Uses of Class
org.apache.sysds.runtime.matrix.data.MatrixBlock
Packages that use MatrixBlock
Package
Description
-
Uses of MatrixBlock in org.apache.sysds.api
Methods in org.apache.sysds.api that return MatrixBlockModifier and TypeMethodDescriptionPythonDMLScript.startReadingMbFromPipe(int id, int rlen, int clen, Types.ValueType type) PythonDMLScript.startReadingMbFromPipes(int[] blockSizes, int rlen, int clen, Types.ValueType type) Methods in org.apache.sysds.api with parameters of type MatrixBlockModifier and TypeMethodDescriptionvoidPythonDMLScript.startWritingMbToPipe(int id, MatrixBlock mb) -
Uses of MatrixBlock in org.apache.sysds.api.jmlc
Methods in org.apache.sysds.api.jmlc that return MatrixBlockModifier and TypeMethodDescriptionConnection.convertToMatrix(InputStream input, int rows, int cols) Converts an input stream of a string matrix in text format into a matrix block.Connection.convertToMatrix(InputStream input, int rows, int cols, String format) Converts an input stream of a string matrix in csv or text format into a matrix block.Connection.convertToMatrix(InputStream input, String meta) Converts an input stream of a string matrix in csv or textcell format into a matrix block.Connection.convertToMatrix(String input, int rows, int cols) Converts an input string representation of a matrix in textcell format into a matrix block.Connection.convertToMatrix(String input, String meta) Converts an input string representation of a matrix in csv or textcell format into a matrix block.ResultVariables.getMatrixBlock(String varname) Obtain the matrix represented by the given output variable.Methods in org.apache.sysds.api.jmlc with parameters of type MatrixBlockModifier and TypeMethodDescriptionvoidPreparedScript.setMatrix(String varname, MatrixBlock matrix, boolean reuse) Binds a matrix object to a registered input variable. -
Uses of MatrixBlock in org.apache.sysds.api.mlcontext
Methods in org.apache.sysds.api.mlcontext that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockMLContextConversionUtil.binaryBlocksToMatrixBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> binaryBlocks, MatrixMetadata matrixMetadata) Convert aJavaPairRDD<MatrixIndexes, MatrixBlock>to aMatrixBlockMatrix.toMatrixBlock()Obtain the matrix as aMatrixBlockMethods in org.apache.sysds.api.mlcontext that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> MLContextConversionUtil.dataFrameToMatrixBinaryBlocks(org.apache.spark.sql.Dataset<org.apache.spark.sql.Row> dataFrame) Convert aDataFrameto aJavaPairRDD<MatrixIndexes, MatrixBlock>binary-block matrix.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> MLContextConversionUtil.dataFrameToMatrixBinaryBlocks(org.apache.spark.sql.Dataset<org.apache.spark.sql.Row> dataFrame, MatrixMetadata matrixMetadata) Convert aDataFrameto aJavaPairRDD<MatrixIndexes, MatrixBlock>binary-block matrix.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> MLContextConversionUtil.matrixObjectToBinaryBlocks(MatrixObject matrixObject, SparkExecutionContext sparkExecutionContext) Convert aMatrixObjectto aJavaPairRDD<MatrixIndexes, MatrixBlock>.org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> Matrix.toBinaryBlocks()Obtain the matrix as aJavaPairRDD<MatrixIndexes, MatrixBlock>Methods in org.apache.sysds.api.mlcontext with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixObjectMLContextConversionUtil.matrixBlockToMatrixObject(String variableName, MatrixBlock matrixBlock, MatrixMetadata matrixMetadata) Convert aMatrixBlockto aMatrixObject.Method parameters in org.apache.sysds.api.mlcontext with type arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockMLContextConversionUtil.binaryBlocksToMatrixBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> binaryBlocks, MatrixMetadata matrixMetadata) Convert aJavaPairRDD<MatrixIndexes, MatrixBlock>to aMatrixBlockstatic MatrixObjectMLContextConversionUtil.binaryBlocksToMatrixObject(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> binaryBlocks) Convert aJavaPairRDD<MatrixIndexes, MatrixBlock>to aMatrixObject.static MatrixObjectMLContextConversionUtil.binaryBlocksToMatrixObject(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> binaryBlocks, MatrixMetadata matrixMetadata) Convert aJavaPairRDD<MatrixIndexes, MatrixBlock>to aMatrixObject.Constructor parameters in org.apache.sysds.api.mlcontext with type arguments of type MatrixBlockModifierConstructorDescriptionMatrix(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> binaryBlocks, MatrixMetadata matrixMetadata) Create a Matrix, specifying the SystemDS binary-block matrix and its metadata. -
Uses of MatrixBlock in org.apache.sysds.hops.estim
