Examples with IntDummyVector com.tencent.angel.ml.math2.vector.IntDummyVector used on opensource projects

Example 16 with IntDummyVector

use of com.tencent.angel.ml.math2.vector.IntDummyVector in project angel by Tencent.

the class MixedBinaryInAllExecutor method apply.

private static Vector apply(CompIntDoubleVector v1, IntDummyVector v2, Binary op) {
    IntDoubleVector[] parts = v1.getPartitions();
    Storage[] resParts = StorageSwitch.applyComp(v1, v2, op);
    if (!op.isKeepStorage()) {
        for (int i = 0; i < parts.length; i++) {
            if (parts[i].getStorage() instanceof IntDoubleSortedVectorStorage) {
                resParts[i] = new IntDoubleSparseVectorStorage(parts[i].getDim(), parts[i].getStorage().getIndices(), parts[i].getStorage().getValues());
            }
        }
    }
    int subDim = (v1.getDim() + v1.getNumPartitions() - 1) / v1.getNumPartitions();
    for (int i = 0; i < v1.getDim(); i++) {
        int pidx = (int) (i / subDim);
        int subidx = i % subDim;
        ((IntDoubleVectorStorage) resParts[pidx]).set(subidx, op.apply(parts[pidx].get(subidx), v2.get(i)));
    }
    IntDoubleVector[] res = new IntDoubleVector[parts.length];
    int i = 0;
    for (IntDoubleVector part : parts) {
        res[i] = new IntDoubleVector(part.getMatrixId(), part.getRowId(), part.getClock(), part.getDim(), (IntDoubleVectorStorage) resParts[i]);
        i++;
    }
    v1.setPartitions(res);
    return v1;
}

Example 17 with IntDummyVector

use of com.tencent.angel.ml.math2.vector.IntDummyVector in project angel by Tencent.

the class DotMatrixExecutor method apply.

private static Vector apply(BlasFloatMatrix mat, boolean trans, IntDummyVector v) {
    int m = mat.getNumRows(), n = mat.getNumCols();
    float[] resArr;
    if (trans) {
        assert m == v.getDim();
        resArr = new float[n];
    } else {
        assert n == v.getDim();
        resArr = new float[m];
    }
    int r = mat.getNumRows(), c = mat.getNumCols();
    float[] data = mat.getData();
    if (trans) {
        for (int j = 0; j < c; j++) {
            int[] idxs = v.getIndices();
            for (int i : idxs) {
                resArr[j] += data[i * c + j];
            }
        }
    } else {
        for (int i = 0; i < r; i++) {
            int[] idxs = v.getIndices();
            for (int j : idxs) {
                resArr[i] += data[i * c + j];
            }
        }
    }
    IntFloatDenseVectorStorage storage = new IntFloatDenseVectorStorage(resArr);
    return new IntFloatVector(v.getMatrixId(), v.getClock(), 0, resArr.length, storage);
}

Example 18 with IntDummyVector

use of com.tencent.angel.ml.math2.vector.IntDummyVector in project angel by Tencent.

the class SimpleBinaryOutAllExecutor method apply.

