truncation_tolerance

Convergence tolerance used to identify the (reduced) dimension of the rotation matrix

Specification

  • Alias: None

  • Arguments: None

  • JSON argument key: None

  • Default: 0.9

Description

This convergence tolerance controls the dimension of the rotation matrix after the truncation. The metric controlled by this convergence tolerance is a quantity varying between [0,1]. If a more aggressive dimension truncation is desired, the default value (0.8) can be reduced to a lower value.

Examples

The following block

model
  id_model = 'SUBSPACE'
  adapted_basis
    truth_model_pointer = 'FULLSPACE'
    sparse_grid_level = 1
    truncation_tolerance = 0.9
{
  "model": [
    {
      "adapted_basis": {
        "id_model": "SUBSPACE",
        "truth_model_pointer": {
          "pointer": "FULLSPACE",
          "basis_control": {
            "sparse_grid_level": 1
          }
        },
        "truncation_tolerance": 0.9
      }
    }
  ]
}

increases the default value (0.8) for the truncation_tolerance to 0.9 to obtain a rotation matrix and, consequently, a reduced model that more closely approximates the full dimensional model.