def execute_operation(self):
max_wavelengths = lambda operands: np.arange(max([o[0].wavelengths[0] for o in operands]), min([o[0].wavelengths[-1] for o in operands]))
datasets = lambda operands, wavelengths: [PlotsMath.__data(o[1], wavelengths) for o in operands]
operands = [(self.operands_model.item(a).data(PlotsMath.FITS_SPECTRUM), self.operands_model.item(a).data(PlotsMath.F_X)) for a in np.arange(self.operands_model.rowCount())]
def divide(operands):
if len(operands) > 2:
print("Division supports only 2 operands, truncating")
wavelengths = max_wavelengths(operands[0:2])
datas = datasets(operands[0:2], wavelengths)
return (wavelengths, datas[0]/datas[1])
def mean(operands):
wavelengths = max_wavelengths(operands)
mean_data = np.zeros(wavelengths.shape)
for data in datasets(operands, wavelengths):
mean_data += data
return (wavelengths, mean_data/len(wavelengths))
operations = { 0: divide, 1: mean }
try:
wavelengths, data = operations[self.ui.operation_type.currentIndex()](operands)
self.spectrum = Spectrum(data, wavelengths)
self.spectrum.normalize_to_max()
self.undo.set_spectrum(self.spectrum)
self.ui.spline_degrees.setValue(5)
self.ui.spline_factor.setValue(0)
self.ui.spline_factor_auto.setChecked(False)
self.draw()
except IndexError:
QMessageBox.warning(None, "Error", "Datasets are not compatible. Maybe you need to calibrate better, or use a different reference file")
def test_get_index_from_wavelength(self):
spectrum = Spectrum(np.array([0.2, 0.3, 0.5, 0.7]),
np.array([3, 6, 9, 12]))
self.assertEqual(1, spectrum.wavelength_index(6))
self.assertEqual(0, spectrum.wavelength_index(5))
self.assertEqual(0, spectrum.wavelength_index(3))
self.assertEqual(2, spectrum.wavelength_index(9))
self.assertEqual(3, spectrum.wavelength_index(12))
def test_empty_wavelengths(self):
spectrum = Spectrum(np.array([0.2, 0.3, 0.5]))
assert_array_equal(spectrum.wavelengths, [0, 1, 2])
spectrum = Spectrum(np.array([0.2, 0.3, 0.5]),
first_wavelength=5,
dispersion=2)
assert_array_equal(spectrum.wavelengths, [5, 7, 9])
def test_calibrate(self):
spectrum = Spectrum(np.array([0.2, 0.3, 0.5]), np.array([3, 4, 5]))
spectrum.calibrate(points=[{'x': 1, 'wavelength': 2}], dispersion=4)
assert_array_equal(spectrum.wavelengths, [-2., 2., 6.])
spectrum = Spectrum(np.array([0.2, 0.3, 0.5, 0.7]), np.array([3, 4,
5]))
spectrum.calibrate(points=[{
'x': 1,
'wavelength': 3.
}, {
'x': 2,
'wavelength': 9.
}])
assert_array_almost_equal(spectrum.wavelengths, [-3., 3., 9., 15.])
def test_calc_dispersion(self):
spectrum = Spectrum(np.array([0.2, 0.3, 0.5]), np.array([0, 1, 2]))
self.assertEqual(spectrum.dispersion(), 1)
spectrum = Spectrum(np.array([0.2, 0.3, 0.5]), np.array([0, 2.5, 5]))
self.assertEqual(spectrum.dispersion(), 2.5)
def test_cut(self):
spectrum = Spectrum(np.array([0.2, 0.3, 0.5, 0.7]),
np.array([3, 6, 9, 12]))
spectrum.cut(start=1)
assert_array_equal(spectrum.wavelengths, [6, 9, 12])
assert_array_equal(spectrum.fluxes, [0.3, 0.5, 0.7])
spectrum = Spectrum(np.array([0.2, 0.3, 0.5, 0.7]),
np.array([3, 6, 9, 12]))
spectrum.cut(end=2)
assert_array_equal(spectrum.wavelengths, [3, 6, 9])
assert_array_equal(spectrum.fluxes, [0.2, 0.3, 0.5])
spectrum = Spectrum(np.array([0.2, 0.3, 0.5, 0.7]),
np.array([3, 6, 9, 12]))
spectrum.cut(start=1, end=2)
assert_array_equal(spectrum.wavelengths, [6, 9])
assert_array_equal(spectrum.fluxes, [0.3, 0.5])
class PlotsMath(QWidget):
F_X = Qt.UserRole + 1
FITS_SPECTRUM = Qt.UserRole + 2
def __init__(self, settings, database, project=None):
super(PlotsMath, self).__init__()
self.ui = Ui_PlotsMath()
self.ui.setupUi(self)
self.settings = settings
self.project=project
self.plot = QtCommons.nestWidget(self.ui.plot, QMathPlotWidget())
self.reference_dialog = ReferenceSpectraDialog(database)
self.reference_dialog.fits_picked.connect(self.open_fits)
self.toolbar = QToolBar('Instrument Response Toolbar')
open_btn = QtCommons.addToolbarPopup(self.toolbar, text="Open...", icon_file=':/new_open_20')
open_file_action = open_btn.menu().addAction('FITS file')
