def spec_ind(indices_file, spec_data, verbose=True):
'''
Second giant monster function that is going to do everything!
Parameters:
*indices_file*
File formatted as comma separated values that contains, in
the following order:
Author&Year,Line,NumMin,NumMax,DenMin,DenMax
*spec_data*
Spectrum as a Python list or array with wavelength in position
0 and flux in position 1.
*verbose*
Boolean: Print warning messages.
Returns:
List including the name & value of the index! (for now)
'''
indices = []
with open(indices_file) as f:
csvreader = csv.reader(f)
for row in csvreader:
if row[0][0] != "#":
indices.append(row)
namelist = []
rangelist = []
for x in range(len(indices)):
namelist.append(indices[x][1])
rangelist.append(map(float,indices[x][2:]))
# Initialize numerator and denominator arrays according to number
# of spectral indices to be measured.
length = len(namelist)
numarray = numpy.zeros((length,1))
denarray = numpy.zeros((length,1))
# Use avg_flux function from astrotools over numerator and
# denominator ranges, then calculate indices
for x in range(len(rangelist)):
numarray[x] = astrotools.avg_flux(rangelist[x][0], rangelist[x][1], spec_data)
denarray[x] = astrotools.avg_flux(rangelist[x][2], rangelist[x][3], spec_data)
indexarray = numarray/denarray
# Put everything into a dictionary.
spec_ind = {}
for x in range(len(namelist)):
spec_ind[namelist[x]] = indexarray[x][0]
return spec_ind
def spec_ind(spec_data):
"""
First giant monster function that is going to do everything!
Spectral indices hard coded in-- using above list.
Parameters:
*wl_array*
The array containing wavelength data.
*flux_array*
The array containing flux data.
Returns:
List including the name & value of the index! (for now)
"""
namelist = ["H2OA07", "Na", "H20J", "H2OH", "CH4K", "FeH-z"]
rangelist = [
[1.55, 1.56, 1.492, 1.502],
[1.15, 1.16, 1.134, 1.144],
[1.140, 1.165, 1.260, 1.285],
[1.480, 1.520, 1.560, 1.600],
[2.215, 2.255, 2.080, 2.120],
[0.965, 0.985, 0.990, 1.01],
]
# Initialize numerator and denominator arrays according to number
# of spectral indices to be measured.
length = len(namelist)
numarray = numpy.zeros((length, 1))
denarray = numpy.zeros((length, 1))
# Use avg_flux function from astrotools over numerator and
# denominator ranges, then calculate indices
for x in range(len(rangelist)):
numarray[x] = at.avg_flux(rangelist[x][0], rangelist[x][1], spec_data)
denarray[x] = at.avg_flux(rangelist[x][2], rangelist[x][3], spec_data)
indexarray = numarray / denarray
# Put everything into a list of dictionaries.
spec_ind = []
for x in range(len(namelist)):
dictionary = {namelist[x]: indexarray[x][0]}
spec_ind.append(dictionary)
return spec_ind
def spec_ind(spec_data):
'''
First giant monster function that is going to do everything!
Spectral indices hard coded in-- using above list.
Parameters:
*wl_array*
The array containing wavelength data.
*flux_array*
The array containing flux data.
Returns:
List including the name & value of the index! (for now)
'''
namelist = ['H2OA07', 'Na', 'H20J', 'H2OH', 'CH4K', 'FeH-z']
rangelist = [[1.55, 1.56, 1.492, 1.502], [1.15, 1.16, 1.134, 1.144],
[1.140, 1.165, 1.260, 1.285], [1.480, 1.520, 1.560, 1.600],
[2.215, 2.255, 2.080, 2.120], [0.965, 0.985, 0.990, 1.01]]