Methods in org.apache.sysds.hops.estim that return MatrixBlockModifier and TypeMethodDescriptionMMNode.getData()EstimatorDensityMap.DensityMap.getMap()EstimatorLayeredGraph.LayeredGraph.toMatrixBlock()Methods in org.apache.sysds.hops.estim that return types with arguments of type MatrixBlockMethods in org.apache.sysds.hops.estim with parameters of type MatrixBlockModifier and TypeMethodDescriptionvoidEstimatorLayeredGraph.LayeredGraph.buildNext(MatrixBlock mb) EstimatorBitsetMM.createBitset(MatrixBlock in) doubleEstimatorBasicAvg.estim(MatrixBlock m, SparsityEstimator.OpCode op) doubleEstimatorBasicAvg.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorBasicAvg.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) doubleEstimatorBasicWorst.estim(MatrixBlock m, SparsityEstimator.OpCode op) doubleEstimatorBasicWorst.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorBasicWorst.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) doubleEstimatorBitsetMM.estim(MatrixBlock m, SparsityEstimator.OpCode op) doubleEstimatorBitsetMM.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorBitsetMM.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) doubleEstimatorDensityMap.estim(MatrixBlock m, SparsityEstimator.OpCode op) doubleEstimatorDensityMap.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorDensityMap.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) doubleEstimatorLayeredGraph.estim(MatrixBlock m, SparsityEstimator.OpCode op) doubleEstimatorLayeredGraph.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorLayeredGraph.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) doubleEstimatorMatrixHistogram.estim(MatrixBlock m1, SparsityEstimator.OpCode op) doubleEstimatorMatrixHistogram.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorMatrixHistogram.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) doubleEstimatorRowWise.estim(MatrixBlock m1, SparsityEstimator.OpCode op) doubleEstimatorRowWise.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorRowWise.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) doubleEstimatorSample.estim(MatrixBlock m, SparsityEstimator.OpCode op) doubleEstimatorSample.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorSample.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) doubleEstimatorSampleRa.estim(MatrixBlock m, SparsityEstimator.OpCode op) doubleEstimatorSampleRa.estim(MatrixBlock m1, MatrixBlock m2) doubleEstimatorSampleRa.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) abstract doubleSparsityEstimator.estim(MatrixBlock m, SparsityEstimator.OpCode op) Estimates the output sparsity for a given unary operation.abstract doubleSparsityEstimator.estim(MatrixBlock m1, MatrixBlock m2) Estimates the output sparsity for a single matrix multiplication.abstract doubleSparsityEstimator.estim(MatrixBlock m1, MatrixBlock m2, SparsityEstimator.OpCode op) Estimates the output sparsity for a given binary operation.static longEstimationUtils.getSelfProductOutputNnz(MatrixBlock m1) This utility function computes the exact output nnz of a self matrix product without need to materialize the output.static longEstimationUtils.getSparseProductOutputNnz(MatrixBlock m1, MatrixBlock m2) voidEstimatorMatrixHistogram.MatrixHistogram.setData(MatrixBlock mb) Constructors in org.apache.sysds.hops.estim with parameters of type MatrixBlockModifierConstructorDescriptionDensityMap(MatrixBlock in, int b) DensityMap(MatrixBlock map, int rlenOrig, int clenOrig, int b, boolean scaled) LayeredGraph(MatrixBlock m, int r, int seed) MatrixHistogram(MatrixBlock in, boolean useExcepts) MMNode(MatrixBlock in) Constructor parameters in org.apache.sysds.hops.estim with type arguments of type MatrixBlock -
Uses of MatrixBlock in org.apache.sysds.hops.rewrite
Methods in org.apache.sysds.hops.rewrite with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic voidHopRewriteUtils.copyLineNumbers(MatrixBlock mb, Hop tread) static DataOpHopRewriteUtils.createTransientRead(String name, MatrixBlock mb) -
Uses of MatrixBlock in org.apache.sysds.runtime.codegen
