Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Simplify axes conversion code #164

Merged
merged 1 commit into from
Dec 6, 2022
Merged

Simplify axes conversion code #164

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
344 changes: 93 additions & 251 deletions lumispy/signals/luminescence_spectrum.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -93,6 +93,78 @@ def _reset_variance_linear_model(self):
UserWarning,
)

def _convert_data(self, newaxis, factor, inplace, jacobian, data2, var2):
"""Utility function to perform the data and variance conversion for
signal unit transformations.
"""
# convert data
if jacobian:
s2data = np.copy(
data2(
self.isig[::-1].data,
factor,
newaxis.axis,
self.axes_manager.signal_axes[0],
)
)
else:
s2data = np.copy(self.isig[::-1].data)

# inplace conversion
if inplace:
s2 = self
s2.data = s2data
# create new signal object with correct data, axes, metadata
else:
s2 = self._deepcopy_with_new_data(
s2data, copy_variance=(not jacobian), copy_learning_results=False
)
# convert axis
oldaxis = self.axes_manager.signal_axes[0]
s2.axes_manager.set_axis(
newaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
)
# convert variance
if self.metadata.has_item("Signal.Noise_properties.variance"):
var = self.get_noise_variance()
if jacobian:
# if variance is a numeric value, cast into signal object
if isinstance(var, (float, int)):
var = self._deepcopy_with_new_data(
np.ones(self.data.shape) * var,
copy_variance=False,
copy_learning_results=False,
)
s2var = s2._deepcopy_with_new_data(
var2(
var.isig[::-1].data,
factor,
newaxis.axis,
oldaxis,
),
copy_variance=False,
copy_learning_results=False,
)
s2.set_noise_variance(s2var)
s2._reset_variance_linear_model()
else:
# variance left unchanged, if it is a number and Jacobian not performed
if isinstance(var, (float, int)):
if not inplace:
s2.set_noise_variance(self.get_noise_variance())
else:
s2.set_noise_variance(
s2._deepcopy_with_new_data(
var.isig[::-1].data,
copy_variance=False,
copy_learning_results=False,
)
)
if not inplace:
return s2
else:
return None

def to_eV(self, inplace=True, jacobian=True):
"""Converts signal axis of 1D signal to non-linear energy axis (eV)
using wavelength dependent refractive index of air. Assumes wavelength
Expand Down Expand Up @@ -128,120 +200,7 @@ def to_eV(self, inplace=True, jacobian=True):

evaxis, factor = axis2eV(self.axes_manager.signal_axes[0])

# in place conversion
if inplace:
# convert data
if jacobian:
self.data = data2eV(
self.isig[::-1].data,
factor,
self.axes_manager.signal_axes[0],
evaxis.axis,
)
else:
self.data = self.isig[::-1].data
# convert axis
oldaxis = self.axes_manager.signal_axes[0]
self.axes_manager.set_axis(
evaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
)
# convert variance
if jacobian:
if self.metadata.has_item("Signal.Noise_properties.variance"):
var = self.get_noise_variance()
# if variance is a numeric value, cast into signal object
if isinstance(var, (float, int)):
self.set_noise_variance(
self._deepcopy_with_new_data(
np.ones(self.data.shape) * var,
copy_variance=False,
copy_learning_results=False,
)
)
var = self.get_noise_variance()
svar = self._deepcopy_with_new_data(
var2eV(
var.isig[::-1].data,
factor,
oldaxis,
evaxis.axis,
),
copy_variance=False,
copy_learning_results=False,
)
self.set_noise_variance(svar)
self._reset_variance_linear_model()
else:
if self.metadata.has_item("Signal.Noise_properties.variance"):
var = self.get_noise_variance()
# variance left unchanged, if it is a number and Jacobian not performed
if not isinstance(var, (float, int)):
self.set_noise_variance(
self._deepcopy_with_new_data(
var.isig[::-1].data,
copy_variance=False,
copy_learning_results=False,
)
)

# create and return new signal
else:
# create new data array
if jacobian:
s2data = np.copy(
data2eV(
self.isig[::-1].data,
factor,
self.axes_manager.signal_axes[0],
evaxis.axis,
)
)
else:
s2data = np.copy(self.isig[::-1].data)

