Examples with Group org.apache.sysml.lops.Group used on opensource projects

Example 36 with Group

use of org.apache.sysml.lops.Group in project incubator-systemml by apache.

the class ParameterizedBuiltinOp method constructLopsRExpand.

private void constructLopsRExpand(HashMap<String, Lop> inputlops, ExecType et) {
    if (et == ExecType.CP || et == ExecType.SPARK) {
        int k = OptimizerUtils.getConstrainedNumThreads(_maxNumThreads);
        ParameterizedBuiltin pbilop = new ParameterizedBuiltin(inputlops, HopsParameterizedBuiltinLops.get(_op), getDataType(), getValueType(), et, k);
        setOutputDimensions(pbilop);
        setLineNumbers(pbilop);
        setLops(pbilop);
    } else if (et == ExecType.MR) {
        ParameterizedBuiltin pbilop = new ParameterizedBuiltin(inputlops, HopsParameterizedBuiltinLops.get(_op), getDataType(), getValueType(), et);
        setOutputDimensions(pbilop);
        setLineNumbers(pbilop);
        Group group1 = new Group(pbilop, Group.OperationTypes.Sort, getDataType(), getValueType());
        setOutputDimensions(group1);
        setLineNumbers(group1);
        Aggregate finalagg = new Aggregate(group1, Aggregate.OperationTypes.Sum, DataType.MATRIX, getValueType(), ExecType.MR);
        setOutputDimensions(finalagg);
        setLineNumbers(finalagg);
        setLops(finalagg);
    }
}

Example 37 with Group

use of org.apache.sysml.lops.Group in project incubator-systemml by apache.

the class QuaternaryOp method constructMRLopsWeightedSquaredLoss.

