Original ONNX Forward TE Curve Verification Pipeline Curve Plotter

Plot TE Curve Verification Pipeline curves from the recovered original paper ONNX forward bank.

class scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.HardcodedOnnxTarget(target_kind, harmonic_order, family_name, model_path, session, input_name)[source]

Bases: object

One hardcoded paper-original ONNX target.

Parameters:

  • target_kind (str)

  • harmonic_order (int)

  • family_name (str)

  • model_path (Path)

  • session (onnxruntime.InferenceSession)

  • input_name (str)

target_kind: str
harmonic_order: int
family_name: str
model_path: Path
session: onnxruntime.InferenceSession
input_name: str
scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.parse_command_line_arguments()[source]

Parse command-line arguments.

Return type:

Namespace

scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.resolve_onnx_variant_configuration(variant_id)[source]

Resolve one named original-ONNX variant configuration.

Parameters:

variant_id (str)

Return type:

dict[str, Any]

scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.load_hardcoded_onnx_target_list(target_configuration_list=None)[source]

Load one hardcoded original paper forward ONNX target list.

Parameters:

target_configuration_list (Sequence[tuple[str, int, str, str]] | None)

Return type:

list[HardcodedOnnxTarget]

scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.build_forward_track2_curve_record_list(config_path)[source]

Build the canonical forward TE Curve Verification Pipeline curve records.

Parameters:

config_path (Path)

Return type:

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

scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.predict_curve_target_dictionary(curve_record, target_list)[source]

Predict the 19 harmonic targets for one curve operating point.

Parameters:
Return type:

dict[tuple[str, int], float]

scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.reconstruct_curve_from_prediction_dictionary(angular_position_deg, selected_harmonic_list, prediction_dictionary)[source]

Reconstruct one TE curve from amplitude and phase predictions.

Parameters:

  • angular_position_deg (ndarray)

  • selected_harmonic_list (list[int])

  • prediction_dictionary (dict[tuple[str, int], float])

Return type:

ndarray

scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.save_or_show_curve_plot(curve_record, predicted_curve_deg, metric_dictionary, plot_path, show_plot)[source]

Save or show one measured-versus-predicted curve plot.

Parameters:

  • curve_record (HarmonicCurveRecord)

  • predicted_curve_deg (ndarray)

  • metric_dictionary (dict[str, float])

  • plot_path (Path)

  • show_plot (bool)

Return type:

None

scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.run_plotter(arguments)[source]

Run the hardcoded original-ONNX curve plotter.

Parameters:

arguments (Namespace)

Return type:

dict[str, Any]

scripts.paper_reimplementation.rcim_ml_compensation.reference_family_vs_feedforward.plot_original_onnx_fw_track2_curves.main()[source]

Run the command-line entry point.

Return type:

None