all_time = []
all_flux = []
for i in range(len(filenames)):
#read into cadence
instr = fits.open(filenames[i])
tstart, tstop, bjdref, cadence = kepio.timekeys(instr, filenames[i])
#reduce lc
intime, nordata = kepreduce.reduce_lc(instr, filenames[i])
#calculte runing stddev
stddev = kepstat.running_frac_std(intime, nordata, 6.5 / 24) * 1.0e6
#cdpp
cdpp = stddev / math.sqrt(6.5 * 3600.0 / cadence)
# filter cdpp
for i in range(len(cdpp)):
if cdpp[i] > np.median(cdpp) * 10.0:
cdpp[i] = cdpp[i - 1]
#calculte RMS cdpp
rms = kepstat.rms(cdpp, np.zeros(len(stddev)))
rs = np.random.RandomState(seed=13)
flux1 = np.zeros(nordata.size)+ np.median(nordata)
errors1 = rms*1e-06*np.ones_like(nordata)
#add gaussian noise
flux1 += errors1*rs.randn(len(nordata))
all_time.append(intime)
all_flux.append(flux1)
instr.close()
t = np.concatenate(all_time)
simflux = np.concatenate(all_flux)
print(t.shape, simflux.shape)
def kepstddev(infile,outfile,datacol,timescale,clobber,verbose,logfile,status,cmdLine=False):
# startup parameters
status = 0
labelsize = 44
ticksize = 36
xsize = 16
ysize = 6
lcolor = '#0000ff'
lwidth = 1.0
fcolor = '#ffff00'
falpha = 0.2
# log the call
hashline = '----------------------------------------------------------------------------'
kepmsg.log(logfile,hashline,verbose)
call = 'KEPSTDDEV -- '
call += 'infile='+infile+' '
call += 'outfile='+outfile+' '
call += 'datacol='+str(datacol)+' '
call += 'timescale='+str(timescale)+' '
overwrite = 'n'
if (clobber): overwrite = 'y'
call += 'clobber='+overwrite+ ' '
chatter = 'n'
if (verbose): chatter = 'y'
call += 'verbose='+chatter+' '
call += 'logfile='+logfile
kepmsg.log(logfile,call+'\n',verbose)
# start time
kepmsg.clock('KEPSTDDEV started at',logfile,verbose)
# test log file
logfile = kepmsg.test(logfile)
# clobber output file
if clobber: status = kepio.clobber(outfile,logfile,verbose)
if kepio.fileexists(outfile):
message = 'ERROR -- KEPSTDDEV: ' + outfile + ' exists. Use clobber=yes'
status = kepmsg.err(logfile,message,verbose)
# open input file
if status == 0:
instr, status = kepio.openfits(infile,'readonly',logfile,verbose)
if status == 0:
tstart, tstop, bjdref, cadence, status = kepio.timekeys(instr,infile,logfile,verbose,status)
if status == 0:
try:
work = instr[0].header['FILEVER']
cadenom = 1.0
except:
cadenom = cadence
# fudge non-compliant FITS keywords with no values
if status == 0:
instr = kepkey.emptykeys(instr,file,logfile,verbose)
# read table structure
if status == 0:
table, status = kepio.readfitstab(infile,instr[1],logfile,verbose)
# filter input data table
if status == 0:
work1 = numpy.array([table.field('time'), table.field(datacol)])
work1 = numpy.rot90(work1,3)
work1 = work1[~numpy.isnan(work1).any(1)]
# read table columns
if status == 0:
intime = work1[:,1] + bjdref
indata = work1[:,0]
# calculate STDDEV in units of ppm
if status == 0:
stddev = running_frac_std(intime,indata,timescale/24) * 1.0e6
astddev = numpy.std(indata) * 1.0e6
cdpp = stddev / sqrt(timescale * 3600.0 / cadence)
