- Train model on the source domain **A** (**s = 0**), we view the full source data as a test set.
```python
cd object/
python MHPL_source.py --trte full --da uda --output ckps/source/ --gpu_id 0 --dset office --max_epoch 100 --s 0 --t 1
```

- Adaptation to other target domains **D and W**, respectively
```python
python MHPL_target.py --beta 0.3 --ratio 0.05 --alpha 3.0 --da uda --output_src ckps/source/ --output ckps/target/ --gpu_id 0 --dset office --s 0 --t 1  
```

- Synthetic-to-real 
```python
cd object/
 python MHPL_source.py --trte full --output ckps/source/ --da uda --gpu_id 0 --dset VISDA-C --net resnet101 --lr 1e-3 --max_epoch 10 --s 0 --t 1
 python MHPL_target.py --beta 0.3 --ratio 0.05 --alpha 25.0 --da uda --dset VISDA-C --gpu_id 0 --s 0 --t 1 --output_src ckps/source/ --output ckps/target/ --net resnet101 --lr 1e-3
 ```