def result_dataset_mapping(result: ResultLike) -> Mapping[str, xr.Dataset]:
"""Convert a ``ResultLike`` object to a per dataset mapping of result like data.
Parameters
----------
result : ResultLike
Data structure which can be converted to a mapping.
Returns
-------
Mapping[str, Dataset]
Per dataset mapping of result like data.
Raises
------
TypeError
If any value of a ``result`` isn't of :class:`DatasetConvertible`.
TypeError
If ``result`` isn't a :class:`ResultLike` object.
"""
result_mapping = {}
if isinstance(result, Result):
return result.data
if isinstance(result, (xr.Dataset, xr.DataArray, Path, str)):
return {"dataset": load_data(result)}
if isinstance(result, Sequence):
for index, value in enumerate(result):
result_mapping[f"dataset{index}"] = load_data(value)
return result_mapping
if isinstance(result, Mapping):
for key, value in result.items():
result_mapping[key] = load_data(value)
return result_mapping
raise TypeError(f"Result needs to be of type {ResultLike!r}, but was {result!r}.")
def plot_overview(
result: DatasetConvertible,
center_λ: float | None = None,
linlog: bool = True,
linthresh: float = 1,
linscale: float = 1,
show_data: bool = False,
main_irf_nr: int = 0,
figsize: tuple[int, int] = (18, 16),
cycler: Cycler | None = PlotStyle().cycler,
figure_only: bool = True,
nr_of_data_svd_vectors: int = 4,
nr_of_residual_svd_vectors: int = 2,
show_data_svd_legend: bool = True,
show_residual_svd_legend: bool = True,
) -> Figure | tuple[Figure, Axes]:
"""Plot overview of the optimization result.
Parameters
----------
result: DatasetConvertible
Result from a pyglotaran optimization as dataset, Path or Result object.
center_λ: float | None
Center wavelength (λ in nm)
linlog: bool
Whether to use 'symlog' scale or not. Defaults to False.
linthresh: float
A single float which defines the range (-x, x), within which the plot is linear.
This avoids having the plot go to infinity around zero. Defaults to 1.
linscale: float
This allows the linear range (-linthresh to linthresh) to be stretched
relative to the logarithmic range.
Its value is the number of decades to use for each half of the linear range.
For example, when linscale == 1.0 (the default), the space used for the
positive and negative halves of the linear range will be equal to one
decade in the logarithmic range. Defaults to 1.
show_data: bool
Whether to show the input data or residual. Defaults to False.
main_irf_nr: int
Index of the main ``irf`` component when using an ``irf``
parametrized with multiple peaks. Defaults to 0.
figsize : tuple[int, int]
Size of the figure (N, M) in inches. Defaults to (18, 16).
cycler : Cycler | None
Plot style cycler to use. Defaults to PlotStyle().cycler.
figure_only: bool
Whether or not to only return the figure.
This is a deprecation helper argument to transition to a consistent return value
consisting of the :class:`Figure` and the :class:`Axes`. Defaults to True.
nr_of_data_svd_vectors: int
Number of data SVD vector to plot. Defaults to 4.
nr_of_residual_svd_vectors: int
Number of residual SVD vector to plot. Defaults to 2.
show_data_svd_legend: bool
Whether or not to show the data SVD legend. Defaults to True.
show_residual_svd_legend: bool
Whether or not to show the residual SVD legend. Defaults to True.
Returns
-------
Figure|tuple[Figure, Axes]
If ``figure_only`` is True, Figure object which contains the plots (deprecated).
If ``figure_only`` is False, Figure object which contains the plots and the Axes.
"""
res = load_data(result)
# Plot dimensions
M = 4
N = 3
fig, axes = plt.subplots(M, N, figsize=figsize, constrained_layout=True)
if center_λ is None: # center wavelength (λ in nm)
center_λ = min(res.dims["spectral"], round(res.dims["spectral"] / 2))
# First and second row: concentrations - SAS/EAS - DAS
plot_concentrations(
res,
axes[0, 0],
center_λ,
linlog=linlog,
linthresh=linthresh,
linscale=linscale,
main_irf_nr=main_irf_nr,
cycler=cycler,
)
plot_spectra(res, axes[0:2, 1:3], cycler=cycler)
plot_svd(
res,
axes[2:4, 0:3],
linlog=linlog,
linthresh=linthresh,
cycler=cycler,
nr_of_data_svd_vectors=nr_of_data_svd_vectors,
nr_of_residual_svd_vectors=nr_of_residual_svd_vectors,
show_data_svd_legend=show_data_svd_legend,
show_residual_svd_legend=show_residual_svd_legend,
)
plot_residual(
res, axes[1, 0], linlog=linlog, linthresh=linthresh, show_data=show_data, cycler=cycler
)
# plt.tight_layout(pad=3, w_pad=4.0, h_pad=4.0)
if figure_only is True:
warn(PyglotaranExtrasApiDeprecationWarning(FIG_ONLY_WARNING), stacklevel=2)
return fig
else:
return fig, axes
def plot_simple_overview(
result: DatasetConvertible,
title: str | None = None,
figsize: tuple[int, int] = (12, 6),
cycler: Cycler | None = PlotStyle().cycler,
figure_only: bool = True,
) -> Figure | tuple[Figure, Axes]:
"""Plot simple overview.
