import numpy as np
import os
import matplotlib.pyplot as plt
from ..lib import util
from ..lib import plots
import warnings
warnings.filterwarnings("ignore", message='Ignoring specified arguments in '
'this call because figure with num')
[docs]
@plots.apply_style
def binned_lightcurve(meta, log, lc, i, white=False):
'''Plot each spectroscopic light curve. (Figs 4102)
Parameters
----------
meta : eureka.lib.readECF.MetaClass
The metadata object.
log : logedit.Logedit
The open log in which notes from this step can be added.
lc : Xarray Dataset
The Dataset object containing light curve and time data.
i : int
The current bandpass number.
white : bool; optional
Is this figure for the additional white-light light curve
'''
fig = plt.figure(4102)
fig.set_size_inches(8, 6, forward=True)
fig.clf()
ax = fig.gca()
if white:
fig.suptitle(f'White-light Bandpass {i}: {meta.wave_min:.3f} - '
f'{meta.wave_max:.3f}')
# Normalize the light curve
norm_lcdata, norm_lcerr = util.normalize_spectrum(
meta, lc.flux_white, lc.err_white,
scandir=getattr(lc, 'scandir', None))
i = 0
fname_tag = 'white'
elif meta.photometry:
fig.suptitle(f'Photometric Lightcurve at {meta.phot_wave} microns')
# Normalize the light curve
norm_lcdata, norm_lcerr = util.normalize_spectrum(
meta, lc['data'][i], lc['err'][i],
scandir=getattr(lc, 'scandir', None))
fname_tag = 'phot'
else:
fig.suptitle(f'Bandpass {i}: {lc.wave_low.values[i]:.3f} - '
f'{lc.wave_hi.values[i]:.3f}')
# Normalize the light curve
norm_lcdata, norm_lcerr = util.normalize_spectrum(
meta, lc['data'][i], lc['err'][i],
scandir=getattr(lc, 'scandir', None))
ch_number = str(i).zfill(int(np.floor(np.log10(meta.nspecchan))+1))
fname_tag = f'ch{ch_number}'
time_modifier = np.floor(np.ma.min(lc.time.values))
# Plot the normalized light curve
if meta.inst == 'wfc3':
for p in range(2):
iscans = np.nonzero(lc.scandir.values == p)[0]
if len(iscans) > 0:
ax.errorbar(lc.time.values[iscans]-time_modifier,
norm_lcdata[iscans]+0.005*p,
norm_lcerr[iscans], fmt='o', color=f'C{p}',
mec=f'C{p}', alpha=0.2)
mad = util.get_mad_1d(norm_lcdata[iscans])
meta.mad_s4_binned.append(mad)
log.writelog(f' MAD = {np.round(mad).astype(int)} ppm')
plt.text(0.05, 0.075+0.05*p,
f"MAD = {np.round(mad).astype(int)} ppm",
transform=ax.transAxes, color=f'C{p}')
else:
plt.errorbar(lc.time.values-time_modifier, norm_lcdata, norm_lcerr,
fmt='o', color=f'C{i}', mec=f'C{i}', alpha=0.2)
mad = util.get_mad_1d(norm_lcdata)
meta.mad_s4_binned.append(mad)
log.writelog(f' MAD = {np.round(mad).astype(int)} ppm')
plt.text(0.05, 0.1, f"MAD = {np.round(mad).astype(int)} ppm",
transform=ax.transAxes, color='k')
plt.ylabel('Normalized Flux')
time_units = lc.data.attrs['time_units']
plt.xlabel(f'Time [{time_units} - {time_modifier}]')
fname = f'figs{os.sep}fig4102_{fname_tag}_1D_LC'+plots.get_filetype()
fig.savefig(meta.outputdir+fname, bbox_inches='tight', dpi=300)
if not meta.hide_plots:
plt.pause(0.2)
[docs]
@plots.apply_style
def binned_background(meta, log, lc, i, white=False):
'''Plot each spectroscopic background light curve. (Figs 4105)
Parameters
----------
meta : eureka.lib.readECF.MetaClass
The metadata object.
log : logedit.Logedit
The open log in which notes from this step can be added.
lc : Xarray Dataset
The Dataset object containing light curve and time data.
i : int
The current bandpass number.