# filter cdpp
if status == 0:
for i in range(len(cdpp)):
if cdpp[i] > median(cdpp) * 10.0: cdpp[i] = cdpp[i-1]
# calculate median STDDEV
if status == 0:
medcdpp = ones((len(cdpp)),dtype='float32') * median(cdpp[:])
# print '\nMedian %.1fhr standard deviation = %d ppm' % (timescale, median(stddev[:]))
print('\nStandard deviation = %d ppm' % astddev)
# calculate median STDDEV
if status == 0:
medcdpp = ones((len(cdpp)),dtype='float32') * median(cdpp[:])
print('Median %.1fhr CDPP = %d ppm' % (timescale, median(cdpp[:])))
# calculate RMS STDDEV
if status == 0:
rms, status = kepstat.rms(cdpp,zeros(len(stddev)),logfile,verbose)
rmscdpp = ones((len(cdpp)),dtype='float32') * rms
print(' RMS %.1fhr CDPP = %d ppm\n' % (timescale, rms))
# clean up x-axis unit
if status == 0:
intime0 = float(int(tstart / 100) * 100.0)
ptime = intime - intime0
xlab = 'BJD $-$ %d' % intime0
# clean up y-axis units
if status == 0:
pout = copy(cdpp)
nrm = math.ceil(math.log10(median(cdpp))) - 1.0
# pout = pout / 10**nrm
# ylab = '%.1fhr $\sigma$ (10$^%d$ ppm)' % (timescale,nrm)
ylab = '%.1fhr $\sigma$ (ppm)' % timescale
# data limits
xmin = ptime.min()
xmax = ptime.max()
ymin = pout.min()
ymax = pout.max()
xr = xmax - xmin
yr = ymax - ymin
ptime = insert(ptime,[0],[ptime[0]])
ptime = append(ptime,[ptime[-1]])
pout = insert(pout,[0],[0.0])
pout = append(pout,0.0)
# plot style
if status == 0:
try:
params = {'backend': 'png',
'axes.linewidth': 2.5,
'axes.labelsize': 36,
'axes.font': 'sans-serif',
'axes.fontweight' : 'bold',
'text.fontsize': 12,
'legend.fontsize': 12,
'xtick.labelsize': 32,
'ytick.labelsize': 36}
pylab.rcParams.update(params)
except:
pass
# define size of plot on monitor screen
pylab.figure(figsize=[xsize,ysize])
# delete any fossil plots in the matplotlib window
pylab.clf()
# position first axes inside the plotting window
ax = pylab.axes([0.07,0.15,0.92,0.83])
# force tick labels to be absolute rather than relative
pylab.gca().xaxis.set_major_formatter(pylab.ScalarFormatter(useOffset=False))
pylab.gca().yaxis.set_major_formatter(pylab.ScalarFormatter(useOffset=False))
ax.yaxis.set_major_locator(MaxNLocator(5))
# rotate y labels by 90 deg
labels = ax.get_yticklabels()
pylab.setp(labels, 'rotation', 90,fontsize=36)
# plot flux vs time
ltime = array([],dtype='float64')
ldata = array([],dtype='float32')
dt = 0
work1 = 2.0 * cadence / 86400
for i in range(1,len(ptime)-1):
dt = ptime[i] - ptime[i-1]
if dt < work1:
ltime = append(ltime,ptime[i])
ldata = append(ldata,pout[i])
else:
pylab.plot(ltime,ldata,color='#0000ff',linestyle='-',linewidth=1.0)
ltime = array([],dtype='float64')
ldata = array([],dtype='float32')
pylab.plot(ltime,ldata,color='#0000ff',linestyle='-',linewidth=1.0)
# plot the fill color below data time series, with no data gaps
pylab.fill(ptime,pout,fc='#ffff00',linewidth=0.0,alpha=0.2)
