mediapipe_model_maker.object_detector.HParams

The hyperparameters for training object detectors.

Inherits From: BaseHParams

mediapipe_model_maker.object_detector.HParams(
    learning_rate: float = 0.3,
    batch_size: int = 8,
    epochs: int = 30,
    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 = '',
    cosine_decay_epochs: Optional[int] = None,
    cosine_decay_alpha: float = 1.0
)

learning_rate Learning rate to use for gradient descent training.
batch_size Batch size for training.
epochs Number of training iterations over the dataset.
cosine_decay_epochs The number of epochs for cosine decay learning rate. See https://www.tensorflow.org/api_docs/python/tf/keras/optimizers/schedules/CosineDecay for more info.
cosine_decay_alpha The alpha value for cosine decay learning rate. See https://www.tensorflow.org/api_docs/python/tf/keras/optimizers/schedules/CosineDecay for more info.
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 8
class_weights None
cosine_decay_alpha 1.0
cosine_decay_epochs None
distribution_strategy 'off'
epochs 30
export_dir '/tmpfs/tmp/tmpnt_h4p9w'
learning_rate 0.3
num_gpus 0
repeat False
shuffle False
steps_per_epoch None
tpu ''