Fields in org.apache.sysds.runtime.codegen declared as MatrixBlockMethods in org.apache.sysds.runtime.codegen that return MatrixBlockModifier and TypeMethodDescriptionSpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k, long rix) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k, long rix) abstract MatrixBlockSpoofOperator.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalars, MatrixBlock out) SpoofOperator.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalars, MatrixBlock out, int k) SpoofOuterProduct.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofOuterProduct.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int numThreads) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, boolean allocTmp, boolean aggIncr, long rix) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) Methods in org.apache.sysds.runtime.codegen with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic voidSpoofMultiAggregate.aggregatePartialResults(SpoofCellwise.AggOp[] aggOps, MatrixBlock c, MatrixBlock b) static SpoofOperator.SideInputCodegenUtils.createSideInput(MatrixBlock in) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k, long rix) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k, long rix) abstract MatrixBlockSpoofOperator.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalars, MatrixBlock out) SpoofOperator.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalars, MatrixBlock out, int k) SpoofOuterProduct.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofOuterProduct.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int numThreads) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, boolean allocTmp, boolean aggIncr, long rix) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) Method parameters in org.apache.sysds.runtime.codegen with type arguments of type MatrixBlockModifier and TypeMethodDescriptionSpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, int k) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, int k, long rix) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) SpoofCellwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k, long rix) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) SpoofMultiAggregate.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k, long rix) SpoofOperator.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalars) SpoofOperator.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalars, int k) abstract MatrixBlockSpoofOperator.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalars, MatrixBlock out) SpoofOperator.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalars, MatrixBlock out, int k) SpoofOuterProduct.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects) SpoofOuterProduct.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, int numThreads) SpoofOuterProduct.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofOuterProduct.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int numThreads) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, int k) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, boolean allocTmp, boolean aggIncr, long rix) SpoofRowwise.execute(ArrayList<MatrixBlock> inputs, ArrayList<ScalarObject> scalarObjects, MatrixBlock out, int k) static longSpoofOperator.getTotalInputNnz(ArrayList<MatrixBlock> inputs) static longSpoofOperator.getTotalInputSize(ArrayList<MatrixBlock> inputs) static booleanSpoofRowwise.hasMatrixSideInput(ArrayList<MatrixBlock> inputs) Constructors in org.apache.sysds.runtime.codegen with parameters of type MatrixBlock -
Uses of MatrixBlock in org.apache.sysds.runtime.compress
Subclasses of MatrixBlock in org.apache.sysds.runtime.compressMethods in org.apache.sysds.runtime.compress that return MatrixBlockModifier and TypeMethodDescriptionCompressedMatrixBlock.aggregateBinaryOperations(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, AggregateBinaryOperator op) CompressedMatrixBlock.aggregateBinaryOperations(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, AggregateBinaryOperator op, boolean transposeLeft, boolean transposeRight) CompressedMatrixBlock.aggregateUnaryOperations(AggregateUnaryOperator op, MatrixValue result, int blen, MatrixIndexes indexesIn, boolean inCP) CompressedMatrixBlock.append(MatrixBlock[] that, MatrixBlock ret, boolean cbind) CompressedMatrixBlock.binaryOperations(BinaryOperator op, MatrixValue thatValue, MatrixValue result) CompressedMatrixBlock.binaryOperationsInPlace(BinaryOperator op, MatrixValue thatValue) CompressedMatrixBlock.binaryOperationsLeft(BinaryOperator op, MatrixValue thatValue, MatrixValue result) CompressedMatrixBlock.chainMatrixMultOperations(MatrixBlock v, MatrixBlock w, MatrixBlock out, MapMultChain.ChainType ctype, int k) CompressedMatrixBlock.colSum()CompressedMatrixBlock.copyShallow(MatrixBlock that) CompressedMatrixBlock.ctableSeqOperations(MatrixValue thatMatrix, double thatScalar, MatrixBlock resultBlock, boolean updateClen) CompressedMatrixBlock.decompress()Decompress block into a MatrixBlockCompressedMatrixBlock.decompress(int k) Decompress block into a MatrixBlockCompressedMatrixBlock.extractTriangular(MatrixBlock ret, boolean lower, boolean diag, boolean