def test_plot(self):
# download data needed to generate the plots
alltime, allflux, allflux_err, all_md, alltimebinned, allfluxbinned, allx1, allx2, ally1, ally2, alltime12, allfbkg, start_sec, end_sec, in_sec, tessmag, teff, srad = utils.download_data(
indir,
sector,
tic,
binfac=5,
test='./LATTE/tests/tic55525572_lc.fits')
X1_list, X4_list, oot_list, intr_list, bkg_list, apmask_list, arrshape_list, t_list, T0_list, tpf_filt_list = utils.download_tpf_mast(
indir,
transit_sec,
transit_list,
tic,
test='./LATTE/tests/tic55525572_tp.fits')
# this function downloads the data and saved plots that show the aperture sizes (for the large and small apertures)
TESS_unbinned_t_l, TESS_binned_t_l, small_binned_t_l, TESS_unbinned_l, TESS_binned_l, small_binned_l, tpf_list = utils.download_tpf_lightkurve(
indir,
transit_list,
sectors,
tic,
test='./LATTE/tests/tic55525572_tp.fits')
# ---------
# create the plots that the normal routine makes
utils.plot_full_md(tic, indir, alltime, allflux, all_md, alltimebinned,
allfluxbinned, transit_list, args)
utils.plot_centroid(tic, indir, alltime12, allx1, ally1, allx2, ally2,
transit_list, args)
utils.plot_background(tic, indir, alltime, allfbkg, transit_list, args)
utils.plot_pixel_level_LC(tic, indir, X1_list, X4_list, oot_list,
intr_list, bkg_list, apmask_list,
arrshape_list, t_list, transit_list, args)
utils.plot_aperturesize(tic, indir, TESS_unbinned_t_l, TESS_binned_t_l,
small_binned_t_l, TESS_unbinned_l,
TESS_binned_l, small_binned_l, transit_list,
args)
# --------
# now check that the plots were actually made!
# in order to be able to run this code multiple times, assert that the files were created in the last minuet - if the code takes lomger than that to run something is goimg wrong...?
# without checking when the file is made, the files would have to be manually deleted after every test
# these are the paths of the files that should have bee created
full_LC_path = '{}/55525572/55525572_fullLC_md.png'.format(indir)
full_centroid = '{}/55525572/55525572_centroids.png'.format(indir)
bkg_path = '{}/55525572/55525572_background.png'.format(indir)
pixel_path = '{}/55525572/55525572_individual_pixel_LCs_0.png'.format(
indir)
ap_LC_path = '{}/55525572/55525572_aperture_size.png'.format(indir)
apertures_path = '{}/55525572/55525572_apertures_0.png'.format(indir)
# Get file's Last modification time stamp only in terms of seconds since epoch
time_created_full_LC = os.path.getmtime(full_LC_path)
time_created_centroid = os.path.getmtime(full_centroid)
time_created_bkg = os.path.getmtime(bkg_path)
time_created_pixel = os.path.getmtime(pixel_path)
time_created_ap_LC = os.path.getmtime(ap_LC_path)
time_created_apertures = os.path.getmtime(apertures_path)
# Convert seconds since epoch to readable timestamp
t_create_full_LC = parser.parse(
time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time_created_full_LC)))
t_create_centroid = parser.parse(
time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time_created_centroid)))
t_create_bkg = parser.parse(
time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time_created_bkg)))
t_create_pixel = parser.parse(
time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time_created_pixel)))
t_create_ap_LC = parser.parse(
time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time_created_ap_LC)))
t_create_apertures = parser.parse(
time.strftime('%Y-%m-%d %H:%M:%S',
time.localtime(time_created_apertures)))
# the time now (needed to get the time since creation)
t_now = (datetime.datetime.now())
# -------
# time difference in minutes
time_since_creation_full_LC = ((t_now - t_create_full_LC).seconds / 60)
time_since_creation_centroid = ((t_now - t_create_centroid).seconds /
60)
time_since_creation_bkg = ((t_now - t_create_bkg).seconds / 60)
time_since_creation_pixel = ((t_now - t_create_pixel).seconds / 60)
time_since_creation_bkg = ((t_now - t_create_ap_LC).seconds / 60)
time_since_creation_pixel = ((t_now - t_create_apertures).seconds / 60)
# check that the expected files were created less than a minute ago
self.assertLess(
time_since_creation_full_LC, 1,
"No (new) full LC plot was made in the last five minutes"
) # a less than b
self.assertLess(
time_since_creation_centroid, 1,
"No (new) centroid plot was made in the last five minutes"
) # a less than b
self.assertLess(
time_since_creation_bkg, 1,
"No (new) background plot was made in the last five minutes"
) # a less than b
self.assertLess(
time_since_creation_pixel, 1,
"No (new) pixel level LC plot was made in the last five minutes"
) # a less than b
self.assertLess(
time_since_creation_bkg, 1,
"No (new) background plot was made in the last five minutes"
) # a less than b
self.assertLess(
time_since_creation_pixel, 1,
"No (new) pixel level LC plot was made in the last five minutes"
) # a less than b
def test_LC_request_response(self):