# create new signal object with correct data, axes, metadata
s2 = self._deepcopy_with_new_data(
s2data, copy_variance=(not jacobian), copy_learning_results=False
)
s2.axes_manager.set_axis(
evaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
)
# convert variance
if self.metadata.has_item("Signal.Noise_properties.variance"):
var = self.get_noise_variance()
if jacobian:
# if variance is a numeric value, cast into signal object
if isinstance(var, (float, int)):
var = self._deepcopy_with_new_data(
np.ones(self.data.shape) * var,
copy_variance=False,
copy_learning_results=False,
)
# convert
s2var = s2._deepcopy_with_new_data(
var2eV(
var.isig[::-1].data,
factor,
self.axes_manager.signal_axes[0],
evaxis.axis,
),
copy_variance=False,
copy_learning_results=False,
)
s2.set_noise_variance(s2var)
else:
if isinstance(var, (float, int)):
s2.set_noise_variance(self.get_noise_variance())
else:
s2.set_noise_variance(
s2._deepcopy_with_new_data(
var.isig[::-1].data,
copy_variance=False,
copy_learning_results=False,
)
)

return s2
return self._convert_data(evaxis, factor, inplace, jacobian, data2eV, var2eV)

TO_INVCM_DOCSTRING = """
The intensity is converted from counts/nm (counts/µm) to counts/cm^-1
Expand Down Expand Up @@ -284,114 +243,9 @@ def to_invcm(self, inplace=True, jacobian=True):

invcmaxis, factor = axis2invcm(self.axes_manager.signal_axes[0])

# in place conversion
if inplace:
# convert data
if jacobian:
self.data = data2invcm(
self.isig[::-1].data,
factor,
invcmaxis.axis,
)
else:
self.data = self.isig[::-1].data
# convert axis
self.axes_manager.set_axis(
invcmaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
)
# convert variance
if jacobian:
if self.metadata.has_item("Signal.Noise_properties.variance"):
var = self.get_noise_variance()
# if variance is a numeric value, cast into signal object
if isinstance(var, (float, int)):
self.set_noise_variance(
self._deepcopy_with_new_data(
np.ones(self.data.shape) * var,
copy_variance=False,
copy_learning_results=False,
)
)
var = self.get_noise_variance()
# convert
svar = self._deepcopy_with_new_data(
var2invcm(
var.isig[::-1].data,
factor,
invcmaxis.axis,
),
copy_variance=False,
copy_learning_results=False,
)
self.set_noise_variance(svar)
self._reset_variance_linear_model()
else:
if self.metadata.has_item("Signal.Noise_properties.variance"):
var = self.get_noise_variance()
# variance left unchanged, if it is a number and Jacobian not performed
if not isinstance(var, (float, int)):
self.set_noise_variance(
self._deepcopy_with_new_data(
var.isig[::-1].data,
copy_variance=False,
copy_learning_results=False,
)
)

# create and return new signal
else:
# create new data array
if jacobian:
s2data = np.copy(
data2invcm(
self.isig[::-1].data,
factor,
invcmaxis.axis,
)
)
else:
s2data = np.copy(self.isig[::-1].data)

# create new signal object with correct data, axes, metadata
s2 = self._deepcopy_with_new_data(
s2data, copy_variance=(not jacobian), copy_learning_results=False
)
s2.axes_manager.set_axis(
invcmaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
)
# convert variance
if self.metadata.has_item("Signal.Noise_properties.variance"):
var = self.get_noise_variance()
if jacobian:
# if variance is a numeric value, cast into signal object
if isinstance(var, (float, int)):
var = self._deepcopy_with_new_data(
np.ones(self.data.shape) * var,
copy_variance=False,
copy_learning_results=False,
)
s2var = s2._deepcopy_with_new_data(
var2invcm(
var.isig[::-1].data,
factor,
invcmaxis.axis,
),
copy_variance=False,
copy_learning_results=False,
)
s2.set_noise_variance(s2var)
else:
if isinstance(var, (float, int)):
s2.set_noise_variance(var)
else:
s2.set_noise_variance(
s2._deepcopy_with_new_data(
var.isig[::-1].data,
copy_variance=False,
copy_learning_results=False,
)
)
return s2
return self._convert_data(
invcmaxis, factor, inplace, jacobian, data2invcm, var2invcm
)

to_invcm.__doc__ %= (TO_INVCM_DOCSTRING, TO_INVCM_EXAMPLE)