public static Vector apply(IntDoubleVector v1, IntDummyVector v2, Binary op) {
    IntDoubleVector res;
    if (v1.isDense()) {
        res = v1.copy();
        double[] resValues = res.getStorage().getValues();
        for (int i = 0; i < v1.getDim(); i++) {
            resValues[i] = op.apply(resValues[i], v2.get(i));
        }
    } else if (v1.isSparse()) {
        if (op.isKeepStorage()) {
            throw new AngelException("operation is not support!");
        } else {
            IntDoubleVectorStorage newStorage = v1.getStorage().emptyDense();
            double[] resValues = newStorage.getValues();
            ObjectIterator<Int2DoubleMap.Entry> iter1 = v1.getStorage().entryIterator();
            while (iter1.hasNext()) {
                Int2DoubleMap.Entry entry = iter1.next();
                int idx = entry.getIntKey();
                resValues[idx] = entry.getDoubleValue();
            }
            for (int i = 0; i < v1.getDim(); i++) {
                resValues[i] = op.apply(resValues[i], v2.get(i));
            }
            res = new IntDoubleVector(v1.getMatrixId(), v1.getRowId(), v1.getClock(), v1.getDim(), newStorage);
        }
    } else {
        // sorted
        if (op.isKeepStorage()) {
            throw new AngelException("operation is not support!");
        } else {
            IntDoubleVectorStorage newStorage = v1.getStorage().emptyDense();
            double[] resValues = newStorage.getValues();
            int[] v1Indices = v1.getStorage().getIndices();
            double[] v1Values = v1.getStorage().getValues();
            int size = v1.size();
            for (int i = 0; i < size; i++) {
                int idx = v1Indices[i];
                resValues[idx] = v1Values[i];
            }
            for (int i = 0; i < v1.getDim(); i++) {
                resValues[i] = op.apply(resValues[i], v2.get(i));
            }
            res = new IntDoubleVector(v1.getMatrixId(), v1.getRowId(), v1.getClock(), v1.getDim(), newStorage);
        }
    }
    return res;
}

Example 19 with IntDummyVector

use of com.tencent.angel.ml.math2.vector.IntDummyVector in project angel by Tencent.

the class SimpleBinaryOutAllExecutor method apply.

public static Vector apply(IntLongVector v1, IntDummyVector v2, Binary op) {
    IntLongVector res;
    if (v1.isDense()) {
        res = v1.copy();
        long[] resValues = res.getStorage().getValues();
        for (int i = 0; i < v1.getDim(); i++) {
            resValues[i] = op.apply(resValues[i], v2.get(i));
        }
    } else if (v1.isSparse()) {
        if (op.isKeepStorage()) {
            throw new AngelException("operation is not support!");
        } else {
            IntLongVectorStorage newStorage = v1.getStorage().emptyDense();
            long[] resValues = newStorage.getValues();
            ObjectIterator<Int2LongMap.Entry> iter1 = v1.getStorage().entryIterator();
            while (iter1.hasNext()) {
                Int2LongMap.Entry entry = iter1.next();
                int idx = entry.getIntKey();
                resValues[idx] = entry.getLongValue();
            }
            for (int i = 0; i < v1.getDim(); i++) {
                resValues[i] = op.apply(resValues[i], v2.get(i));
            }
            res = new IntLongVector(v1.getMatrixId(), v1.getRowId(), v1.getClock(), v1.getDim(), newStorage);
        }
    } else {
        // sorted
        if (op.isKeepStorage()) {
            throw new AngelException("operation is not support!");
        } else {
            IntLongVectorStorage newStorage = v1.getStorage().emptyDense();
            long[] resValues = newStorage.getValues();
            int[] v1Indices = v1.getStorage().getIndices();
            long[] v1Values = v1.getStorage().getValues();
            int size = v1.size();
            for (int i = 0; i < size; i++) {
                int idx = v1Indices[i];
                resValues[idx] = v1Values[i];
            }
            for (int i = 0; i < v1.getDim(); i++) {
                resValues[i] = op.apply(resValues[i], v2.get(i));
            }
            res = new IntLongVector(v1.getMatrixId(), v1.getRowId(), v1.getClock(), v1.getDim(), newStorage);
        }
    }
    return res;
}

Example 20 with IntDummyVector

use of com.tencent.angel.ml.math2.vector.IntDummyVector in project angel by Tencent.

the class SimpleBinaryOutNonZAExecutor method apply.

