zcut=None)
predictions, time_steps = classification.get_predictions(
light_curve_list, return_predictions_at_obstime=False)
print(predictions)
if plot:
# try:
classification.plot_light_curves_and_classifications(
step=True,
use_interp_flux=False,
figdir=figdir,
plot_matrix_input=True)
# except Exception as e:
# print(e)
# classification.plot_light_curves_and_classifications(step=False, use_interp_flux=True)
# classification.plot_light_curves_and_classifications(step=False, use_interp_flux=False)
# classification.plot_classification_animation_step()
# classification.plot_classification_animation()
return classification.orig_lc, classification.timesX, classification.y_predict
1.9793906e+02, -1.3370536e+01, -3.9001358e+01, 7.4040916e+01, -1.7343750e+00, 2.7844931e+01, 6.0861992e+01,
4.2057487e+01, 7.1565346e+01, -2.6085690e-01, -6.8435440e+01, 17.573107, 41.445435, -110.72664, 111.328964,
-63.48336, 352.44907, 199.59058, 429.83075, 338.5255, 409.94604, 389.71262, 195.63905, 267.13318, 123.92461,
200.3431, 106.994514, 142.96387, 56.491238, 55.17521, 97.556946, -29.263103, 142.57687, -20.85057, -0.67210346,
63.353024, -40.02601]
fluxerr = [42.784702, 43.83665, 99.98704, 45.26248, 43.040398, 44.00679, 41.856007, 49.354336, 105.86439, 114.0044,
45.697918, 44.15781, 60.574158, 93.08788, 66.04482, 44.26264, 91.525085, 42.768955, 43.228336, 44.178196,
62.15593, 109.270035, 174.49638, 72.6023, 48.021034, 44.86118, 48.659588, 100.97703, 148.94061, 44.98218,
139.11194, 71.4585, 47.766987, 45.77923, 45.610615, 60.50458, 105.11658, 71.41217, 43.945534, 45.154167,
43.84058, 52.93122, 44.722775, 44.250145, 43.95989, 68.101326, 127.122025, 124.1893, 49.952255, 54.50728,
114.91599]
passband = ['g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g', 'g',
'g', 'g', 'g', 'g', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r',
'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r', 'r']
photflag = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4096, 4096, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4096,
6144, 4096, 4096, 4096, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]
objid = 'transient_1'
ra = 3.75464531293933
dec = 0.205076187109334
redshift = 0.233557
mwebv = 0.0228761
light_curve_list = [(mjd, flux, fluxerr, passband, photflag, ra, dec, objid, redshift, mwebv)]
classification = Classify()
predictions, time_steps = classification.get_predictions(light_curve_list)
print(predictions)
classification.plot_light_curves_and_classifications()
classification.plot_classification_animation()
class RAPIDModel:
"""Wrapper for the model outlined by the RAPID paper (Muthukrishna et al 2019)"""
def __init__(self, model: str = None):
"""
Creates a new instance of RAPID
Parameters
----------
model: str, optional
The filepath of the model to be loaded, if any.
"""
if model is not None:
self.classifier = Classify(known_redshift=True,
model_filepath=model)
else:
self.classifier = Classify(known_redshift=True)
def set_metadata(self, metadata: pd.DataFrame):
"""Sets the loaded metadata
Parameters
----------
metadata: pd.DataFrame :
The new metadata to be loaded
"""
assert isinstance(metadata, pd.DataFrame)
self._metadata = metadata
def set_curves(self, curves: pd.DataFrame):
"""Sets the loaded transient data
Parameters
----------
curves: pd.DataFrame :
The new transient data to be loaded
"""
assert isinstance(curves, pd.DataFrame)
self._curves = curves
def set_data(self, curves: pd.DataFrame, metadata: pd.DataFrame):
"""Set both the transient data and the metadata
Parameters
----------
curves: pd.DataFrame :
The new transient data to be loaded
metadata: pd.DataFrame :
The new metadata to be loaded
"""
self.set_curves(curves)
self.set_metadata(metadata)
def _get_custom_data(self, class_num, data_dir, save_dir, passbands,
known_redshift, nprocesses, redo):
"""Function for traning purposes.
Notes
-----
See astrorapid for API usage.
"""
light_list, target_list = p2r.convert(self._curves, self._metadata)
# now we need to preprocess
return read_multiple_light_curves(light_list)
def train(self,
curves: pd.DataFrame,
metadata: pd.DataFrame,
*,
class_map: dict = p2r.class_map,
band_map: dict = p2r.band_map,
save_path: str = "models/",
file_name: str = None,
load_model: bool = False):
"""Train the model on the loaded data.
Parameters
----------
curves : pd.DataFrame
The curve data to train on.
metadata : pd.DataFrame
The metadata for each curve.
class_map : dict, optional
The class mapping from PLAsTiCC to the model.
band_map : dict, optional
The bands and their mapping from PLAsTiCC to models.
save_path : str, optional
Where the model should be saved. Default "models/"
file_name : str, optional
Override the filename of the model. Defaults to RAPIDModel_yyyy_mm_dd_hh_mm_ss.hdf5
load_model : bool, optional
Whether the model should be loaded into the class after training. Default False.
"""
# we need to create a new model and replace the classifier with it.
# use currying to return a function.
def _get_data(class_num, data_dir, save_dir, passbands, known_redshift,
nprocesses, redo, calculate_t0):
# This is a function RAPID needs to call to get the data.
# get the class num tuple
class_map = {
key: value
for (key, value) in p2r.class_map.items() if value is class_num
}
band_map = {
key: value
for (key, value) in p2r.band_map.items() if key in passbands
}
light_list, target_list = p2r.convert(curves,
metadata,
classes=class_map,
bands=band_map)
# now we need to preprocess
return read_multiple_light_curves(light_list)
#
create_custom_classifier(
get_data_func=_get_data,
data_dir='data/',
class_nums=tuple(class_map.values()),
passbands=tuple(band_map.keys()),
save_dir=save_path,
)
pass
def test(self,
curves: pd.DataFrame,
metadata: pd.DataFrame,
return_probabilities: bool = False) -> (list, list):
"""Tests the model on the currently loaded data.
Parameters
----------
curves : pd.DataFrame
The transient data for each object.
metadata : pd.DataFrame
The metadata for each object.
return_probabilities : bool, optional
If the predictions should be a probability of arrays as opposed to the most likely class
Returns
-------
target_list : list of int
A list of true classes, ordered.
predictions_list : list of int or list of list of int
A list of predictions, ordered. Depending on the value of `return_probabilities`,
can either be an int
"""
logger.info("testing model")
light_list, target_list = p2r.convert(curves, metadata)
predictions, steps = self.classifier.get_predictions(light_list)
assert len(target_list) == len(predictions)
target_list = np.add(target_list, 1)
logger.info("done testing model")
if return_probabilities:
return target_list, predictions
predictions_list = []
for index, pred in enumerate(predictions):
pred_class = np.argmax(pred[-1])
# if pred_class is forbidden (0 or 9) don't bother
# if pred_class < 0 or pred_class > 8:
# continue
predictions_list.append(pred_class)
return target_list, predictions_list
