Graph Windows

Context

• A Graph Window is a container for network graphs, which represent qualitative and quantitative information regarding a domain:

  • Qualitative refers to the structure of the network graph, i.e., nodes and arcs.
  • Quantitative refers to the Conditional Probability Tables that define probabilistic relationships and any dataset that can be associated with the Graph Window.

• Each Graph Window is a distinct entity and is associated with a corresponding file in BayesiaLab’s XBL format.

  • The filename associated with each Graph Window is displayed in the upper-left corner of the Graph Window frame.
  • In BayesiaLab, you can keep multiple Graph Windows open simultaneously and work on them in parallel.

• You can switch between Graph Windows and arrange them on your screen using the Graph Bar at the bottom of the Graph Window.

Illustration

Usage

A Graph Window can be in two modes:

• Modeling Mode F4
• Validation Mode F5

as indicated by their corresponding buttons in the lower-left corner.

Modeling Mode F4

• In Modeling Mode F4, you can conduct all modeling activities, such as learning and editing network graphs.
• In this mode, only the Graph Panel is visible and accessible inside the Graph Window, i.e., it fills the Graph Window entirely.

Validation Mode F5

• In Validation Mode F5, both the Graph Panel and the Monitor Panel are visible.
• Model validation, analysis, and simulation are performed exclusively in this mode.

Switching Between Modeling Mode & Validation Mode

• You can switch between these modes by clicking on the respective buttons or using keyboard shortcuts:
  • Modeling Mode F4
  • Validation Mode F5

Workflow Animation