# Initialize numerator and denominator arrays according to number
# of spectral indices to be measured.
length = len(namelist)
numarray = numpy.zeros((length, 1))
denarray = numpy.zeros((length, 1))
# Use avg_flux function from astrotools over numerator and
# denominator ranges, then calculate indices
for x in range(len(rangelist)):
numarray[x] = at.avg_flux(rangelist[x][0], rangelist[x][1], spec_data)
denarray[x] = at.avg_flux(rangelist[x][2], rangelist[x][3], spec_data)
indexarray = numarray / denarray
# Put everything into a list of dictionaries.
spec_ind = []
for x in range(len(namelist)):
dictionary = {namelist[x]: indexarray[x][0]}
spec_ind.append(dictionary)
return spec_ind
def spec_ind(indices_file, specData, objname, plot=True):
'''
Measure the spectral indices as specified by an input file. The
output is a dictionary containing the name of the spectral index
and its value.
*indices_file*
File formatted as comma separated values that contains, in
the following order:
Author&Year,Line,NumMin,NumMax,DenMin,DenMax
*specData*
Spectrum as a Python list or array with wavelength in position
0 and flux in position 1. Use either read_spec (fits files) or
pull_data (from database) to get these arrays.
*targetinfo*
Target instance of the object, from the BDNYC Python Database.
*plot*
Boolean: Save output plot.
'''
# Read indices file.
indices = []
try:
with open(indices_file, 'rb') as f:
csvreader = csv.reader(f)
for row in csvreader:
if row[0][0] != "#":
indices.append(row)
f.close()
except IOError:
print str(indices_file) + ' not found.'
return
# Store indices info in list of names and numerator/denominator ranges.
index_name = []
index_range = []
for x in range(len(indices)):
index_name.append(indices[x][1])
index_range.append(map(float, indices[x][2:]))
for y in range(len(specData)):
# Initialize numerator and denominator arrays according to number
# of spectral indices to be measured.
length = len(index_name)
numarray = numpy.zeros((length, 1))
denarray = numpy.zeros((length, 1))
signum = [0] * length
sigden = [0] * length
sig_ind = []
# Use avg_flux function (outputs [avgflux,stddev]) over numerator and
# denominator ranges, then calculate indices by dividing.
for x in range(len(index_range)):
[numarray[x],
signum[x]] = astrotools.avg_flux(index_range[x][0],
index_range[x][1], specData[y])
[denarray[x],
sigden[x]] = astrotools.avg_flux(index_range[x][2],
index_range[x][3], specData[y])
sig_ind.append(
numpy.sqrt((signum[x] / denarray[x])**2 +
(numarray[x] * sigden[x])**2 / denarray[x]**4))
indexarray = numarray / denarray
#Polynomials:
H2O_A07 = indexarray[index_name.index('H2O_A07')][0]
sig_H2O_A07 = sig_ind[index_name.index('H2O_A07')]
H2OJ = indexarray[index_name.index('H2OJ')][0]
sig_H2OJ = sig_ind[index_name.index('H2OJ')]
H2OH = indexarray[index_name.index('H2OH')][0]
sig_H2OH = sig_ind[index_name.index('H2OH')]
CH4K = indexarray[index_name.index('CH4K')][0]
sig_CH4K = sig_ind[index_name.index('CH4K')]
#Allers07
spt_H2O_A07 = (H2O_A07 - 0.77) / 0.04
sig_spt_H2O_A07 = sig_H2O_A07 / 0.04
#Burgasser07
spt_H2OJ = 1.949e1 - 3.919e1 * H2OJ + 1.312e2 * H2OJ**2 - 2.156e2 * H2OJ**3 + 1.038e2 * H2OJ**4 + 10