private void constructMRLopsWeightedSquaredLoss(WeightsType wtype) {
    // NOTE: the common case for wsloss are factors U/V with a rank of 10s to 100s; the current runtime only
    // supports single block outer products (U/V rank <= blocksize, i.e., 1000 by default); we enforce this
    // by applying the hop rewrite for Weighted Squared Loss only if this constraint holds.
    Hop X = getInput().get(0);
    Hop U = getInput().get(1);
    Hop V = getInput().get(2);
    Hop W = getInput().get(3);
    // MR operator selection, part1
    // size U
    double m1Size = OptimizerUtils.estimateSize(U.getDim1(), U.getDim2());
    // size V
    double m2Size = OptimizerUtils.estimateSize(V.getDim1(), V.getDim2());
    boolean isMapWsloss = (!wtype.hasFourInputs() && m1Size + m2Size < OptimizerUtils.getRemoteMemBudgetMap(true));
    if (// broadcast
    !FORCE_REPLICATION && isMapWsloss) {
        // partitioning of U
        boolean needPartU = !U.dimsKnown() || U.getDim1() * U.getDim2() > DistributedCacheInput.PARTITION_SIZE;
        Lop lU = U.constructLops();
        if (needPartU) {
            // requires partitioning
            lU = new DataPartition(lU, DataType.MATRIX, ValueType.DOUBLE, (m1Size > OptimizerUtils.getLocalMemBudget()) ? ExecType.MR : ExecType.CP, PDataPartitionFormat.ROW_BLOCK_WISE_N);
            lU.getOutputParameters().setDimensions(U.getDim1(), U.getDim2(), getRowsInBlock(), getColsInBlock(), U.getNnz());
            setLineNumbers(lU);
        }
        // partitioning of V
        boolean needPartV = !V.dimsKnown() || V.getDim1() * V.getDim2() > DistributedCacheInput.PARTITION_SIZE;
        Lop lV = V.constructLops();
        if (needPartV) {
            // requires partitioning
            lV = new DataPartition(lV, DataType.MATRIX, ValueType.DOUBLE, (m2Size > OptimizerUtils.getLocalMemBudget()) ? ExecType.MR : ExecType.CP, PDataPartitionFormat.ROW_BLOCK_WISE_N);
            lV.getOutputParameters().setDimensions(V.getDim1(), V.getDim2(), getRowsInBlock(), getColsInBlock(), V.getNnz());
            setLineNumbers(lV);
        }
        // map-side wsloss always with broadcast
        Lop wsloss = new WeightedSquaredLoss(X.constructLops(), lU, lV, W.constructLops(), DataType.MATRIX, ValueType.DOUBLE, wtype, ExecType.MR);
        wsloss.getOutputParameters().setDimensions(1, 1, X.getRowsInBlock(), X.getColsInBlock(), -1);
        setLineNumbers(wsloss);
        Group grp = new Group(wsloss, Group.OperationTypes.Sort, DataType.MATRIX, ValueType.DOUBLE);
        grp.getOutputParameters().setDimensions(1, 1, X.getRowsInBlock(), X.getColsInBlock(), -1);
        setLineNumbers(grp);
        Aggregate agg1 = new Aggregate(grp, HopsAgg2Lops.get(AggOp.SUM), DataType.MATRIX, ValueType.DOUBLE, ExecType.MR);
        // aggregation uses kahanSum
        agg1.setupCorrectionLocation(CorrectionLocationType.NONE);
        agg1.getOutputParameters().setDimensions(1, 1, X.getRowsInBlock(), X.getColsInBlock(), -1);
        setLineNumbers(agg1);
        UnaryCP unary1 = new UnaryCP(agg1, HopsOpOp1LopsUS.get(OpOp1.CAST_AS_SCALAR), getDataType(), getValueType());
        unary1.getOutputParameters().setDimensions(0, 0, 0, 0, -1);
        setLineNumbers(unary1);
        setLops(unary1);
    } else // general case
    {
        // MR operator selection part 2
        boolean cacheU = !FORCE_REPLICATION && (m1Size < OptimizerUtils.getRemoteMemBudgetReduce());
        boolean cacheV = !FORCE_REPLICATION && ((!cacheU && m2Size < OptimizerUtils.getRemoteMemBudgetReduce()) || (cacheU && m1Size + m2Size < OptimizerUtils.getRemoteMemBudgetReduce()));
        Group grpX = new Group(X.constructLops(), Group.OperationTypes.Sort, DataType.MATRIX, ValueType.DOUBLE);
        grpX.getOutputParameters().setDimensions(X.getDim1(), X.getDim2(), X.getRowsInBlock(), X.getColsInBlock(), -1);
        setLineNumbers(grpX);
        Lop grpW = W.constructLops();
        if (grpW.getDataType() == DataType.MATRIX) {
            grpW = new Group(W.constructLops(), Group.OperationTypes.Sort, DataType.MATRIX, ValueType.DOUBLE);
            grpW.getOutputParameters().setDimensions(W.getDim1(), W.getDim2(), W.getRowsInBlock(), W.getColsInBlock(), -1);
            setLineNumbers(grpW);
        }
        Lop lU = constructLeftFactorMRLop(U, V, cacheU, m1Size);
        Lop lV = constructRightFactorMRLop(U, V, cacheV, m2Size);
        // reduce-side wsloss w/ or without broadcast
        Lop wsloss = new WeightedSquaredLossR(grpX, lU, lV, grpW, DataType.MATRIX, ValueType.DOUBLE, wtype, cacheU, cacheV, ExecType.MR);
        wsloss.getOutputParameters().setDimensions(1, 1, X.getRowsInBlock(), X.getColsInBlock(), -1);
        setLineNumbers(wsloss);
        Group grp = new Group(wsloss, Group.OperationTypes.Sort, DataType.MATRIX, ValueType.DOUBLE);
        grp.getOutputParameters().setDimensions(1, 1, X.getRowsInBlock(), X.getColsInBlock(), -1);
        setLineNumbers(grp);
        Aggregate agg1 = new Aggregate(grp, HopsAgg2Lops.get(AggOp.SUM), DataType.MATRIX, ValueType.DOUBLE, ExecType.MR);
        // aggregation uses kahanSum
        agg1.setupCorrectionLocation(CorrectionLocationType.NONE);
        agg1.getOutputParameters().setDimensions(1, 1, X.getRowsInBlock(), X.getColsInBlock(), -1);
        setLineNumbers(agg1);
        UnaryCP unary1 = new UnaryCP(agg1, HopsOpOp1LopsUS.get(OpOp1.CAST_AS_SCALAR), getDataType(), getValueType());
        unary1.getOutputParameters().setDimensions(0, 0, 0, 0, -1);
        setLineNumbers(unary1);
        setLops(unary1);
    }
}

Example 38 with Group

use of org.apache.sysml.lops.Group in project incubator-systemml by apache.

the class QuaternaryOp method constructMRLopsWeightedSigmoid.

