Demonstration project that uses Databricks Connect to run run local code on a Databricks cluster
Make sure you have Databricks Connect for Python installed
Example code from the link above
python3.11 -m venv ./venv
pip3 show pyspark
pip3 uninstall pyspark
# This example uses Databricks Runtime 14.3 LTS
pip3 install --upgrade "databricks-connect==14.3.*"
You should have a table to query in your Databricks environment. On your cluster compute you can add both spark variables and environment variable to parameterize