https://github.com/anirbanbasu/fhenn
Fully Homomorphic Encryption and neural networks experiments
https://github.com/anirbanbasu/fhenn
fhe fully-homomorphic-encryption neural-networks
Last synced: 12 months ago
JSON representation
Fully Homomorphic Encryption and neural networks experiments
- Host: GitHub
- URL: https://github.com/anirbanbasu/fhenn
- Owner: anirbanbasu
- License: mit
- Created: 2023-10-13T13:49:56.000Z (over 2 years ago)
- Default Branch: master
- Last Pushed: 2024-12-31T14:00:19.000Z (over 1 year ago)
- Last Synced: 2025-06-01T23:53:38.702Z (about 1 year ago)
- Topics: fhe, fully-homomorphic-encryption, neural-networks
- Language: Python
- Homepage:
- Size: 136 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Fully Homomorphic Encryption (FHE) and neural networks
This repository contains haphazard experiments with applying fully homomorphic encryption (FHE) on neural networks.
To begin, you must have `poetry` installed. See [installation instructions](https://python-poetry.org/docs/#installation).
You will need [`gettext`](https://www.gnu.org/software/gettext/) on your system to support localisation. Once installed, assuming that you have `msgfmt` on your system, run `./format_messages.sh` to convert the PO files for localisation to their MO equivalents.
Then, run `poetry install` in the directory where you cloned this repository. This will automatically create a virtual environment. Switch to it by invoking `poetry shell`. Then, run `fhenn --help` to see the rather self-explanatory commands available. Alternatively, without switching to the Poetry shell, you can run `poetry run fhenn --help` for the same result.
Note that during the course of running the commands, e.g., training, datasets will be downloaded in the `data` directory relative to the repository root.