public virtual void TestGetSummaryForInstance(GraphicalModel[] dataset, ConcatVector weights)
{
LogLikelihoodDifferentiableFunction fn = new LogLikelihoodDifferentiableFunction();
foreach (GraphicalModel model in dataset)
{
double goldLogLikelihood = LogLikelihood(model, (ConcatVector)weights);
ConcatVector goldGradient = DefinitionOfDerivative(model, (ConcatVector)weights);
ConcatVector gradient = new ConcatVector(0);
double logLikelihood = fn.GetSummaryForInstance(model, (ConcatVector)weights, gradient);
NUnit.Framework.Assert.AreEqual(logLikelihood, Math.Max(1.0e-3, goldLogLikelihood * 1.0e-2), goldLogLikelihood);
// Our check for gradient similarity involves distance between endpoints of vectors, instead of elementwise
// similarity, b/c it can be controlled as a percentage
ConcatVector difference = goldGradient.DeepClone();
difference.AddVectorInPlace(gradient, -1);
double distance = Math.Sqrt(difference.DotProduct(difference));
// The tolerance here is pretty large, since the gold gradient is computed approximately
// 5% still tells us whether everything is working or not though
if (distance > 5.0e-2)
{
System.Console.Error.WriteLine("Definitional and calculated gradient differ!");
System.Console.Error.WriteLine("Definition approx: " + goldGradient);
System.Console.Error.WriteLine("Calculated: " + gradient);
}
NUnit.Framework.Assert.AreEqual(distance, 5.0e-2, 0.0);
}
}
public virtual void TestOptimizeLogLikelihood(AbstractBatchOptimizer optimizer, GraphicalModel[] dataset, ConcatVector initialWeights, double l2regularization)
{
AbstractDifferentiableFunction <GraphicalModel> ll = new LogLikelihoodDifferentiableFunction();
ConcatVector finalWeights = optimizer.Optimize((GraphicalModel[])dataset, ll, (ConcatVector)initialWeights, (double)l2regularization, 1.0e-9, true);
System.Console.Error.WriteLine("Finished optimizing");
double logLikelihood = GetValueSum((GraphicalModel[])dataset, finalWeights, ll, (double)l2regularization);
// Check in a whole bunch of random directions really nearby that there is no nearby point with a higher log
// likelihood
Random r = new Random(42);
for (int i = 0; i < 1000; i++)
{
int size = finalWeights.GetNumberOfComponents();
ConcatVector randomDirection = new ConcatVector(size);
for (int j = 0; j < size; j++)
{
double[] dense = new double[finalWeights.IsComponentSparse(j) ? finalWeights.GetSparseIndex(j) + 1 : finalWeights.GetDenseComponent(j).Length];
for (int k = 0; k < dense.Length; k++)
{
dense[k] = (r.NextDouble() - 0.5) * 1.0e-3;
}
randomDirection.SetDenseComponent(j, dense);
}
ConcatVector randomPerturbation = finalWeights.DeepClone();
randomPerturbation.AddVectorInPlace(randomDirection, 1.0);
double randomPerturbedLogLikelihood = GetValueSum((GraphicalModel[])dataset, randomPerturbation, ll, (double)l2regularization);
// Check that we're within a very small margin of error (around 3 decimal places) of the randomly
// discovered value
if (logLikelihood < randomPerturbedLogLikelihood - (1.0e-3 * Math.Max(1.0, Math.Abs(logLikelihood))))
{
System.Console.Error.WriteLine("Thought optimal point was: " + logLikelihood);
System.Console.Error.WriteLine("Discovered better point: " + randomPerturbedLogLikelihood);
}
NUnit.Framework.Assert.IsTrue(logLikelihood >= randomPerturbedLogLikelihood - (1.0e-3 * Math.Max(1.0, Math.Abs(logLikelihood))));
}
}