Installation#
This guide will help you install the necessary software to run the library. The library has been tested on Linux and MacOS, and may work on Windows as well. We also provide a Docker file, which can be used to run the library in a container or for development.
Prerequisites#
The library requires Python 3.11 or later.
You need git to clone the repository.
To log experiment data, you will need a Weights & Biases account.
To run experiments with OpenAI models, you need an OpenAI API key. You can get one by signing up at OpenAI. Note that in general the use of the OpenAI API is not free.
Installation Steps#
Clone the repository:
git clone https://github.com/SamAdamDay/neural-interactive-proofs.git --branch v2.1.1 --depth 1
git clone https://github.com/SamAdamDay/neural-interactive-proofs.git
Alternatively, you may wish to fork the repository and clone your fork.
Change to the repository directory:
cd neural-interactive-proofs
Install the requirements:
uv sync --no-dev
uv sync
uv sync --extra lm-server
pip -m venv .venv source .venv/bin/activate pip install wheel pip install torch==2.7.0 torchvision==0.22.0 --index-url https://download.pytorch.org/whl/cu118 pip install -e --group dev .
Log in to Weights & Biases:
wandb login
Copy the template secrets file:
cp .env.template .env
Edit the
.env
file and fill in the necessary information for your use case. The comments in the file should guide you on what to fill in.
Next Steps#
See the Running Experiments guide for information on how to run experiments.