Parameters
----------
result: DatasetConvertible
Result from a pyglotaran optimization as dataset, Path or Result object.
title: str | None
Title of the figure. Defaults to None.
figsize : tuple[int, int]
Size of the figure (N, M) in inches. Defaults to (18, 16).
cycler : Cycler | None
Plot style cycler to use. Defaults to PlotStyle().cycler.
figure_only: bool
Whether or not to only return the figure.
This is a deprecation helper argument to transition to a consistent return value
consisting of the :class:`Figure` and the :class:`Axes`. Defaults to True.
Returns
-------
Figure|tuple[Figure, Axes]
If ``figure_only`` is True, Figure object which contains the plots (deprecated).
If ``figure_only`` is False, Figure object which contains the plots and the Axes.
"""
res = load_data(result)
fig, axes = plt.subplots(2, 3, figsize=figsize, constrained_layout=True)
for ax in axes.flatten():
add_cycler_if_not_none(ax, cycler)
if title:
fig.suptitle(title, fontsize=16)
sas = res.species_associated_spectra
traces = res.species_concentration
if "spectral" in traces.coords:
traces.sel(spectral=res.spectral.values[0], method="nearest").plot.line(
x="time", ax=axes[0, 0]
)
else:
traces.plot.line(x="time", ax=axes[0, 0])
sas.plot.line(x="spectral", ax=axes[0, 1])
rLSV = res.residual_left_singular_vectors
rLSV.isel(left_singular_value_index=range(min(2, len(rLSV)))).plot.line(
x="time", ax=axes[1, 0]
)
axes[1, 0].set_title("res. LSV")
rRSV = res.residual_right_singular_vectors
rRSV.isel(right_singular_value_index=range(min(2, len(rRSV)))).plot.line(
x="spectral", ax=axes[1, 1]
)
axes[1, 1].set_title("res. RSV")
res.data.plot(x="time", ax=axes[0, 2])
axes[0, 2].set_title("data")
res.residual.plot(x="time", ax=axes[1, 2])
axes[1, 2].set_title("residual")
if figure_only is True:
warn(PyglotaranExtrasApiDeprecationWarning(FIG_ONLY_WARNING), stacklevel=2)
return fig
else:
return fig, axes
rRSV.isel(right_singular_value_index=range(min(2, len(rRSV)))).plot.line(
x="spectral", ax=axes[1, 1]
)
axes[1, 1].set_title("res. RSV")
res.data.plot(x="time", ax=axes[0, 2])
axes[0, 2].set_title("data")
res.residual.plot(x="time", ax=axes[1, 2])
axes[1, 2].set_title("residual")
if figure_only is True:
warn(PyglotaranExtrasApiDeprecationWarning(FIG_ONLY_WARNING), stacklevel=2)
return fig
else:
return fig, axes
if __name__ == "__main__":
import sys
result_path = Path(sys.argv[1])
res = load_data(result_path)
print(res)
fig, plt.axes = plot_overview(res, figure_only=False)
if len(sys.argv) > 2:
fig.savefig(sys.argv[2], bbox_inches="tight")
print(f"Saved figure to: {sys.argv[2]}")
else:
plt.show(block=False)
input("press <ENTER> to continue")
def plot_doas(
result: DatasetConvertible,
figsize: tuple[int, int] = (25, 25),
cycler: Cycler | None = PlotStyle().cycler,
figure_only: bool = True,
) -> Figure | tuple[Figure, Axes]:
"""Plot Damped oscillation associated spectra (DOAS).
Parameters
----------
result: DatasetConvertible
Result from a pyglotaran optimization as dataset, Path or Result object.
figsize : tuple[int, int]
Size of the figure (N, M) in inches. Defaults to (18, 16).
cycler : Cycler | None
Plot style cycler to use. Defaults to PlotStyle().cycler.
figure_only: bool
Whether or not to only return the figure.
This is a deprecation helper argument to transition to a consistent return value
consisting of the :class:`Figure` and the :class:`Axes`. Defaults to True.
Returns
-------
Figure|tuple[Figure, Axes]
If ``figure_only`` is True, Figure object which contains the plots (deprecated).
If ``figure_only`` is False, Figure object which contains the plots and the Axes.
"""
dataset = load_data(result)
# Create M x N plotting grid
M = 6
N = 3
fig, axes = plt.subplots(M, N, figsize=figsize)
for ax in axes.flatten():
add_cycler_if_not_none(ax, cycler)
# Plot data
dataset.species_associated_spectra.plot.line(x="spectral", ax=axes[0, 0])
dataset.decay_associated_spectra.plot.line(x="spectral", ax=axes[0, 1])
if "spectral" in dataset.species_concentration.coords:
dataset.species_concentration.isel(spectral=0).plot.line(x="time",
ax=axes[1, 0])
else:
dataset.species_concentration.plot.line(x="time", ax=axes[1, 0])
axes[1, 0].set_xscale("symlog", linthreshx=1)
if "dampened_oscillation_associated_spectra" in dataset:
dataset.dampened_oscillation_cos.isel(spectral=0).sel(
time=slice(-1, 10)).plot.line(x="time", ax=axes[1, 1])
dataset.dampened_oscillation_associated_spectra.plot.line(x="spectral",
ax=axes[2,
0])
dataset.dampened_oscillation_phase.plot.line(x="spectral",
ax=axes[2, 1])
dataset.residual_left_singular_vectors.isel(
left_singular_value_index=0).plot(ax=axes[0, 2])
dataset.residual_singular_values.plot.line("ro-",
yscale="log",
ax=axes[1, 2])
dataset.residual_right_singular_vectors.isel(
right_singular_value_index=0).plot(ax=axes[2, 2])
interval = int(dataset.spectral.size / 11)
for i in range(0):
axi = axes[i % 3, int(i / 3) + 3]
index = (i + 1) * interval
dataset.data.isel(spectral=index).plot(ax=axi)
dataset.residual.isel(spectral=index).plot(ax=axi)
dataset.fitted_data.isel(spectral=index).plot(ax=axi)
plt.tight_layout(pad=5, w_pad=2.0, h_pad=2.0)
if figure_only is True:
warn(PyglotaranExtrasApiDeprecationWarning(FIG_ONLY_WARNING),
stacklevel=2)
return fig
else:
return fig, axes
def plot_data_overview(
dataset: DatasetConvertible,
title: str = "Data overview",
linlog: bool = False,
linthresh: float = 1,
figsize: tuple[int, int] = (15, 10),
nr_of_data_svd_vectors: int = 4,
show_data_svd_legend: bool = True,
) -> tuple[Figure, Axes]:
"""Plot data as filled contour plot and SVD components.
Parameters
----------
dataset : DatasetConvertible
Dataset containing data and SVD of the data.
title : str
Title to add to the figure. Defaults to "Data overview".
linlog : bool
Whether to use 'symlog' scale or not. Defaults to False.
linthresh : float
A single float which defines the range (-x, x), within which the plot is linear.
This avoids having the plot go to infinity around zero. Defaults to 1.
figsize : tuple[int, int]
Size of the figure (N, M) in inches. Defaults to (15, 10).
nr_of_data_svd_vectors: int
Number of data SVD vector to plot. Defaults to 4.
show_data_svd_legend: bool
Whether or not to show the data SVD legend. Defaults to True.
Returns
-------
tuple[Figure, Axes]
Figure and axes which can then be refined by the user.
"""
dataset = load_data(dataset)
fig = plt.figure(figsize=figsize)
data_ax = cast(
Axis, plt.subplot2grid((4, 3), (0, 0), colspan=3, rowspan=3, fig=fig))
lsv_ax = cast(Axis, plt.subplot2grid((4, 3), (3, 0), fig=fig))
sv_ax = cast(Axis, plt.subplot2grid((4, 3), (3, 1), fig=fig))
rsv_ax = cast(Axis, plt.subplot2grid((4, 3), (3, 2), fig=fig))
if len(dataset.data.time) > 1:
dataset.data.plot(x="time", ax=data_ax, center=False)
else:
dataset.data.plot(ax=data_ax)
add_svd_to_dataset(dataset=dataset, name="data")
plot_lsv_data(dataset,
lsv_ax,
indices=range(nr_of_data_svd_vectors),
show_legend=show_data_svd_legend)
plot_sv_data(dataset, sv_ax)
plot_rsv_data(dataset,
rsv_ax,
indices=range(nr_of_data_svd_vectors),
show_legend=show_data_svd_legend)
fig.suptitle(title, fontsize=16)
fig.tight_layout()
if linlog:
data_ax.set_xscale("symlog", linthresh=linthresh)
return fig, (data_ax, lsv_ax, sv_ax, rsv_ax)