# plot median CDPP
# pylab.plot(intime - intime0,medcdpp / 10**nrm,color='r',linestyle='-',linewidth=2.0)
# pylab.plot(intime - intime0,medcdpp,color='r',linestyle='-',linewidth=2.0)
# plot RMS CDPP
# pylab.plot(intime - intime0,rmscdpp / 10**nrm,color='r',linestyle='--',linewidth=2.0)
# define plot x and y limits
pylab.xlim(xmin - xr * 0.01, xmax + xr * 0.01)
if ymin - yr * 0.01 <= 0.0:
pylab.ylim(1.0e-10, ymax + yr * 0.01)
else:
pylab.ylim(ymin - yr * 0.01, ymax + yr * 0.01)
# plot labels
pylab.xlabel(xlab, {'color' : 'k'})
pylab.ylabel(ylab, {'color' : 'k'})
# make grid on plot
pylab.grid()
# render plot
if status == 0:
if cmdLine:
pylab.show(block=True)
else:
pylab.ion()
pylab.plot([])
pylab.ioff()
# add NaNs back into data
if status == 0:
n = 0
work1 = array([],dtype='float32')
instr, status = kepio.openfits(infile,'readonly',logfile,verbose)
table, status = kepio.readfitstab(infile,instr[1],logfile,verbose)
for i in range(len(table.field(0))):
if isfinite(table.field('time')[i]) and isfinite(table.field(datacol)[i]):
work1 = append(work1,cdpp[n])
n += 1
else:
work1 = append(work1,nan)
# write output file
if status == 0:
status = kepkey.new('MCDPP%d' % (timescale * 10.0),medcdpp[0],
'Median %.1fhr CDPP (ppm)' % timescale,
instr[1],outfile,logfile,verbose)
status = kepkey.new('RCDPP%d' % (timescale * 10.0),rmscdpp[0],
'RMS %.1fhr CDPP (ppm)' % timescale,
instr[1],outfile,logfile,verbose)
colname = 'CDPP_%d' % (timescale * 10)
col1 = pyfits.Column(name=colname,format='E13.7',array=work1)
cols = instr[1].data.columns + col1
instr[1] = pyfits.new_table(cols,header=instr[1].header)
instr.writeto(outfile)
# comment keyword in output file
if status == 0:
status = kepkey.history(call,instr[0],outfile,logfile,verbose)
# close FITS
if status == 0:
status = kepio.closefits(instr,logfile,verbose)
# end time
if (status == 0):
message = 'KEPSTDDEV completed at'
else:
message = '\nKEPSTDDEV aborted at'
kepmsg.clock(message,logfile,verbose)
def leastsquare(functype,pinit,xdata,ydata,yerr,logfile,verbose):
status = 0
coeffs = []
# functional form
if (functype == 'poly0'): fitfunc = kepfunc.poly0()
if (functype == 'poly1'): fitfunc = kepfunc.poly1()
if (functype == 'poly2'): fitfunc = kepfunc.poly2()
if (functype == 'poly3'): fitfunc = kepfunc.poly3()
if (functype == 'poly4'): fitfunc = kepfunc.poly4()
if (functype == 'poly5'): fitfunc = kepfunc.poly5()
if (functype == 'poly6'): fitfunc = kepfunc.poly6()
if (functype == 'poly7'): fitfunc = kepfunc.poly7()
if (functype == 'poly8'): fitfunc = kepfunc.poly8()
if (functype == 'poly1con'): fitfunc = kepfunc.poly1con()
if (functype == 'gauss'): fitfunc = kepfunc.gauss()
if (functype == 'gauss0'): fitfunc = kepfunc.gauss0()
if (functype == 'congauss'): fitfunc = kepfunc.congauss()
if (functype == 'sine'): fitfunc = kepfunc.sine()
if (functype == 'moffat0'): fitfunc = kepfunc.moffat0()
if (functype == 'conmoffat'): fitfunc = kepfunc.conmoffat()
# define error coefficent calculation
errfunc = lambda p, x, y, err: (y - fitfunc(p, x)) / err
# if no data errors, substitude rms of fit
if (yerr == None):
yerr = []
rerr = []
for i in range(len(ydata)):
rerr.append(1.e10)
try:
