IterativeInversionResult

class ctis.inverters.IterativeInversionResult(solutions, success, images, inverter, message, num_iteration, mean_chi_squared, correlation_residual)

Bases: AbstractInversionResult

The results of an iterative inversion attempt.

Attributes

correlation_residual	The correlation between the predicted images and the residuals for each iteration.
images	The observed images on which the inversion was performed.
inverter	The inversion algorithm instance that produced these results.
iteration	The iteration value for each iteration.
mean_chi_squared	The mean chi squared statistic for each iteration.
message	Any message from the inversion routine concerning the results.
num_iteration	The number of iterations performed by the inverter.
solution	The reconstructed scene found by the inversion.
success	A boolean flag indicating whether the inversion was successful.
solutions	Intermediate solutions from each iteration.

Methods

__init__(solutions, success, images, ...)
plot_moments(truth, axis[, num_bins, ...])	Plot column-normalized 2D histograms of the true vs. reconstructed radiance, median, and interquartile range of the spectral line profile.

Inheritance Diagram

Parameters:

• solutions (FunctionArray[SpectralPositionalVectorArray, ScalarArray])

• success (bool)

• images (FunctionArray[SpectralPositionalVectorArray, ScalarArray])

• inverter (AbstractInverter)

• message (str)

• num_iteration (int)

• mean_chi_squared (ScalarArray)

• correlation_residual (ScalarArray)

plot_moments(truth, axis, num_bins=50, range_radiance=None, range_median=None, range_iqr=None, percentile_radiance=0)

Plot column-normalized 2D histograms of the true vs. reconstructed radiance, median, and interquartile range of the spectral line profile.

Parameters:

• truth (FunctionArray[AbstractDopplerPositionalVectorArray, ScalarArray]) – The true scene which will be compared to the reconstructed scene.

• axis (str) – The axis along which to compute the radiance, median, and interquartile range.

• num_bins (int) – The number of bins along each axis of the histogram.

• range_radiance (None | tuple[Quantity, Quantity]) – The domain of the radiance histogram.

• range_median (None | tuple[Quantity, Quantity]) – The domain of the median histogram.

• range_iqr (None | tuple[Quantity, Quantity]) – The domain of the interquartile range histogram.

• percentile_radiance (float) – Spatial locations below this threshold are excluded from the histogram. The default is to not exclude any pixels.

Return type:

tuple[Figure, ndarray]

correlation_residual: ScalarArray = <dataclasses._MISSING_TYPE object>

The correlation between the predicted images and the residuals for each iteration.

images: FunctionArray[SpectralPositionalVectorArray, ScalarArray] = <dataclasses._MISSING_TYPE object>

The observed images on which the inversion was performed.

inverter: AbstractInverter = <dataclasses._MISSING_TYPE object>

The inversion algorithm instance that produced these results.

property iteration: ScalarArray

The iteration value for each iteration.

mean_chi_squared: ScalarArray = <dataclasses._MISSING_TYPE object>

The mean chi squared statistic for each iteration.

message: str = <dataclasses._MISSING_TYPE object>

Any message from the inversion routine concerning the results.

num_iteration: int = <dataclasses._MISSING_TYPE object>

The number of iterations performed by the inverter.

property solution: FunctionArray[SpectralPositionalVectorArray, ScalarArray]

The reconstructed scene found by the inversion.

solutions: FunctionArray[SpectralPositionalVectorArray, ScalarArray]

Intermediate solutions from each iteration.

If AbstractIterativeInverter.intermediate is set to True, this has up to num_iteration elements along the axis_iteration logical axis. Otherwise this has only one element along the axis_iteration axis.

success: bool = <dataclasses._MISSING_TYPE object>

A boolean flag indicating whether the inversion was successful.