Package gov.sandia.cognition.math.matrix

Examples of gov.sandia.cognition.math.matrix.Matrix

242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
      final LogitMixParticle particle = particleEntry.getKey();

      final MultivariateGaussian predictivePrior = particle.getLinearState().clone();
      KalmanFilter kf = particle.getRegressionFilter();
      final Matrix G = kf.getModel().getA();
      final Matrix F = data.getObservedData();
      predictivePrior.setMean(G.times(predictivePrior.getMean()));
      predictivePrior.setCovariance(
          G.times(predictivePrior.getCovariance()).times(G.transpose())
            .plus(kf.getModelCovariance()));
      final Vector betaMean = predictivePrior.getMean();

      final int particleCount;
      if (particleEntry.getValue() instanceof MutableDoubleCount) {
        particleCount = ((MutableDoubleCount)particleEntry.getValue()).count;
      } else {
        particleCount = 1;
      }
      for (int p = 0; p < particleCount; p++) {
        for (int j = 0; j < 10; j++) {
          final LogitMixParticle predictiveParticle =
              particle.clone();

          predictiveParticle.setPreviousParticle(particle);
          predictiveParticle.setBetaSample(betaMean);
          predictiveParticle.setLinearState(predictivePrior);

          final UnivariateGaussian componentDist =
              filter.evDistribution.getDistributions().get(j);

          predictiveParticle.setEVcomponent(componentDist);
         
          /*
           * Update the observed data for the regression component.
           */
          predictiveParticle.getRegressionFilter().getModel().setC(F);

          final Matrix compVar = MatrixFactory.getDefault().copyArray(
              new double[][] {{componentDist.getVariance()}});
          predictiveParticle.getRegressionFilter().setMeasurementCovariance(compVar);
         
          final double compPredPriorObsMean =
               F.times(betaMean).getElement(0)
View Full Code Here
53
54
55
56
57
58
59
60
61
62
63
      final Vector dataSample = VectorFactory.getDenseDefault().copyArray(new
        double[] {1d});
      final double phi = Math.exp(-trueGlobalMean - dataSample.dotProduct(trueBetas));
      final double pi = 1d / (1d + phi);
      final Vector y = VectorFactory.getDenseDefault().createVector1D(rng.nextDouble() <= pi ? 1d : 0d);
      final Matrix dataDesign = MatrixFactory.getDenseDefault().copyRowVectors(dataSample);
      observations.add(ObservedValue.create(i, y, dataDesign));
    }

    /*
     * Instantiate PL filter by first providing prior parameters/distributions.
View Full Code Here
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
    final MultivariateGaussian initialPrior = new MultivariateGaussian(
        VectorFactory.getDefault().copyValues(0d, 0d),
        MatrixFactory.getDefault().copyArray(new double[][] {
            {10d, 0d},
            {0d, 10d}}));
    final Matrix F = MatrixFactory.getDefault().copyArray(new double[][] {
        {1d, 1d}});
    final Matrix G = MatrixFactory.getDefault().copyArray(new double[][] {
        {1d, 0d},
        {0d, 1d}});
    final Matrix modelCovariance = MatrixFactory.getDefault().copyArray(new double[][] {
        {0d, 0d},
        {0d, 0d}});
    final LogitFSWFFilter plFilter =
        new LogitFSWFFilter(initialPrior, F, G, modelCovariance, rng);
    plFilter.setNumParticles(10000);
View Full Code Here
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
      KalmanFilter currentFilter = this.stateFilters.get(currentClass);
      if (currentState == null) {
        currentState = currentFilter.createInitialLearnedObject().sample(random);
      } else {
        final Matrix G = currentFilter.getModel().getA();
        Matrix modelCovSqrt = CholeskyDecompositionMTJ.create(
            (DenseMatrix) currentFilter.getModelCovariance()).getR();
        currentState = MultivariateGaussian.sample(G.times(currentState), modelCovSqrt, random);
      }
      currentState.plusEquals(currentFilter.getModel().getB().times(
          currentFilter.getCurrentInput()));