sig_spt_H2OJ = numpy.absolute(
sig_H2OJ * (-3.919e1 + 1.312e2 * 2 * H2OJ - 2.156e2 * 3 * H2OJ**2 +
1.038e2 * 4 * H2OJ**3))
spt_H2OH = 2.708e1 - 8.45e1 * H2OH + 2.424e2 * H2OH**2 - 3.381e2 * H2OH**3 + 1.491e2 * H2OH**4 + 10
sig_spt_H2OH = numpy.absolute(
sig_H2OH * (-8.45e1 + 2.424e2 * 2 * H2OH - 3.381e2 * 3 * H2OH**2 +
1.491e2 * 4 * H2OH**3))
spt_CH4K = 1.885e1 - 2.246e1 * CH4K + 2.534e1 * CH4K**2 - 4.734 * CH4K**3 - 1.259e1 * CH4K**4 + 10
sig_spt_CH4K = numpy.absolute(
sig_CH4K * (-2.246e1 + 2.534e1 * 2 * CH4K - 4.734 * 3 * CH4K**2 -
1.259e1 * 4 * CH4K**3))
spt_ind = [0] * length
spt_ind[index_name.index('H2OJ')] = spt_H2OJ
spt_ind[index_name.index('H2O_A07')] = spt_H2O_A07
spt_ind[index_name.index('H2OH')] = spt_H2OH
spt_ind[index_name.index('CH4K')] = spt_CH4K
sig_spt = [0] * length
sig_spt[index_name.index('H2OJ')] = sig_spt_H2OJ[0]
sig_spt[index_name.index('H2O_A07')] = sig_spt_H2O_A07[0]
sig_spt[index_name.index('H2OH')] = sig_spt_H2OH[0]
sig_spt[index_name.index('CH4K')] = sig_spt_CH4K[0]
spt_avg = numpy.mean([spt_H2OJ, spt_H2O_A07, spt_H2OH])
spt_stddev = numpy.std([spt_H2OJ, spt_H2O_A07, spt_H2OH])
spt_sigavg = numpy.mean([sig_spt_H2OJ, sig_spt_H2O_A07, sig_spt_H2OH])
# if len(targetinfo[y].sptype) >= 3 and targetinfo[y].sptype[-3] in ('p',':'):
# spnum = float(targetinfo[y].sptype[1:-3])
# elif targetinfo[y].sptype[-2]==':':
# spnum = float(targetinfo[y].sptype[1:-2])
# elif targetinfo[y].sptype[-1] in ('g','b',':','d'):
# spnum = float(targetinfo[y].sptype[1:-1])
# else:
# spnum = float(targetinfo[y].sptype[1:])
# if targetinfo[y].sptype[0] == 'L':
# sptype = spnum + 10
# if targetinfo[y].sptype[0] == 'T':
# sptype = spnum + 20
# if targetinfo[y].sptype[0] == 'M':
# sptype = spnum
specind = [[0] * len(index_name)] * len(specData)
for x in range(len(index_name)):
specind[y][x] = ([index_name[x], indexarray[x][0], spt_ind[x]])
#note, add sptype back to end of specind
#Make Plots
colors = [
'blue', 'darkblue', 'dodgerblue', 'darkcyan', 'darkgreen', 'green',
'darkred', 'red'
]
fig = plt.figure(figsize=(19, 11))
plt.plot(specData[y][0], specData[y][1], c='k')
plt.xlim(specData[y][0][0], specData[y][0][-1])
patches = []
lo = plt.ylim()[1] / 12
for x in range(len(index_range)):
# Make rectangle for numerator range.
yindex1 = numpy.where(specData[y][0] >= index_range[x][0])
yindex2 = numpy.where(specData[y][0] <= index_range[x][1])
ymedian = yindex1[0][0] + (yindex2[0][-1] - yindex1[0][0]) / 2
numx = index_range[x][0]
numy = specData[y][1][ymedian] - signum[x]
numwidth = index_range[x][1] - index_range[x][0]
numheight = signum[x] * 2
numrec = Rectangle((numx, numy),
numwidth,
numheight,
color=colors[x])
patches.append(numrec)
plt.gca().add_patch(numrec)
# Make rectangle for denominator range.
yindex1 = numpy.where(specData[y][0] >= index_range[x][2])
yindex2 = numpy.where(specData[y][0] <= index_range[x][3])
ymedian = yindex1[0][0] + (yindex2[0][-1] - yindex1[0][0]) / 2
denx = index_range[x][2]
deny = specData[y][1][ymedian] - sigden[x]
denwidth = index_range[x][3] - index_range[x][2]
denheight = sigden[x] * 2
denrec = Rectangle((denx, deny),
denwidth,
denheight,
color=colors[x])
patches.append(denrec)
plt.gca().add_patch(denrec)
