Python API for generating adapted and unique neighbourhoods for searching for adversarial examples.
Project description
Python API for generating adapted and unique neighbourhoods for searching for adversarial examples
Installation & usage
This work is released on PyPi. Installation, therefore, is as simple as installing the package with pip:
python3 -m pip install adaptive-neighbourhoods
At this point, you're free to start generating neighbourhoods for your own dataset:
from adaptive_neighbourhoods import epsilon_expand
neighbourhoods = epsilon_expand(
x, # your input data
y) # the integer encoded labels for your data
Move information on the variable parameters and general guidance on using this package can be found at: https://jaypmorgan.github.io/adaptive-neighbourhoods/
Contributing
All contributions and feedback are welcome!
There are three main remote mirrors used for hosting this project. If you would like to contribute, please submit an issue/pull-request/patch-request to any of these mirrors:
- Github: https://github.com/jaypmorgan/adaptive-neighbourhoods
- Gitlab: https://gitlab.com/jaymorgan/adaptive-neighbourhoods
- Source Hut: https://git.sr.ht/~jaymorgan/adaptive-neighbourhoods
Citing this work
If you use this work in your research, please consider referencing our article using the following bibtex entry:
@article{DBLP:journals/corr/abs-2101-09108,
author = {Jay Morgan and
Adeline Paiement and
Arno Pauly and
Monika Seisenberger},
title = {Adaptive Neighbourhoods for the Discovery of Adversarial Examples},
journal = {CoRR},
volume = {abs/2101.09108},
year = {2021},
url = {https://arxiv.org/abs/2101.09108},
eprinttype = {arXiv},
eprint = {2101.09108},
timestamp = {Sat, 30 Jan 2021 18:02:51 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2101-09108.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for adaptive_neighbourhoods-0.0.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | e32a50c028bf8445e5e0fda1f896803930dc723381a3707aeefc31aa3f431273 |
|
MD5 | 163fe675560426885d1dbabc15175882 |
|
BLAKE2b-256 | cb9430de35767f8bec3026ba35153223820a36982f9f3f8f7d79d61f709aa6f5 |
Hashes for adaptive_neighbourhoods-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | ce94c52df2fd47a60b62742f9897fb28ed9bb087b107483e98d73b3992f4e775 |
|
MD5 | e4c414e69e9381d126a05cda842655ba |
|
BLAKE2b-256 | 48325136b23750181b2bacc40b933355a6cfe756cb96a1381509a8018980eda0 |