def plot(fname):
setup_text_plots(fontsize=10, usetex=True)
data = dl.read_dat(fname,',')
n = len(data[0]) - 4
f,plots = plt.subplots(int(np.ceil(n/2))+n%2,2, sharex=False,sharey=False)
sigmas = []
for sigma in range(0,n):
sigmas.append(dl.get_column(data,4+sigma))
for j in range(0,len(sigmas)):
sigma = sigmas[j]
good_sigmas = []
for i in range(0,len(sigma)):
if sigma[i] != 'nan':
good_sigmas.append((float(sigma[i])))
good_sigmas = np.sort(good_sigmas)
good_sigmas = good_sigmas[0:int(.8*len(good_sigmas))]
hist = np.histogram(good_sigmas,bins = 175)
plots[np.ceil((j)/2)][(j)%2].bar(hist[1][:-1],hist[0],width=hist[1][1]-hist[1][0])
plots[np.ceil((j)/2)][(j)%2].set_xlabel('$\sigma_{'+str(j+1)+'}$',fontsize=40)
plots[np.ceil((j)/2)][(j)%2].set_ylabel('$n$',fontsize=40)
plots[np.ceil((j)/2)][(j)%2].text(plots[np.ceil((j)/2)][(j)%2].get_xlim()[1]*0.9,plots[np.ceil((j)/2)][(j)%2].get_ylim()[1]*0.75,'$\sigma_{'+str(j+1)+'}$',fontsize=40)
plots[np.ceil((j)/2)][(j)%2].tick_params(axis='both', which='major', labelsize=20)
f.set_size_inches(32,20)
f.savefig('histograms/sigma_histo_' + fname+'.png', dpi = 300)
def detect(fname): data = dl.read_dat(fname,',') n = len(data[0]) - 4 sigmas = [] stds = [] ids = dl.get_column(data,0) for sigma in range(0,n): sigmas.append(dl.get_column(data,4+sigma)) for star in range(len(sigmas[0])): star_sigmas = [] for sigma in range(0,n): if sigmas[sigma][star] != 'nan' and sigmas[sigma][star] != '': star_sigmas.append(float(sigmas[sigma][star])) stds.append(np.std(star_sigmas)) stds_median = np.median(stds) stds_std = np.std(stds) bad_stars = [] for i in range(len(stds)): if abs(stds[i]-stds_median) > 3*stds_std: bad_stars.append(ids[i]) out = '' for star in bad_stars: out = out + star + ',' print fname + ' : ' + out
def plot(file): print file data = dl.read_dat(file,',') time = dl.get_column_numerized(data,0) flux = dl.get_column_numerized(data,1) plt.plot(time,flux) plt.show()
def plot(fname):
setup_text_plots(fontsize=10, usetex=True)
data = dl.read_dat(fname,',')
print data[0]
n = len(data[0]) - 4
print int(np.ceil(n/2))+n%2
f,plots = plt.subplots(int(np.ceil(n/2))+n%2,2, sharex=False,sharey=False)
dwarf_flags = dl.get_column(data,1)
kepmags = dl.get_column(data,3)
Teffs = dl.get_column(data,2)
sigmas = []
for sigma in range(0,n):
sigmas.append(dl.get_column(data,4+sigma))
length = len(dl.get_column(data,0))
for i in range(0,length):
done = []
for seg in range(0,n):
done.append(seg)
is_dwarf = dwarf_flags[i]
if is_dwarf == '1.0' or is_dwarf == '1':
symbol = 'd'
elif is_dwarf == '0.0' or is_dwarf == '0':
symbol = 'o'
else:
symbol = 'x'
x = int(np.ceil((seg)/2))
y = (seg)%2
plots[x][y].set_ylabel('$\log{\sigma_{' + str(seg+1) +'}}$',fontsize=40)
plots[x][y].set_xlabel('Kepler band magnitude',fontsize=20)
plots[x][y].tick_params(axis='both', which='major', labelsize=20)
plots[x][y].grid(True)
plots[x][y].set_xlim([8.25,17])
plots[x][y].set_ylim([-4.5,0])
plots[x][y].text(15.25,-0.5,'Segment ' + str(seg+1),fontsize=30)
if sigma != 'nan' and kepmags[i] != 'nan' and sigmas[seg][i] != 'nan' and sigmas[seg][i] != '':
plots[x][y].scatter(float(kepmags[i]),np.log10(float(sigmas[seg][i])),marker=symbol,color=color_T(Teffs[i]))
print fname + ' : ' + str(done) + ' : ' + str((i*100.0)/length) + '%'
f.set_size_inches(32,32)
f.savefig(fname+'.png', dpi = 300)
def plot(fname):
setup_text_plots(fontsize=10, usetex=True)
data = dl.read_dat(fname,',')
mags = dl.get_column(data,3)
for i in range(0,len(mags)):
if mags[i] == 'nan':
del mags[i]
else:
mags[i] = float(mags[i])
hist = np.histogram(mags,bins=100)
plt.bar(hist[1][:-1],hist[0],width=hist[1][1]-hist[1][0])
plt.xlabel('Kepler band magnitude',fontsize=50)
plt.ylabel('$n$',fontsize=50)
plt.tick_params(axis='both', which='major', labelsize=40)
plt.gcf().set_size_inches(32,20)
plt.gcf().savefig(fname.replace('evaluations/','histograms/mag/mag_histo_') +'.png', dpi = 300)
def medians(fname): setup_text_plots(fontsize=10, usetex=True) data = dl.read_dat(fname,',') n = len(data[0]) - 4 sigmas = [] for sigma in range(0,n): sigmas.append(dl.get_column(data,4+sigma)) medians = [] for j in range(0,len(sigmas)): sigma = sigmas[j] good_sigmas = [] for i in range(0,len(sigma)): if sigma[i] != 'nan': good_sigmas.append((float(sigma[i]))) good_sigmas = np.sort(good_sigmas) good_sigmas = good_sigmas[0:int(.8*len(good_sigmas))] medians.append(np.median(good_sigmas)) return medians
import datalib as dl
import numpy as np
import matplotlib.pyplot as plt
data = dl.read_dat('evaluation.csv',',')
sigmas = []
for sigma in range(4,9):
sigmas.append(dl.get_column(data,sigma))
medians = []
for sigma in sigmas:
good_sigmas = []
for i in range(1,len(sigma)):
if sigma[i] != 'nan':
if np.log10(float(sigma[i])):
good_sigmas.append(float(sigma[i]))
good_sigmas = np.sort(good_sigmas)
good_sigmas = good_sigmas[0:int(0.8*len(good_sigmas))]
medians.append(np.median(good_sigmas))
plt.plot([1,2,3,4,5],medians)
plt.show()
targets = datalib.get_column(
K2.read_target_file('K2Campaign0targets.csv - K2Campaign0targets.csv'), 0)
del targets[0]
result = [[
'EPIC', 'is_dwarf', 'teff', 'kepmag', 'sigma_1', 'sigma_2', 'sigma_3',
'sigma_4', 'sigma_5'
]]
errors = []
quality = [[
'#EPIC', 'seg_1_ratio', 'seg_1_flag', 'seg_2_ratio', 'seg_2_flag',
'seg_3_ratio', 'seg_3_flag', 'seg_4_ratio', 'seg_4_flag', 'seg_5_ratio',
'seg_5_flag'
]]
tess_data = datalib.read_dat('k2tess.csv', ',')
del tess_data[0]
faint_data = datalib.read_dat('k2faint.csv', ',')
del faint_data[0]
for i in range(0, len(targets)):
target = targets[i]
print target + ' : ' + str(int((100.0 * i) / len(targets))) + '%'
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
return [0, 0, 0]
Teff = float(Teff)
x = (Teff - 2265) / 5375.0
return [-x * x + 1, -4 * x * (x - 1), -x * (x - 2)]
def label_d(is_dwarf):
if is_dwarf == '1.0':
return 'Dwarf'
elif is_dwarf == '0.0':
symbol = 'Giant'
else:
symbol = 'Unkown'
data = dl.read_dat('evaluation.csv', ',')
f, plots = plt.subplots(3, 2, sharex=False, sharey=False)
dwarf_flags = dl.get_column(data, 1)
kepmags = dl.get_column(data, 3)
Teffs = dl.get_column(data, 2)
sigmas = []
for sigma in range(4, 9):
sigmas.append(dl.get_column(data, sigma))
length = len(dl.get_column(data, 0))
for i in range(1, length):
for seg in range(4, 9):
is_dwarf = dwarf_flags[i]
if is_dwarf == '1.0':
KEPLER_CADENCE = 29.42
def get_data(star_id,data):
for row in data:
if float(star_id) == float(row[0]):
return row[1:4]
return ['nan']*3
targets = datalib.get_column(K2.read_target_file('K2Campaign0targets.csv - K2Campaign0targets.csv'),0)
del targets[0]
result = [['EPIC','is_dwarf','teff','kepmag','sigma_1','sigma_2','sigma_3','sigma_4','sigma_5']]
errors = []
quality = [['#EPIC','seg_1_ratio','seg_1_flag','seg_2_ratio','seg_2_flag','seg_3_ratio','seg_3_flag','seg_4_ratio','seg_4_flag','seg_5_ratio','seg_5_flag']]
tess_data = datalib.read_dat('k2tess.csv',',')
del tess_data[0]
faint_data = datalib.read_dat('k2faint.csv',',')
del faint_data[0]
for i in range(0,len(targets)):
target = targets[i]
print target + ' : ' + str(int((100.0*i)/len(targets))) + '%'
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
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',',')