Installation¶

This section provides instructions on how to install nessy.

Prerequisites¶

Before you can install the Wind Data Engineering Framework, you need to meet the following prerequisites:

To use the framework to its full extent, you need a Spark session
Some features of the framework may not work in Fabric, due to limitations of the platform
The environment must have Python version 3.11 or 3.12 installed

Installation from the initions PyPI repository¶

Installing nessy from the public initions PyPI repository is easy. You can install the package using the following code:

pip install cloe_nessy # specify the version if needed

Warning

In Databricks make sure to restart the python kernel after the installation. This is required, because nessy installs some dependencies in other versions than the default ones on the Databricks clusters.

dbutils.library.restartPython()

or

%restart_python

In Fabric this is handled automatically.

Installation from Source¶

It can be advantageous to install nessy from source (e.g. a Databricks repository), if you are actively developing the framework or if you want to use a version that is not yet available in the PyPI repository.

This can be done by cloning the repository and installing the package from the local source:

pip install /path/to/nessy # e.g. /Workspace/Repos/user@domain/cloe-nessy-py

Mind, that you need to restart the python kernel after the installation, see above.