public static Vector apply(IntIntVector v1, IntDummyVector v2, Binary op) {
    IntIntVectorStorage newStorage = (IntIntVectorStorage) StorageSwitch.apply(v1, v2, op);
    if (v1.isDense()) {
        int[] resValues = newStorage.getValues();
        int[] v2Indices = v2.getIndices();
        for (int idx : v2Indices) {
            resValues[idx] = op.apply(resValues[idx], 1);
        }
    } else if (v1.isSparse()) {
        int[] v2Indices = v2.getIndices();
        if (((v1.size() + v2.size()) * Constant.intersectionCoeff > Constant.sparseDenseStorageThreshold * v1.getDim())) {
            int[] resValues = newStorage.getValues();
            ObjectIterator<Int2IntMap.Entry> iter = v1.getStorage().entryIterator();
            while (iter.hasNext()) {
                Int2IntMap.Entry entry = iter.next();
                newStorage.set(entry.getIntKey(), entry.getIntValue());
            }
            for (int idx : v2Indices) {
                newStorage.set(idx, op.apply(v1.get(idx), 1));
            }
        } else {
            // to avoid multi-rehash
            int capacity = 1 << (32 - Integer.numberOfLeadingZeros((int) (v1.size() / 0.75)));
            if (v1.size() + v2.size() < 1.5 * capacity) {
                int size = v2.size();
                for (int i = 0; i < size; i++) {
                    int idx = v2Indices[i];
                    newStorage.set(idx, op.apply(v1.get(idx), 1));
                }
            } else {
                ObjectIterator<Int2IntMap.Entry> iter1 = v1.getStorage().entryIterator();
                while (iter1.hasNext()) {
                    Int2IntMap.Entry entry = iter1.next();
                    int idx = entry.getIntKey();
                    newStorage.set(idx, entry.getIntValue());
                }
                int size = v2.size();
                for (int i = 0; i < size; i++) {
                    int idx = v2Indices[i];
                    newStorage.set(idx, op.apply(v1.get(idx), 1));
                }
            }
        }
    } else {
        // sorted
        int[] v1Indices = v1.getStorage().getIndices();
        int[] v2Indices = v2.getIndices();
        if (!op.isKeepStorage() && ((v1.size() + v2.size()) * Constant.intersectionCoeff > Constant.sortedDenseStorageThreshold * v1.getDim())) {
            int[] v1Values = v1.getStorage().getValues();
            int size = v1.size();
            for (int i = 0; i < size; i++) {
                newStorage.set(v1Indices[i], v1Values[i]);
            }
            size = v2.size();
            for (int i = 0; i < size; i++) {
                int idx = v2Indices[i];
                newStorage.set(idx, op.apply(newStorage.get(idx), 1));
            }
        } else {
            int v1Pointor = 0;
            int v2Pointor = 0;
            int size1 = v1.size();
            int size2 = v2.size();
            int[] v1Values = v1.getStorage().getValues();
            while (v1Pointor < size1 || v2Pointor < size2) {
                if ((v1Pointor < size1 && v2Pointor < size2) && v1Indices[v1Pointor] == v2Indices[v2Pointor]) {
                    newStorage.set(v1Indices[v1Pointor], op.apply(v1Values[v1Pointor], 1));
                    v1Pointor++;
                    v2Pointor++;
                } else if ((v1Pointor < size1 && v2Pointor < size2) && v1Indices[v1Pointor] < v2Indices[v2Pointor] || (v1Pointor < size1 && v2Pointor >= size2)) {
                    newStorage.set(v1Indices[v1Pointor], v1Values[v1Pointor]);
                    v1Pointor++;
                } else if (((v1Pointor < size1 && v2Pointor < size2) && v1Indices[v1Pointor] >= v2Indices[v2Pointor]) || (v1Pointor >= size1 && v2Pointor < size2)) {
                    newStorage.set(v2Indices[v2Pointor], op.apply(0, 1));
                    v2Pointor++;
                }
            }
        }
    }
    return new IntIntVector(v1.getMatrixId(), v1.getRowId(), v1.getClock(), v1.getDim(), newStorage);
}