Installation

• Requirements before install
• Installing oppy (for user)
• Installing oppy (for developers)
• Jupyter
• Requirements before work with the docs

Requirements before install

We strongly recommend to work with virtual environments. These offer the advantage that separate projects can be isolated form each other and overlaps can be avoided. In the following we use Conda, which is an open-source environment and package manager. For more information see the documentation page from conda about virtual environments. Alternatively you can also work with pip and virtual environments. For more information see the website.

Now we create a new virtual environment 'oppy' in the first step, run:

conda create -n oppy -y

Instead of 'oppy' you can use any other name for the environment. With

conda create -n oppy python=3.9 -y

you create an environment that contains Python 3.9. To verify all environments, use

conda info --envs

In the next step we need to activate the new environment with

conda activate oppy

Our setup installs the packages numpy, scipy and matplotlib. For sure you can install them on your own with

conda install numpy scipy matplotlib

Furthermore, we recommend using an integrated development environment (IDE). Common IDEs are e.g. Spyder and PyCharm. To install Spyder use

conda install spyder -y

Installing oppy (for users)

To install oppy we need pip. The command

conda list

shows if pip is already installed. Otherwise one can install pip with

conda install -c anaconda pip

The command

pip install git+ssh://git@gitlab.inf.uni-konstanz.de/ag-volkwein/oppy.git

will install the latest version of oppy from the master-branch in your environment. This step requires access from the developers. For access request or more questions contact .

Jupyter notebooks

In our tutorials we work with Jupyter Notebooks, for more details see Jupyter. These can also be found on the git repository. If you want to use them, run

conda install -c conda-forge notebook

in the active environment to install. After this you can start it with

jupyter notebook

Installing oppy (for developers)

If you want to work with oppy as a developer, make sure you have read the How To’s carefully and especially the How to use git for oppy while working with oppy, and you have already created your own branch based on the instructions. To install oppy we actually need pip. You can use

conda list

to check if pip is already installed. With

conda install -c anaconda pip

you can install pip. The command

pip install git+ssh://git@gitlab.inf.uni-konstanz.de/ag-volkwein/oppy.git@dev

will install the latest version of oppy from the dev-branch in your environment. Actually you need an access for the installation. Make sure you have one. For more questions contact us via .

In the next step you should run the tests in order to make sure that everything is set up correctly, see therefore How to Unittests in oppy.

Requirements before working with the docs

To work with the documentation we first need to install sphinx. Assume you have a virtual environment oppy_dev where you have already installed oppy and you use conda. If you use pip you can find the necessary packages on PyPi. Then run

conda install -c conda-forge sphinx

For the theme we use Read the Docs, therefore run

conda install -c conda-forge sphinx_rtd_theme

In order to run Jupyter Notebooks and Markdown files, we run

conda install -c conda-forge myst-parser
conda install -c conda-forge myst-nb

In the last step we need some extra packages, which are installed via

conda install -c anaconda sphinxcontrib
conda install -c conda-forge sphinxcontrib-napoleon
conda install -c conda-forge sphinxcontrib-bibtex
pip install sphinxcontrib-email

To work with Jupyter Notebooks you need to follow the instructions in Juypter notebooks. In the requirements.txt file you will also find a current list of packages which are required for oppy.