A Python library to (bene)fit Gaussian Process Emulators.
Status: Actively developed
Type: Personal project
Development years: 2020 - Present
Authors: stelong, ShadowTemplate
- Python3 (>=3.10)
- virtualenv (optional)
- Pull the source code from the project repository:
git clone https://github.com/stelong/GPErks.git
cd GPErks/
- (optional) Create a Python3 virtual environment:
python3 -m venv .venv
source .venv/bin/activate
pip install --upgrade pip setuptools wheel
- (optional) Install the PyTorch package first in order to satisfy your installation requirements. This is especially relevant for CUDA installations, where the specific version needs to match your machine's NVIDIA driver version. For example, for CUDA 11.8:
pip install torch --index-url https://download.pytorch.org/whl/cu118
Note: for CPU-only installations, you can skip step (3). For CUDA installations, please check the PyTorch website to choose the correct version.
- Install GPErks:
pip install .
- (optional) Install Jupyter Notebook to be able to run the example notebooks:
pip install notebook
The full documentation is under construction. For the moment, please refer to the example notebooks and tutorial while noting that this is based on the old GPErks v0.1.0 API (it will still provide a good overview on all package functionalities). The notebooks are also available as plain Python scripts.
stelong and ShadowTemplate are the only maintainers. Any contribution is welcome!
This project is licensed under the MIT license. Please refer to the LICENSE file for more details.
This README.md complies with this project template. Feel free to adopt it and reuse it.