FeatureAlphaDropout

Randomly masks out entire channels (a channel is a feature map) given a batch of multi-channel input.

Abstract Signature:

FeatureAlphaDropout(input: Tensor, p: float = 0.5, training: bool = False, inplace: bool = False)

PyTorchJAX (Core)NumPyKerasTensorFlowApple MLXFlax NNXPaxML / Praxis

PyTorch

API: torch.feature_alpha_dropout

Strategy: Direct Mapping

Official Docs ↗

JAX (Core)

API: —

Strategy: Custom / Partial

NumPy

API: —

Strategy: Custom / Partial

Keras

API: keras.layers.AlphaDropout

Strategy: Direct Mapping

Official Docs ↗

TensorFlow

API: —

Strategy: Custom / Partial

Apple MLX

API: mlx.nn.AlphaDropout

Strategy: Direct Mapping

Official Docs ↗

Flax NNX

API: —

Strategy: Custom / Partial

PaxML / Praxis

API: —

Strategy: Custom / Partial