Package org.ejml.simple

Examples of org.ejml.simple.SimpleMatrix.transpose()

414
415
416
417
418
419
420
421
422
423
424
    }
    SimpleMatrix currentVector = nodeVectors.get(tree);
    SimpleMatrix currentVectorDerivative = NeuralUtils.elementwiseApplyTanhDerivative(currentVector);

    SimpleMatrix scoreW = dvModel.getScoreWForNode(tree);
    currentVectorDerivative = currentVectorDerivative.elementMult(scoreW.transpose());

    // the delta that is used at the current nodes
    SimpleMatrix deltaCurrent = deltaUp.plus(currentVectorDerivative);
    SimpleMatrix W = dvModel.getWForNode(tree);
    SimpleMatrix WTdelta = W.transpose().mult(deltaCurrent);
View Full Code Here
211
212
213
214
215
216
217
218
219
220
221
     * Check by performing the calculation manually
     */
    @Test
    public void computeY() {
        SimpleMatrix A = SimpleMatrix.random(2*r+2,2*r+2,-1,1,rand);
        A = A.transpose().mult(A); // needs to be symmetric to pass
        SimpleMatrix Vo = SimpleMatrix.random(r,A.numCols(),-1,1,rand);

        for( int row = 0; row < r; row++ ) {
            SimpleMatrix AA = A.copy();
            SimpleMatrix u = A.extractVector(true,row).transpose();
View Full Code Here
243
244
245
246
247
248
249
250
251
252
253
                        U.set(j,i,0);
                    }
                    U.set(i+1,i,1);
                }

                y = AA.plus(U.mult(V.transpose())).plus(V.mult(U.transpose())).mult(u);
            } else {
                y = AA.mult(u);
            }

            y = y.scale(-gamma);
View Full Code Here
71
72
73
74
75
76
77
78
79
80
81
        SimpleMatrix I = SimpleMatrix.identity(width-w);
        SimpleMatrix Q = I.minus(U.mult(U.transpose()).scale(qr.getGamma()));


        // check the expected properties of Q
        assertTrue(Q.isIdentical(Q.transpose(),1e-6));
        assertTrue(Q.isIdentical(Q.invert(),1e-6));

        SimpleMatrix result = Q.mult(A.extractMatrix(w,width,w,width));

        for( int i = 1; i < width-w; i++ ) {
View Full Code Here
71
72
73
74
75
76
77
78
79
80
81
        SimpleMatrix I = SimpleMatrix.identity(width-w);
        SimpleMatrix Q = I.minus(U.mult(U.transpose()).scale(qr.getGamma()));


        // check the expected properties of Q
        assertTrue(Q.isIdentical(Q.transpose(),1e-6));
        assertTrue(Q.isIdentical(Q.invert(),1e-6));

        SimpleMatrix result = Q.mult(A.extractMatrix(w,width,w,width));

        for( int i = 1; i < width-w; i++ ) {
View Full Code Here
312
313
314
315
316
317
318
319
320
321
322
    int size = deltaFull.getNumElements();
    SimpleTensor Wt_df = new SimpleTensor(size*2, size*2, size);
    // TODO: combine this concatenation with computeTensorDeltaDown?
    SimpleMatrix fullVector = NeuralUtils.concatenate(leftVector, rightVector);
    for (int slice = 0; slice < size; ++slice) {
      Wt_df.setSlice(slice, fullVector.scale(deltaFull.get(slice)).mult(fullVector.transpose()));
    }
    return Wt_df;
  }

  /**
 
View Full Code Here
104
105
106
107
108
109
110
111
112
113
114
            SimpleMatrix u = new SimpleMatrix(QR.numRows(),1);
            u.insertIntoThis(j,0,QR.extractMatrix(j, END,j,j+1));
            u.set(j,1.0);

            // A = (I - &gamma;*u*u<sup>T</sup>)*A<br>
            Q = Q.plus(-gammas[j],u.mult(u.transpose()).mult(Q));
        }

        return Q;
    }

View Full Code Here
25
26
27
28
29
30
31
32
33
34
35
   
    assertEquals(0.35000000000000003, NeuralUtils.dot(vector1, vector2), 1e-5);
    assertEquals(0.6363636363636364, NeuralUtils.cosine(vector1, vector2), 1e-5);
   
    vector1 = vector1.transpose();
    vector2 = vector2.transpose();
    assertEquals(0.35000000000000003, NeuralUtils.dot(vector1, vector2), 1e-5);
    assertEquals(0.6363636363636364, NeuralUtils.cosine(vector1, vector2), 1e-5);
  }

  public void testIsZero() {
View Full Code Here
206
207
208
209
210
211
212
213
214
215
216
      for (int trainingExample = 0; trainingExample < centroids.get(centroid).trainingExamples.size(); trainingExample++) {
        errorMatrix[(centroids.get(centroid).trainingExamples.get(trainingExample))] = centroids.get(centroid).features;
      }
    }   
    SimpleMatrix errorSimpleMatrix = new SimpleMatrix(errorMatrix).minus(datasetMatrix);
    errorSimpleMatrix = errorSimpleMatrix.mult(errorSimpleMatrix.transpose());
    double error = errorSimpleMatrix.elementSum() / (noFeatures * noTrainingExamples);
   
    LOGGER.info("Clustering completed - Error: " + Math.floor(error * 100) + "%");     

    return centroids;
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.