GL2U is an open source python package for computing upper and lower probabilities for non-separately specified, non-binary credal nets. The framework of GL2U combines the local specification, the binarization and the loopy 2U inference process.

Requirements

To run GL2U, you should have Java runtime environment, release 6.0 or higher, and Python interpreter, version 2.5 or higher installed on your computer. The latest Java virtual machine can be download here, and Python interpreter can be found here.

Downloads

The latest GL2U package (version 0.94), including code, documentation and examples, can be downloaded here.

A brief description can be downloaded here.

The detailed description can be downloaded here. For more information, please refer to the paper on GL2U.

User Manual

To lunch GL2U, just go to the GL2U directory and type:

python gl2u.py

Then follow the command line instructions. For detailed explanation of command line argument, please refer to the documentation.

References

GL2U uses the 2UBayes package to perform the loopy 2U propagation. 2UBayes employs the same definitions and formats used in JavaBayes.

GL2U also use the LRS package for extracting vertices of polytopes from interval specifications of credal nets.

The executable files of 2UBayes and LRS can be found in extern folder.