out = optimize.leastsq(errfunc,pinit,args=(xdata,ydata,rerr),full_output=1)
except:
message = 'ERROR -- KEPFIT.LEASTSQUARE: failed to fit data'
status = kepmsg.err(logfile,message,verbose)
if functype == 'poly0':
out = [numpy.mean(ydata),sqrt(numpy.mean(ydata))]
if (functype == 'poly0' or functype == 'sineCompareBinPSF'):
coeffs.append(out[0])
else:
coeffs = out[0]
if (len(coeffs) > 1):
fit = fitfunc(coeffs,xdata)
else:
fit = numpy.zeros(len(xdata))
for i in range(len(fit)):
fit[i] = coeffs[0]
sigma, status = kepstat.rms(ydata,fit,logfile,verbose)
for i in range(len(ydata)):
yerr.append(sigma)
# fit data
try:
out = optimize.leastsq(errfunc, pinit, args=(xdata, ydata, yerr), full_output=1)
except:
message = 'ERROR -- KEPFIT.LEASTSQUARE: failed to fit data'
status = kepmsg.err(logfile,message,verbose)
if functype == 'poly0':
out = [numpy.mean(ydata),sqrt(numpy.mean(ydata))]
# define coefficients
coeffs = []
covar = []
if (functype == 'poly0' or functype == 'poly1con' or
functype == 'sineCompareBinPSF'):
coeffs.append(out[0])
covar.append(out[1])
else:
coeffs = out[0]
covar = out[1]
# calculate 1-sigma error on coefficients
errors = []
if (covar == None):
message = 'WARNING -- KEPFIT.leastsquare: NULL covariance matrix'
# kepmsg.log(logfile,message,verbose)
for i in range(len(coeffs)):
if (covar != None and len(coeffs) > 1):
errors.append(sqrt(covar[i][i]))
else:
errors.append(coeffs[i])
# generate fit points for rms calculation
if (len(coeffs) > 1):
fit = fitfunc(coeffs,xdata)
else:
fit = numpy.zeros(len(xdata))
for i in range(len(fit)):
fit[i] = coeffs[0]
sigma, status = kepstat.rms(ydata,fit,logfile,verbose)
# generate fit points for plotting
dx = xdata[len(xdata)-1] - xdata[0]
plotx = linspace(xdata.min(),xdata.max(),10000)
ploty = fitfunc(coeffs,plotx)
if (len(coeffs) == 1):
ploty = []
for i in range(len(plotx)):
ploty.append(coeffs[0])
ploty = numpy.array(ploty)
# reduced chi^2 calculation
chi2 = 0
dof = len(ydata) - len(coeffs)
for i in range(len(ydata)):
chi2 += (ydata[i] - fit[i])**2 / yerr[i]
chi2 /= dof
return coeffs, errors, covar, sigma, chi2, dof, fit, plotx, ploty, status
def kepstddev(infile,outfile,datacol,timescale,clobber,verbose,logfile,status,cmdLine=False):
# startup parameters
status = 0
labelsize = 44
ticksize = 36
xsize = 16
ysize = 6
lcolor = '#0000ff'
lwidth = 1.0
fcolor = '#ffff00'
falpha = 0.2
# log the call
hashline = '----------------------------------------------------------------------------'
kepmsg.log(logfile,hashline,verbose)
call = 'KEPSTDDEV -- '
call += 'infile='+infile+' '
call += 'outfile='+outfile+' '
call += 'datacol='+str(datacol)+' '
call += 'timescale='+str(timescale)+' '
overwrite = 'n'
if (clobber): overwrite = 'y'
call += 'clobber='+overwrite+ ' '
chatter = 'n'
if (verbose): chatter = 'y'
call += 'verbose='+chatter+' '
call += 'logfile='+logfile
kepmsg.log(logfile,call+'\n',verbose)
# start time
kepmsg.clock('KEPSTDDEV started at',logfile,verbose)
# test log file
logfile = kepmsg.test(logfile)
