Pytorch implementation for NIPS2021 paper Implicit Semantic Response Alignment for Partial Domain Adaptation.
- python == 3.6.8
- pytorch ==1.1.0
- orchvision == 0.3.0
- numpy, scipy, PIL, argparse, tqdm, pandas
The datasets are set up with the same data protocol as PADA. Please download Office31, Office-Home and ImageNet-Caltech datasets and change the path in the image list files (*.txt) in the './data/' directory.
Run the code for PDA on Office-Home for Task (A -> C)
python run_partial.py --s 0 --t 1 --dset office_home --net ResNet50 --cot_weight 1. --output run1 --gpu_id 0 --rcwt 1. --alwt 1. --fdim 512 --edim 256
This project is built on the open-source implementation BA3US