Skip to content

readme.md

Latest commit

 

History

History
30 lines (19 loc) · 773 Bytes

readme.md

File metadata and controls

30 lines (19 loc) · 773 Bytes

Code for paper "SF(DA)2: Source-free Domain Adaptation Through the Lens of Data Augmentation"

Packages

We conducted experiments on the following versions of pakages:

  • cudatoolkit == 11.3.1

  • pytorch == 1.10.0

  • python == 3.8

Data Preparation

  • Please download the VisDA dataset from the official website and place in the path ./data/visda-2017.

Source Pretrained Model

  • We used the source pretrained model parameters provided in the github repository of SHOT.

Adaptation

  • VisDA: sh visda.sh

Note: We also include log files (for both linux and windows) generated with three different seeds in ./log/{linux or windows} More code will be included in our Github repository.