drop_block3d¶
- torchvision.ops.drop_block3d(input: Tensor, p: float, block_size: int, inplace: bool = False, eps: float = 1e-06, training: bool = True) Tensor[原始碼]¶
實現了 DropBlock3d,來自 “DropBlock: A regularization method for convolutional networks” <https://arxiv.org/abs/1810.12890>。
- 引數:
- 返回:
Dropblock 後隨機歸零的張量。
- 返回型別:
Tensor[N, C, D, H, W]