white : bool; optional
Is this figure for the additional white-light light curve
'''
fig = plt.figure(4105)
fig.set_size_inches(8, 6, forward=True)
fig.clf()
ax = fig.gca()
if white:
fig.suptitle(f'White-light Bandpass {i}: {meta.wave_min:.3f} - '
f'{meta.wave_max:.3f}')
# Normalize the light curve
norm_bgdata, norm_bgerr = util.normalize_spectrum(
meta, lc.skylev_white, lc.skyerr_white,
scandir=getattr(lc, 'scandir', None))
i = 0
fname_tag = 'white'
elif meta.photometry:
fig.suptitle(f'Photometric Lightcurve at {meta.phot_wave} microns')
# Normalize the light curve
norm_bgdata, norm_bgerr = util.normalize_spectrum(
meta, lc['skylev'][i], lc['skyerr'][i],
scandir=getattr(lc, 'scandir', None))
fname_tag = 'phot'
else:
fig.suptitle(f'Bandpass {i}: {lc.wave_low.values[i]:.3f} - '
f'{lc.wave_hi.values[i]:.3f}')
# Normalize the light curve
norm_bgdata, norm_bgerr = util.normalize_spectrum(
meta, lc['skylev'][i], lc['skyerr'][i],
scandir=getattr(lc, 'scandir', None))
ch_number = str(i).zfill(int(np.floor(np.log10(meta.nspecchan))+1))
fname_tag = f'ch{ch_number}'
time_modifier = np.floor(np.ma.min(lc.time.values))
# Plot the normalized light curve
if meta.inst == 'wfc3':
for p in range(2):
iscans = np.nonzero(lc.scandir.values == p)[0]
if len(iscans) > 0:
ax.errorbar(lc.time.values[iscans]-time_modifier,
norm_bgdata[iscans]+0.005*p,
norm_bgerr[iscans], fmt='o', color=f'C{p}',
mec=f'C{p}', alpha=0.2)
mad = util.get_mad_1d(norm_bgdata[iscans])
meta.mad_s4_binned.append(mad)
log.writelog(f' MAD = {np.round(mad).astype(int)} ppm')
plt.text(0.05, 0.075+0.05*p,
f"MAD = {np.round(mad).astype(int)} ppm",
transform=ax.transAxes, color=f'C{p}')
else:
plt.errorbar(lc.time.values-time_modifier, norm_bgdata, norm_bgerr,
fmt='o', color=f'C{i}', mec=f'C{i}', alpha=0.2)
mad = util.get_mad_1d(norm_bgdata)
meta.mad_s4_binned_bg.append(mad)
plt.text(0.05, 0.1, f"MAD = {np.round(mad).astype(int)} ppm",
transform=ax.transAxes, color='k')
plt.ylabel('Normalized Background')
time_units = lc.data.attrs['time_units']
plt.xlabel(f'Time [{time_units} - {time_modifier}]')
fname = f'figs{os.sep}fig4105_{fname_tag}_1D_BG'+plots.get_filetype()
fig.savefig(meta.outputdir+fname, bbox_inches='tight', dpi=300)
if not meta.hide_plots:
plt.pause(0.2)
[docs]
@plots.apply_style
def driftxpos(meta, lc):
'''Plot the 1D drift/jitter results. (Fig 4103)
Parameters
----------
meta : eureka.lib.readECF.MetaClass
The metadata object.
lc : Xarray Dataset
The light curve object containing drift arrays.
'''
fig = plt.figure(4103)
fig.set_size_inches(8, 4, forward=True)
fig.clf()
mask = lc.driftmask
plt.plot(np.arange(meta.n_int)[~mask],
lc.centroid_x[~mask], '.',
label='Good Drift Points')
plt.plot(np.arange(meta.n_int)[mask],
lc.centroid_x[mask], '.',
label='Interpolated Drift Points')
plt.ylabel('Spectrum Drift Along x')
plt.xlabel('Frame Number')
plt.legend(loc='best')
fname = 'figs'+os.sep+'fig4103_DriftXPos'+plots.get_filetype()
plt.savefig(meta.outputdir+fname, bbox_inches='tight', dpi=300)
if not meta.hide_plots:
plt.pause(0.2)
[docs]
@plots.apply_style
def driftxwidth(meta, lc):
'''Plot the 1D drift width results. (Fig 4104)
Parameters
----------
meta : eureka.lib.readECF.MetaClass
The metadata object.
lc : Xarray Dataset
The light curve object containing drift arrays.