# clobber output file
if clobber: status = kepio.clobber(outfile,logfile,verbose)
if kepio.fileexists(outfile):
message = 'ERROR -- KEPSTDDEV: ' + outfile + ' exists. Use clobber=yes'
status = kepmsg.err(logfile,message,verbose)
# open input file
if status == 0:
instr, status = kepio.openfits(infile,'readonly',logfile,verbose)
if status == 0:
tstart, tstop, bjdref, cadence, status = kepio.timekeys(instr,infile,logfile,verbose,status)
if status == 0:
try:
work = instr[0].header['FILEVER']
cadenom = 1.0
except:
cadenom = cadence
# fudge non-compliant FITS keywords with no values
if status == 0:
instr = kepkey.emptykeys(instr,file,logfile,verbose)
# read table structure
if status == 0:
table, status = kepio.readfitstab(infile,instr[1],logfile,verbose)
# filter input data table
if status == 0:
work1 = numpy.array([table.field('time'), table.field(datacol)])
work1 = numpy.rot90(work1,3)
work1 = work1[~numpy.isnan(work1).any(1)]
# read table columns
if status == 0:
intime = work1[:,1] + bjdref
indata = work1[:,0]
# calculate STDDEV in units of ppm
if status == 0:
stddev = running_frac_std(intime,indata,timescale/24) * 1.0e6
astddev = numpy.std(indata) * 1.0e6
cdpp = stddev / sqrt(timescale * 3600.0 / cadence)
# filter cdpp
if status == 0:
for i in range(len(cdpp)):
if cdpp[i] > median(cdpp) * 10.0: cdpp[i] = cdpp[i-1]
# calculate median STDDEV
if status == 0:
medcdpp = ones((len(cdpp)),dtype='float32') * median(cdpp[:])
# print '\nMedian %.1fhr standard deviation = %d ppm' % (timescale, median(stddev[:]))
print '\nStandard deviation = %d ppm' % astddev
# calculate median STDDEV
if status == 0:
medcdpp = ones((len(cdpp)),dtype='float32') * median(cdpp[:])
print 'Median %.1fhr CDPP = %d ppm' % (timescale, median(cdpp[:]))
# calculate RMS STDDEV
if status == 0:
rms, status = kepstat.rms(cdpp,zeros(len(stddev)),logfile,verbose)
rmscdpp = ones((len(cdpp)),dtype='float32') * rms
print ' RMS %.1fhr CDPP = %d ppm\n' % (timescale, rms)
# clean up x-axis unit
if status == 0:
intime0 = float(int(tstart / 100) * 100.0)
ptime = intime - intime0
xlab = 'BJD $-$ %d' % intime0
# clean up y-axis units
if status == 0:
pout = copy(cdpp)
nrm = math.ceil(math.log10(median(cdpp))) - 1.0
# pout = pout / 10**nrm
# ylab = '%.1fhr $\sigma$ (10$^%d$ ppm)' % (timescale,nrm)
ylab = '%.1fhr $\sigma$ (ppm)' % timescale
# data limits
xmin = ptime.min()
xmax = ptime.max()
ymin = pout.min()
ymax = pout.max()
xr = xmax - xmin
yr = ymax - ymin
ptime = insert(ptime,[0],[ptime[0]])
ptime = append(ptime,[ptime[-1]])
pout = insert(pout,[0],[0.0])
pout = append(pout,0.0)
# plot style
if status == 0:
try:
params = {'backend': 'png',
'axes.linewidth': 2.5,
'axes.labelsize': 36,
'axes.font': 'sans-serif',
'axes.fontweight' : 'bold',
'text.fontsize': 12,
'legend.fontsize': 12,
'xtick.labelsize': 32,
'ytick.labelsize': 36}
pylab.rcParams.update(params)
except:
pass
# define size of plot on monitor screen
pylab.figure(figsize=[xsize,ysize])
# delete any fossil plots in the matplotlib window
pylab.clf()
