double GetTestQuadraticLoss_bn ( tester_bn*  test,   node_bn*  node )

Returns the "quadratic loss" of node under the tests previously performed.

The "quadratic loss" (also known as "Brier score") is defined as:  MOAC [1 - 2 * pc + sum[j=1 to n] (pj2)]

  where MOAC stands for the mean (average) over all cases (i.e., all cases
  for which the case file provides a value for the node in question),

  where pc is the probability predicted for the state that turns out to be correct,

  where pj is the probability predicted for state j, and

  where n is the number of states of the node.

Values for quadratic loss vary from 0 to 2, with 0 being a perfect score.

node is required to have been in the test_nodes list originally passed to NewNetTester_bn.


Versions 2.08 and later have this function.

See also:

GetTestConfusion_bn    Get elements of the confusion matrix
GetTestErrorRate_bn    Get the fraction of test cases for which the prediction failed
GetTestLogLoss_bn    Get the "logarithmic loss" score of the test
NewNetTester_bn    Construct the tester_bn object