      final Matrix F = currentFilter.getModel().getC();
      Vector observationMean = F.times(currentState);
      Matrix measurementCovSqrt = CholeskyDecompositionMTJ.create(
          (DenseMatrix) currentFilter.getMeasurementCovariance()).getR();
      Vector observation = MultivariateGaussian.sample(observationMean,
          measurementCovSqrt, random);
 
      results.add(new SimHmmObservedValue<Vector, Vector>(i, currentClass,
View Full Code Here
101
102
103
104
105
106
107
108
109
110
111
  @Override
  public MultivariateGaussian getEmissionFunction(MultivariateGaussian state, int classId) {
    KalmanFilter kf = this.stateFilters.get(classId);
    final Vector mean = kf.getModel().getC().times(state.getMean());
    final Matrix cov = kf.getModel().getC().times(state.getCovariance())
        .times(kf.getModel().getC().transpose())
        .plus(kf.getMeasurementCovariance());
    final MultivariateGaussian likelihood = new MultivariateGaussian(
        mean, cov);
    return likelihood;
View Full Code Here
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
       * Fruewirth-Schnatter's method for upper utility sampling, where
       * instead of sampling the predictors, we use the mean.
       */
      final MultivariateGaussian predictivePrior = particle.getLinearState().clone();
      KalmanFilter kf = particle.getRegressionFilter();
      final Matrix G = kf.getModel().getA();
      final Matrix F = data.getObservedData();
      predictivePrior.setMean(G.times(predictivePrior.getMean()));
      predictivePrior.setCovariance(
          G.times(predictivePrior.getCovariance()).times(G.transpose())
            .plus(kf.getModelCovariance()));
      final Vector betaSample =
          predictivePrior.sample(getRandom());
//          predictivePrior.getMean();
      final double predPriorObsMean =
            F.times(betaSample).getElement(0);

      final int particleCount;
      if (particleEntry.getValue() instanceof MutableDoubleCount) {
        particleCount = ((MutableDoubleCount)particleEntry.getValue()).count;
      } else {
        particleCount = 1;
      }
      for (int p = 0; p < particleCount; p++) {
        final double dSampledAugResponse = sampleAugResponse(predPriorObsMean, isOne);

        Vector sampledAugResponse = VectorFactory.getDefault().copyValues(dSampledAugResponse);

        for (int j = 0; j < 10; j++) {
          final LogitMixParticle predictiveParticle = particle.clone();
          predictiveParticle.setPreviousParticle(particle);
          predictiveParticle.setBetaSample(betaSample);
          predictiveParticle.setAugResponseSample(sampledAugResponse);
          predictiveParticle.setLinearState(predictivePrior);

          final UnivariateGaussian componentDist =
              this.evDistribution.getDistributions().get(j);

          predictiveParticle.setEVcomponent(componentDist);
         
          /*
           * Update the observed data for the regression component.
           */
          predictiveParticle.getRegressionFilter().getModel().setC(F);

          // TODO would be great to have a 1x1 matrix class here...
          final Matrix compVar = MatrixFactory.getDefault().copyArray(
              new double[][] {{componentDist.getVariance()}});
          predictiveParticle.getRegressionFilter().setMeasurementCovariance(compVar);
         
          final double compPredPriorObsMean =
               F.times(betaSample).getElement(0)
View Full Code Here
147
148
149
150
151
152
153
154
155
156
157
158
      final MultivariateGaussian postBeta = particle.getPriorBeta().clone();

      final ScaledInverseGammaCovDistribution postObsCov = particle.getPriorObsCov();

      final Matrix augCovLassoSample = this.augLassoDist.sample(random);
      final Matrix obsCovSample = postObsCov.sample(random);
     