'''
fig = plt.figure(4104)
fig.set_size_inches(8, 4, forward=True)
fig.clf()
mask = lc.driftmask
plt.plot(np.arange(meta.n_int)[~mask],
lc.centroid_sx[~mask], '.',
label='Good Drift Points')
plt.plot(np.arange(meta.n_int)[mask],
lc.centroid_sx[mask], '.',
label='Interpolated Drift Points')
plt.ylabel('Spectrum Drift CC Width Along x')
plt.xlabel('Frame Number')
plt.legend(loc='best')
fname = 'figs'+os.sep+'fig4104_DriftXWidth'+plots.get_filetype()
plt.savefig(meta.outputdir+fname, bbox_inches='tight', dpi=300)
if not meta.hide_plots:
plt.pause(0.2)
[docs]
@plots.apply_style
def lc_driftcorr(meta, wave_1d, optspec_in, optmask=None, scandir=None):
'''Plot a 2D light curve with drift correction. (Fig 4101)
Parameters
----------
meta : eureka.lib.readECF.MetaClass
The metadata object.
wave_1d : ndarray
Wavelength array with trimmed edges depending on xwindow and ywindow
which have been set in the S3 ecf.
optspec_in : Xarray DataArray
The optimally extracted spectrum.
optmask : Xarray DataArray; optional
A mask array to use if optspec is not a masked array. Defaults to None
in which case only the invalid values of optspec will be masked.
scandir : ndarray; optional
For HST spatial scanning mode, 0=forward scan and 1=reverse scan.
Defaults to None which is fine for JWST data, but must be provided
for HST data (can be all zero values if not spatial scanning mode).
'''
optspec = np.ma.masked_invalid(optspec_in.values)
optspec = np.ma.masked_where(optmask.values, optspec)
wmin = meta.wave_min
wmax = meta.wave_max
iwmin = np.nanargmin(np.abs(wave_1d-wmin))
iwmax = np.nanargmin(np.abs(wave_1d-wmax))
# Normalize the light curve
norm_lcdata = util.normalize_spectrum(meta, optspec[:, iwmin:iwmax],
scandir=scandir)
if meta.time_axis not in ['y', 'x']:
print("WARNING: meta.time_axis is not one of ['y', 'x']!"
" Using 'y' by default.")
meta.time_axis = 'y'
cmap = plt.cm.RdYlBu_r
fig = plt.figure(4101)
fig.set_size_inches(8, 8, forward=True)
fig.clf()
if meta.time_axis == 'y':
plt.pcolormesh(wave_1d[iwmin:iwmax], np.arange(meta.n_int)+0.5,
norm_lcdata, vmin=meta.vmin, vmax=meta.vmax,
cmap=cmap)
plt.xlim(meta.wave_min, meta.wave_max)
plt.ylim(0, meta.n_int)
plt.ylabel('Integration Number')
plt.xlabel(r'Wavelength ($\mu m$)')
plt.colorbar(label='Normalized Flux')
if len(meta.wave) > 1 and len(wave_1d) != meta.nspecchan:
# Insert vertical dashed lines at spectroscopic channel edges
secax = plt.gca().secondary_xaxis('top')
xticks = np.unique(np.concatenate([meta.wave_low, meta.wave_hi]))
xticks_labels = [f'{np.round(xtick, 3):.3f}' for xtick in xticks]
secax.set_xticks(xticks, xticks_labels, rotation=90,
fontsize='xx-small')
plt.vlines(xticks, 0, meta.n_int, '0.3', 'dashed')
else:
plt.pcolormesh(np.arange(meta.n_int), wave_1d[iwmin:iwmax],
norm_lcdata.swapaxes(0, 1), vmin=meta.vmin,
vmax=meta.vmax, cmap=cmap)
plt.ylim(meta.wave_min, meta.wave_max)
plt.xlim(0, meta.n_int)
plt.ylabel(r'Wavelength ($\mu m$)')
plt.xlabel('Integration Number')
plt.colorbar(label='Normalized Flux', pad=0.075)
if len(meta.wave_low) > 1 and len(wave_1d) != meta.nspecchan:
# Insert vertical dashed lines at spectroscopic channel edges
secax = plt.gca().secondary_yaxis('right')
yticks = np.unique(np.concatenate([meta.wave_low, meta.wave_hi]))
yticks_labels = [f'{np.round(ytick, 3):.3f}' for ytick in yticks]
secax.set_yticks(yticks, yticks_labels, rotation=0,
fontsize='xx-small')
plt.hlines(yticks, 0, meta.n_int, '0.3', 'dashed')
plt.minorticks_on()
plt.title(f"MAD = {np.round(meta.mad_s4).astype(int)} ppm")
fname = 'figs'+os.sep+'fig4101_2D_LC'+plots.get_filetype()
plt.savefig(meta.outputdir+fname, bbox_inches='tight', dpi=300)
if meta.hide_plots:
plt.close()
else:
plt.pause(0.2)
return
[docs]
@plots.apply_style
def mad_outliers(meta, pp):
'''Plot spectroscopic MAD values and identify outliers. (Figs 4106)
Outliers will be appended to `mask_columns` in the Stage 4 ECF.
