PDG extends the previous work "Diversity Probe" (ACM MM2023) by incorporating the
Paper Link: https://dl.acm.org/doi/abs/10.1145/3581783.3612375
This paper appears in: IEEE Transactions on Multimedia (TMM)
- python=3.9.16
- torch==2.0.1
- torchvision==0.15.2
- munkres=1.1.4
- numpy==1.24.1
- opencv-python==4.7.0.72
- scikit-learn=1.2.2
- pandas==2.0.1
Note: You need to download the data if you wish to train your own model.
Download the digits dataset from this link[BaiDuYunDisk] and its extracted code: xcl3
. Please extract it inside the data
directory
cd data
unzip digits.zip
cd ..
Pretrained task model is available at this link[BaiDuYunDisk] and its extracted code:2mz0
. Download and extract it in the models_pth
directory.
In train.py
:
- Specify the output directory to save the results in
--dir
. - Turn on the evaluation in
--eval
- Run
python train.py --dir SAVE_DIR --eval True
We thank the following authors for releasing their source code, data and models: