This repo contains the implementation of DECUN and some useful files to reuse its building blocks.
-
data contains data samples for testing. The full data can be download from Here.
-
trained_models contains trained for testing.
-
test.py is the main test file to be run.
-
option.py contains the running option for test.py .
-
decovNet.py contains the DECUN network .
All required packages are found in requirements.txt.
- Creating the conda env for "DECUN"
conda create env -n "DECUN" python=3.9
- Activate "DECUN" conda env
conda activate DECUN
- Install PyTorch
pip install torch==1.12.1+cu116 torchvision==0.13.1+cu116 torchaudio==0.12.1 --extra-index-url https://download.pytorch.org/whl/cu116
- Install required library
pip install -r requirements.txt
For more running option can see in the option.py
python test.py
@ARTICLE{Yanan24, author = {Yanan Zhao and Yuelong Li and Haichuan Zhang and Vishal Monga and Yonina C. Eldar}, title = {Deep, convergent, unrolled half-quadratic splitting for image deconvolution}, journal={arXiv preprint arXiv:2402.12872}, year = {2024}, }