Skip to content

pytorch implement for the paper Few-Shot Adversarial Domain Adaptation

Notifications You must be signed in to change notification settings

HX-idiot/FADA-Pytorch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FADA-Pytorch

pytorch implement for the NIPS 2017 paper Few-Shot Adversarial Domain Adaptation

Install dependencies

  • This code has been tested on Python 3.6. pytorch 0.4.
  • Install PyTorch, torchvision and SciPy: pip install -r requirements.txt

Set up the dataset

  • Run python main.pyand get MNIST and SVHN dataset in ./data/

Train the model

  • Run python main.py.
    • Step 1 refers to pretrain model using MNIST dataset
    • Setp 2 refers to pretrain DCD using G1,G2,G3,G4.
    • Step 3 refers to train g,h and DCD iteratively.

Evaluate

  • Using 7 shot in target domain , training step 3 after 30 epoches ,I got the same result 47% acc as shown in papers results

About

pytorch implement for the paper Few-Shot Adversarial Domain Adaptation

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages