Examples with ManhattanDistance org.nd4j.linalg.api.ops.impl.accum.distances.ManhattanDistance used on opensource projects

Example 1 with ManhattanDistance

use of org.nd4j.linalg.api.ops.impl.accum.distances.ManhattanDistance in project nd4j by deeplearning4j.

the class CrashTest method op.

protected void op(INDArray x, INDArray y, int i) {
    // broadcast along row & column
    INDArray row = Nd4j.ones(64);
    INDArray column = Nd4j.ones(1024, 1);
    x.addiRowVector(row);
    x.addiColumnVector(column);
    // casual scalar
    x.addi(i * 2);
    // reduction along all dimensions
    float sum = x.sumNumber().floatValue();
    // index reduction
    Nd4j.getExecutioner().exec(new IMax(x), Integer.MAX_VALUE);
    // casual transform
    Nd4j.getExecutioner().exec(new Sqrt(x, x));
    // dup
    INDArray x1 = x.dup(x.ordering());
    INDArray x2 = x.dup(x.ordering());
    INDArray x3 = x.dup('c');
    INDArray x4 = x.dup('f');
    // vstack && hstack
    INDArray vstack = Nd4j.vstack(x, x1, x2, x3, x4);
    INDArray hstack = Nd4j.hstack(x, x1, x2, x3, x4);
    // reduce3 call
    Nd4j.getExecutioner().exec(new ManhattanDistance(x, x2));
    // flatten call
    INDArray flat = Nd4j.toFlattened(x, x1, x2, x3, x4);
    // reduction along dimension: row & column
    INDArray max_0 = x.max(0);
    INDArray max_1 = x.max(1);
    // index reduction along dimension: row & column
    INDArray imax_0 = Nd4j.argMax(x, 0);
    INDArray imax_1 = Nd4j.argMax(x, 1);
    // logisoftmax, softmax & softmax derivative
    Nd4j.getExecutioner().exec(new OldSoftMax(x));
    Nd4j.getExecutioner().exec(new SoftMaxDerivative(x));
    Nd4j.getExecutioner().exec(new LogSoftMax(x));
    // BooleanIndexing
    BooleanIndexing.replaceWhere(x, 5f, Conditions.lessThan(8f));
    // assing on view
    BooleanIndexing.assignIf(x, x1, Conditions.greaterThan(-1000000000f));
    // std var along all dimensions
    float std = x.stdNumber().floatValue();
    // std var along row & col
    INDArray xStd_0 = x.std(0);
    INDArray xStd_1 = x.std(1);
    // blas call
    float dot = (float) Nd4j.getBlasWrapper().dot(x, x1);
    // mmul
    for (boolean tA : paramsA) {
        for (boolean tB : paramsB) {
            INDArray xT = tA ? x.dup() : x.dup().transpose();
            INDArray yT = tB ? y.dup() : y.dup().transpose();
            Nd4j.gemm(xT, yT, tA, tB);
        }
    }
    // specially for views, checking here without dup and rollover
    Nd4j.gemm(x, y, false, false);
    System.out.println("Iteration passed: " + i);
}

Example 2 with ManhattanDistance

use of org.nd4j.linalg.api.ops.impl.accum.distances.ManhattanDistance in project nd4j by deeplearning4j.

the class LongTests method testLongTadOp1.

@Test
public void testLongTadOp1() {
    double exp = Transforms.manhattanDistance(Nd4j.create(1000).assign(1.0), Nd4j.create(1000).assign(2.0));
    INDArray hugeX = Nd4j.create(2200000, 1000).assign(1.0);
    INDArray hugeY = Nd4j.create(1, 1000).assign(2.0);
    for (int x = 0; x < hugeX.rows(); x++) {
        assertEquals("Failed at row " + x, 1000, hugeX.getRow(x).sumNumber().intValue());
    }
    INDArray result = Nd4j.getExecutioner().exec(new ManhattanDistance(hugeX, hugeY, hugeX.lengthLong()), 1);
    for (int x = 0; x < hugeX.rows(); x++) {
        assertEquals(exp, result.getDouble(x), 1e-5);
    }
}

Example 3 with ManhattanDistance

use of org.nd4j.linalg.api.ops.impl.accum.distances.ManhattanDistance in project nd4j by deeplearning4j.

the class CudaReduce3Tests method testPinnedManhattanDistance.