# Call the service, which will send a request to the server.
TESS_unbinned_t_l, TESS_binned_t_l, small_binned_t_l, TESS_unbinned_l, TESS_binned_l, small_binned_l, tpf_list = utils.download_tpf_lightkurve(indir, transit_list, sectors, tic, test = './LATTE/tests/tic55525572_tp.fits')
print(float(TESS_unbinned_t_l[0]))
print(float(TESS_binned_t_l[0]))
print(float(small_binned_t_l[0]))
print(float(TESS_unbinned_l[0]))
print(float(TESS_binned_l[0]))
print(float(small_binned_l[0]))
# If the request is sent successfully, then I expect a response to be returned.
self.assertAlmostEqual(float(TESS_unbinned_t_l[0]), float(1437.9924102871835), msg='TP lighkurve data is not what it should be.')
self.assertAlmostEqual(float(TESS_binned_t_l[0]), float(1437.9972713631325), places=5)
self.assertAlmostEqual(float(small_binned_t_l[0]), float(1437.9972713631325), places=5)
self.assertAlmostEqual(float(TESS_unbinned_l[0]), float(0.9975904130977179), places=5)
self.assertAlmostEqual(float(TESS_binned_l[0]), float(0.9986087992155438), places=5)
self.assertAlmostEqual(float(small_binned_l[0]), float(0.9971978343909532), places=5)
def brew_LATTE(tic, indir, syspath, transit_list, simple, BLS, model, save, DV,
sectors, sectors_all, alltime, allflux, allflux_err, all_md,
alltimebinned, allfluxbinned, allx1, allx2, ally1, ally2,
alltime12, allfbkg, start_sec, end_sec, in_sec, tessmag, teff,
srad, ra, dec, args):
'''
This function combines all the results from LATTE and calls all the different functions -
it makes the plots, saves them, runs the BLS model and the pyaneti model before making a PHT DV report (if this option is selected.)
Parameters
----------
tic : str
target TIC ID
indir : str
path to directory where all the plots and data will be saved.
transit_list : list
list of the transit-like events
simple : boolean
whether or not to run the simple version
BLS : boolean
whether or not to run the BLS routine
model : boolean
whether or not to model the transit using pyaneti
save : boolean
whether or not to save the figures and data
DV : boolean
whether or not to write and save a DV report
sectors_all : list
all the sectors in which the target has been/ will be observed
alltime : list
times (not binned)
allflux : list
normalized flux (not binned)
allflux_err : list
normalized flux errors (not binned)
all_md : list
times of the momentum dumps
alltimebinned : list
binned time
allfluxbinned : list
normalized binned flux
allx1 : list
CCD column position of target’s flux-weighted centroid. In x direction
allx2 : list
The CCD column local motion differential velocity aberration (DVA), pointing drift, and thermal effects. In x direction
ally1 : list
CCD column position of target’s flux-weighted centroid. In y direction
ally2 : list
The CCD column local motion differential velocity aberration (DVA), pointing drift, and thermal effects. In y direction
alltimel2 : list
time used for the x and y centroid position plottin
allfbkg : list
background flux
start_sec : list
times of the start of the sector
end_sec : list
times of the end of the sector
in_sec : list
the sectors for which data was downloaded
tessmag : list
TESS magnitude of the target star
teff : float
effective temperature of the tagret star (K)
srad : float
radius of the target star (solar radii)
ra : float
the right ascension of the target stars
dec : float
the declination of the target star
'''
# -------------------
# SAVE THE DATA FILES
# -------------------
if (save == True) or (DV == True):
save = True
# if this folder doesn't exist then create it. These are the folder where the images, data and reports for each TIC ID will be stored.
