Watson Machine Learning is getting an upgrade and it’s time for you to move to Python 3.6.

In Watson Machine Learning, users can deploy and train machine learning models with different versions of Python runtimes. Due to a recent security vulnerability for multiple Python versions (including Python 3.5) and end-of-life schedule of Python 2.7, we decide to update Python versions to 3.6.8 and later for all of Watson Machine Learning deployment and training runtimes. We will also be deprecating Python 3.5 and Python 2.7 runtimes.

The new Python 3.6 environments do not only differ in the Python language version. Open source library versions for packages you may be using also may have changed. This might affect your ability to run the code without modification in the future. You might be required to make minor modifications upfront to ensure a smooth transition on August 30, 2019.

Here are the dates you need to know

• Python 3.6 runtimes available: July 23, 2019
• Deprecate older version announcement: July 23, 2019
• End of life for Python 3.5 and 2.7: August 30, 2019 

You can read more about working with Watson Machine Learning runtimes, including the new Python 3.6 runtimes, in our documentation.