      /*
       * Update sufficient stats.
       * TODO FIXME not done
       */
 
View Full Code Here
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
      final Vector y = observation.getObservedValue();
      /*
       * Now, we compute the predictive dist, so that we can evaluate the likelihood.
       */
      final Matrix X = observation.getObservedData();
      final Vector priorPredObsMean = X.times(particle.getPriorBeta().getMean());
      Matrix priorPredObsCov = X.times(
          particle.getPriorBeta().getCovariance())
          .times(X.transpose()).plus(particle.augLassoSample);
      priorPredObsCov.times(particle.priorObsCovSample);

      final MultivariateGaussian priorPredictiveObsDist=
          new MultivariateGaussian(priorPredObsMean, priorPredObsCov);

      return priorPredictiveObsDist.getProbabilityFunction().logEvaluate(y);
View Full Code Here
76
77
78
79
80
81
82
83
84
85
86
87
    public DataDistribution<LassoRegressionDistribution> createInitialParticles(int numParticles) {

      final DefaultDataDistribution<LassoRegressionDistribution> initialParticles =
          new DefaultDataDistribution<LassoRegressionDistribution>(numParticles);
      for (int i = 0; i < numParticles; i++) {
        final Matrix priorBetaCovSmpl = this.priorBetaCov.sample(this.rng);
        final Matrix augLassoSmpl = this.augLassoDist.sample(this.rng);
        final LassoRegressionDistribution particleMvgDPDist =
            new LassoRegressionDistribution(this.priorBeta.clone(),
                this.priorBetaCov.clone(), augLassoSmpl, priorBetaCovSmpl);
        initialParticles.increment(particleMvgDPDist);
      }
View Full Code Here
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
    for (Entry<LogitMixParticle, ? extends Number> particleEntry : target.asMap().entrySet()) {
      final LogitMixParticle particle = particleEntry.getKey();

      final MultivariateGaussian predictivePrior = particle.getLinearState().clone();
      KalmanFilter kf = particle.getRegressionFilter();
      final Matrix G = kf.getModel().getA();
      final Matrix F = data.getObservedData();
      predictivePrior.setMean(G.times(predictivePrior.getMean()));
      predictivePrior.setCovariance(
          G.times(predictivePrior.getCovariance()).times(G.transpose())
            .plus(kf.getModelCovariance()));
      final Vector betaMean = predictivePrior.getMean();

      final int particleCount;
      if (particleEntry.getValue() instanceof MutableDoubleCount) {
        particleCount = ((MutableDoubleCount)particleEntry.getValue()).count;
      } else {
        particleCount = 1;
      }
      for (int p = 0; p < particleCount; p++) {

        for (int j = 0; j < 10; j++) {
          final LogitMixParticle predictiveParticle = particle.clone();
          predictiveParticle.setPreviousParticle(particle);
          predictiveParticle.setBetaSample(betaMean);
          predictiveParticle.setLinearState(predictivePrior);

          final UnivariateGaussian componentDist =
              this.evDistribution.getDistributions().get(j);

          predictiveParticle.setEVcomponent(componentDist);
         
          /*
           * Update the observed data for the regression component.
           */
          predictiveParticle.getRegressionFilter().getModel().setC(F);

          // TODO would be great to have a 1x1 matrix class here...
          final Matrix compVar = MatrixFactory.getDefault().copyArray(
              new double[][] {{componentDist.getVariance()}});
          predictiveParticle.getRegressionFilter().setMeasurementCovariance(compVar);
         
          final double compPredPriorObsMean =
               F.times(betaMean).getElement(0)
View Full Code Here
TOP

Related Classes of gov.sandia.cognition.math.matrix.Matrix

  • plm.clustering.MvGaussianDPPLFilter
  • plm.clustering.MvGaussianDPPLFilter$MultivariateGaussianDPPLUpdater
  • plm.gaussian.GaussianArHpWfPlFilter
  • plm.gaussian.GaussianArHpWfPlFilter$GaussianArHpWfPlUpdater
  • plm.gaussian.GaussianArHpWfPlFilterTest
  • plm.gaussian.GaussianArHpWfPlFilterV2$GaussianArHpWfPlUpdater
  • plm.hmm.DlmHiddenMarkovModel
  • plm.hmm.gaussian.GaussianArHmmMkfFilterTest
  • plm.hmm.gaussian.GaussianArHpHmmFilterTest
  • plm.hmm.gaussian.GaussianArHpHmmPLFilter$GaussianArHpHmmPlUpdater
    • Copyright © 2018 www.massapicom. All rights reserved.
    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.