kitcar_ml.traffic_sign_detection package

Subpackages

• kitcar_ml.traffic_sign_detection.fasterrcnn package
  • Submodules
  • kitcar_ml.traffic_sign_detection.fasterrcnn.evaluate module
  • kitcar_ml.traffic_sign_detection.fasterrcnn.export module
  • kitcar_ml.traffic_sign_detection.fasterrcnn.inference module
  • kitcar_ml.traffic_sign_detection.fasterrcnn.model module
    • Model
      • Model.__get_prediction_indices()
      • Model.predict()
      • Model.DEFAULT_OPTIMIZER_KWARGS
      • Model.fit()
      • Model.save()
      • Model.load()
      • Model.export_to_onnx()
      • Model.half()
      • Model._abc_impl
  • kitcar_ml.traffic_sign_detection.fasterrcnn.train module
  • Module contents

Submodules

kitcar_ml.traffic_sign_detection.detection_model module

Classes:

DetectionModel()

class DetectionModel

Bases: ABC

Methods:

fit(data_loader, val_data_loader[, epochs])	Train the model on the given data_loader.
predict(images, **kwargs)	Take in a list of images and return predictions for object locations.
evaluate(data_loader[, evaluator_type])	Run evaluator on data loader.
export_to_onnx(output_file)	Export this model into a onnx format.
save(file)	Save the internal model weights to a file.
load(file)	Load a model from a .pth file containing the model weights.

Attributes:

_abc_impl

abstract fit(data_loader: DataLoader, val_data_loader: DataLoader, epochs: int = 10, **kwargs) → None

Train the model on the given data_loader.

abstract predict(images: List[ndarray], **kwargs) → List[Tuple[List[ndarray], List[str], List[float]]]

Take in a list of images and return predictions for object locations.

evaluate(data_loader: ~kitcar_ml.utils.data.data_loader.bbox_data_loader.BBoxDataLoader, evaluator_type: ~typing.Type[~kitcar_ml.utils.evaluation.evaluator.Evaluator] = <class 'kitcar_ml.utils.evaluation.simple_evaluator.SimpleEvaluator'>, **kwargs) → Evaluator

Run evaluator on data loader.

The evaluator_type can be used to run a different evaluator than SimpleEvaluator. :returns: The evaluator with results.

abstract export_to_onnx(output_file: str)

Export this model into a onnx format.

Parameters

output_file – Path to the output file

abstract save(file: str)

Save the internal model weights to a file.

Parameters

file – The name of the file. Should have a .pth file extension.

abstract classmethod load(file: str)

Load a model from a .pth file containing the model weights.

Parameters

file – The path to the .pth file containing the saved model.

Returns

The model loaded from the file.

_abc_impl = <_abc_data object>

Module contents

Traffic Sign Detection Network.