values) CompressedMatrixBlock.getCachedDecompressed()Get the cached decompressed matrix (if it exists otherwise null).CompressedMatrixBlock.getUncompressed()CompressedMatrixBlock.getUncompressed(int k) CompressedMatrixBlock.getUncompressed(String operation) CompressedMatrixBlock.getUncompressed(String operation, int k) static MatrixBlockCompressedMatrixBlock.getUncompressed(MatrixValue mVal) static MatrixBlockCompressedMatrixBlock.getUncompressed(MatrixValue mVal, String message) static MatrixBlockCompressedMatrixBlock.getUncompressed(MatrixValue mVal, String message, int k) CompressedMatrixBlock.groupedAggOperations(MatrixValue tgt, MatrixValue wghts, MatrixValue ret, int ngroups, Operator op, int k) CompressedMatrixBlock.leftIndexingOperations(ScalarObject scalar, int rl, int cl, MatrixBlock ret, MatrixObject.UpdateType update) CompressedMatrixBlock.leftIndexingOperations(MatrixBlock rhsMatrix, int rl, int ru, int cl, int cu, MatrixBlock ret, MatrixObject.UpdateType update) CompressedMatrixBlock.merge(MatrixBlock that, boolean appendOnly, boolean par, boolean deep) CompressedMatrixBlock.quaternaryOperations(QuaternaryOperator qop, MatrixBlock um, MatrixBlock vm, MatrixBlock wm, MatrixBlock out, int k) CompressedMatrixBlock.randOperationsInPlace(RandomMatrixGenerator rgen, org.apache.commons.math3.random.Well1024a bigrand, long bSeed) CompressedMatrixBlock.randOperationsInPlace(RandomMatrixGenerator rgen, org.apache.commons.math3.random.Well1024a bigrand, long bSeed, int k) CompressedMatrixBlock.removeEmptyOperations(MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) CompressedMatrixBlock.reorgOperations(ReorgOperator op, MatrixValue ret, int startRow, int startColumn, int length) CompressedMatrixBlock.replaceOperations(MatrixValue result, double pattern, double replacement, int k) CompressedMatrixBlock.reshape(int rows, int cols, boolean byRow) CompressedMatrixBlock.rexpandOperations(MatrixBlock ret, double max, boolean rows, boolean cast, boolean ignore, int k) CompressedMatrixBlock.scalarOperations(ScalarOperator sop, MatrixValue result) CompressedMatrixBlock.seqOperationsInPlace(double from, double to, double incr) CompressedMatrixBlock.slice(int rl, int ru, int cl, int cu, boolean deep, MatrixBlock ret) CompressedMatrixBlock.sortOperations(MatrixValue weights, MatrixBlock result, int k) CompressedMatrixBlock.sparseToDense(int k) CompressedMatrixBlock.ternaryOperations(TernaryOperator op, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret) CompressedMatrixBlock.transpose(int k) CompressedMatrixBlock.transposeSelfMatrixMultOperations(MatrixBlock out, MMTSJ.MMTSJType tstype, int k) CompressedMatrixBlock.uaggouterchainOperations(MatrixBlock mbLeft, MatrixBlock mbRight, MatrixBlock mbOut, BinaryOperator bOp, AggregateUnaryOperator uaggOp) CompressedMatrixBlock.unaryOperations(UnaryOperator op, MatrixValue result) CompressedMatrixBlock.zeroOutOperations(MatrixValue result, IndexRange range) Methods in org.apache.sysds.runtime.compress that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionCompressedMatrixBlock.allocateBlockAsync()static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb) Default sequential compression with no parallelizationstatic org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CompressionSettingsBuilder compSettings) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CompressionSettingsBuilder compSettings, ACostEstimate costEstimator) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CompressionSettingsBuilder compSettings, CostEstimatorBuilder csb) The main method for compressing the input matrix.static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CompressionSettingsBuilder compSettings, WTreeRoot root) The main method for compressing the input matrix.static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, ACostEstimate costEstimator) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CostEstimatorBuilder csb) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, InstructionTypeCounter ins) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, WTreeRoot root) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, CompressionSettingsBuilder customSettings) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, ACostEstimate costEstimator) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, CostEstimatorBuilder csb) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, InstructionTypeCounter ins) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, WTreeRoot root) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, ScalarObject sf, int k, WTreeRoot root) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, MatrixBlock