# Put object info in top left corner.
# objname = targetinfo[y].unum
# plt.figtext(0.1115,0.86,objname)
# plt.figtext(0.15,0.84,targetinfo[y].name)
# plt.figtext(0.15,0.82,targetinfo[y].sptype+", "+"%.1f"%sptype)
# Put NIR index predicted spectral type info in top right corner.
plt.text(denx, deny - 0.6 * lo, index_name[x], color=colors[x])
plt.text(denx,
deny - lo,
"%.2f" % specind[y][x][1],
color=colors[x])
heights = [0.8, 0, 0.78, 0.76, 0.74]
plt.figtext(0.75, 0.82, "NIR Index Predicted Type", color='r')
if spt_ind[x] != 0:
plt.text(denx,
deny - 1.3 * lo,
"%.2f" % spt_ind[x],
color=colors[x])
plt.figtext(0.75,
heights[x],
index_name[x] + ' = ' + "%.2f" % spt_ind[x] +
"$\pm$" + "%.2f" % sig_spt[x],
color=colors[x])
plt.figtext(0.75, 0.72, 'Overall Average = ' + "%.2f" % spt_avg)
plt.figtext(0.75, 0.7, 'Std Dev = ' + "%.2f" % spt_stddev)
plt.figtext(0.75, 0.68, 'Uncertainty = ' + "%.2f" % spt_sigavg)
plt.title(objname)
print y
if plot:
# Save plot with UNum.pdf as name.
figpath = '/Users/Joci/Research/Data/Templates/' + objname + '.pdf'
plt.savefig(figpath)
# Save index data with UNum_specind.txt as name.
filepath = '/Users/Joci/Research/Data/Templates/' + objname + '_specind.txt'
with open(filepath, 'wb') as f:
writer = csv.writer(f)
writer.writerow([
'Index Name', 'Index Value', 'NIR Index Predicted Type',
'Optical Type'
])
writer.writerows(specind[y])
def spec_ind(indices_file, specData, objname, plot=True):
"""
Measure the spectral indices as specified by an input file. The
output is a dictionary containing the name of the spectral index
and its value.
*indices_file*
File formatted as comma separated values that contains, in
the following order:
Author&Year,Line,NumMin,NumMax,DenMin,DenMax
*specData*
Spectrum as a Python list or array with wavelength in position
0 and flux in position 1. Use either read_spec (fits files) or
pull_data (from database) to get these arrays.
*targetinfo*
Target instance of the object, from the BDNYC Python Database.
*plot*
Boolean: Save output plot.
"""
# Read indices file.
indices = []
try:
with open(indices_file, "rb") as f:
csvreader = csv.reader(f)
for row in csvreader:
if row[0][0] != "#":
indices.append(row)
f.close()
except IOError:
print str(indices_file) + " not found."
return
# Store indices info in list of names and numerator/denominator ranges.
index_name = []
index_range = []
for x in range(len(indices)):
index_name.append(indices[x][1])
index_range.append(map(float, indices[x][2:]))
for y in range(len(specData)):
# Initialize numerator and denominator arrays according to number
# of spectral indices to be measured.
length = len(index_name)
numarray = numpy.zeros((length, 1))
denarray = numpy.zeros((length, 1))
signum = [0] * length
sigden = [0] * length
sig_ind = []
