frequency_mix_style

Name	Description
`FrequencyMixStyle2D`	2D frequency mix style augmentation

Name

Description

FrequencyMixStyle2D

2D frequency mix style augmentation

FrequencyMixStyle2D(probability: float = 0.5, alpha: float = 1.0, epsilon: float = 1e-06, **kwargs={})

Apply frequency mix style augmentation to the 2D input.

Example:

    x = np.random.rand(4, 4, 3)
    lyr = FrequencyMixStyle2D(probability=1.0, alpha=1.0)
    y = lyr(x, training=True)

Parameters

Parameters of FrequencyMixStyle2D
Name	Type	Default	Description
`probability`	`float`	0.5	Probability of applying the augmentation.
`alpha`	`float`	1.0	Mixup alpha value.
`epsilon`	`float`	1e-06	Epsilon value for numerical stability.

probability: float = probability

alpha: float = alpha

epsilon: float = epsilon

get_random_transformations(input_shape: tuple[int, int, int]) -> dict

Generate noise distortion tensor

Parameters

Parameters of get_random_transformations
Name	Type	Default	Description
`input_shape`	`tuple[int, ...]`	Required	Input shape.

Returns

Returns of get_random_transformations
Value	Type	Description
`dict`	`dict`	Dictionary containing the noise tensor.

apply_mixstyle(x: keras.KerasTensor, lmda: keras.KerasTensor, perm: keras.KerasTensor) -> keras.KerasTensor

Apply mixstyle augmentation

Parameters

Parameters of apply_mixstyle
Name	Type	Default	Description
`x`	`tf.Tensor`	Required	Input tensor
`lmda`	`tf.Tensor`	Required	Lambda tensor
`perm`	`tf.Tensor`	Required	Permutation tensor

Returns

Returns of apply_mixstyle
Type	Description
`keras.KerasTensor`	tf.Tensor: Augmented tensor

augment_samples(inputs) -> keras.KerasTensor

Augment samples

Parameters

Parameters of augment_samples
Name	Type	Default	Description
`inputs`	`tf.Tensor`	Required	Input tensor

Returns

Returns of augment_samples
Type	Description
`keras.KerasTensor`	tf.Tensor: Augmented tensor

get_config()

Frequency Mix Style Layer API