Examples of CurveFitter

ca.nengo.math.CurveFitter
Finds a Function that fits a set of a example points in some sense (e.g. least-squares). For example, least-squares polynomial approximation and spline interpolation are possibly implementations. @author Bryan Tripp
org.apache.commons.math.optimization.fitting.CurveFitter
Fitter for parametric univariate real functions y = f(x).
When a univariate real function y = f(x) does depend on some unknown parameters p₀, p₁ ... p_n-1, this class can be used to find these parameters. It does this by fitting the curve so it remains very close to a set of observed points (x₀, y₀), (x₁, y₁) ... (x_k-1, y_k-1). This fitting is done by finding the parameters values that minimizes the objective function ∑(y_i-f(x_i))². This is really a least squares problem.
@version $Revision: 1073158 $ $Date: 2011-02-21 22:46:52 +0100 (lun. 21 févr. 2011) $ @since 2.0

Examples of ca.nengo.math.CurveFitter

  /* 
   * Test method for 'ca.nengo.math.impl.LinearCurveFitter.fit()'
   */
  public void testFindCoefficients() {
    Function target = new Polynomial(new float[]{1f,4f,-3f,0.5f});
    CurveFitter lcf = new LinearCurveFitter();
    float[][] values = new float[2][10];
    
    for (int i=0; i<values[0].length; i++) {
      values[0][i] = -9 + i * 2;
      values[1][i] = target.map(new float[]{values[0][i]});
    }
    
    Function fitted = lcf.fit(values[0], values[1]);
    
//    Plotter.plot(target, -10, 0.1f, 10, "target");
//    Plotter.plot(fitted, -10, 0.5f, 10, "fitted");
    
    float targetVal = 0f;

View Full Code Here

Examples of ca.nengo.math.CurveFitter

    for (int i = 0; i < myCurrents.length; i++) {
      countSpikes(myCurrents[i], dt, myTransientTime);
      rates[i] = countSpikes(myCurrents[i], dt, simTime) / simTime;
    }


    CurveFitter cf = new LinearCurveFitter();
    Function result = cf.fit(myCurrents, rates);
    myConstantRateFunction = result;
    myConstantRateFunctionOK = true;
    myMode = mode;
  }

View Full Code Here

Examples of ca.nengo.math.CurveFitter

    myMode = SimulationMode.DEFAULT;
    mySupportedModes = new SimulationMode[]{SimulationMode.DEFAULT, SimulationMode.RATE, SimulationMode.CONSTANT_RATE};
  }


  private Function getSteadyStateVmFunction() {
    CurveFitter fitter = new LinearCurveFitter();


    float[] current = new float[]{0f, .25f, .5f, .75f, 1f, 1.25f, 1.5f, 1.75f, 2f, 2.25f, 2.5f, 2.75f, 3f, 3.5f, 4f, 5f, 6f, 8f, 10f, 15f, 20f, 30f, 40f, 50f, 60f};
    float[] Vm = new float[current.length];
    float[] rt = new float[current.length];
    for (int i = 0; i < current.length; i++) {
      TimeSeries input = new TimeSeriesImpl(new float[]{0, 0.5f},
          new float[][]{new float[]{current[i], 1.0f}, new float[]{current[i], 1.0f}}, new Units[]{Units.uAcm2, Units.UNK});
      TimeSeries output = myIntegrator.integrate(myDynamics, input);
      Vm[i] = output.getValues()[output.getValues().length - 1][0];
      reset(false);
//      Plotter.plot(output, "simulation "+i);
      rt[i] = getRefreactoryTime(Vm[i]);
    }


    Function result = fitter.fit(current, Vm);
//    Plotter.plot(result, 0, .1f, 60, "current -> Vm");
//    Plotter.plot(current, rt, "current -> rt");
    return result;
  }

View Full Code Here

Examples of org.apache.commons.math.optimization.fitting.CurveFitter

     * Constructs an instance using the specified optimizer.
     *
     * @param optimizer optimizer to use for the fitting
     */
    public GaussianFitter(DifferentiableMultivariateVectorialOptimizer optimizer) {
        fitter = new CurveFitter(optimizer);
    }

View Full Code Here

Examples of org.apache.commons.math.optimization.fitting.CurveFitter

     * @throws FunctionEvaluationException in the event of a test case error
     */
    @Test
    public void testFit01()
    throws OptimizationException, FunctionEvaluationException {
        CurveFitter fitter = new CurveFitter(new LevenbergMarquardtOptimizer());
        addDatasetToCurveFitter(DATASET1, fitter);
        double[] parameters = fitter.fit(new ParametricGaussianFunction(),
                                         new double[] {8.64753e3, 3.483323e6, 4.06322, 1.946857e-2});
        assertEquals(99200.94715858076, parameters[0], 1e-4);
        assertEquals(3410515.221897707, parameters[1], 1e-4);
        assertEquals(4.054928275257894, parameters[2], 1e-4);
        assertEquals(0.014609868499860, parameters[3], 1e-4);

View Full Code Here

Examples of org.apache.commons.math.optimization.fitting.CurveFitter

     * @throws FunctionEvaluationException in the event of a test case error
     */
    @Test
    public void testFit02()
    throws OptimizationException, FunctionEvaluationException {
        CurveFitter fitter = new CurveFitter(new LevenbergMarquardtOptimizer());
        addDatasetToCurveFitter(DATASET1, fitter);
        double[] parameters = fitter.fit(new ParametricGaussianFunction(),
                                         new double[] {500000.0, 3500000.0, 4.055, 0.025479654});
        assertEquals(99200.81836264656, parameters[0], 1e-4);
        assertEquals(3410515.327151986, parameters[1], 1e-4);
        assertEquals(4.054928275377392, parameters[2], 1e-4);
        assertEquals(0.014609869119806, parameters[3], 1e-4);

View Full Code Here

TOP

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.