Installation

The following contains information on installing tsml-eval for users and developers with write access. Those who wish to contribute to tsml-eval without write access will need to create a fork, see the aeon and sklearn documentation on contributing and developer installation for guidance.

We recommend setting up a fresh virtual environment or the conda equivalent before installing tsml-eval. See the aeon guide for setup information.

Install from PyPi

The easiest way to install tsml-eval is using pip:

pip install tsml-eval

Some estimators require additional dependencies. You can install these individually as required, or install all of them using the all_extras extra dependency set:

pip install tsml-eval[all_extras]

Note

If this results in a "no matches found" error, it may be due to how your shell
handles special characters. Try surrounding the dependency portion with quotes i.e.

pip install tsml-eval"[all_extras]"

All extra dependency sets can be found in the pyproject.toml file [project.optional-dependencies] options.

To install a specific release version, specify the version number when installing:

pip install tsml-eval==0.1.0

pip install tsml-eval[all_extras]==0.1.0

Install from conda-forge

tsml-eval is also available on conda-forge.

conda create -n tsml-env -c conda-forge tsml-eval
conda activate tsml-env

Currently for conda installations, optional dependencies must be installed separately.

Install fixed dependency versions for a publication

tsml-eval publications contain a static_publication_reqs.txt file that lists the versions of all dependencies used to generate results at the time of the release.

To install the dependencies using this file, run:

pip install -r static_publication_reqs.txt

Install the latest in-development version from GitHub

The latest development version of tsml-eval can be installed directly from GitHub using pip:

pip install git+https://github.com/time-series-machine-learning/tsml-eval.git@main

The latest development version of dependencies can be installed this way, i.e. for aeon

pip install git+https://github.com/aeon-toolkit/aeon.git@main

If you have a different version of tsml-eval or a dependency installed, you must uninstall it first before installing the development version.

Install for developers with write access

To install tsml-eval for development, first clone the GitHub repository:

git clone https://github.com/time-series-machine-learning/tsml-eval.git

Then install the package in editable mode with developer dependencies:

pip install --editable .[dev]

We recommend setting up pre-commit hooks to automatically format code and check for common issues before committing:

pre-commit install