mediapipe_model_maker.gesture_recognizer.HParams

The hyperparameters for training gesture recognizer.

Inherits From: BaseHParams

mediapipe_model_maker.gesture_recognizer.HParams(
    learning_rate: float = 0.001,
    batch_size: int = 2,
    epochs: int = 10,
    steps_per_epoch: Optional[int] = None,
    class_weights: Optional[Mapping[int, float]] = None,
    shuffle: bool = False,
    repeat: bool = False,
    export_dir: str = tempfile.mkdtemp(),
    distribution_strategy: str = 'off',
    num_gpus: int = 0,
    tpu: str = '',
    lr_decay: float = 0.99,
    gamma: int = 2
)

learning_rate Learning rate to use for gradient descent training.
batch_size Batch size for training.
epochs Number of training iterations over the dataset.
lr_decay Learning rate decay to use for gradient descent training.
gamma Gamma parameter for focal loss.
steps_per_epoch Dataclass field
class_weights Dataclass field
shuffle Dataclass field
repeat Dataclass field
export_dir Dataclass field
distribution_strategy Dataclass field
num_gpus Dataclass field
tpu Dataclass field

Methods

get_strategy

View source

get_strategy()

__eq__

__eq__(
    other
)

batch_size 2
class_weights None
distribution_strategy 'off'
epochs 10
export_dir '/tmpfs/tmp/tmpnt_h4p9w'
gamma 2
learning_rate 0.001
lr_decay 0.99
num_gpus 0
repeat False
shuffle False
steps_per_epoch None
tpu ''