Installing julearn
Requirements
julearn requires the following packages:
Python >= 3.7
Running the examples requires:
Depending on the installation method, this packages might be installed automatically.
Installing
Note
Currently, there is a big difference betweek the latest release and developer version. Therefore, it might make more sense for you to use the developer version until we release the next julearn version.
There are different ways to install julearn:
Install the Latest release. This is the most suitable approach for most end users.
Install the Latest Development Version. This version will have the latest features. However, it is still under development and not yet officially released. Some features might still change before the next stable release.
Install from Local git repository (for developers). This is mostly suitable for developers that want to have the latest version and yet edit the code.
Either way, we strongly recommend using virtual environments:
Latest release
We have packaged julearn and published it in PyPi, so you can just install it with pip.
pip install -U julearn
Latest Development Version
First, make sure that you have all the dependencies installed:
pip install -U scikit-learn pandas numpy
OR:
conda install scikit-learn pandas numpy
Then, install julearn from TestPypi
pip install --index-url https://test.pypi.org/simple/ -U julearn --pre
Local git repository (for developers)
First, make sure that you have all the dependencies installed:
pip install -U scikit-learn pandas numpy
OR:
conda install scikit-learn pandas numpy
Then, clone julearn Github repository in a folder of your choice:
git clone https://github.com/juaml/julearn.git
Install development mode requirements:
cd julearn
pip install -r dev-requirements.txt
Finally, install in development mode:
python setup.py develop
Note
Every time that you run setup.py develop
, the version is going to
be automatically set based on the git history. Nevertheless, this change
should not be committed (changes to _version.py
). Running git stash
at this point will forget the local changes to _version.py
.