v0.1   an initial version that seems to be fairly stable but is missing functionality compared to the original versions

v0.2   New class functionality

v0.3   New class functionality

v0.4   New class functionality

v0.5   New class functionality: LP linear programming

v0.6   added new class: RelativeValueIteration

v0.7   class changes in PolicyIteration

v0.8   PolicyIteration class now stable Fixes to make the use of policy iteration now possible. This includes some unit tests for this class. It has been lightly tested for usefulness, buty much more testing needs to be done.

v0.9   stability improvements to ValueIterationGS the Gauss-Seidel value iteration algorithm is now in a useful state. The documentation still needs cleaning up and expanding but the functionality is there. The target for the next tagged version is to have the RelativeValueIteration class able to be used.

v0.10   RelvativeValueIteration is ready for use, however its documentation still needs cleaning up

v0.11   QLearning code is stable, the doc is also mostly all there

v0.12   PolicyIterationModified has been completed

v0.13   finite horizon algorithm fixed

v0.14   LP class is ready for use

v4.0alpha1   tagging first 4.0 alpha release“

v4.0a2   stability enhancements

v4.0a4

v4.0b1   tag first beta release

v4.0b2

v4.0b3