MLPo-BF/OA-Control: Classic and online-adaptive closed-loop control

With this release, we introduce classic and online-adaptive closed-loop control into MLPro. Release highlights are:

• MLPro-BF-Control: New sub-framework for classic closed-loop control
• MLPro-OA-Control: New sub-framework for online-adaptive closed-loop control
• RLPID: An online-adaptive PID controller auto-tuned by a reinforcement learning algorithm

We have furthermore cleaned up the entire API to avoid circular imports and reduce the overall import effort at program startup time. Please note that we made some incompatible changes, requiring a more restrictive and precise import of MLPro's classes. Particularly, the classes Log, Mode, PlotSettings (and a couple of further classes) are affected. These can now easily be imported from mlpro.bf. Please see the IDE documentation of mlpro.bf for further details. We apologize for any inconvenience.

What's Changed

MLPro-OA - Online Adaptivity

• BF/OA-Streams: Enhancements by @detlefarend in #1244
• Refact: MLPro-OA - API cleanup - Export lists all by @detlefarend in #1255
• BF/OA: Online adaptive closed-loop control by @detlefarend in #1053
• OA-Streams: Extensions of class OAStreamHelper by @detlefarend in #1256
• BF/OA: Online adaptive PID controllers by @detlefarend in #1055

Full Changelog: v2.0.3...v2.1.0

MLPro-OA: Bugfixes

What's Changed

MLPro-OA - Online Adaptivity

• BF/OA Streams: Bugfixes by @detlefarend in #1240

Documentation

• DOC: RSS Feed #1230 by @detlefarend in #1231
• DOC: RSS Feed #1230 by @detlefarend in #1232

Other Changes

• Rss feed by @detlefarend in #1233

Full Changelog: v2.0.2...v2.0.3

MLPro-OA: Bugfixes and enhancements

What's Changed

MLPro-OA - Online Adaptivity

• BF/OA: Bugfixes and enhancements on OADSP by @detlefarend in #1190

Other Changes

• Update Extension Hub by @mlpro-admin in #1184
• Update Extension Hub by @mlpro-admin in #1185
• Update Extension Hub by @mlpro-admin in #1218
• Update Extension Hub by @mlpro-admin in #1226

Full Changelog: v2.0.1...v2.0.2

MLPro-OA: Online machine learning (Hotfix)

In the release, we fixed a bug in the design pattern for cluster analysis. Furthermore, we ended MLPro's presence in Anaconda due to the low number of downloads and the administrative effort.

What's Changed

MLPro-OA - Online Adaptivity

• Hotfix v2.0.1 by @detlefarend in #1181

Full Changelog: v2.0.0...v2.0.1

MLPro-OA: Online machine learning

It's done! Our new sub-framework, MLPro-OA, for online machine learning is complete.

We are still working on the online documentation, but our new major journal paper MLPro 2.0 - Online machine learning in Python and our new ChatGPT MLPro are available to answer your questions!

The entire MLPro team wishes you lots of fun and success with MLPro-OA!

What's Changed

MLPro-GT - Game Theory

• GT: Gradient-based Learning in SbPGs by @steveyuwono in #1177

MLPro-OA - Online Adaptivity

• Final changes in v2.0.0 (10) by @detlefarend in #1168

Full Changelog: v1.9.8...v2.0.0

MLPro-BF: Fix for Pickling Compatibility with PyTorch 2.4+

This release addresses an issue with object serialization in PyTorch versions later than 2.4.0. To ensure compatibility, we have set the recurse=True parameter in the dill library, allowing objects to be broken down into smaller components for proper pickling. This fix has been tested with MLPro-Int-SB3 and is now merged into the main repository. With this update, serialization in MLPro should work across different PyTorch versions.

What's Changed

MLPro-BF - Basic Functions

• Bug SB3: Unit Test Failed due to Pickling Serialized Objects by @steveyuwono in #1165

Full Changelog: v1.9.7...v1.9.8

MLPro-OA: Pre-release for early adopters (6)

What's Changed

MLPro-BF - Basic Functions

• Final changes in v2.0.0 (9) by @detlefarend in #1157

Full Changelog: v1.9.6...v1.9.7

MLPro-OA: Pre-release for early adopters (5)

We are approaching the next main release 2.0.0. Technically, the target functionality has been achieved. The online documentation is not yet finished, but it is worth a first look: a new homepage, an AI-generated podcast and our chatGPT 'MLPRO', which already knows much about our project. There is already a lot to discover...

What's Changed

MLPro-OA - Online Adaptivity

• Final changes in v2.0.0 (5) by @detlefarend in #1136
• Final changes in v2.0.0 (7) by @detlefarend in #1143
• Final changes in v2.0.0 (8) by @detlefarend in #1147

Documentation

• Final changes in v2.0.0 (4) by @detlefarend in #1102
• Final changes in v2.0.0 (6) by @detlefarend in #1142
• Final changes v1.9.5 for MLPro-RL Documentation by @steveyuwono in #1148

Full Changelog: v1.9.4...v1.9.5

MLPro-OA: Pre-release for early adopters (4)

On our way to establishing online machine learning in MLPro, we did some last-minute refactorings...

What's Changed

MLPro-OA - Online Adaptivity

• Final changes in v2.0.0 (3) by @detlefarend in #1100

Full Changelog: v1.9.3...v1.9.4

MLPro-OA: Pre-release for early adopters (4)

In our fourth pre-release for early adopters of our new sub-framework MLPro-OA, we fixed a bug in the context of group anomalies as part of our upcoming systematics for anomaly detection.

What's Changed

MLPro-OA - Online Adaptivity

• Final changes in v2.0.0 (2) by @detlefarend in #1095

Full Changelog: v1.9.2...v1.9.3