sf, int k, WTreeRoot root) CompressedMatrixBlock.getSoftReferenceToDecompressed()Methods in org.apache.sysds.runtime.compress with parameters of type MatrixBlockModifier and TypeMethodDescriptionCompressedMatrixBlock.aggregateBinaryOperations(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, AggregateBinaryOperator op) CompressedMatrixBlock.aggregateBinaryOperations(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, AggregateBinaryOperator op, boolean transposeLeft, boolean transposeRight) CompressedMatrixBlock.append(MatrixBlock[] that, MatrixBlock ret, boolean cbind) voidCompressedMatrixBlock.appendRowToSparse(SparseBlock dest, MatrixBlock src, int i, int rowoffset, int coloffset, boolean deep) CompressedMatrixBlock.chainMatrixMultOperations(MatrixBlock v, MatrixBlock w, MatrixBlock out, MapMultChain.ChainType ctype, int k) CompressedMatrixBlock.cmOperations(CMOperator op, MatrixBlock weights) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb) Default sequential compression with no parallelizationstatic org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CompressionSettingsBuilder compSettings) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CompressionSettingsBuilder compSettings, ACostEstimate costEstimator) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CompressionSettingsBuilder compSettings, CostEstimatorBuilder csb) The main method for compressing the input matrix.static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CompressionSettingsBuilder compSettings, WTreeRoot root) The main method for compressing the input matrix.static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, ACostEstimate costEstimator) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, CostEstimatorBuilder csb) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, InstructionTypeCounter ins) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, int k, WTreeRoot root) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, CompressionSettingsBuilder customSettings) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, ACostEstimate costEstimator) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, CostEstimatorBuilder csb) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, InstructionTypeCounter ins) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, WTreeRoot root) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, ScalarObject sf, int k, WTreeRoot root) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,CompressionStatistics> CompressedMatrixBlockFactory.compress(MatrixBlock mb, MatrixBlock sf, int k, WTreeRoot root) voidCompressedMatrixBlock.copy(int rl, int ru, int cl, int cu, MatrixBlock src, boolean awareDestNZ) CompressedMatrixBlock.copyShallow(MatrixBlock that) CompressedMatrixBlock.covOperations(COVOperator op, MatrixBlock that) CompressedMatrixBlock.covOperations(COVOperator op, MatrixBlock that, MatrixBlock weights) voidCompressedMatrixBlock.ctableOperations(Operator op, double scalar, double scalar2, CTableMap resultMap, MatrixBlock resultBlock) voidCompressedMatrixBlock.ctableOperations(Operator op, double scalar, MatrixValue that, CTableMap resultMap, MatrixBlock resultBlock) voidCompressedMatrixBlock.ctableOperations(Operator op, MatrixIndexes ix1, double scalar, boolean left, int brlen, CTableMap resultMap, MatrixBlock resultBlock) voidCompressedMatrixBlock.ctableOperations(Operator op, MatrixValue that, double scalar, boolean ignoreZeros, CTableMap resultMap, MatrixBlock resultBlock) voidCompressedMatrixBlock.ctableOperations(Operator op, MatrixValue that, MatrixValue that2, CTableMap resultMap, MatrixBlock resultBlock) CompressedMatrixBlock.ctableSeqOperations(MatrixValue thatMatrix, double thatScalar, MatrixBlock resultBlock, boolean updateClen) CompressedMatrixBlock.extractTriangular(MatrixBlock ret, boolean lower, boolean diag, boolean values) static CompressedMatrixBlockCompressedMatrixBlockFactory.genUncompressedCompressedMatrixBlock(MatrixBlock mb) Generate a CompressedMatrixBlock Object that contains a single uncompressed matrix block column group.CompressedMatrixBlock.leftIndexingOperations(ScalarObject scalar, int rl, int cl, MatrixBlock ret, MatrixObject.UpdateType update) CompressedMatrixBlock.leftIndexingOperations(MatrixBlock rhsMatrix, int rl, int ru, int cl, int cu, MatrixBlock ret, MatrixObject.UpdateType update) CompressedMatrixBlock.merge(MatrixBlock that, boolean appendOnly, boolean par, boolean deep) voidCompressedMatrixBlock.putInto(MatrixBlock target, int rowOffset, int colOffset, boolean sparseCopyShallow) CompressedMatrixBlock.quaternaryOperations(QuaternaryOperator