Skip to content

Latest commit

 

History

History
177 lines (107 loc) · 6.71 KB

CHANGELOG.md

File metadata and controls

177 lines (107 loc) · 6.71 KB

v0.6.2

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Changes:

  • Fixed some bugs reported by @gunjanthesystem

Release Date: 07/03/2017 Further Information: Project Web Page,JavaDoc

v0.6.1

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Changes:

  • Unified loading streams names
  • Fixed some bugs

Release Date: 03/01/2017 Further Information: Project Web Page,JavaDoc

v0.6.0

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Changes:

  • Added sparklink module implementing the integration with Apache Spark. More information here.
  • Fluent pattern in latent-variable-models
  • Predefined model implementing the concept drift detection
  • Fixed some bugs

Release Date: 14/10/2016 Further Information: Project Web Page,JavaDoc

v0.6.0-alpha

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Changes:

  • Added sparklink module implementing the integration with Apache Spark. More information here.
  • Fixed some bugs

Release Date: 14/09/2016 Further Information: Project Web Page,JavaDoc

v0.5.1

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Changes:

  • Fixed some bugs

Release Date: 15/07/2016 Further Information: Project Web Page,JavaDoc

v0.5.0

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Added functionalities:

  • Support to Flink for distributed learning of probabilistic models.
  • Support for Latent Dirichlet Allocation Models

Release Date: 06/07/2016 Further Information: Project Web Page,JavaDoc

v0.4.3

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Added functionalities:

  • Bugs fixed
  • Link to the Weka

Minor changes:

  • Module standardmodels has been renamed as latent-variable-models

Release Date: 01/06/2016 Further Information: Project Web Page, JavaDoc

v0.4.2

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Added functionalities:

  • A wide range of latent variable models coded in the toolbox as a proof-of-concept of the flexibility of our toolbox.

Latent Variable Models

Release Date: 02/05/2016 Further Information: Project Web Page, JavaDoc

v0.4.1

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Added Functionalities:

  • Support for multi-core parallel Bayesian learning using Java streams.

Release Date: 31/12/2015 Further Information: Deliverable 4.4, JavaDoc

v0.4

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Added Functionalities:

  • Support for approximate inference in dynamic Bayesian networks through the Factored Frontier algorithm.
  • Support for MAP and MPE inference in static Bayesian networks.
  • Link with MOA software

Release Date: 30/11/2015 Further Information: Deliverable 3.3

v0.3

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning with probabilistic graphical models from local and distributed (streaming) data.

Added Functionalities:

  • Support for Bayesian parameter learning in both static and dynamic Bayesian networks.
  • Support for scalable Importance sampling for performing probabilistic queries.
  • Link to Hugin

Release Date: 31/06/2015 Further Information: Deliverable 3.2

v0.2

This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning of both static and dynamic Bayesian networks from streaming data.

Added Functionalities:

  • Support for representing dynamic Bayesian networks.
  • Support for loading data sets with dynamic data instances.

Release Date: 31/03/2015 Further Information: Deliverable 2.3

v0.1

This is first release of the toolbox. This toolbox aims to offers a collection of scalable and parallel algorithms for inference and learning of both static and dynamic Bayesian networks from streaming data.

Functionalities:

  • Support for representing static Bayesian networks.
  • Support for loading streaming data sets.

Release Date: 31/12/2014 Further Information: Deliverable 4.1