Skip to content

Latest commit

 

History

History
41 lines (25 loc) · 853 Bytes

README.md

File metadata and controls

41 lines (25 loc) · 853 Bytes

nn-hyperplane-bounds

Create an environment, for instance in anaconda do:

conda create -n bounds python=3

and activate your environment:

conda activate bounds

You need to install numpy, pytorch==1.10.1 and torchvision==0.11.2. Find more information to install pytorch here.

To run the experiments on MNIST and CIFAR, run:

python Experiments.py

Experiments will run for a large set of combinations of parameters, consider selecting the parameters within the code to run a specific subset of experiments.

Cite us

The paper is available from arxiv here.

@article{jimeno2022bounds,
  title={Hyperplane bounds for neural feature mappings},
  author={Jimeno Yepes, Antonio},
  journal={arXiv preprint arXiv:2201.05799},
  year={2022}
}