qop, MatrixBlock um, MatrixBlock vm, MatrixBlock wm, MatrixBlock out, int k) CompressedMatrixBlock.removeEmptyOperations(MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) CompressedMatrixBlock.rexpandOperations(MatrixBlock ret, double max, boolean rows, boolean cast, boolean ignore, int k) CompressedMatrixBlock.slice(int rl, int ru, int cl, int cu, boolean deep, MatrixBlock ret) CompressedMatrixBlock.sortOperations(MatrixValue weights, MatrixBlock result, int k) CompressedMatrixBlock.ternaryOperations(TernaryOperator op, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret) static booleanCompressedMatrixBlockFactory.transposeHeuristics(int nGroups, MatrixBlock mb) CompressedMatrixBlock.transposeSelfMatrixMultOperations(MatrixBlock out, MMTSJ.MMTSJType tstype, int k) CompressedMatrixBlock.uaggouterchainOperations(MatrixBlock mbLeft, MatrixBlock mbRight, MatrixBlock mbOut, BinaryOperator bOp, AggregateUnaryOperator uaggOp) -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.bitmap
Methods in org.apache.sysds.runtime.compress.bitmap with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic ABitmapBitmapEncoder.extractBitmap(IColIndex colIndices, MatrixBlock rawBlock, boolean transposed, int estimatedNumberOfUniques, boolean sortedEntries) Generate uncompressed bitmaps for a set of columns in an uncompressed matrix block.static ABitmapBitmapEncoder.extractBitmap(IColIndex colIndices, MatrixBlock rawBlock, boolean transposed, int estimatedNumberOfUniques, boolean sortedEntries, double[] scaleFactors) Generate quantization-fused uncompressed bitmaps for a set of columns in an uncompressed matrix block.static ABitmapBitmapEncoder.extractBitmap(IColIndex colIndices, MatrixBlock rawBlock, int estimatedNumberOfUniques, CompressionSettings cs) -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.colgroup
Methods in org.apache.sysds.runtime.compress.colgroup that return MatrixBlockMethods in org.apache.sysds.runtime.compress.colgroup with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic voidColGroupUtils.addMatrixToResult(MatrixBlock tmp, MatrixBlock result, IColIndex colIndexes, int rl, int ru) ColGroupFactory.compressColGroups(MatrixBlock in, CompressedSizeInfo csi, CompressionSettings cs) The actual compression method, that handles the logic of compressing multiple columns together.ColGroupFactory.compressColGroups(MatrixBlock in, CompressedSizeInfo csi, CompressionSettings cs, int k) The actual compression method, that handles the logic of compressing multiple columns together.ColGroupFactory.compressColGroups(MatrixBlock in, CompressedSizeInfo csi, CompressionSettings cs, ACostEstimate ce, int k) static voidColGroupUtils.copyValuesColGroupMatrixBlocks(AColGroup lhs, AColGroup rhs, MatrixBlock tmpResult, MatrixBlock result) Copy values from tmpResult into correct positions of result (according to colIndexes in lhs and rhs)static AColGroupColGroupUncompressed.create(IColIndex colIndexes, MatrixBlock rawBlock, boolean transposed) Main constructor for Uncompressed ColGroup.static AColGroupColGroupUncompressed.create(MatrixBlock data) static AColGroupColGroupUncompressed.create(MatrixBlock mb, IColIndex colIndexes) Create an Uncompressed Matrix Block, where the columns are offset by col indexes.static AColGroupColGroupUncompressed.createQuantized(IColIndex colIndexes, MatrixBlock rawBlock, boolean transposed, double[] scaleFactors) Main constructor for a quantization-fused uncompressed ColGroup.static AColGroupColGroupUncompressed.createQuantized(MatrixBlock mb, IColIndex colIndexes, double[] scaleFactors) Create ana quantization-fused uncompressed Matrix Block, where the columns are offset by col indexes.static double[]ColGroupUtils.extractMostCommonValueInColumns(MatrixBlock mb) abstract voidAPreAgg.leftMMIdentityPreAggregateDense(MatrixBlock that, MatrixBlock ret, int rl, int ru, int cl, int cu) voidColGroupDDC.leftMMIdentityPreAggregateDense(MatrixBlock that, MatrixBlock ret, int rl, int ru, int cl, int cu) voidColGroupDDCLZW.leftMMIdentityPreAggregateDense(MatrixBlock that, MatrixBlock ret, int rl, int ru, int cl, int cu) voidColGroupOLE.leftMMIdentityPreAggregateDense(MatrixBlock that, MatrixBlock ret, int rl, int ru, int cl, int cu) voidColGroupRLE.leftMMIdentityPreAggregateDense(MatrixBlock that, MatrixBlock ret, int rl, int ru, int cl, int cu) voidColGroupSDCSingleZeros.leftMMIdentityPreAggregateDense(MatrixBlock that, MatrixBlock ret, int rl, int ru, int cl, int cu) voidColGroupSDCZeros.leftMMIdentityPreAggregateDense(MatrixBlock that, MatrixBlock ret, int rl, int ru, int cl, int cu) abstract voidAColGroup.leftMultByAColGroup(AColGroup