open_btn.menu().addAction('Reference library', self.reference_dialog.show)
self.blackbody_menu = blackbody.BlackBodyAction(self.blackbody, open_btn.menu())
if project:
save_result = QtCommons.addToolbarPopup(self.toolbar, text='Save', icon_file=':/save_20')
save_result.menu().addAction('As File', lambda: QtCommons.save_file('Save Operation Result...', FITS_EXTS, lambda f: self.save(f[0]), project.path))
save_result.menu().addAction('As Instrument Response', self.save_project_instrument_response)
open_file_action.triggered.connect(lambda: QtCommons.open_file('Open FITS Spectrum',FITS_EXTS, lambda f: self.open_fits(f[0]), project.path))
else:
open_file_action.triggered.connect(lambda: open_file_sticky('Open FITS Spectrum',FITS_EXTS, lambda f: self.open_fits(f[0]), self.settings, CALIBRATED_PROFILE, [RAW_PROFILE]))
self.toolbar.addAction(QIcon(':/save_20'), 'Save', lambda: save_file_sticky('Save Operation Result...', 'FITS file (.fit)', lambda f: self.save(f[0]), self.settings, MATH_OPERATION, [CALIBRATED_PROFILE]))
self.toolbar.addAction('Set operand', self.set_operand)
self.toolbar.addSeparator()
self.toolbar.addAction(self.ui.actionZoom)
self.ui.actionZoom.triggered.connect(self.start_zoom)
self.toolbar.addAction(self.ui.actionReset_Zoom)
self.ui.actionReset_Zoom.triggered.connect(self.reset_zoom)
self.toolbar.addSeparator()
self.operands_model = QStandardItemModel()
self.ui.operands_listview.setModel(self.operands_model)
remove_btn = QtCommons.addToolbarPopup(self.toolbar, text='Remove...')
remove_btn.menu().addAction(self.ui.actionSelectPointsToRemove)
remove_btn.menu().addAction("Before point", lambda: spectrum_trim_dialog(self.spectrum, 'before', self.plot.axes, lambda: self.draw(), self, before_removal=self.undo.save_undo))
remove_btn.menu().addAction("After point", lambda: spectrum_trim_dialog(self.spectrum, 'after', self.plot.axes, lambda: self.draw(), self, before_removal=self.undo.save_undo))
self.ui.clear_operands.clicked.connect(self.operands_model.clear)
self.ui.remove_operand.clicked.connect(lambda: self.operands_model.removeRows(self.ui.operands_listview.selectionModel().selectedRows()[0].row(), 1))
self.operands_model.rowsInserted.connect(lambda: self.ui.clear_operands.setEnabled(self.operands_model.rowCount() > 0) )
self.operands_model.rowsRemoved.connect(lambda: self.ui.clear_operands.setEnabled(self.operands_model.rowCount() > 0) )
self.ui.operands_listview.selectionModel().selectionChanged.connect(lambda s, u: self.ui.remove_operand.setEnabled(len(s)))
self.ui.actionSelectPointsToRemove.triggered.connect(self.pick_rm_points)
self.undo = Undo(None, self.draw)
self.undo.add_actions(self.toolbar)
self.ui.spline_factor.valueChanged.connect(self.factor_valueChanged)
self.ui.spline_degrees.valueChanged.connect(lambda v: self.draw())
self.ui.spline_factor_auto.toggled.connect(lambda v: self.draw())
self.ui.spline_factor_auto.toggled.connect(lambda v: self.ui.spline_factor.setEnabled(not v))
self.ui.execute.clicked.connect(self.execute_operation)
self.plot.figure.tight_layout()
def blackbody(self, blackbody):
self.spectrum = blackbody.spectrum()
self.spectrum_name = "Blackbody radiation for {0}".format(blackbody.kelvin)
self.undo.set_spectrum(self.spectrum)
self.spectrum.normalize_to_max()
self.draw()
def open_fits(self, filename):
fits_file = fits.open(filename)
fits_spectrum = FitsSpectrum(fits_file)
self.spectrum_name = fits_spectrum.name()
self.spectrum = fits_spectrum.spectrum
self.undo.set_spectrum(self.spectrum)
self.spectrum.normalize_to_max()
if self.spectrum.dispersion() <0.4:
print("dispersion too high ({}), reducing spectrum resolution".format(self.spectrum.dispersion()))
self.spectrum.resample(self.spectrum.dispersion() / 0.4)
self.draw()
@pyqtSlot(float)
def factor_valueChanged(self, f):
self.draw()
def pick_rm_points(self):
self.plot.rm_element('zoom')
self.plot.add_span_selector('pick_rm_points', lambda min,max: self.rm_points(min,max+1),direction='horizontal')
