Dynamic pruning for neural networks. Implemented in PyTorch. Supports Dynamic Pruning Rates(A pruning rate that changes based on model time) and dynamic pruning iterations. More information can be seen in the paper:
This will install the specific version of pyTorch used, including CUDA 11.3. The CUDA version could be higher, however the PyTorch implementation uses some experimnetal features which may have change din newer versions.
pip install torch==1.12.0+cu113 torchvision==0.13.0+cu113 torchaudio==0.12.0 --extra-index-url https://download.pytorch.org/whl/cu113
pip install torchdata==0.4.0
pip install torchtext==0.13.0
pip install torchviz
pip install torchsummary
pip install matplotlib
pip install pandas
pip install tqdm
pip install ipywidgets widgetsnbextension pandas-profiling
pip install scikit-learn
pip install torchmetrics
An example script is provided, which should contain all necessary instructions on how to use the work. Further the code presented is documented. Any questions can be submitted as an issue.