deeplc.calibration

Calibration utilities.

class deeplc.calibration.Calibration(*args, **kwargs)

Bases: ABC

Initialize the calibration model.

abstract property is_fitted: bool: Indicates whether the calibration model has been fitted.

abstractmethod fit(target, source)

Fit the calibration from source to target.

Parameters:

target (ndarray)
source (ndarray)

Return type:

None

abstractmethod transform(source)

Transform source values into the calibrated target space.

Parameters:: source (ndarray)
Return type:: ndarray

class deeplc.calibration.IdentityCalibration(*args, **kwargs)

Bases: Calibration

Initialize the calibration model.

property is_fitted: bool: Always fitted; identity calibration requires no fitting.

fit(target, source)

No-op; identity calibration does not fit.

Parameters:

target (ndarray)
source (ndarray)

Return type:

None

transform(source)

Return source unchanged.

Parameters:: source (ndarray)
Return type:: ndarray

class deeplc.calibration.PiecewiseLinearCalibration(number_of_splits=10, extrapolate=True, use_median=False, min_samples_per_segment=20)

Bases: Calibration

Piece-wise linear calibration based on per-split anchors.

Parameters:

number_of_splits (int) – Number of segments to split the source value range into. More segments allow more flexibility but may lead to overfitting.
extrapolate (bool) – If True, allows extrapolation outside the fitted source value range. If False, clips input values to the fitted range.
use_median (bool) – If True, uses the median of each segment to define anchors. If False, uses the mean.
min_samples_per_segment (int) – Minimum number of samples required for a segment to contribute an anchor. Segments with fewer samples are skipped, which helps avoid unstable anchors in sparse regions when using many splits.

property is_fitted: bool: True if the calibration model has been fitted.

property calibrate_min: float | None: Minimum source value seen during fitting.

property calibrate_max: float | None: Maximum source value seen during fitting.

fit(target, source)

Fit a piece-wise linear model mapping source to target values.

Parameters:

target (ndarray)
source (ndarray)

Return type:

None

transform(source)

Transform source values using the fitted piece-wise linear model.

Parameters:: source (ndarray)
Return type:: ndarray

get_calibration_curve()

Return the calibration anchors as two arrays (x, y).

Return type:: tuple[ndarray, ndarray]

class deeplc.calibration.SplineTransformerCalibration

Bases: Calibration

Initialize SplineTransformerCalibration.

property is_fitted: bool: True if the calibration model has been fitted.

fit(target, source)

Fit a spline-based model mapping source to target values.

Parameters:

target (ndarray)
source (ndarray)

Return type:

None

transform(source)

Transform source values using the fitted spline model.

Parameters:: source (ndarray)
Return type:: ndarray