@Test
public void testPinnedManhattanDistance() throws Exception {
    // simple way to stop test if we're not on CUDA backend here
    INDArray array1 = Nd4j.create(new float[] { 0.0f, 1.0f, 2.0f, 3.0f, 4.0f });
    INDArray array2 = Nd4j.create(new float[] { 0.5f, 1.5f, 2.5f, 3.5f, 4.5f });
    double result = Nd4j.getExecutioner().execAndReturn(new ManhattanDistance(array1, array2)).getFinalResult().doubleValue();
    System.out.println("Distance: " + result);
    assertEquals(2.5, result, 0.01);
    System.out.println("Array1: " + array1);
    System.out.println("Array2: " + array2);
}

Example 4 with ManhattanDistance

use of org.nd4j.linalg.api.ops.impl.accum.distances.ManhattanDistance in project nd4j by deeplearning4j.

the class NativeOpExecutionerTest method testEuclideanManhattanDistanceAlongDimension_Rank4.

@Test
public void testEuclideanManhattanDistanceAlongDimension_Rank4() {
    Nd4j.getRandom().setSeed(12345);
    INDArray firstOneExample = Nd4j.rand('c', new int[] { 1, 2, 2, 2 });
    INDArray secondOneExample = Nd4j.rand('c', new int[] { 1, 2, 2, 2 });
    double[] d1 = firstOneExample.data().asDouble();
    double[] d2 = secondOneExample.data().asDouble();
    double sumSquaredDiff = 0.0;
    double expManhattanDistance = 0.0;
    for (int i = 0; i < d1.length; i++) {
        double diff = d1[i] - d2[i];
        sumSquaredDiff += diff * diff;
        expManhattanDistance += Math.abs(diff);
    }
    double expected = Math.sqrt(sumSquaredDiff);
    System.out.println("Expected, Euclidean: " + expected);
    System.out.println("Expected, Manhattan: " + expManhattanDistance);
    int mb = 2;
    INDArray firstOrig = Nd4j.create(mb, 2, 2, 2);
    INDArray secondOrig = Nd4j.create(mb, 2, 2, 2);
    for (int i = 0; i < mb; i++) {
        firstOrig.put(new INDArrayIndex[] { NDArrayIndex.point(i), NDArrayIndex.all(), NDArrayIndex.all(), NDArrayIndex.all() }, firstOneExample);
        secondOrig.put(new INDArrayIndex[] { NDArrayIndex.point(i), NDArrayIndex.all(), NDArrayIndex.all(), NDArrayIndex.all() }, secondOneExample);
    }
    for (char order : new char[] { 'c', 'f' }) {
        INDArray first = firstOrig.dup(order);
        INDArray second = secondOrig.dup(order);
        assertEquals(firstOrig, first);
        assertEquals(secondOrig, second);
        INDArray out = Nd4j.getExecutioner().exec(new EuclideanDistance(first, second), 1, 2, 3);
        INDArray outManhattan = Nd4j.getExecutioner().exec(new ManhattanDistance(first, second), 1, 2, 3);
        System.out.println("\n\nOrder: " + order);
        System.out.println("Euclidean:");
        System.out.println(Arrays.toString(out.getRow(0).dup().data().asDouble()));
        System.out.println(Arrays.toString(out.getRow(1).dup().data().asDouble()));
        assertEquals(out.getRow(0), out.getRow(1));
        System.out.println("Manhattan:");
        System.out.println(Arrays.toString(outManhattan.getRow(0).dup().data().asDouble()));
        System.out.println(Arrays.toString(outManhattan.getRow(1).dup().data().asDouble()));
        assertEquals(expected, out.getRow(0).getDouble(0), 1e-5);
        assertEquals(expManhattanDistance, outManhattan.getRow(0).getDouble(0), 1e-5);
    }
}

Example 5 with ManhattanDistance

use of org.nd4j.linalg.api.ops.impl.accum.distances.ManhattanDistance in project nd4j by deeplearning4j.

the class NativeOpExecutionerTest method testPinnedManhattanDistance2.

@Test
public void testPinnedManhattanDistance2() throws Exception {
    // simple way to stop test if we're not on CUDA backend here
    INDArray array1 = Nd4j.linspace(1, 1000, 1000);
    INDArray array2 = Nd4j.linspace(1, 900, 1000);
    double result = Nd4j.getExecutioner().execAndReturn(new ManhattanDistance(array1, array2)).getFinalResult().doubleValue();
    assertEquals(50000.0, result, 0.001f);
}