mediapipe_model_maker.model_util.get_steps_per_epoch

Gets the estimated training steps per epoch.

Main aliases

mediapipe_model_maker.face_stylizer.face_stylizer.face_stylizer_options.model_opt.loss_functions.model_util.get_steps_per_epoch, mediapipe_model_maker.face_stylizer.face_stylizer.loss_functions.model_util.get_steps_per_epoch, mediapipe_model_maker.face_stylizer.face_stylizer.model_opt.loss_functions.model_util.get_steps_per_epoch, mediapipe_model_maker.face_stylizer.face_stylizer.model_util.get_steps_per_epoch, mediapipe_model_maker.face_stylizer.face_stylizer_options.model_opt.loss_functions.model_util.get_steps_per_epoch, mediapipe_model_maker.face_stylizer.model_options.loss_functions.model_util.get_steps_per_epoch

 
mediapipe_model_maker.model_util.get_steps_per_epoch(
    steps_per_epoch: Optional[int] = None,
    batch_size: Optional[int] = None,
    train_data: Optional[mediapipe_model_maker.model_util.dataset.Dataset] = None
) -> int
  1. If steps_per_epoch is set, returns steps_per_epoch directly.
  2. Else if we can get the length of training data successfully, returns train_data_length // batch_size.

steps_per_epoch int, training steps per epoch.
batch_size int, batch size.
train_data training data.

Estimated training steps per epoch.

ValueError if both steps_per_epoch and train_data are not set.