Skip to main content

Implementation of DropBlock: A regularization method for convolutional networks in PyTorch.

Project description



Implementation of DropBlock: A regularization method for convolutional networks in PyTorch.


Deep neural networks often work well when they are over-parameterized and trained with a massive amount of noise and regularization, such as weight decay and dropout. Although dropout is widely used as a regularization technique for fully connected layers, it is often less effective for convolutional layers. This lack of success of dropout for convolutional layers is perhaps due to the fact that activation units in convolutional layers are spatially correlated so information can still flow through convolutional networks despite dropout. Thus a structured form of dropout is needed to regularize convolutional networks. In this paper, we introduce DropBlock, a form of structured dropout, where units in a contiguous region of a feature map are dropped together. We found that applying DropBlock in skip connections in addition to the convolution layers increases the accuracy. Also, gradually increasing number of dropped units during training leads to better accuracy and more robust to hyperparameter choices. Extensive experiments show that DropBlock works better than dropout in regularizing convolutional networks. On ImageNet classification, ResNet-50 architecture with DropBlock achieves 78.13% accuracy, which is more than 1.6% improvement on the baseline. On COCO detection, DropBlock improves Average Precision of RetinaNet from 36.8% to 38.4%.


Install directly from PyPI:

pip install dropblock

or the bleeding edge version from github:

pip install git+

NOTE: Implementation and tests were done in Python 3.6, if you have problems with other versions of python please open an issue.


For 2D inputs (DropBlock2D):

import torch
from dropblock import DropBlock2D

# (bsize, n_feats, height, width)
x = torch.rand(100, 10, 16, 16)

drop_block = DropBlock2D(block_size=3, drop_prob=0.3)
regularized_x = drop_block(x)

For 3D inputs (DropBlock3D):

import torch
from dropblock import DropBlock3D

# (bsize, n_feats, depth, height, width)
x = torch.rand(100, 10, 16, 16, 16)

drop_block = DropBlock3D(block_size=3, drop_prob=0.3)
regularized_x = drop_block(x)

Scheduled Dropblock:

import torch
from dropblock import DropBlock2D, LinearScheduler

# (bsize, n_feats, depth, height, width)
loader = [torch.rand(20, 10, 16, 16) for _ in range(10)]

drop_block = LinearScheduler(
                DropBlock2D(block_size=3, drop_prob=0.),

probs = []
for x in loader:
    regularized_x = drop_block(x)


The drop probabilities will be:

>>> [0.    , 0.0625, 0.125 , 0.1875, 0.25, 0.25, 0.25, 0.25, 0.25, 0.25]

The user should include the step() call at the start of the batch loop, or at the the start of a model's forward call.

Check examples/ to see an implementation example.

Implementation details

We use drop_prob instead of keep_prob as a matter of preference, and to keep the argument consistent with pytorch's dropout. Regardless, everything else should work similarly to what is described in the paper.


Refer to


[Ghiasi et al., 2018] DropBlock: A regularization method for convolutional networks


  • [x] Scheduled DropBlock
  • [x] Get benchmark numbers
  • [x] Extend the concept for 3D images

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
dropblock-0.3.0-py3-none-any.whl (5.4 kB) Copy SHA256 hash SHA256 Wheel py3
dropblock-0.3.0.tar.gz (4.1 kB) Copy SHA256 hash SHA256 Source None

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN SignalFx SignalFx Supporter DigiCert DigiCert EV certificate StatusPage StatusPage Status page