lhs, MatrixBlock result, int nRows) Left side matrix multiplication with a column group that is transposed.final voidAMorphingMMColGroup.leftMultByAColGroup(AColGroup lhs, MatrixBlock result, int nRows) final voidAPreAgg.leftMultByAColGroup(AColGroup lhs, MatrixBlock result, int nRows) voidColGroupConst.leftMultByAColGroup(AColGroup lhs, MatrixBlock result, int nRows) voidColGroupEmpty.leftMultByAColGroup(AColGroup lhs, MatrixBlock c, int nRows) voidColGroupLinearFunctional.leftMultByAColGroup(AColGroup lhs, MatrixBlock result, int nRows) voidColGroupUncompressed.leftMultByAColGroup(AColGroup lhs, MatrixBlock result, int nRows) voidColGroupUncompressedArray.leftMultByAColGroup(AColGroup lhs, MatrixBlock result, int nRows) abstract voidAColGroup.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) Left multiply with this column group.final voidAMorphingMMColGroup.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) final voidASDCZero.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupConst.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupDDC.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupDDCLZW.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupDeltaDDC.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupEmpty.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupLinearFunctional.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupOLE.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupRLE.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupUncompressed.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidColGroupUncompressedArray.leftMultByMatrixNoPreAgg(MatrixBlock matrix, MatrixBlock result, int rl, int ru, int cl, int cu) voidAPreAgg.mmWithDictionary(MatrixBlock preAgg, MatrixBlock tmpRes, MatrixBlock ret, int k, int rl, int ru) final voidAPreAgg.preAggregate(MatrixBlock m, double[] preAgg, int rl, int ru) Pre aggregate a matrix block into a pre aggregate target (first step of left matrix multiplication)abstract voidAPreAgg.preAggregateDense(MatrixBlock m, double[] preAgg, int rl, int ru, int cl, int cu) Pre aggregate a dense matrix block into a pre aggregate target (first step of left matrix multiplication)voidColGroupDDC.preAggregateDense(MatrixBlock m, double[] preAgg, int rl, int ru, int cl, int cu) voidColGroupDDCLZW.preAggregateDense(MatrixBlock m, double[] preAgg, int rl, int ru, int cl, int cu) voidColGroupDeltaDDC.preAggregateDense(MatrixBlock m, double[] preAgg, int rl, int ru, int cl, int cu) voidColGroupOLE.preAggregateDense(MatrixBlock m, double[] preAgg, int rl, int ru, int cl, int cu) voidColGroupRLE.preAggregateDense(MatrixBlock m, double[] preAgg, int rl, int ru, int cl, int cu) voidColGroupSDCSingleZeros.preAggregateDense(MatrixBlock m, double[] preAgg, int rl, int ru, int cl, int cu) voidColGroupSDCZeros.preAggregateDense(MatrixBlock m, double[] preAgg, int rl, int ru, int cl, int cu) voidAColGroup.rightDecompressingMult(MatrixBlock right, MatrixBlock ret, int rl, int ru, int nRows, int crl, int cru) Right side Matrix multiplication, iterating though this column group and adding to the retvoidColGroupDDC.rightDecompressingMult(MatrixBlock right, MatrixBlock ret, int rl, int ru, int nRows, int crl, int cru) voidColGroupDeltaDDC.rightDecompressingMult(MatrixBlock right, MatrixBlock ret, int rl, int ru, int nRows, int crl, int cru) final AColGroupAColGroup.rightMultByMatrix(MatrixBlock right) Right matrix multiplication with this column group.abstract AColGroupAColGroup.rightMultByMatrix(MatrixBlock right, IColIndex allCols, int k) Right matrix multiplication with this column group.final AColGroupADictBasedColGroup.rightMultByMatrix(MatrixBlock right, IColIndex allCols, int k) ColGroupEmpty.rightMultByMatrix(MatrixBlock right, IColIndex allCols, int k) ColGroupLinearFunctional.rightMultByMatrix(MatrixBlock right, IColIndex allCols, int k) ColGroupUncompressed.rightMultByMatrix(MatrixBlock right, IColIndex allCols, int k) ColGroupUncompressedArray.rightMultByMatrix(MatrixBlock right, IColIndex allCols, int k) final voidAColGroup.selectionMultiply(MatrixBlock selection, ColGroupUtils.P[] points, MatrixBlock ret, int rl, int ru) Selection (left matrix multiply)voidColGroupDDC.sparseSelection(MatrixBlock selection, ColGroupUtils.P[] points, MatrixBlock ret, int rl, int ru) voidColGroupDDCLZW.sparseSelection(MatrixBlock selection, ColGroupUtils.P[] points, MatrixBlock ret, int rl, int ru) voidColGroupLinearFunctional.sparseSelection(MatrixBlock selection, ColGroupUtils.P[] points, MatrixBlock ret, int rl, int ru) voidColGroupSDCFOR.sparseSelection(MatrixBlock selection, ColGroupUtils.P[] points, MatrixBlock