sakura.model_controllers.extractor_controller.ExtractorController.save_checkpoint

ExtractorController.save_checkpoint(save_config=False, save_model_arch=False)

Save the current training state of the extractor.

Serialize current configurations, epoch data, loss weights, model state, and optimizer state for resuming training or evaluating the model later.

Parameters:

• save_config (bool, optional) – Whether to include configuration settings in the checkpoint, defaults to False

• save_model_arch (bool, optional) – Whether to include the model architecture in the checkpoint, defaults to False

Returns:

A dictionary containing the current state of the extractor

Return type:

dict