It is safest to install expertsystem within a virtual environment, so that all Python dependencies are contained within there. This is helpful in case something goes wrong with the dependencies: you can just trash the environment and recreate it. There are two options: Conda
<install/virtual-environment:Conda environment>
or Python's venv
<install/virtual-environment:Python venv>
.
Conda can be installed without administrator rights, see instructions on this page. Once installed, you can use the environment.yml file from the local repository <get-the-source-code>
to create a Conda environment for the expertsystem installation:
conda env create -f environment.yml
This command will take a few minutes, because all required dependencies will be installed in one go through pip. After Conda finishes creating the environment, you can activate it with as follows:
conda activate tw
You need to have the tw
environment activated whenever you want to run expertsystem.
Alternatively, you can use Python's venv. All you have to do, is navigate into local repository <get-the-source-code>
and run:
python3 -m venv ./venv
This creates a folder called venv
where all Python packages will be contained. You first have to activate the environment, and will have to do so whenever you want to run those Python packages.
source ./venv/bin/activate
Now you can safely install the required Python requirements through pip:
pip install -r requirements.txt
That's it, now you're all set to install expertsystem <install>
!