DropPath/drop_path 是一种正则化手段，其效果是将深度学习模型中的多分支结构随机”删除“，python中实现如下所示：

def drop_path(x, drop_prob: float = 0., training: bool = False):
    if drop_prob == 0. or not training:
        return x
    keep_prob = 1 - drop_prob
    shape = (x.shape[0],) + (1,) * (x.ndim - 1)  
    random_tensor = keep_prob + torch.rand(shape, dtype=x.dtype, device=x.device)
    random_tensor.floor_()  # binarize
    output = x.div(keep_prob) * random_tensor
    return output


class DropPath(nn.Module):
    def __init__(self, drop_prob=None):
        super(DropPath, self).__init__()
        self.drop_prob = drop_prob

    def forward(self, x):
        return drop_path(x, self.drop_prob, self.training)

调用如下：

self.drop_path = DropPath(drop_prob) if drop_prob > 0. else nn.Identity()

x = x + self.drop_path(self.token_mixer(self.norm1(x)))
x = x + self.drop_path(self.mlp(self.norm2(x)))

看起来似乎有点迷茫，这怎么就随机删除了分支呢
实验如下：

import torch

drop_prob = 0.2
keep_prob = 1 - drop_prob
x = torch.randn(4, 3, 2, 2)
shape = (x.shape[0],) + (1,) * (x.ndim - 1)
random_tensor = keep_prob + torch.rand(shape, dtype=x.dtype, device=x.device)
random_tensor.floor_()
output = x.div(keep_prob) * random_tensor

输出：

x.size():[4,3,2,2]
x：
tensor([[[[ 1.3833, -0.3703],
          [-0.4608,  0.6955]],
         [[ 0.8306,  0.6882],
          [ 2.2375,  1.6158]],
         [[-0.7108,  1.0498],
          [ 0.6783,  1.5673]]],

        [[[-0.0258, -1.7539],
          [-2.0789, -0.9648]],
         [[ 0.8598,  0.9351],
          [-0.3405,  0.0070]],
         [[ 0.3069, -1.5878],
          [-1.1333, -0.5932]]],

        [[[ 1.0379,  0.6277],
          [ 0.0153, -0.4764]],
         [[ 1.0115, -0.0271],
          [ 1.6610, -0.2410]],
         [[ 0.0681, -2.0821],
          [ 0.6137,  0.1157]]],

        [[[ 0.5350, -2.8424],
          [ 0.6648, -1.6652]],
         [[ 0.0122,  0.3389],
          [-1.1071, -0.6179]],
         [[-0.1843, -1.3026],
          [-0.3247,  0.3710]]]])

random_tensor.size():[4, 1, 1, 1]
random_tensor：
tensor([[[[0.]]],
        [[[1.]]],
        [[[1.]]],
        [[[1.]]]])

output.size():[4,3,2,2]
output：
tensor([[[[ 0.0000, -0.0000],
          [-0.0000,  0.0000]],
         [[ 0.0000,  0.0000],
          [ 0.0000,  0.0000]],
         [[-0.0000,  0.0000],
          [ 0.0000,  0.0000]]],

        [[[-0.0322, -2.1924],
          [-2.5986, -1.2060]],
         [[ 1.0748,  1.1689],
          [-0.4256,  0.0088]],
         [[ 0.3836, -1.9848],
          [-1.4166, -0.7415]]],

        [[[ 1.2974,  0.7846],
          [ 0.0192, -0.5955]],
         [[ 1.2644, -0.0339],
          [ 2.0762, -0.3012]],
         [[ 0.0851, -2.6027],
          [ 0.7671,  0.1446]]],

        [[[ 0.6687, -3.5530],
          [ 0.8310, -2.0815]],
         [[ 0.0152,  0.4236],
          [-1.3839, -0.7723]],
         [[-0.2303, -1.6282],
          [-0.4059,  0.4638]]]])

  random_tensor作为是否保留分支的直接置0项，若drop_path的概率设为0.2，random_tensor中的数有0.2的概率为0，而output中被保留概率为0.8。

  结合drop_path的调用，若x为输入的张量，其通道为[B,C,H,W]，那么drop_path的含义为在一个Batch_size中，随机有drop_prob的样本，不经过主干，而直接由分支进行恒等映射。