newpath = '{}/{}'.format(indir, tic)
if not exists(newpath):
os.makedirs(newpath)
# save the data used as a text file - these often come in use later for a quick re-analysis.
with open('{}/{}/{}_data.txt'.format(indir, tic, tic), "w") as f:
# get rid of nan values first using a mask
good_mask = np.isfinite(np.array(alltime)) * np.isfinite(
np.array(allflux)) * np.isfinite(np.array(allflux_err))
alltime_ar = np.array(alltime)[good_mask]
allflux_ar = np.array(allflux)[good_mask]
allflux_err_ar = np.array(allflux_err)[good_mask]
# save
writer = csv.writer(f, delimiter='\t')
writer.writerow(['time', 'flux', 'flux_err'])
writer.writerows(zip(alltime_ar, allflux_ar, allflux_err_ar))
'''
if the modelling option was also chose, save another data file with slightly different formatting to be called by Pyaneti.
Pyaneti requires a very specific data format.
Furhermore, in order for Pyaneti to run more efficiently (it has a Bayesian backend which scales with number of data points)
we create a cutout of the times around the times of the marked transit events.
'''
if model == True:
with open('{}/{}/{}_data_pyaneti.dat'.format(indir, tic, tic),
"w") as f:
writer = csv.writer(f, delimiter='\t')
writer.writerow(['#time', 'flux', 'flux_err'])
# If the dip separations are too small, then don't create cut outs and save the whole dataset
if (len(transit_list) > 1) and (
(transit_list[1] - transit_list[0]) < 2
): # if there are LOTS of transit events on short period (if so it's probably a TOI but let's keep it here as a condition)
with open('{}/{}/{}_data_pyaneti.dat'.format(indir, tic, tic),
"a") as f:
writer = csv.writer(f, delimiter='\t')
writer.writerows(
zip(alltime_ar, allflux_ar,
allflux_err_ar)) # save all the data
# else create a cut out of the data around the time of the transit events
else:
for transit in transit_list:
# save the data
# get rid of nan values first - this is used for the pyaneti code
pyaneti_mask = (alltime_ar >
(transit - 1)) * (alltime_ar <
(transit + 1))
with open(
'{}/{}/{}_data_pyaneti.dat'.format(
indir, tic, tic), "a") as f:
writer = csv.writer(f, delimiter='\t')
writer.writerows(
zip(alltime_ar[pyaneti_mask],
allflux_ar[pyaneti_mask],
allflux_err_ar[pyaneti_mask]))
# -----------------------------------
# START PLOTTING - calls functions from LATTEutils.py
# -----------------------------------
# create a plot of the fulllighcurves with the momentum dumps (MDs) marked and a zoom-in of the marked transits
# this plit is saved but not shown (as already shown in the interact part fo the code)
utils.plot_full_md(tic, indir, alltime, allflux, all_md, alltimebinned,
allfluxbinned, transit_list, args)
# Get a list of the sectors that have transit marked in them
# this is so that we no longer have to loop through all of the sectors, and can focus on the ones which are important.
transit_sec = utils.transit_sec(in_sec, start_sec, end_sec, transit_list)
# -----------
# plot how the centroids moved during the transit event
utils.plot_centroid(tic, indir, alltime12, allx1, ally1, allx2, ally2,
transit_list, args)
# plot the background flux at the time of the transit event.
utils.plot_background(tic, indir, alltime, allfbkg, transit_list, args)
print("Centroid and background plots... done.")
