How to install pyodbc driver on MLflow deployment.

Question

My model is utilising pyodbc which requires odbc driver to work. I am deploying it using mlflow serving (on azure databricks). How can I ensure that ODBC driver is installed on resulting endpoint?

Answer 1

Hello @Mateusz Macias ,

Thanks for the question and using MS Q&A platform.

Are you using model serving v1 or v2?

Go to your experiments, and navigate the the experiment run of interest. Expand ">Artifacts" and check the model_packaged folder, depending on your environment you will see a conda.yaml or python_env.yaml. These files will contain your dependencies. Example below:

Hope this will help. Please let us know if any further queries.

------------------------------

• Please don't forget to click on or upvote button whenever the information provided helps you. Original posters help the community find answers faster by identifying the correct answer. Here is how
• Want a reminder to come back and check responses? Here is how to subscribe to a notification
• If you are interested in joining the VM program and help shape the future of Q&A: Here is jhow you can be part of Q&A Volunteer Moderators