pycsamt.ai.inversion.schema#

Parameter-validation schemas for AI inversion interfaces.

These schemas describe public call-boundary constraints for DUHI and its geometry-aware mapper. Numerical array contents, broadcasting, geometry consistency, and cross-parameter relationships are validated by the implementation after the compatibility decorator accepts the top-level argument types and scalar ranges.

Module Attributes

`OBSERVATION_RELIABILITY_SCHEMA`	Constraints for reliability-weighted observation errors.
`DIMENSIONALITY_RELIABILITY_SCHEMA`	Constraints for phase-tensor dimensionality reliability.
`RELIABILITY_COMBINATION_SCHEMA`	Constraints for combined observation reliability.
`DUHI_INIT_SCHEMA`	Constraints for `DUHIInverter2D`.
`DUHI_PREPARE_SCHEMA`	Constraints for `DUHIInverter2D.prepare`.
`GRID_MAPPING_SCHEMA`	Constraints for geometry-aware AI-to-Occam grid mapping.

pycsamt.ai.inversion.schema.DUHI_INIT_SCHEMA = {'grid_mapper': [<built-in function callable>, None], 'lambda_ai': [<sklearn.utils._param_validation.Interval object>], 'prejudice_filename': [<class 'str'>], 'reliability_floor': [<sklearn.utils._param_validation.Interval object>], 'sigma_ai_floor': [<sklearn.utils._param_validation.Interval object>], 'verbose': ['verbose']}: Constraints for DUHIInverter2D.

pycsamt.ai.inversion.schema.DUHI_PREPARE_SCHEMA = {'ai_initialize': ['boolean'], 'ai_mean': ['array-like'], 'ai_std': ['array-like'], 'ai_x': ['array-like', None], 'ai_z': ['array-like', None], 'builder': [<class 'object'>], 'observation_reliability': ['array-like']}: Constraints for DUHIInverter2D.prepare.

pycsamt.ai.inversion.schema.DIMENSIONALITY_RELIABILITY_SCHEMA = {'beta_deg': ['array-like'], 'beta_scale_deg': [<sklearn.utils._param_validation.Interval object>], 'minimum': [<sklearn.utils._param_validation.Interval object>]}: Constraints for phase-tensor dimensionality reliability.

pycsamt.ai.inversion.schema.GRID_MAPPING_SCHEMA = {'grid': ['array-like'], 'mesh': [<class 'object'>], 'model': [<class 'object'>], 'x_coordinates': ['array-like', None], 'z_coordinates': ['array-like', None]}: Constraints for geometry-aware AI-to-Occam grid mapping.

pycsamt.ai.inversion.schema.OBSERVATION_RELIABILITY_SCHEMA = {'errors': ['array-like'], 'reliability': ['array-like'], 'reliability_floor': [<sklearn.utils._param_validation.Interval object>]}: Constraints for reliability-weighted observation errors.

pycsamt.ai.inversion.schema.RELIABILITY_COMBINATION_SCHEMA = {'dimensionality': ['array-like'], 'measurement': ['array-like'], 'minimum': [<sklearn.utils._param_validation.Interval object>]}: Constraints for combined observation reliability.