
See recommendation for installing PyTorch for MacOS users above.

Git clone https: // /facebook/ax.git -depth 1 Pip3 install git+https: // /cornellius-gp/gpytorch.git When contributing to Ax, we recommend cloning the repository and installing all optional dependencies: # bleeding edge versions of GPyTorch + BoTorch are recommended If storing Ax experiments via SQLAlchemy in MySQL or SQLite: pip install 'git+ #egg=ax-platform' Development Install it first: pip install 'git+ #egg=ax-platform' To use Ax with a notebook environment, you will need Jupyter. Pip3 install git+https: // /pytorch/botorch.git We therefore recommend installing those from Git as well: pip3 install git+https: // /cornellius-gp/gpytorch.git See also the recommendation for installing PyTorch for MacOS users above.Īt times, the bleeding edge for Ax can depend on bleeding edge versions of BoTorch (or GPyTorch). You can install the latest (bleeding edge) version from GitHub: pip install 'git+ #egg=ax-platform' If you want to store the experiments in MySQL, you will need SQLAlchemy: pip3 install SQLAlchemy Please consult the PyTorch installation instructions above. If you need CUDA on MacOS, you will need to build PyTorch from source. This will result in up to an order-of-magnitude speed-up for Bayesian optimization, whereas installing PyTorch from pip does not link against MKL.

Installing from Anaconda will link against MKL (a library that optimizes mathematical computation for Intel processors). However, we recommend you install PyTorch manually before installing Ax, using the Anaconda package manager. Recommendation for MacOS users: PyTorch is a required dependency of BoTorch, and can be automatically installed via pip. If you're using a Unix-based OS, you can use which pip3 to check. Note: Make sure the pip3 being used to install ax-platform is actually the one from the newly created Conda environment. Installation will use Python wheels from PyPI, available for OSX, Linux, and Windows. We recommend installing Ax via pip (even if using Conda environment): conda install pytorch torchvision -c pytorch # OSX only (details below)