Parameters
----------
meta : eureka.lib.readECF.MetaClass
The metadata object.
pp : Dictionary
A dictionary of plotting parameters from outliers.get_outliers().
'''
# Unpack dictionary of plotting parameters
x = pp["x"]
x_mask = pp["x_mask"]
x_mad_outliers = pp["x_mad_outliers"]
x_dev_outliers = pp["x_dev_outliers"]
mad = pp["mad"]
dev = pp["dev"]
masked_mad = pp["masked_mad"]
masked_dev = pp["masked_dev"]
smoothed_mad = pp["smoothed_mad"]
residual_mad = pp["residual_mad"]
smoothed_dev = pp["smoothed_dev"]
residual_dev = pp["residual_dev"]
# Plot spectroscopic MAD values
alpha = 0.5
fig = plt.figure(4106)
fig.set_size_inches(8, 8, forward=True)
fig.clf()
plt.subplot(211)
plt.plot(x, mad, '.', color='b', zorder=1,
label="Unbinned LC MAD", alpha=alpha)
plt.plot(x, dev, '.', color='gold', zorder=2,
label="Deviation from white LC (Scaled)", alpha=alpha)
plt.plot(x_mask, smoothed_mad, '-', color='r', lw=2, zorder=5,
label="Unbinned LC MAD (Smoothed)")
plt.plot(x_mask, smoothed_dev, '--', color='0.3', lw=2, zorder=6,
label="Deviation from white LC (Smoothed)")
plt.ylabel('MAD Value (ppm)')
plt.legend(loc='best')
plt.subplot(212)
plt.plot(x_mask, residual_mad, '.', color='b', zorder=1,
label="Unbinned LC MAD", alpha=alpha)
plt.plot(x_mask, residual_dev, '.', color='gold', zorder=2,
label="Deviation from white LC", alpha=alpha)
plt.plot(x_mad_outliers, residual_mad[masked_mad.mask], '.', color='C3',
zorder=5)
plt.plot(x_dev_outliers, residual_dev[masked_dev.mask], '.', color='C3',
zorder=5, label=rf"{meta.mad_sigma}$\sigma$ Outliers")
plt.legend(loc='best')
plt.ylabel('Residuals (ppm)')
plt.xlabel('Detector Column Number')
fname = 'figs'+os.sep+'fig4106_MAD_Outliers'+plots.get_filetype()
plt.savefig(meta.outputdir+fname, bbox_inches='tight', dpi=300)
if not meta.hide_plots:
plt.pause(0.1)
return
[docs]
@plots.apply_style
def cc_spec(meta, ref_spec, fit_spec, n):
'''Compare the spectrum used for cross-correlation with the current
spectrum (Fig 4301).
Parameters
----------
meta : eureka.lib.readECF.MetaClass
The metadata object.
ref_spec : ndarray (1D)
The reference spectrum used for cross-correlation.
fit_spec : ndarray (1D)
The extracted spectrum for the current integration.
n : int
The current integration number.
'''
fig = plt.figure(4301)
fig.set_size_inches(8, 8, forward=True)
fig.clf()
plt.title(f'Cross Correlation - Spectrum {n}')
nx = len(ref_spec)
plt.plot(np.arange(nx), ref_spec, '-', label='Reference Spectrum')
plt.plot(np.arange(meta.drift_range, nx-meta.drift_range), fit_spec, '-',
label='Current Spectrum')
plt.legend(loc='best')
int_number = str(n).zfill(int(np.floor(np.log10(meta.n_int))+1))
fname = ('figs'+os.sep+f'fig4301_int{int_number}_CC_Spec' +
plots.get_filetype())
plt.savefig(meta.outputdir+fname, bbox_inches='tight', dpi=300)
if not meta.hide_plots:
plt.pause(0.2)
[docs]
@plots.apply_style
def cc_vals(meta, vals, n):
'''Make the cross-correlation strength plot (Fig 4302).
Parameters
----------
meta : eureka.lib.readECF.MetaClass
The metadata object.
vals : ndarray (1D)
The cross-correlation strength.
n : int
The current integration number.
'''
fig = plt.figure(4302)
fig.set_size_inches(8, 8, forward=True)
fig.clf()
plt.title(f'Cross Correlation - Values {n}')
plt.plot(np.arange(-meta.drift_range, meta.drift_range+1), vals, '.')
int_number = str(n).zfill(int(np.floor(np.log10(meta.n_int))+1))
fname = ('figs'+os.sep+f'fig4302_int{int_number}_CC_Vals' +
plots.get_filetype())
plt.savefig(meta.outputdir+fname, bbox_inches='tight', dpi=300)
if not meta.hide_plots:
plt.pause(0.2)