DropBlock2d¶
- class
dragon.vm.torch.nn.
DropBlock2d
(
p=0.1,
block_size=7,
inplace=False
)[source]¶ Set the blocks to zero randomly. [Ghiasi et.al, 2018].
The DropBlock function is defined as:
\[\text{DropBlock}(x_{ijk}) = x_{ijk} * (r_{ik} \sim \mathcal{B}(1, 1 - \gamma)) \\ \quad \\ \text{where}\quad \gamma = \frac{p}{\text{block\_size}^{n}} \frac{\text{feat\_size}^{n}}{(\text{feat\_size} - \text{block\_size} + 1)^n} \]Examples:
x = torch.ones(1, 3, 4, 4) m = torch.nn.DropBlock2d(block_size=3) y = m(x)
See also
__init__¶
DropBlock2d.
__init__
(
p=0.1,
block_size=7,
inplace=False
)[source]¶Create a
DropBlock2d
module.- Parameters:
- p (float, optional, default=0.1) – The dropping ratio.
- block_size (int, optional, default=7) – The size of a spatial block.
- inplace (bool, optional, default=False) – Whether to do the operation in-place.