public void AbsoluteConvergenceConstructorTest()
{
var criteria = new AbsoluteConvergence(iterations: 10, tolerance: 0.1);
int progress = 1;
do
{
// Do some processing...
// Update current iteration information:
criteria.NewValue = 12345.6 / progress++;
} while (!criteria.HasConverged);
// The method will converge after reaching the
// maximum of 10 iterations with a final value
// of 1371.73:
int iterations = criteria.CurrentIteration; // 10
double value = criteria.OldValue; // 1371.7333333
Assert.AreEqual(10, criteria.CurrentIteration);
Assert.AreEqual(1371.7333333333333, criteria.OldValue);
}
public void AbsoluteConvergenceConstructorTest4()
{
var criteria = new AbsoluteConvergence(iterations: 0, tolerance: 1e-5, startValue: 1);
criteria.CurrentIteration = -2;
do
{
criteria.NewValue /= 10.0;
} while (!criteria.HasConverged);
Assert.AreEqual(4, criteria.CurrentIteration);
Assert.AreEqual(-5, Math.Log10(criteria.OldValue));
Assert.AreEqual(-6, Math.Log10(criteria.NewValue));
}
/// <summary>
/// Initializes a new instance of the <see cref="ResilientBackpropagation"/> class.
/// </summary>
///
/// <param name="parameters">The number of parameters in the function to be optimized.</param>
///
public ResilientBackpropagation(int parameters)
{
convergence = new AbsoluteConvergence();
solution = new double[parameters];
gradient = new double[parameters];
previousGradient = new double[parameters];
weightsUpdates = new double[parameters];
// Initialize steps
Reset(initialStep);
for (int i = 0; i < solution.Length; i++)
{
solution[i] = Accord.Math.Tools.Random.NextDouble() * 2.0 - 1.0;
}
}
/// <summary>
/// Initializes a new instance of the <see cref="BaseBaumWelchLearning"/> class.
/// </summary>
///
protected BaseBaumWelchLearning(IHiddenMarkovModel model)
{
this.convergence = new AbsoluteConvergence();
this.model = model;
}
/// <summary>
/// Creates a new <see cref="LowerBoundNewtonRaphson"/>.
/// </summary>
///
public LowerBoundNewtonRaphson()
{
convergence = new AbsoluteConvergence();
}
/// <summary>
/// Initializes a new instance of the <see cref="ResilientBackpropagation"/> class.
/// </summary>
///
/// <param name="parameters">The number of parameters in the function to be optimized.</param>
///
public ResilientBackpropagation(int parameters)
{
convergence = new AbsoluteConvergence();
solution = new double[parameters];
gradient = new double[parameters];
previousGradient = new double[parameters];
weightsUpdates = new double[parameters];
// Initialize steps
Reset(initialStep);
for (int i = 0; i < solution.Length; i++)
solution[i] = Accord.Math.Tools.Random.NextDouble() * 2.0 - 1.0;
}
/// <summary>
/// Initializes a new instance of the <see cref="ResilientBackpropagation"/> class.
/// </summary>
///
/// <param name="parameters">The number of parameters in the function to be optimized.</param>
///
public ResilientBackpropagation(int parameters)
{
convergence = new AbsoluteConvergence();
solution = new double[parameters];
gradient = new double[parameters];
previousGradient = new double[parameters];
weightsUpdates = new double[parameters];
// Intialize steps
Reset(initialStep);
}
/// <summary>
/// Creates a new instance of the Viterbi learning algorithm.
/// </summary>
///
public ViterbiLearning(HiddenMarkovModel model)
{
this.convergence = new AbsoluteConvergence();
this.mle = new MaximumLikelihoodLearning(model);
}
