Need help with pytorch-domain-adaptation?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

376 Stars 85 Forks MIT License 5 Commits 6 Opened issues


A collection of implementations of adversarial domain adaptation algorithms

Services available


Need anything else?

Contributors list

# 135,593
5 commits

Pytorch Adversarial Domain Adaptation

A collection of implementations of adversarial unsupervised domain adaptation algorithms.

Domain adaptation

The goal of domain adaptation is to transfer the knowledge of a model to a different but related data distribution. The model is trained on a source dataset and applied to a target dataset (usually unlabeled). In this case, the model is trained on regular MNIST images, but we want to get good performance on MNIST with random color (without any labels).

In adversarial domain adaptation, this problem is usually solved by training an auxiliary model called the domain discriminator. The goal of this model is to classify examples as coming from the source or target distribution. The original classifier will then try to maximize the loss of the domain discriminator, comparable to the GAN training procedure.

Implemented papers

Paper: Unsupervised Domain Adaptation by Backpropagation, Ganin & Lemptsky (2014)
Description: Negates the gradient of the discriminator for the feature extractor to train both networks simultaneously.

Paper: Adversarial Discriminative Domain Adaptation, Tzeng et al. (2017)
Description: Adapts the weights of a classifier pretrained on source data to produce similar features on the target data.

Paper: Wasserstein Distance Guided Representation Learning, Shen et al. (2017)
Description: Uses a domain critic to minimize the Wasserstein Distance (with Gradient Penalty) between domains.



Accuracy on MNIST-M Parameters
Source only 0.33
RevGrad 0.74 default
ADDA 0.76 default
WDGRL 0.78

--k-clf 10 --wd-clf 0.1


  1. Download the BSDS500 dataset and extract it somewhere. Point the
    variable in
    to this location.
  2. In a Python 3.6 environment, run:
    $ conda install pytorch torchvision numpy -c pytorch
    $ pip install tqdm opencv-python
  3. Train a model on the source dataset with
    $ python
  4. Choose an algorithm and pass it the pretrained network, for example:
    $ python trained_models/

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.