Skip to content
MDP 3.6 is released!

After some years maintenance work, MDP development has started again at full throttle!

In this release, the last one to support Python 2, you will find a new online training mode,
many new nodes, and compatibility with the newest versions of numpy, scipy and scikit-learn.

Some highlights:

- Added online mode to enable use of MDP in reinforcement learning.
- New nodes and flows: OnlineNode, OnlineFlow, CircularFlow,
  OnlineCenteringNode, OnlineTimeDiffNode, OnlineFlowNode,
  CircularOnlineFlowNode, OnlineLayer, CloneOnlineLayer, SameInputOnlineLayer,
  MCANode, CCIPCANode, IncSFANode.
- Added SFA-based supervised learning, specifically graph based SFA nodes
  GSFANode and iGSFANode and hierarchical GSFA (HGSFA).
- Added solvers to SFA-node that are robust against rank deficit in the
  covariance matrix.
- Added a new family of expansion nodes, including Legendre, Hermite and
  Chebyshev polynomials.
- VartimeSFANode supports SFA on data with non-constant time increments
- Docs have moved to https://mdpdocs.readthedocs.io

More details in the CHANGES file!

If you are interested in MDP development, join us on the MDP mailing list:
https://mail.python.org/mm3/mailman3/lists/mdp-toolkit.python.org