Harmonic-Wise Support

This page documents the repository-owned helper functions used by the paper-faithful RCIM harmonic-wise reimplementation branch.

Support utilities for the harmonic-wise offline comparison pipeline.

scripts.paper_reimplementation.rcim_ml_compensation.harmonic_wise_support.load_harmonic_pipeline_config(config_path)

Load one harmonic-wise pipeline configuration file.

Parameters:

config_path (str | Path) – Harmonic-wise YAML configuration path.

Returns:

Parsed training-configuration dictionary.

Return type:

dict[str, Any]

scripts.paper_reimplementation.rcim_ml_compensation.harmonic_wise_support.resolve_selected_harmonic_list(training_config)

Resolve and validate the configured harmonic list.

Parameters:

training_config (dict[str, Any])

Return type:

list[int]

scripts.paper_reimplementation.rcim_ml_compensation.harmonic_wise_support.build_split_record_bundle(training_config)

Build train, validation, and test curve bundles for the pipeline.

Parameters:

training_config (dict[str, Any])

Return type:

tuple[dict[str, list[HarmonicCurveRecord]], dict[str, int], dict[str, int], Path]

scripts.paper_reimplementation.rcim_ml_compensation.harmonic_wise_support.fit_harmonic_target_models(feature_matrix, target_matrix, target_name_list, model_configuration)

Fit one estimator per harmonic coefficient target.

Parameters:

• feature_matrix (ndarray)

• target_matrix (ndarray)

• target_name_list (list[str])

• model_configuration (dict[str, Any])

Return type:

dict[str, Any]

scripts.paper_reimplementation.rcim_ml_compensation.harmonic_wise_support.evaluate_curve_record_split(curve_record_list, harmonic_model_dictionary, selected_harmonic_list, target_name_list, percentage_error_denominator, engineered_feature_term_list)

Evaluate one split by reconstructing TE curves from predicted harmonics.

Parameters:

• curve_record_list (list[HarmonicCurveRecord])

• harmonic_model_dictionary (dict[str, Any])

• selected_harmonic_list (list[int])

• target_name_list (list[str])

• percentage_error_denominator (str)

• engineered_feature_term_list (list[str])

Return type:

dict[str, Any]

scripts.paper_reimplementation.rcim_ml_compensation.harmonic_wise_support.run_motion_profile_playback(training_config, harmonic_model_dictionary, selected_harmonic_list, target_name_list)

Run offline Robot and Cycloidal style playback from the predicted harmonics.

Parameters:

• training_config (dict[str, Any])

• harmonic_model_dictionary (dict[str, Any])

• selected_harmonic_list (list[int])

• target_name_list (list[str])

Return type:

dict[str, Any]