M-LOOP 3.0.0

• Documentation for neural net learner.
• Several improvements to visualization features.
• Documentation fixes.
• Support for TensorFlow 2.

M-LOOP 2.2.0

• Made a variety of small bug fixes with visualisations.
• The find local minima option has been removed, as it is rarely used, slow, and of little value when attempting to learn about the cost landscape.
• Neural nets have now been added as a machine learning option. Functionality is still undocumented, will is available to interested users. Full documentation will be in the 3.0 series release.

M-LOOP 2.1.1

Fixed some issues with:
*Halting conditions
*Appropriately processing bad runs.

Documentation has been updated.

M-LOOP 2.1.0

M-LOOP now some new options and has a different default behaviour.

• A differential evolution algorithm has been added. This can now be used on its own or as a trainer for the gaussian process (GP). It is now the default trainer for the GP.
• Added a shell based interface.
• Improved the documentation. Added a tutorial for using M-LOOP as a python API and also improved the installation instructions. New features have also been appropriately documented.

M-LOOP 2.0.3

Added support for python 2.

M-LOOP 2.0

M-LOOP has been significantly upgraded.

The new version allows you to automatically optimize your quantum experiments or numerical simulations faster and easier than before. New features include:

• Multithreaded execution with no downtime between runs.
• Visualizations of your optimization process and cost landscape.
• Documentation with easy to follow installation instructions.

Go to http://m-loop.readthedocs.io/ and follow the installation instructions to get the latest version.

Pre-release of M-LOOP

More complete documentation. Testing new documentation hosting.

Pre-release of M-LOOP

Preliminary release of M-LOOP 2, now with threading, automated testing and documentation.