# position first axes inside the plotting window
ax = pylab.axes([0.07,0.15,0.92,0.83])
# force tick labels to be absolute rather than relative
pylab.gca().xaxis.set_major_formatter(pylab.ScalarFormatter(useOffset=False))
pylab.gca().yaxis.set_major_formatter(pylab.ScalarFormatter(useOffset=False))
ax.yaxis.set_major_locator(MaxNLocator(5))
# rotate y labels by 90 deg
labels = ax.get_yticklabels()
pylab.setp(labels, 'rotation', 90,fontsize=36)
# plot flux vs time
ltime = array([],dtype='float64')
ldata = array([],dtype='float32')
dt = 0
work1 = 2.0 * cadence / 86400
for i in range(1,len(ptime)-1):
dt = ptime[i] - ptime[i-1]
if dt < work1:
ltime = append(ltime,ptime[i])
ldata = append(ldata,pout[i])
else:
pylab.plot(ltime,ldata,color='#0000ff',linestyle='-',linewidth=1.0)
ltime = array([],dtype='float64')
ldata = array([],dtype='float32')
pylab.plot(ltime,ldata,color='#0000ff',linestyle='-',linewidth=1.0)
# plot the fill color below data time series, with no data gaps
pylab.fill(ptime,pout,fc='#ffff00',linewidth=0.0,alpha=0.2)
# plot median CDPP
# pylab.plot(intime - intime0,medcdpp / 10**nrm,color='r',linestyle='-',linewidth=2.0)
# pylab.plot(intime - intime0,medcdpp,color='r',linestyle='-',linewidth=2.0)
# plot RMS CDPP
# pylab.plot(intime - intime0,rmscdpp / 10**nrm,color='r',linestyle='--',linewidth=2.0)
# define plot x and y limits
pylab.xlim(xmin - xr * 0.01, xmax + xr * 0.01)
if ymin - yr * 0.01 <= 0.0:
pylab.ylim(1.0e-10, ymax + yr * 0.01)
else:
pylab.ylim(ymin - yr * 0.01, ymax + yr * 0.01)
# plot labels
pylab.xlabel(xlab, {'color' : 'k'})
pylab.ylabel(ylab, {'color' : 'k'})
# make grid on plot
pylab.grid()
# render plot
if status == 0:
if cmdLine:
pylab.show(block=True)
else:
pylab.ion()
pylab.plot([])
pylab.ioff()
# add NaNs back into data
if status == 0:
n = 0
work1 = array([],dtype='float32')
instr, status = kepio.openfits(infile,'readonly',logfile,verbose)
table, status = kepio.readfitstab(infile,instr[1],logfile,verbose)
for i in range(len(table.field(0))):
if isfinite(table.field('time')[i]) and isfinite(table.field(datacol)[i]):
work1 = append(work1,cdpp[n])
n += 1
else:
work1 = append(work1,nan)
# write output file
if status == 0:
status = kepkey.new('MCDPP%d' % (timescale * 10.0),medcdpp[0],
'Median %.1fhr CDPP (ppm)' % timescale,
instr[1],outfile,logfile,verbose)
status = kepkey.new('RCDPP%d' % (timescale * 10.0),rmscdpp[0],
'RMS %.1fhr CDPP (ppm)' % timescale,
instr[1],outfile,logfile,verbose)
colname = 'CDPP_%d' % (timescale * 10)
col1 = pyfits.Column(name=colname,format='E13.7',array=work1)
cols = instr[1].data.columns + col1
instr[1] = pyfits.new_table(cols,header=instr[1].header)
instr.writeto(outfile)
# comment keyword in output file
if status == 0:
status = kepkey.history(call,instr[0],outfile,logfile,verbose)
# close FITS
if status == 0:
status = kepio.closefits(instr,logfile,verbose)
# end time
if (status == 0):
message = 'KEPSTDDEV completed at'
else:
message = '\nKEPSTDDEV aborted at'
kepmsg.clock(message,logfile,verbose)