Expand Down Expand Up @@ -452,40 +306,28 @@ def to_invcm_relative(self, laser=None, inplace=True, jacobian=False):
invcmaxis.name = "Raman Shift"

# replace signal axis after conversion
# in place conversion (using absolute scale for Jacobian)
# conversion (using absolute scale for Jacobian)
if inplace:
self.to_invcm(inplace=inplace, jacobian=jacobian)
self.axes_manager.set_axis(
invcmaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
)
self.data = self.isig[::-1].data
# replace variance axis
if self.metadata.has_item(
"Signal.Noise_properties.variance"
) and not isinstance(self.get_noise_variance(), (float, int)):
self.metadata.Signal.Noise_properties.variance.axes_manager.set_axis(
invcmaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
)
self.metadata.Signal.Noise_properties.variance.data = (
self.metadata.Signal.Noise_properties.variance.isig[::-1].data
)
# create and return new signal
s2 = self
else:
s2 = self.to_invcm(inplace=inplace, jacobian=jacobian)
s2.axes_manager.set_axis(
invcmaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
# replace axis
s2.axes_manager.set_axis(
invcmaxis, self.axes_manager.signal_axes[0].index_in_axes_manager
)
s2.data = s2.isig[::-1].data
# replace variance axis
if s2.metadata.has_item("Signal.Noise_properties.variance") and not isinstance(
s2.get_noise_variance(), (float, int)
):
s2.metadata.Signal.Noise_properties.variance.axes_manager.set_axis(
invcmaxis, s2.axes_manager.signal_axes[0].index_in_axes_manager
)
s2.data = s2.isig[::-1].data
# replace variance axis
if s2.metadata.has_item(
"Signal.Noise_properties.variance"
) and not isinstance(s2.get_noise_variance(), (float, int)):
s2.metadata.Signal.Noise_properties.variance.axes_manager.set_axis(
invcmaxis, s2.axes_manager.signal_axes[0].index_in_axes_manager
)
s2.metadata.Signal.Noise_properties.variance.data = (
s2.metadata.Signal.Noise_properties.variance.isig[::-1].data
)
s2.metadata.Signal.Noise_properties.variance.data = (
s2.metadata.Signal.Noise_properties.variance.isig[::-1].data
)
if not inplace:
return s2

to_invcm_relative.__doc__ %= (TO_INVCM_DOCSTRING, TO_INVCMREL_EXAMPLE)
Expand Down
6 changes: 3 additions & 3 deletions lumispy/tests/utils/test_axis_conversion.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -75,19 +75,19 @@ def test_data2eV():
data = 100 * ones(20)
ax0 = DataAxis(axis=arange(200, 400, 10), units="nm")
evaxis, factor = axis2eV(ax0)
evdata = data2eV(data, factor, ax0, evaxis.axis)
evdata = data2eV(data, factor, evaxis.axis, ax0)
assert_allclose(evdata[0], 12.271168)
ax0 = DataAxis(axis=arange(0.2, 0.4, 0.01), units="µm")
evaxis, factor = axis2eV(ax0)
evdata = data2eV(data, factor, ax0, evaxis.axis)
evdata = data2eV(data, factor, evaxis.axis, ax0)
assert_allclose(evdata[0], 12.271168e-3)


def test_var2eV():
data = 100 * ones(20)
ax0 = DataAxis(axis=arange(200, 400, 10), units="nm")
evaxis, factor = axis2eV(ax0)
evvar = var2eV(data, factor, ax0, evaxis.axis)
evvar = var2eV(data, factor, evaxis.axis, ax0)
assert_allclose(evvar[0], 1.5058156)


Expand Down
Loading