nn.functional.dropout2d

lucid.nn.functional.dropout2d(input_: Tensor, p: float = 0.5, training: bool = True) Tensor

The dropout2d function randomly zeroes entire channels of the input tensor with a probability p during training for 2D data.

Function Signature

def dropout2d(input_: Tensor, p: float = 0.5, training: bool = True) -> Tensor

Parameters

  • input_ (Tensor):

    The input tensor of shape (N, C, H, W), where N is the batch size, C is the number of channels, H is the height, and W is the width.

  • p (float, optional):

    Probability of an element to be zeroed. Default: 0.5.

  • training (bool, optional):

    Apply dropout if True; do nothing if False. Default: True.

Returns

  • Tensor:

    A tensor of the same shape as the input, with entire channels randomly zeroed with probability p if training is True.

Examples

Applying dropout2d during training:

>>> input_ = Tensor([[[[1.0, 2.0], [3.0, 4.0]]]])  # Shape: (1, 1, 2, 2)
>>> out = F.dropout2d(input_, p=0.5, training=True)
>>> print(out)
# Example output: Tensor([[[[0.0, 0.0], [0.0, 0.0]]]])