private void constructMRLopsWeightedSigmoid(WSigmoidType wtype) {
    // NOTE: the common case for wsigmoid are factors U/V with a rank of 10s to 100s; the current runtime only
    // supports single block outer products (U/V rank <= blocksize, i.e., 1000 by default); we enforce this
    // by applying the hop rewrite for Weighted Sigmoid only if this constraint holds.
    Hop X = getInput().get(0);
    Hop U = getInput().get(1);
    Hop V = getInput().get(2);
    // MR operator selection, part1
    // size U
    double m1Size = OptimizerUtils.estimateSize(U.getDim1(), U.getDim2());
    // size V
    double m2Size = OptimizerUtils.estimateSize(V.getDim1(), V.getDim2());
    boolean isMapWsig = (m1Size + m2Size < OptimizerUtils.getRemoteMemBudgetMap(true));
    if (// broadcast
    !FORCE_REPLICATION && isMapWsig) {
        // partitioning of U
        boolean needPartU = !U.dimsKnown() || U.getDim1() * U.getDim2() > DistributedCacheInput.PARTITION_SIZE;
        Lop lU = U.constructLops();
        if (needPartU) {
            // requires partitioning
            lU = new DataPartition(lU, DataType.MATRIX, ValueType.DOUBLE, (m1Size > OptimizerUtils.getLocalMemBudget()) ? ExecType.MR : ExecType.CP, PDataPartitionFormat.ROW_BLOCK_WISE_N);
            lU.getOutputParameters().setDimensions(U.getDim1(), U.getDim2(), getRowsInBlock(), getColsInBlock(), U.getNnz());
            setLineNumbers(lU);
        }
        // partitioning of V
        boolean needPartV = !V.dimsKnown() || V.getDim1() * V.getDim2() > DistributedCacheInput.PARTITION_SIZE;
        Lop lV = V.constructLops();
        if (needPartV) {
            // requires partitioning
            lV = new DataPartition(lV, DataType.MATRIX, ValueType.DOUBLE, (m2Size > OptimizerUtils.getLocalMemBudget()) ? ExecType.MR : ExecType.CP, PDataPartitionFormat.ROW_BLOCK_WISE_N);
            lV.getOutputParameters().setDimensions(V.getDim1(), V.getDim2(), getRowsInBlock(), getColsInBlock(), V.getNnz());
            setLineNumbers(lV);
        }
        // map-side wsig always with broadcast
        Lop wsigmoid = new WeightedSigmoid(X.constructLops(), lU, lV, DataType.MATRIX, ValueType.DOUBLE, wtype, ExecType.MR);
        setOutputDimensions(wsigmoid);
        setLineNumbers(wsigmoid);
        setLops(wsigmoid);
    // in contrast to wsloss no aggregation required
    } else // general case
    {
        // MR operator selection part 2
        boolean cacheU = !FORCE_REPLICATION && (m1Size < OptimizerUtils.getRemoteMemBudgetReduce());
        boolean cacheV = !FORCE_REPLICATION && ((!cacheU && m2Size < OptimizerUtils.getRemoteMemBudgetReduce()) || (cacheU && m1Size + m2Size < OptimizerUtils.getRemoteMemBudgetReduce()));
        Group grpX = new Group(X.constructLops(), Group.OperationTypes.Sort, DataType.MATRIX, ValueType.DOUBLE);
        grpX.getOutputParameters().setDimensions(X.getDim1(), X.getDim2(), X.getRowsInBlock(), X.getColsInBlock(), X.getNnz());
        setLineNumbers(grpX);
        Lop lU = constructLeftFactorMRLop(U, V, cacheU, m1Size);
        Lop lV = constructRightFactorMRLop(U, V, cacheV, m2Size);
        // reduce-side wsig w/ or without broadcast
        Lop wsigmoid = new WeightedSigmoidR(grpX, lU, lV, DataType.MATRIX, ValueType.DOUBLE, wtype, cacheU, cacheV, ExecType.MR);
        setOutputDimensions(wsigmoid);
        setLineNumbers(wsigmoid);
        setLops(wsigmoid);
    // in contrast to wsloss no aggregation required
    }
}

Example 39 with Group

use of org.apache.sysml.lops.Group in project incubator-systemml by apache.

the class QuaternaryOp method constructMRLopsWeightedUMM.

