LIONoso, a non-profit tool for Machine Learning and Intelligent Optimization

LIONoso 2.1, developed by LIONlab for non-profit research and academic use, offers flexible training and experimentation for researchers who know what they are doing and want freedom in their ML experimentation.

By Roberto Battiti, Nov 2014.

LIONosoLIONoso 2.1, developed by LIONlab, is a comprehensive Machine Learning and Intelligent Optimization tool exclusively for non-profit research and academic use.

Some checks are omitted for complete freedom in training and experimentation, as appropriate in academic and research environments.

It is appropriate for researchers and practitioners who "know what they are doing" and like to push the envelope of the LION way.

In addition to the ML and optimization tools, advanced users adopt it for orchestrating heterogeneous components. Orchestration deals with automating processes, with the arrangement, coordination, and management of complex software components connecting data, experiments, simulators, models, decisions.

In particular

  • Support for the R language is now available. R is a language for statistical computing developed at Bell Laboratories (formerly AT&T, now Lucent Technologies) by John Chambers and colleagues. Commanding R scripts with inputs from LIONoso or receiving data from R calculations can lead to powerful combinations.
  • Python and additional scripting languages can be used in a seamless manner.

Many use cases and samples from the LION community pages take advantage of this software's powerful and simple interface.

Orchestration of tools occurs via the Parametric table component, which allows advanced users to add support for additional languages with limited effort.

LIONoso also supports the "Learning from Data" course by Caltech professor Yaser Abu-Mostafa, including sample exercises.

You can download and run LIONoso 2.1 at