# Use avg_flux function (outputs [avgflux,stddev]) over numerator and
# denominator ranges, then calculate indices by dividing.
for x in range(len(index_range)):
[numarray[x], signum[x]] = astrotools.avg_flux(index_range[x][0], index_range[x][1], specData[y])
[denarray[x], sigden[x]] = astrotools.avg_flux(index_range[x][2], index_range[x][3], specData[y])
sig_ind.append(
numpy.sqrt((signum[x] / denarray[x]) ** 2 + (numarray[x] * sigden[x]) ** 2 / denarray[x] ** 4)
)
indexarray = numarray / denarray
# Polynomials:
H2O_A07 = indexarray[index_name.index("H2O_A07")][0]
sig_H2O_A07 = sig_ind[index_name.index("H2O_A07")]
H2OJ = indexarray[index_name.index("H2OJ")][0]
sig_H2OJ = sig_ind[index_name.index("H2OJ")]
H2OH = indexarray[index_name.index("H2OH")][0]
sig_H2OH = sig_ind[index_name.index("H2OH")]
CH4K = indexarray[index_name.index("CH4K")][0]
sig_CH4K = sig_ind[index_name.index("CH4K")]
# Allers07
spt_H2O_A07 = (H2O_A07 - 0.77) / 0.04
sig_spt_H2O_A07 = sig_H2O_A07 / 0.04
# Burgasser07
spt_H2OJ = 1.949e1 - 3.919e1 * H2OJ + 1.312e2 * H2OJ ** 2 - 2.156e2 * H2OJ ** 3 + 1.038e2 * H2OJ ** 4 + 10
sig_spt_H2OJ = numpy.absolute(
sig_H2OJ * (-3.919e1 + 1.312e2 * 2 * H2OJ - 2.156e2 * 3 * H2OJ ** 2 + 1.038e2 * 4 * H2OJ ** 3)
)
spt_H2OH = 2.708e1 - 8.45e1 * H2OH + 2.424e2 * H2OH ** 2 - 3.381e2 * H2OH ** 3 + 1.491e2 * H2OH ** 4 + 10
sig_spt_H2OH = numpy.absolute(
sig_H2OH * (-8.45e1 + 2.424e2 * 2 * H2OH - 3.381e2 * 3 * H2OH ** 2 + 1.491e2 * 4 * H2OH ** 3)
)
spt_CH4K = 1.885e1 - 2.246e1 * CH4K + 2.534e1 * CH4K ** 2 - 4.734 * CH4K ** 3 - 1.259e1 * CH4K ** 4 + 10
sig_spt_CH4K = numpy.absolute(
sig_CH4K * (-2.246e1 + 2.534e1 * 2 * CH4K - 4.734 * 3 * CH4K ** 2 - 1.259e1 * 4 * CH4K ** 3)
)
spt_ind = [0] * length
spt_ind[index_name.index("H2OJ")] = spt_H2OJ
spt_ind[index_name.index("H2O_A07")] = spt_H2O_A07
spt_ind[index_name.index("H2OH")] = spt_H2OH
spt_ind[index_name.index("CH4K")] = spt_CH4K
sig_spt = [0] * length
sig_spt[index_name.index("H2OJ")] = sig_spt_H2OJ[0]
sig_spt[index_name.index("H2O_A07")] = sig_spt_H2O_A07[0]
sig_spt[index_name.index("H2OH")] = sig_spt_H2OH[0]
sig_spt[index_name.index("CH4K")] = sig_spt_CH4K[0]
spt_avg = numpy.mean([spt_H2OJ, spt_H2O_A07, spt_H2OH])
spt_stddev = numpy.std([spt_H2OJ, spt_H2O_A07, spt_H2OH])
spt_sigavg = numpy.mean([sig_spt_H2OJ, sig_spt_H2O_A07, sig_spt_H2OH])
# if len(targetinfo[y].sptype) >= 3 and targetinfo[y].sptype[-3] in ('p',':'):
# spnum = float(targetinfo[y].sptype[1:-3])
# elif targetinfo[y].sptype[-2]==':':
# spnum = float(targetinfo[y].sptype[1:-2])
# elif targetinfo[y].sptype[-1] in ('g','b',':','d'):
# spnum = float(targetinfo[y].sptype[1:-1])
# else:
# spnum = float(targetinfo[y].sptype[1:])
# if targetinfo[y].sptype[0] == 'L':
# sptype = spnum + 10
# if targetinfo[y].sptype[0] == 'T':
# sptype = spnum + 20
# if targetinfo[y].sptype[0] == 'M':
# sptype = spnum
specind = [[0] * len(index_name)] * len(specData)
for x in range(len(index_name)):
specind[y][x] = [index_name[x], indexarray[x][0], spt_ind[x]]
# note, add sptype back to end of specind
# Make Plots
colors = ["blue", "darkblue", "dodgerblue", "darkcyan", "darkgreen", "green", "darkred", "red"]
fig = plt.figure(figsize=(19, 11))
plt.plot(specData[y][0], specData[y][1], c="k")
plt.xlim(specData[y][0][0], specData[y][0][-1])
patches = []
lo = plt.ylim()[1] / 12
for x in range(len(index_range)):