private void constructMRLopsWeightedUMM(WUMMType wtype) {
    // NOTE: the common case for wumm are factors U/V with a rank of 10s to 100s; the current runtime only
    // supports single block outer products (U/V rank <= blocksize, i.e., 1000 by default); we enforce this
    // by applying the hop rewrite for Weighted UnaryMM  only if this constraint holds.
    Unary.OperationTypes uop = _uop != null ? HopsOpOp1LopsU.get(_uop) : _sop == OpOp2.POW ? Unary.OperationTypes.POW2 : Unary.OperationTypes.MULTIPLY2;
    Hop X = getInput().get(0);
    Hop U = getInput().get(1);
    Hop V = getInput().get(2);
    // MR operator selection, part1
    // size U
    double m1Size = OptimizerUtils.estimateSize(U.getDim1(), U.getDim2());
    // size V
    double m2Size = OptimizerUtils.estimateSize(V.getDim1(), V.getDim2());
    boolean isMapWumm = (m1Size + m2Size < OptimizerUtils.getRemoteMemBudgetMap(true));
    if (// broadcast
    !FORCE_REPLICATION && isMapWumm) {
        // partitioning of U
        boolean needPartU = !U.dimsKnown() || U.getDim1() * U.getDim2() > DistributedCacheInput.PARTITION_SIZE;
        Lop lU = U.constructLops();
        if (needPartU) {
            // requires partitioning
            lU = new DataPartition(lU, DataType.MATRIX, ValueType.DOUBLE, (m1Size > OptimizerUtils.getLocalMemBudget()) ? ExecType.MR : ExecType.CP, PDataPartitionFormat.ROW_BLOCK_WISE_N);
            lU.getOutputParameters().setDimensions(U.getDim1(), U.getDim2(), getRowsInBlock(), getColsInBlock(), U.getNnz());
            setLineNumbers(lU);
        }
        // partitioning of V
        boolean needPartV = !V.dimsKnown() || V.getDim1() * V.getDim2() > DistributedCacheInput.PARTITION_SIZE;
        Lop lV = V.constructLops();
        if (needPartV) {
            // requires partitioning
            lV = new DataPartition(lV, DataType.MATRIX, ValueType.DOUBLE, (m2Size > OptimizerUtils.getLocalMemBudget()) ? ExecType.MR : ExecType.CP, PDataPartitionFormat.ROW_BLOCK_WISE_N);
            lV.getOutputParameters().setDimensions(V.getDim1(), V.getDim2(), getRowsInBlock(), getColsInBlock(), V.getNnz());
            setLineNumbers(lV);
        }
        // map-side wumm always with broadcast
        Lop wumm = new WeightedUnaryMM(X.constructLops(), lU, lV, DataType.MATRIX, ValueType.DOUBLE, wtype, uop, ExecType.MR);
        setOutputDimensions(wumm);
        setLineNumbers(wumm);
        setLops(wumm);
    // in contrast to wsloss no aggregation required
    } else // general case
    {
        // MR operator selection part 2
        boolean cacheU = !FORCE_REPLICATION && (m1Size < OptimizerUtils.getRemoteMemBudgetReduce());
        boolean cacheV = !FORCE_REPLICATION && ((!cacheU && m2Size < OptimizerUtils.getRemoteMemBudgetReduce()) || (cacheU && m1Size + m2Size < OptimizerUtils.getRemoteMemBudgetReduce()));
        Group grpX = new Group(X.constructLops(), Group.OperationTypes.Sort, DataType.MATRIX, ValueType.DOUBLE);
        grpX.getOutputParameters().setDimensions(X.getDim1(), X.getDim2(), X.getRowsInBlock(), X.getColsInBlock(), X.getNnz());
        setLineNumbers(grpX);
        Lop lU = constructLeftFactorMRLop(U, V, cacheU, m1Size);
        Lop lV = constructRightFactorMRLop(U, V, cacheV, m2Size);
        // reduce-side wumm w/ or without broadcast
        Lop wumm = new WeightedUnaryMMR(grpX, lU, lV, DataType.MATRIX, ValueType.DOUBLE, wtype, uop, cacheU, cacheV, ExecType.MR);
        setOutputDimensions(wumm);
        setLineNumbers(wumm);
        setLops(wumm);
    // in contrast to wsloss no aggregation required
    }
}

Example 40 with Group

use of org.apache.sysml.lops.Group in project incubator-systemml by apache.

the class QuaternaryOp method constructLeftFactorMRLop.

private Lop constructLeftFactorMRLop(Hop U, Hop V, boolean cacheU, double m1Size) {
    Lop lU = null;
    if (cacheU) {
        // partitioning of U for read through distributed cache
        boolean needPartU = !U.dimsKnown() || U.getDim1() * U.getDim2() > DistributedCacheInput.PARTITION_SIZE;
        lU = U.constructLops();
        if (needPartU) {
            // requires partitioning
            lU = new DataPartition(lU, DataType.MATRIX, ValueType.DOUBLE, (m1Size > OptimizerUtils.getLocalMemBudget()) ? ExecType.MR : ExecType.CP, PDataPartitionFormat.ROW_BLOCK_WISE_N);
            lU.getOutputParameters().setDimensions(U.getDim1(), U.getDim2(), getRowsInBlock(), getColsInBlock(), U.getNnz());
            setLineNumbers(lU);
        }
    } else {
        // replication of U for shuffle to target block
        // ncol of t(V) -> nrow of V determines num replicates
        Lop offset = createOffsetLop(V, false);
        lU = new RepMat(U.constructLops(), offset, true, V.getDataType(), V.getValueType());
        lU.getOutputParameters().setDimensions(U.getDim1(), U.getDim2(), U.getRowsInBlock(), U.getColsInBlock(), U.getNnz());
        setLineNumbers(lU);
        Group grpU = new Group(lU, Group.OperationTypes.Sort, DataType.MATRIX, ValueType.DOUBLE);
        grpU.getOutputParameters().setDimensions(U.getDim1(), U.getDim2(), U.getRowsInBlock(), U.getColsInBlock(), -1);
        setLineNumbers(grpU);
        lU = grpU;
    }
    return lU;
}