s-chh / PyTorch-MDC

Unofficial PyTorch implementation of Maximum Domain Confusion loss for Unsupervised Domain Adaptation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

PyTorch-MDC

Unofficial PyTorch implementation of Maximum Domain Confusion loss from Simultaneous Deep Transfer Across Domains and Tasks for Unsupervised Domain Adaptation.



Results:
Dataset Source Only MDC
SVHN ⟶ MNIST 61.2 80.2
MNIST ⟶ USPS 81.8 96.2
USPS ⟶ MNIST 49.4 97.5
SyDigits ⟶ SVHN 87.9 91.5
SySigns ⟶ GTSRB 91.0 93.3

About

Unofficial PyTorch implementation of Maximum Domain Confusion loss for Unsupervised Domain Adaptation


Languages

Language:Python 99.2%Language:Shell 0.8%