def rmsestimation(quarter, timescale):
"""
do stastitical analysis of rms cdpp(expected snr) of certain quarter
"""
#read data from keplerstellar catalog and match lightcurve in lc database
cata_path = '../catalog/cumulative.csv'
data_path = '/scratch/kepler_data/'
kepid = kepio.get_id(cata_path)
all_rms = []
own_kid = []
for i, k_id in enumerate(kepid[:100]):
#print('This is '+str(kepid[i]))
#lc path here
file_dir = kepio.pathfinder(k_id, data_path, quarter)
try:
filename = glob.glob(file_dir)
name = filename[0]
#open file and read time keys
instr = fits.open(name)
tstart, tstop, bjdref, cadence = kepio.timekeys(instr, filename)
#read lc
hdu = instr[1]
time = hdu.data.TIME
time = time + bjdref - 2454900
flux = hdu.data.PDCSAP_FLUX
#filter data
work1 = np.array([time, flux])
work1 = np.rot90(work1, 3)
work1 = work1[~np.isnan(work1).any(1)]
intime = work1[:, 1]
indata = work1[:, 0]
#split lc
intime, indata = keputils.split(intime, indata, gap_width=0.75)
#calculate breaking points
bkspaces = np.logspace(np.log10(0.5), np.log10(20), num=20)
#calculate spline to every data points
spline = kepspline.choose_kepler_spline(intime,
indata,
bkspaces,
penalty_coeff=1.0,
verbose=False)[0]
if spline is None:
raise ValueError("faied to fit spline")
#flatten the data array
intime = np.concatenate(intime).ravel()
indata = np.concatenate(indata).ravel()
spline = np.concatenate(spline).ravel()
#normalized flux using spline
nordata = indata / spline
#calculte runing stddev
stddev = kepstat.running_frac_std(intime, nordata,
timescale / 24) * 1.0e6
#cdpp
cdpp = stddev / math.sqrt(timescale * 3600.0 / cadence)
# filter cdpp
for i in range(len(cdpp)):
if cdpp[i] > np.median(cdpp) * 10.0:
cdpp[i] = cdpp[i - 1]
#calculte RMS cdpp
rms = kepstat.rms(cdpp, np.zeros(len(stddev)))
rmscdpp = np.ones((len(cdpp)), dtype='float32') * rms
if rms > 200:
plt.figure(figsize=(10, 8))
#plt.hist(cdpp, bins =25, color = 'gold', fill = True, edgecolor = 'black', linewidth = 2.0)
#plt.ylabel('Count')
#plt.xlabel('CDPP/6.5hrs(ppm)')
#plt.savefig("./test/KOIQ"+str(quarter)+str(k_id)+"cdpp.png")
plt.scatter(intime, nordata, marker='.')
plt.savefig("./test/KOIQ" + str(quarter) + str(k_id) +
"display_lc.png")
# print('%d has RMS %.1fhr CDPP = %d ppm\n' % (k_id,timescale, rms))
all_rms.append(rms)
own_kid.append(k_id)
except IndexError:
pass
c, low, upp = sigmaclip(all_rms, 3, 3)
plt.figure(figsize=(10, 8))
plt.hist(c,
bins=25,
range=(low, upp),
fill=True,
edgecolor='black',
linewidth=2.0)
plt.ylabel('Count')
plt.xlabel('RMSCDPP(ppm)')
#plt.savefig("./result/KOIQ"+str(quarter)+"rms.png")
result = np.transpose([own_kid, all_rms])
def leastsquare(functype, pinit, xdata, ydata, yerr, logfile, verbose):
status = 0
coeffs = []
# functional form
if (functype == 'poly0'): fitfunc = kepfunc.poly0()
if (functype == 'poly1'): fitfunc = kepfunc.poly1()
if (functype == 'poly2'): fitfunc = kepfunc.poly2()
if (functype == 'poly3'): fitfunc = kepfunc.poly3()
if (functype == 'poly4'): fitfunc = kepfunc.poly4()
