Getting Started with Snowpark Gets Faster with New Python Worksheets in Snowflake

Want to try Snowpark?

As a Python developer, you’ve always had a lot of flexibility in terms of which IDE/notebook you can use to get started; all you need to do is install the Snowpark client API and connect to your Snowflake account. But to make it even faster to get started and remove the need to setup Python environments or install open-source libraries on your client, you can now try Snowpark from within Snowsight, Snowflake’s web UI, using a new feature called Python worksheets, now available in public preview.

With Python worksheets you can write, run, and deploy Snowpark for Python programs from the browser. No installation, no setup, and no need to setup a connection from your development environment to Snowflake. Just open Snowsight, create a Python worksheet, and start experimenting with Snowpark.

Below are some highlights of Python worksheets and how they help you stay productive:

Python Language support: Syntax highlighting, type-sensitive autocomplete for keywords and most supported Anaconda libraries, and diagnostics such as undeclared variables or invalid method usage.

Use open-source libraries available through the Anaconda integration or libraries from a stage: At the top of the worksheet, using the Packages drop-down menu, you can select packages from our Anaconda channel using the UI or import packages from a named stage.

Debug your script using Py Output: Use show() or print() commands and see the results right inside Snowsight.

Inspect Snowpark DataFrames: Use return to see a preview of Snowpark DataFrame output in tabular form. You can also return a string or variant value.

Deploy Python code: Turn your Snowpark program into a Python stored procedure that can be scheduled for automated processing using the new Deploy option.

Getting Started

To help you get started with Snowpark, today we are launching a new Snowpark trial experience that has a pre-populated Python worksheet with code samples that will help you go from zero to deploying your first set of DataFrame operations on a dataset with ~300K records all from within Snowflake.

Get started with a trial account that will walk you through your first data transformations.