info = K2.get_K2_info(int(target),poly_fit_deg = 8)
if info['error'] == 'None':
star = []
star_data_tess = get_data(target,tess_data)
if star_data_tess != ['nan']*3:
star = [target] + star_data_tess
else:
star = [target] + get_data(target,faint_data)
star_qual = [target]
data_by_segs = info['subtraction_by_segments']
for seg in data_by_segs:
t0 = seg[0][0]*1440
t1 = seg[0][-1]*1440
num_i = (t1-t0)/KEPLER_CADENCE
if len(seg[0]) < .5*num_i:
star_qual.append('BAD')
star.append('nan')
else:
star_qual.append('GOOD')
sigma = np.std(seg[1])
star.append(sigma)
star_qual.append(len(seg[0])/float(num_i))
result.append(star)
quality.append(star_qual)
else:
errors.append(target)
datalib.write_dat(result,'evaluation.csv',',')
datalib.write_dat(errors,'bad_data.csv','')
datalib.write_dat(quality,'quality.csv',',')
info = K2.get_K2_info(int(target), poly_fit_deg=8)
if info['error'] == 'None':
star = []
star_data_tess = get_data(target, tess_data)
if star_data_tess != ['nan'] * 3:
star = [target] + star_data_tess
else:
star = [target] + get_data(target, faint_data)
star_qual = [target]
data_by_segs = info['subtraction_by_segments']
for seg in data_by_segs:
t0 = seg[0][0] * 1440
t1 = seg[0][-1] * 1440
num_i = (t1 - t0) / KEPLER_CADENCE
if len(seg[0]) < .5 * num_i:
star_qual.append('BAD')
star.append('nan')
else:
star_qual.append('GOOD')
sigma = np.std(seg[1])
star.append(sigma)
star_qual.append(len(seg[0]) / float(num_i))
result.append(star)
quality.append(star_qual)
else:
errors.append(target)
datalib.write_dat(result, 'evaluation.csv', ',')
datalib.write_dat(errors, 'bad_data.csv', '')
datalib.write_dat(quality, 'quality.csv', ',')
def evaluate(folder_name):
#ignores NumPy warnings
warnings.simplefilter('ignore', np.RankWarning)
#loads K2 tess and K2 faint catalogs
tess_data = dl.read_dat('k2tess.csv',',')
del tess_data[0]
tess_data_ids = dl.get_column(tess_data,0)
tess_data_is_dwarf = dl.get_column(tess_data,1)
tess_data_teff = dl.get_column(tess_data,2)
tess_data_mag = dl.get_column(tess_data,3)
faint_data = dl.read_dat('k2faint.csv',',')
del faint_data[0]
faint_data_ids = dl.get_column(faint_data,0)
faint_data_is_dwarf = dl.get_column(faint_data,1)
faint_data_teff = dl.get_column(faint_data,2)
faint_data_mag = dl.get_column(faint_data,3)
output = [] #2d array that will be stored in a .csv file. [[id,is_dwarf,teff,mag,sigma1,sigma2...],[id,is_dwarf,teff,mag,sigma1,sigma2...]...]
not_on_catalog = [] #list of the stars that could not be found in the tess and faint catalogs
dirlist = os.listdir(folder_name)
for n in range(0,len(dirlist)):
fname = dirlist[n]
data = dl.read_dat(folder_name + '/' + fname,',') #2d array containing the light curve. [[time,flux,segment],[time,flux,segment],...]
#Searching on K2 catalogs for info
star_id = ''
is_dwarf = ''
teff = 0
mag = 0
for i in range(0,len(tess_data_ids)):
if str(tess_data_ids[i]).replace('.0','') in fname:
star_id = tess_data_ids[i]
is_dwarf = tess_data_is_dwarf[i]
teff = tess_data_teff[i]
mag = tess_data_mag[i]
break
del tess_data_ids[i],tess_data_mag[i],tess_data_teff[i],tess_data_is_dwarf[i]
if star_id == '': #if it was not found in the tess catalog, it searches the faint catalog
for i in range(0,len(faint_data_ids)):
if str(faint_data_ids[i]).replace('.0','') in fname:
star_id = faint_data_ids[i]
is_dwarf = faint_data_is_dwarf[i]
teff = faint_data_teff[i]
mag = faint_data_mag[i]
break
del faint_data_ids[i],faint_data_mag[i],faint_data_teff[i],faint_data_is_dwarf[i]
print star_id + ' : ' + str((n*100.0/len(dirlist))) + '%' #To know how much is already done
if star_id != '' and len(data) > 1: #The code only runs if the star is on the catalog and there is data on the lightcurve
#
#Computes standard deviations for each segment:
#
time = []
flux = []
segment = []
for row in data:
time.append(float(row[0]))
flux.append(float(row[1]))
segment.append(row[2])
#Deletes >3sigma points and nan segments
sigma = np.std(flux)
avg = np.average(flux)
i = 0
end = len(time)
while i in range(0,end):
if abs(flux[i]-avg) > 3*sigma or str(segment[i]) == 'nan':
del time[i],flux[i],segment[i]
end = len(time)
i+=1
#Splits into segments
n = 0 #Number of segments
for i in range(1,len(segment)):
if str(segment[-i]) != 'nan':
n = int(float(segment[-i]))+1
break
time_seg = []
flux_seg = []
for i in range(0,n):
time_seg.append([])
flux_seg.append([])
for i in range(0,len(time)):
if segment[i] != 'nan':
time_seg[int(float(segment[i]))].append(time[i])
flux_seg[int(float(segment[i]))].append(flux[i])
#Subtracts the polynomial fit (4 deg) in each segment and computes the standard deviation
sigmas = []
for i in range(0,len(time_seg)):
if time_seg[i] != [] and len(time_seg[i]) > .5*(time_seg[i][0]-time_seg[i][-1])/KEPLER_CADENCE: #This conditional checks if there is data in the segment and also that the segment have more than the 50% of the expected points
poly = np.poly1d(np.polyfit(time_seg[i],flux_seg[i],4))
for j in range(0,len(time_seg[i])):
flux_seg[i][j] -= poly(time_seg[i][j])
sigmas.append(np.std(flux_seg[i]))
else:
sigmas.append('nan')
output.append([star_id,is_dwarf,teff,mag]+sigmas)
else:
not_on_catalog.append(fname)
print not_on_catalog
dl.write_dat(output,'evaluations/evaluation_of_' + folder_name + '.csv',',')
