The hyperparameters for training face stylizers.
Inherits From: BaseHParams
mediapipe_model_maker.face_stylizer.HParams(
learning_rate: float = 0.0008,
batch_size: int = 4,
epochs: int = 100,
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 = ''
)
Attributes |
learning_rate
|
Learning rate to use for gradient descent training.
|
batch_size
|
Batch size for training.
|
epochs
|
Number of training iterations.
|
beta_1
|
beta_1 used in tf.keras.optimizers.Adam.
|
beta_2
|
beta_2 used in tf.keras.optimizers.Adam.
|
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
)
Class Variables |
batch_size
|
4
|
beta_1
|
0.0
|
beta_2
|
0.99
|
class_weights
|
None
|
distribution_strategy
|
'off'
|
epochs
|
100
|
export_dir
|
'/tmpfs/tmp/tmpnt_h4p9w'
|
learning_rate
|
0.0008
|
num_gpus
|
0
|
repeat
|
False
|
shuffle
|
False
|
steps_per_epoch
|
None
|
tpu
|
''
|