A simple hook based implementation of "Deep Networks with Stochastic Depth" for torchvision resnets.
Project description
Stochastic Depth with PyTorch Hooks
A simple hook based implementation of Deep Networks with Stochastic Depth for torchvision resnets.
Example
import torch
import torchvision.models as models
resnet = models.resnet152(pretrained=False)
resnet.train()
from stochdepth import uniform
hooks = uniform(resnet, p=0.2)
x = torch.zeros((8, 3, 224, 224), dtype=torch.float32)
y = resnet(x)
# remove hooks
for h in hooks:
h.remove()
from stochdepth import resnet_linear
hooks = resnet_linear(resnet)
y = resnet(x)
# remove hooks
for h in hooks:
h.remove()
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
stochdepth-0.5.0.tar.gz
(3.0 kB
view hashes)