# -----------
# if the 'simple' option is chosen in the GUI, then the code will end here - this is designed to provide a quick analysis requiring no TPFs.
if simple == True:
print("Simple option was selected, therefore end analysis here.")
sys.exit('')
# -----------
# call function to extract the Target Pixel File information
# this is needed in order to extract the LCs in different aperture sizes.
# the data is extracted using the open source Lightkurve package as they a built in function to extract LCs using different aperture sizes
TESS_unbinned_t_l, TESS_binned_t_l, small_binned_t_l, TESS_unbinned_l, TESS_binned_l, small_binned_l, tpf_list = utils.download_tpf_lightkurve(
indir, transit_list, sectors, tic)
# if the TPF wasn't corrupt then make the TPF files (only very ocassionally corrupt but don't want code to crash if it is corrrupt)
if (TESS_unbinned_t_l != -111):
tpf_corrupt = False
# plot the LCs using two different aperture sizes.
utils.plot_aperturesize(tic, indir, TESS_unbinned_t_l, TESS_binned_t_l,
small_binned_t_l, TESS_unbinned_l,
TESS_binned_l, small_binned_l, transit_list,
args)
print("Aperture size plots... done.")
# ------------
'''
Plot the average pixel brightness of the cut-out around the target star and the corresponding SDSS field of view.
Both are oriented so that North is pointing upwards.
The former also shows the nearby stars with TESS magnitude brighter than 17. Queried from GAIA using astroquery.
The function returns the mass of the star (also output from astroquery)- this is a useful input for the Pyaneti modelling
'''
if args.mpi == False:
test_astroquery, _, _, mstar = utils.plot_TESS_stars(
tic, indir, transit_list, transit_sec, tpf_list, args)
else:
test_astroquery, _, _, mstar = utils.plot_TESS_stars_not_proj(
tic, indir, transit_list, transit_sec, tpf_list, args)
# keep track of whether astroquery is working (sometimes the site is down and we don't want this to stop us from running the code)
astroquery_corrupt = False
if test_astroquery == -999:
astroquery_corrupt = True
print("Star Aperture plots... failed.")
else:
print("Star Aperture plots... done.")
# ------------
# Download the Target Pixel File using the raw MAST data - this comes in a different format as the TPFs extracted using Lightkurve
# This data is then corrected using Principal Component Analysis is orderto get rid of systematics.
X1_list, X4_list, oot_list, intr_list, bkg_list, apmask_list, arrshape_list, t_list, T0_list, tpf_filt_list = utils.download_tpf_mast(
indir, transit_sec, transit_list, tic)
# ------------
'''
plot the in and out of transit flux comparison.
By default the images are NOT oriented north - this is because the reprojection takes longer to run and for a simple
analysis to check whether the brightest pixel moves during the transit this is not required.
The orientation towards north can be defined in the command line with '--north'.
'''
if args.north == True:
utils.plot_in_out_TPF_proj(tic, indir, X4_list, oot_list, t_list,
intr_list, T0_list, tpf_filt_list,
tpf_list, args)
print(
"In and out of aperture flux comparison with reprojection... done. "
)
else:
utils.plot_in_out_TPF(tic, indir, X4_list, oot_list, t_list,
intr_list, T0_list, tpf_filt_list, args)
print("In and out of aperture flux comparison... done.")
# ------------
# For each pixel in the TPF, extract and plot a lightcurve around the time of the marked transit event.
utils.plot_pixel_level_LC(tic, indir, X1_list, X4_list, oot_list,
intr_list, bkg_list, apmask_list,
arrshape_list, t_list, T0_list, args)
print("Pixel level LCs plot... done.")
# ------------
else:
tpf_corrupt = True
mstar = 1 # need to define mstar otherwise pyaneti will complain - just make it one as an approximation.
astroquery_corrupt = True
# ------------
# end of plots that require target pixel files
# ------------
# If more than one transit has been marked by the user, the LC is phase folded based on the period of the separation of the first two maarked peaks.