def start_zoom(self):
self.plot.rm_element('pick_rm_points')
self.plot.select_zoom()
def draw(self):
self.ui.spline_degrees_value.setText("{}".format(self.ui.spline_degrees.value()))
spline_factor = self.ui.spline_factor.value() if not self.ui.spline_factor_auto.isChecked() else None
spline = UnivariateSpline(self.spectrum.wavelengths, self.spectrum.fluxes, k=self.ui.spline_degrees.value(), s=spline_factor)
self.f_x = lambda x: spline(x)
self.plot.plot(None, self.spectrum.wavelengths, self.spectrum.fluxes, '--', self.spectrum.wavelengths, spline(self.spectrum.wavelengths), '-')
self.plot.figure.tight_layout()
self.plot.figure.canvas.draw()
def rm_points(self, wmin, wmax):
self.undo.save_undo()
x_min = self.spectrum.wavelength_index(max(self.spectrum.wavelengths[0], wmin))
x_max = self.spectrum.wavelength_index(min(self.spectrum.wavelengths[-1], wmax))
m=(self.spectrum.fluxes[x_max]-self.spectrum.fluxes[x_min])/(x_max-x_min)
q = self.spectrum.fluxes[x_min]
f = lambda x: x * m + q
self.spectrum.fluxes[x_min:x_max] = np.fromfunction(f, self.spectrum.fluxes[x_min:x_max].shape)
self.draw()
def trim(self, direction):
point = QInputDialog.getInt(None, 'Trim curve', 'Enter wavelength for trimming', self.spectrum.wavelengths[0] if direction == 'before' else self.spectrum.wavelengths[-1], self.spectrum.wavelengths[0], self.spectrum.wavelengths[-1])
if not point[1]:
return
self.undo.save_undo()
if direction == 'before':
self.spectrum.cut(start=self.spectrum.wavelength_index(point[0]))
else:
self.spectrum.cut(end=self.spectrum.wavelength_index(point[0]))
self.reset_zoom()
self.draw()
def set_operand(self):
item = QStandardItem(self.spectrum_name)
item.setData(self.f_x, PlotsMath.F_X)
item.setData(self.spectrum, PlotsMath.FITS_SPECTRUM)
self.operands_model.appendRow(item)
def execute_operation(self):
max_wavelengths = lambda operands: np.arange(max([o[0].wavelengths[0] for o in operands]), min([o[0].wavelengths[-1] for o in operands]))
datasets = lambda operands, wavelengths: [PlotsMath.__data(o[1], wavelengths) for o in operands]
operands = [(self.operands_model.item(a).data(PlotsMath.FITS_SPECTRUM), self.operands_model.item(a).data(PlotsMath.F_X)) for a in np.arange(self.operands_model.rowCount())]
def divide(operands):
if len(operands) > 2:
print("Division supports only 2 operands, truncating")
wavelengths = max_wavelengths(operands[0:2])
datas = datasets(operands[0:2], wavelengths)
return (wavelengths, datas[0]/datas[1])
def mean(operands):
wavelengths = max_wavelengths(operands)
mean_data = np.zeros(wavelengths.shape)
for data in datasets(operands, wavelengths):
mean_data += data
return (wavelengths, mean_data/len(wavelengths))
operations = { 0: divide, 1: mean }
try:
wavelengths, data = operations[self.ui.operation_type.currentIndex()](operands)
self.spectrum = Spectrum(data, wavelengths)
self.spectrum.normalize_to_max()
self.undo.set_spectrum(self.spectrum)
self.ui.spline_degrees.setValue(5)
self.ui.spline_factor.setValue(0)
self.ui.spline_factor_auto.setChecked(False)
self.draw()
except IndexError:
QMessageBox.warning(None, "Error", "Datasets are not compatible. Maybe you need to calibrate better, or use a different reference file")
def save_project_instrument_response(self):
name = QInputDialog.getText(self, 'Enter Name', 'Enter new instrument response name for saving')
if name[1]:
self.project.add_file(Project.INSTRUMENT_RESPONSES, lambda f: self.save(f), bare_name=name[0])
def save(self, filename):
hdu = fits.PrimaryHDU( PlotsMath.__data(self.f_x, self.spectrum.wavelengths))
#hdu = fits.PrimaryHDU( self.spectrum.fluxes)
fits_file = fits.HDUList([hdu])
hdu.header['CRPIX1'] = 1
hdu.header['CRVAL1'] = self.spectrum.wavelengths[0]
hdu.header['CDELT1'] = self.spectrum.dispersion()
hdu.writeto(filename, clobber=True)
def reset_zoom(self):
self.plot.reset_zoom(self.spectrum.wavelengths, self.spectrum.fluxes.min(), self.spectrum.fluxes.max())
def __data(f_x, xrange):
return np.fromfunction(lambda x: f_x(x+xrange[0]), xrange.shape)