/pruning

Implementation of post-hoc weight pruning for sparse neural networks.

Primary LanguageJupyter Notebook

pruning

To build and run:

Ensure that you have TensorFlow installed on your machine.

Run commands:
pip3 install -e .
cd pruning

To train model from scratch and evaluate pruned networks, run:
python run.py run

To load pre-trained model and evaluate pruned networks, run:
python run.py run --load-trained

To see results and analysis, open experiments.ipynb.