# These plots are saved but do not feature in the DV report.
if len(
transit_list
) > 1: # needs to know a period so can only do this if more than one transit has been marked.
period = transit_list[1] - transit_list[0]
t0 = transit_list[0] # time of the first marking
# calculate the phase
phased = np.array([
-0.5 + ((t - t0 - 0.5 * period) % period) / period
for t in alltimebinned
])
fig, ax = plt.subplots(figsize=(5.55, 5))
ax.plot(phased,
allfluxbinned,
marker='.',
color='k',
alpha=1,
lw=0,
markersize=4,
label='None',
markerfacecolor='k')
#ax.plot(phased, allflux,marker='o',color = 'navy', alpha = 0.7, lw = 0, markersize = 2, label = 'binning = 7', markerfacecolor='white')
plt.title("Phase folded LC")
ax.set_xlabel("Phase (days)")
ax.set_ylabel("Normalized Flux")
plt.plot()
if save == True:
plt.savefig('{}/{}/{}_phase_folded.png'.format(indir, tic, tic),
format='png')
if args.noshow == False:
plt.show()
print("Phase folded plot... done.")
else:
print(
"\n Only one transit marked - therefore can't be phase folded. \n")
# ------------
'''
Plot LCs of the six closest TESS target stars. This allows us to check whether the transit-like events
also appear in other nearby LCs which would be a sign that this is caused by a background event.
'''
# get the tic IDs of the six nearest stars
ticids, distance, target_ra, target_dec = utils.nn_ticids(
indir, transit_sec, tic)
# download the data for these stars
alltime_nn, allflux_nn, all_md_nn, alltimebinned_nn, allfluxbinned_nn, outtics, tessmag_list, distance = utils.download_data_neighbours(
indir, transit_sec[0], ticids, distance)
# plot the LCs
utils.plot_nn(tic, indir, alltime_nn, allflux_nn, alltimebinned_nn,
allfluxbinned_nn, transit_list, outtics, tessmag_list,
distance, args)
print("Nearest neighbour plot... done.")
# ------------
# if the BLS option is chose, a BLS search is run. The LCs are first detrended and smoothed using a moving average.
# The corrected and uncorrected LCs are saves as a single plot for comparison and to verify that the correction worked well - saved but do not feature in the DV report.
if BLS == True:
print("Running BLS algorithm...", end=" ")
bls_stats1, bls_stats2 = utils.data_bls(tic, indir, alltime, allflux,
allfluxbinned, alltimebinned,
args)
print("done.")
# ------------
# SKIP FROM HERE....
'''
NOTE: CURRENTLY ONLY WORKS ON NORA'S COMPUTER - WILL BE AVAILABLE IN NEXT RELEASE SO PLEASE SCIP THIS PART OF THE CODE
If the modelling option is selected (in the GUI), model the transit event using Pyaneti (Barragan et al 2018)
which uses an Bayesian approach with an MCMC sampling to best fit and model the transit.
The code runs slightly differently depending on whether one or multiple transits have been marked.
This is because with multiple transits the code has information about the possible orbital period.
Need to ensure that the code has compiled correctly on the users computer.
'''
# First check if Pyaneti is installed...
if os.path.exists("{}/pyaneti_LATTE.py".format(syspath)):
if model == True:
print(
"Running Pyaneti modelling - this could take a while so be patient..."
)
transit_list_model = (
"{}".format(str(np.asarray(transit_list)))[1:-1]
) # change the list into a string and get rid of the brackets
# the code is usually run through the command line so call it using the os.system function.
os.system(
"python3 {}/pyaneti_LATTE.py {} {} {} {} {} {} {}".format(
syspath, tic, indir, syspath, mstar, teff, srad,
transit_list_model))
else:
print(
"Pyaneti has not been installed so you can't model anything yet. Contact Nora or Oscar for the LATTE version of the Pyaneti code."
)
model = False
# ... UNTIL HERE
# ------------
# Finally, create a DV report which summarises all of the plots and tables.
if DV == True:
from LATTE import LATTE_DV as ldv
if BLS == True:
ldv.LATTE_DV(tic,
indir,
syspath,
transit_list,
sectors_all,
target_ra,
target_dec,
tessmag,
teff,
srad,
bls_stats1,
bls_stats2,
tpf_corrupt,
astroquery_corrupt,
FFI=False,
bls=True,
model=model,
mpi=args.mpi)
else:
ldv.LATTE_DV(tic,
indir,
syspath,
transit_list,
sectors_all,
target_ra,
target_dec,
tessmag,
teff,
srad, [0], [0],
tpf_corrupt,
astroquery_corrupt,
FFI=False,
bls=False,
model=model,
mpi=args.mpi)
