otx.algorithms.action.task#

Task of OTX Video Recognition.

Classes

OTXActionTask(task_environment[, output_path])

Task class for OTX action.
class otx.algorithms.action.task.OTXActionTask(task_environment: TaskEnvironment, output_path: str | None = None)[source]#

Bases: OTXTask, ABC

Task class for OTX action.

evaluate(output_resultset: ResultSetEntity, evaluation_metric: str | None = None)[source]#

Evaluate function of OTX Action Task.

explain(dataset: DatasetEntity, explain_parameters: ExplainParameters | None = None) → DatasetEntity[source]#

Main explain function of OTX Task.

export(export_type: ExportType, output_model: ModelEntity, precision: ModelPrecision = ModelPrecision.FP32, dump_features: bool = True)[source]#

Export function of OTX Task.

infer(dataset: DatasetEntity, inference_parameters: InferenceParameters | None = None) → DatasetEntity[source]#

Main infer function.

save_model(output_model: ModelEntity)[source]#

Save best model weights in ActionTrainTask.

train(dataset: DatasetEntity, output_model: ModelEntity, train_parameters: TrainParameters | None = None, seed: int | None = None, deterministic: bool = False)[source]#

Train function for OTX action task.

Actual training is processed by _train_model fucntion