Installation

spectral_galerkin_heat uses uv to manage its environment.

Prerequisites

Python: 3.12 or higher.
uv: A fast Python package installer and resolver.

For GPU support (optional):

CUDA Toolkit: Compatible with CuPy.
NVIDIA Driver: Matching your CUDA version.

Installing the Project

Clone the repository:

git clone https://github.com/theo-andrieux/spectral_galerkin_heat.git
cd spectral_galerkin_heat

Sync the environment (CPU):
```
uv sync
```
This creates an isolated virtual environment and installs the dependencies defined in pyproject.toml.
(Optional) Sync for GPU: For a compatible NVIDIA GPU:
```
uv sync --extra gpu
```

Running Simulations

The project is driven by YAML configuration files.

# General invocation
uv run python simulations/main.py <path_to_config.yaml>

# Run the quickstart example (finishes in seconds)
uv run python simulations/main.py simulations/examples/01_quickstart.yaml

See Examples for the full walkthrough of the examples.

Building the Documentation

Requires make (available via your system package manager: apt install make, brew install make, etc.).

uv run make -C docs html

The HTML output is written to docs/_build/html/. Open docs/_build/html/index.html in a browser to view it.