calibration_validation

calibration_validation ¤

CalibrationValidationEvents ¤

CalibrationValidationEvents(
    dataframe: DataFrame,
    signal_uuid: str,
    *,
    reference_uuid: str | None = None,
    reference_value: float | None = None,
    value_column: str = "value_double",
    event_uuid: str = "quality:calibration",
    time_column: str = "systime"
)

Bases: Base

Quality: Calibration Validation & Linearity

Multi-point calibration validation for inline sensors. Assesses linearity across measurement range, detects calibration degradation, and determines when recalibration is needed.

Methods: - linearity_assessment: Residual analysis at calibration points. - calibration_curve: Polynomial fit to calibration data. - detect_recalibration_need: Monitor bias/drift for recal triggers. - calibration_stability: Bias stability scoring over time.

linearity_assessment `staticmethod` ¤

linearity_assessment(
    calibration_points: dict[float, list[float]],
) -> pd.DataFrame

Assess linearity at multiple calibration points.

Parameters:

Name	Type	Description	Default
`calibration_points`	`dict[float, list[float]]`	Dict mapping reference values to lists of measured values, e.g. {10.0: [10.1, 9.9, 10.0]}.	required

Returns:

Type	Description
`DataFrame`	DataFrame with columns: reference, measured_mean,
`DataFrame`	measured_std, bias, linearity_error.

calibration_curve `staticmethod` ¤

calibration_curve(
    calibration_points: dict[float, list[float]],
    degree: int = 1,
) -> pd.DataFrame

Fit polynomial calibration curve.

Parameters:

Name	Type	Description	Default
`calibration_points`	`dict[float, list[float]]`	Dict mapping reference values to measured value lists.	required
`degree`	`int`	Polynomial degree (1=linear, 2=quadratic, etc).	`1`

Returns:

Type	Description
`DataFrame`	DataFrame with columns: coefficient_index, coefficient_value,
`DataFrame`	r_squared, max_residual.

detect_recalibration_need ¤

detect_recalibration_need(
    window: str = "1D",
    max_bias: float | None = None,
    max_drift_rate: float | None = None,
) -> pd.DataFrame

Monitor bias and drift to detect recalibration need.

Parameters:

Name	Type	Description	Default
`window`	`str`	Resample window size.	`'1D'`
`max_bias`	`float \| None`	Maximum acceptable absolute bias.	`None`
`max_drift_rate`	`float \| None`	Maximum acceptable bias change per window.	`None`

Returns:

Type	Description
`DataFrame`	Summary DataFrame with columns: start, end, duration_seconds
`DataFrame`	(the window bounds), window_start (the resample label; a
`DataFrame`	deprecated twin of `start`), bias, drift_rate,
`DataFrame`	recalibration_needed, reason.

calibration_stability ¤

calibration_stability(
    window: str = "8h", lookback: int = 10
) -> pd.DataFrame

Track calibration bias stability over rolling windows.

Parameters:

Name	Type	Description	Default
`window`	`str`	Resample window size.	`'8h'`
`lookback`	`int`	Number of previous windows for trend assessment.	`10`

Returns:

Type	Description
`DataFrame`	Summary DataFrame with columns: start, end, duration_seconds
`DataFrame`	(the window bounds), window_start (the resample label; a
`DataFrame`	deprecated twin of `start`), bias_mean, bias_std,
`DataFrame`	bias_trend, stability_score.

calibration_validation