Configure Model Validation Settings

The Data tab of the Property Panel provides options for configuring the datasets used in the Model Validation components. The Weight tab, when available, provides options for configuring weighting parameters used in calculations.

Up to three data sets can be selected for comparison with the sample point data, which are configured using the Block 1, Block 2 and Block 3 tabs.

  • When the Model Validation component is added to point (assay) data, the selected datasets are compared with the sample point data.
  • When the Model Validation component is added to a block model, the selected datasets are compared directly with each other.

The Weight tab allows applying weighting to all Model Validation charts and statistics. It is only applicable when the Model Validation component is inserted below a point data component.

Important: When selecting a block model for comparison, the block data component must have the same domain filters as the point data component. However, the block data component can also have additional filters, not present on the point data.

  1. Select the Model Validation component in the Project Tree.
  2. Select the Data tab.
  3. Select the Block 1, Block 2 or Block 3 tab.
  4. Configure the following options as required.
    • Source – All available block model data sources that have been imported into Supervisor.
    • Assay – All assay columns in the selected Source. Only available once a Source has been selected.
    • Data – Displays the currently selected block data component. This value is set automatically when an Assay is selected. Read-only.
    • Density – Columns available for use as a block density weighting. Can be overwritten with a manually specified value.
    • XInc – Columns available for use as the block size in X direction. Can be overwritten with a manually specified value.
    • YInc
    • ZInc
    • Colour – Block data colour to display on model validation trend plots.
    • Clear – Clear all fields in the current tab.

    5. Note: After selecting a Source and then Assay for comparison, Supervisor attempts to automatically discover a data component within the block data with matching domain filters. An error notification displays if no matching data can be found. In this case, re-group and/or filter the data to ensure there is a matching data component.

    When the Model Validation component is added to point (assay) data, the Weight tab allows weighting to be applied to all Model Validation charts and statistics.

  5. Select the Weight tab.
  6. Choose between the following weighting options:

    • No weight variable – No additional weighting is applied to the point data. All samples contribute equally to the Model Validation calculations.

    • Use declustering – Applies weights defined in the Declustering component. If declustering weights are available in the parent node, this option is automatically selected by default.

    • Weight variable selection – Any available numerical variable can be selected from the dropdown list and used as a weighting factor in the Model Validation calculations. When a weight variable is selected, the name of the selected variable is displayed in parentheses next to the Model Validation component in the Project Tree.

    7. When the Model Validation component is added to a block model, weighting is not available.