if (functype == 'poly5'): fitfunc = kepfunc.poly5()
if (functype == 'poly6'): fitfunc = kepfunc.poly6()
if (functype == 'poly7'): fitfunc = kepfunc.poly7()
if (functype == 'poly8'): fitfunc = kepfunc.poly8()
if (functype == 'poly1con'): fitfunc = kepfunc.poly1con()
if (functype == 'gauss'): fitfunc = kepfunc.gauss()
if (functype == 'gauss0'): fitfunc = kepfunc.gauss0()
if (functype == 'congauss'): fitfunc = kepfunc.congauss()
if (functype == 'sine'): fitfunc = kepfunc.sine()
if (functype == 'moffat0'): fitfunc = kepfunc.moffat0()
if (functype == 'conmoffat'): fitfunc = kepfunc.conmoffat()
# define error coefficent calculation
errfunc = lambda p, x, y, err: (y - fitfunc(p, x)) / err
# if no data errors, substitude rms of fit
if (yerr == None):
yerr = []
rerr = []
for i in range(len(ydata)):
rerr.append(1.e10)
try:
out = optimize.leastsq(errfunc,
pinit,
args=(xdata, ydata, rerr),
full_output=1)
except:
message = 'ERROR -- KEPFIT.LEASTSQUARE: failed to fit data'
status = kepmsg.err(logfile, message, verbose)
if functype == 'poly0':
out = [numpy.mean(ydata), sqrt(numpy.mean(ydata))]
if (functype == 'poly0' or functype == 'sineCompareBinPSF'):
coeffs.append(out[0])
else:
coeffs = out[0]
if (len(coeffs) > 1):
fit = fitfunc(coeffs, xdata)
else:
fit = numpy.zeros(len(xdata))
for i in range(len(fit)):
fit[i] = coeffs[0]
sigma, status = kepstat.rms(ydata, fit, logfile, verbose)
for i in range(len(ydata)):
yerr.append(sigma)
# fit data
try:
out = optimize.leastsq(errfunc,
pinit,
args=(xdata, ydata, yerr),
full_output=1)
except:
message = 'ERROR -- KEPFIT.LEASTSQUARE: failed to fit data'
status = kepmsg.err(logfile, message, verbose)
if functype == 'poly0':
out = [numpy.mean(ydata), sqrt(numpy.mean(ydata))]
# define coefficients
coeffs = []
covar = []
if (functype == 'poly0' or functype == 'poly1con'
or functype == 'sineCompareBinPSF'):
coeffs.append(out[0])
covar.append(out[1])
else:
coeffs = out[0]
covar = out[1]
# calculate 1-sigma error on coefficients
errors = []
if (covar == None):
message = 'WARNING -- KEPFIT.leastsquare: NULL covariance matrix'
# kepmsg.log(logfile,message,verbose)
for i in range(len(coeffs)):
if (covar != None and len(coeffs) > 1):
errors.append(sqrt(abs(covar[i][i])))
else:
errors.append(coeffs[i])
# generate fit points for rms calculation
if (len(coeffs) > 1):
fit = fitfunc(coeffs, xdata)
else:
fit = numpy.zeros(len(xdata))
for i in range(len(fit)):
fit[i] = coeffs[0]
sigma, status = kepstat.rms(ydata, fit, logfile, verbose)
# generate fit points for plotting
dx = xdata[len(xdata) - 1] - xdata[0]
plotx = linspace(xdata.min(), xdata.max(), 10000)
ploty = fitfunc(coeffs, plotx)
if (len(coeffs) == 1):
ploty = []
for i in range(len(plotx)):
ploty.append(coeffs[0])
ploty = numpy.array(ploty)
# reduced chi^2 calculation
chi2 = 0
dof = len(ydata) - len(coeffs)
for i in range(len(ydata)):
chi2 += (ydata[i] - fit[i])**2 / yerr[i]
chi2 /= dof
return coeffs, errors, covar, sigma, chi2, dof, fit, plotx, ploty, status
