Data Lab Python Versions

Default Python Version Update (October 2024):

• The default Python version on our platform will transition from 3.8 to 3.11 in October 2024. This aligns with the Python ecosystem's focus on newer and more secure versions.

Compatibility Testing is Crucial: It's vital to thoroughly test your existing scripts with Python 3.11 to identify and address any potential compatibility issues. This proactive approach ensures a smooth transition and prevents disruptions in your workflow.

Overview

Seeq Data Lab now supports multiple Python versions, enhancing your development experience. Here's a detailed breakdown of the key points:

Multiple Versions of Python:

You now have the freedom to choose between Python 3.8 (current default) and the recently introduced Python 3.11. This lets you leverage the latest features and libraries offered by 3.11 while maintaining compatibility with your existing projects that might rely on 3.8.

Switch Versions Within Your Notebook:

Seeq Data Lab provides a user-friendly way to switch Python versions within your notebook. Simply navigate to the kernels dropdown menu situated in the top right corner of your notebook (1). This will open a dropdown that list the available Python versions to select from (2). The Python version you choose for a particular notebook cell is automatically persisted after execution. This eliminates the need to repeatedly select the version for each cell within the same notebook.

The Python version for the notebook is only saved after executing a cell and saving the notebook.

Guidance for Testing: Duplicate

• To prevent unintended changes to your original notebook, consider creating a duplicate version before testing with a different Python version. This safeguard ensures you have a pristine copy of your work in case any compatibility issues arise.

Switch Version Within Terminal

Since a terminal can only utilize a single Python environment you must activate your desired Python version. We’ve introduced a simple command, pyversion which empowers you to manage Python versions directly from the terminal. To switch to Python 3.11, simply execute the command pyversion 3.11. This ensures that subsequent pip commands (such as package installations using pip install or listing local packages with pip list) target the newly selected Python version (3.11 in this case). To switch back to Python 3.8 execute the command pyversion 3.8.

Seamless Integration with Add-on Tools:

The Python version you select for your notebook also governs the functionality of any Add-on Tools you deploy. This ensures compatibility and optimal performance between your chosen Python version and the Add-on Tools you utilize.

Pre-installed Packages for Python 3.11:

The new Python 3.11 environment will come with a minimal selection of pre-installed packages. These are packages that are required for SPy and Data Lab to operate effectively.

Addressing Missing Packages:

If you run your notebook and there are failed imports this is most likely due to packages not being installed. These issues can be addressed the following ways.

Using the Notebook

To leverage libraries not pre-installed in 3.11, simply run your notebook in a 3.11 kernel and employ the familiar pip install command from a cell to install the missing packages. This approach equips your 3.11 environment with the necessary libraries for your project. Please see Installing Python Modules for more information about installing Python Packages.

Using the Terminal

Open a terminal and activate Python 3.11 environment with pyversion 3.11. You can now proceed to install any missing packages using pip install <package>

Considerations for Packaged Add-ons

Data Lab Functions for Add-ons accept a python version as part of the configuration. Please see Data Lab Functions for more information.