# Make rectangle for numerator range.
yindex1 = numpy.where(specData[y][0] >= index_range[x][0])
yindex2 = numpy.where(specData[y][0] <= index_range[x][1])
ymedian = yindex1[0][0] + (yindex2[0][-1] - yindex1[0][0]) / 2
numx = index_range[x][0]
numy = specData[y][1][ymedian] - signum[x]
numwidth = index_range[x][1] - index_range[x][0]
numheight = signum[x] * 2
numrec = Rectangle((numx, numy), numwidth, numheight, color=colors[x])
patches.append(numrec)
plt.gca().add_patch(numrec)
# Make rectangle for denominator range.
yindex1 = numpy.where(specData[y][0] >= index_range[x][2])
yindex2 = numpy.where(specData[y][0] <= index_range[x][3])
ymedian = yindex1[0][0] + (yindex2[0][-1] - yindex1[0][0]) / 2
denx = index_range[x][2]
deny = specData[y][1][ymedian] - sigden[x]
denwidth = index_range[x][3] - index_range[x][2]
denheight = sigden[x] * 2
denrec = Rectangle((denx, deny), denwidth, denheight, color=colors[x])
patches.append(denrec)
plt.gca().add_patch(denrec)
# Put object info in top left corner.
# objname = targetinfo[y].unum
# plt.figtext(0.1115,0.86,objname)
# plt.figtext(0.15,0.84,targetinfo[y].name)
# plt.figtext(0.15,0.82,targetinfo[y].sptype+", "+"%.1f"%sptype)
# Put NIR index predicted spectral type info in top right corner.
plt.text(denx, deny - 0.6 * lo, index_name[x], color=colors[x])
plt.text(denx, deny - lo, "%.2f" % specind[y][x][1], color=colors[x])
heights = [0.8, 0, 0.78, 0.76, 0.74]
plt.figtext(0.75, 0.82, "NIR Index Predicted Type", color="r")
if spt_ind[x] != 0:
plt.text(denx, deny - 1.3 * lo, "%.2f" % spt_ind[x], color=colors[x])
plt.figtext(
0.75,
heights[x],
index_name[x] + " = " + "%.2f" % spt_ind[x] + "$\pm$" + "%.2f" % sig_spt[x],
color=colors[x],
)
plt.figtext(0.75, 0.72, "Overall Average = " + "%.2f" % spt_avg)
plt.figtext(0.75, 0.7, "Std Dev = " + "%.2f" % spt_stddev)
plt.figtext(0.75, 0.68, "Uncertainty = " + "%.2f" % spt_sigavg)
plt.title(objname)
print y
if plot:
# Save plot with UNum.pdf as name.
figpath = "/Users/Joci/Research/Data/Templates/" + objname + ".pdf"
plt.savefig(figpath)
# Save index data with UNum_specind.txt as name.
filepath = "/Users/Joci/Research/Data/Templates/" + objname + "_specind.txt"
with open(filepath, "wb") as f:
writer = csv.writer(f)
writer.writerow(["Index Name", "Index Value", "NIR Index Predicted Type", "Optical Type"])
writer.writerows(specind[y])
