sfc-gh-ahkhan / snowpark-azure-ml-workshop

A step-by-step guide on how to use Snowflake Snowpark with Azure ML

Workshop: Data Engineering and Machine Learning Using Snowpark & Azure ML

SETUP

1. Open the terminal and clone the workshop repository: git clone https://github.com/filanthropic/snowpark-azure-ml-workshop

2. Login to Azure ML Studio and select an existing workspace or create a new one: https://portal.azure.com/#home

3. From the menu on the left, click on Compute and create a new compute instance. Give it a name. For the instance size, a small memory intensive instance would suffice for this lab.

4. Switch over to Notebooks from the left menu, click on the (+) button and select Upload folder to upload the workshop code folder (snowpark-azure-ml-workshop) that you cloned in step 1.

   

5. Create a new notebook.

6. Once, the notebook comes up, make sure the compute instance you create in step 3 is selected (and is running). Make sure to select Python 3.8 - AzureML as the kernel.

   

7. Install all the dependencies in the kernel by pasting this command into the first cell: !pip install -r snowpark-azure-ml-workshop/requirements.txt

LAB 1: Data Engineering & Model Deployment Using Snowpark

8. Open Lab1_credit_fraud_detector.ipynb, select Python 3.8 - AzureML kernel and follow the instructions in the notebook.

LAB 2: End-to-End ML Using Snowpark

9. Open Lab2_1_snowpark_housing_data_ingest.ipynb, run all cells to ingest data first and then open Lab2_2_snowpark_end_to_end_ml.ipynb. Again, make sure to switch to the Python 3.8 - AzureML kernel before you start running these notebooks.