All changes are very well documented for each release [1]. The biggest changes I've seen were, as expected, upgrading from 0.x to 1.x which were mostly function

All changes are very well documented for each release [1]. The biggest changes I've seen were, as expected, upgrading from 0.x to 1.x which were mostly function attribute names changes that were made to comply with numpy's attribute names (a welcome change). They did provide tools that automatically convert 0.x code to 1.x so that was helpful. It is not necessary to upgrade if you're satisfied with the current version and I sure won't deploy a different version to production than training. Upgrading is mainly worth it if there are performance improvements or new features/tools that make a big impact (and those are the changes we're mostly interested in and look forward to every release). [1] https://github.com/tensorflow/tensorflow/releases https://github.com/tensorflow/tensorflow/releases