Local#

DeepHyper installation requires Python>=3.7 and <3.10. By default, only hyperparameter search features will be installed.

Conda environment#

This installation procedure shows you how to create your own Conda virtual environment and install DeepHyper in it.

Linux#

Install Miniconda, and create the dh environment:

$ conda create -n dh python=3.8 -y
$ conda activate dh
$ conda install gxx_linux-64 gcc_linux-64

Finally install DeepHyper in the previously created dh environment:

$ pip install pip --upgrade
$ pip install deephyper["analytics"]

MacOS#

Install Xcode command line tools:

xcode-select --install

Check you current platform:

python3 -c "import platform; print(platform.platform());"

x86_64#

If your architecture is x86_64 install Miniconda, and create the dh environment:

$ conda create -n dh python=3.8 -y
$ conda activate dh

Then install DeepHyper in the previously created dh environment:

$ pip install pip --upgrade
$ pip install deephyper["analytics"]

arm64#

If your architecture is arm64 download MiniForge then install it:

chmod +x ~/Downloads/Miniforge3-MacOSX-arm64.sh
sh ~/Downloads/Miniforge3-MacOSX-arm64.sh

After installing MiniForge clone the DeepHyper repo and install the package:

git clone https://github.com/deephyper/deephyper.git
cd deephyper/
conda env create -f install/environment.macOS.arm64.yml

Jupyter Notebooks#

To create a custom Jupyter kernel run the following from your activated Conda environment:

$ python -m ipykernel install --user --name deephyper --display-name "Python (deephyper)"

Now when you open a Jupyter notebook the Python (deephyper) kernel will be available.

Docker Image (CPU)#

A Docker image with DeepHyper is provided. Assuming Docker is installed on the system you are using you can access the image with the following commands:

$ docker pull ghcr.io/deephyper/deephyper:0.3.3
$ docker run -i -t ghcr.io/deephyper/deephyper:0.3.3 /bin/bash

Developer Installation#

Follow the Conda environment installation and replace pip install deephyper[analytics] by:

$ git clone https://github.com/deephyper/deephyper.git
$ cd deephyper/ && git checkout develop
$ pip install -e ".[dev,analytics]"

Local

Contents