Grid

Purpose

Grid builds a Cartesian grid from per-variable bin counts and evaluates all grid points.

Full Sampling Section Keys

• Sampling.Method (required): must be Grid.
• Sampling.Variables (required, array):
• name (required, string)
• description (required, string)
• distribution.type (required, string)
• distribution.parameters (required, object)
• Runtime requirement: every variable must provide num (grid count).
• Runtime-safe parameter sets:
  • Flat: min, max, num
  • Log: min, max, num
  • Normal: mean, stddev, num
  • Log-Normal: mean, stddev, num
  • Logit: location, scale, num
• Sampling.LogLikelihood (required, array): {name, expression}
• Sampling.selection (optional in schema): currently not applied by Grid runtime.

Full Skeleton

Sampling:
  Method: "Grid"
  Variables:
    - name: p1
      description: parameter 1
      distribution:
        type: Flat
        parameters:
          min: 0.0
          max: 1.0
          num: 50
  LogLikelihood:
    - name: L_total
      expression: "-0.5*((obs-100.0)/10.0)^2"
  selection: "p1 > 0"

Example

Sampling:
  Method: "Grid"
  Variables:
    - name: xx
      description: eggbox x
      distribution:
        type: Flat
        parameters:
          min: 0.0
          max: 31.4159
          num: 80
    - name: yy
      description: eggbox y
      distribution:
        type: Flat
        parameters:
          min: 0.0
          max: 31.4159
          num: 80
  LogLikelihood:
    - name: L_z
      expression: "-0.5*((z-100.0)/10.0)^2"