ret, int rl, int ru) voidColGroupSDCZeros.sparseSelection(MatrixBlock selection, ColGroupUtils.P[] points, MatrixBlock ret, int rl, int ru) voidColGroupUncompressed.sparseSelection(MatrixBlock selection, ColGroupUtils.P[] points, MatrixBlock ret, int rl, int ru) abstract voidAColGroup.tsmm(MatrixBlock ret, int nRows) Do a transposed self matrix multiplication on the left side t(x) %*% x.final voidAColGroupCompressed.tsmm(MatrixBlock ret, int nRows) final voidColGroupUncompressed.tsmm(MatrixBlock ret, int nRows) voidColGroupUncompressedArray.tsmm(MatrixBlock ret, int nRows) abstract voidAColGroup.tsmmAColGroup(AColGroup other, MatrixBlock result) Matrix multiply with this other column group, but: 1.final voidAMorphingMMColGroup.tsmmAColGroup(AColGroup other, MatrixBlock result) final voidAPreAgg.tsmmAColGroup(AColGroup other, MatrixBlock result) voidColGroupConst.tsmmAColGroup(AColGroup other, MatrixBlock result) voidColGroupEmpty.tsmmAColGroup(AColGroup other, MatrixBlock result) voidColGroupLinearFunctional.tsmmAColGroup(AColGroup other, MatrixBlock result) voidColGroupUncompressed.tsmmAColGroup(AColGroup lhs, MatrixBlock result) voidColGroupUncompressedArray.tsmmAColGroup(AColGroup other, MatrixBlock result) -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.colgroup.dictionary
Methods in org.apache.sysds.runtime.compress.colgroup.dictionary that return MatrixBlockMethods in org.apache.sysds.runtime.compress.colgroup.dictionary with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockDictionaryMatrixBlockDictionary.create(MatrixBlock mb) static MatrixBlockDictionaryMatrixBlockDictionary.create(MatrixBlock mb, boolean check) voidADictionary.MMDict(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidDictionary.MMDict(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidIDictionary.MMDict(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) Matrix multiplication of dictionaries note the left is this, and it is supposed to be treated as if it is transposed while it is not physically transposed.voidMatrixBlockDictionary.MMDict(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidADictionary.MMDictDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidDictionary.MMDictDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidIdentityDictionary.MMDictDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidIDictionary.MMDictDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) Matrix multiplication of dictionaries left side dense and transposed.voidMatrixBlockDictionary.MMDictDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) static voidDictLibMatrixMult.MMDicts(IDictionary left, IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) Matrix Multiply the two dictionaries, note that the left side is considered transposed but not allocated transposed making the multiplication a: t(left) %*% rightvoidADictionary.MMDictScaling(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidDictionary.MMDictScaling(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidIDictionary.MMDictScaling(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) Matrix multiplication of dictionaries, note the left is this, and it is supposed to be treated as if it is transposed while it is not physically transposed.voidMatrixBlockDictionary.MMDictScaling(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidADictionary.MMDictScalingDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidDictionary.MMDictScalingDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidIdentityDictionary.MMDictScalingDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidIDictionary.MMDictScalingDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) Matrix multiplication of dictionaries left side dense and transposed.voidMatrixBlockDictionary.MMDictScalingDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidADictionary.MMDictScalingSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidDictionary.MMDictScalingSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidIDictionary.MMDictScalingSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) Matrix multiplication of dictionaries left side sparse and transposed right side is this.voidMatrixBlockDictionary.MMDictScalingSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result, int[] scaling) voidADictionary.MMDictSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidDictionary.MMDictSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidIDictionary.MMDictSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) Matrix multiplication of dictionaries left side sparse and transposed right side is this.voidMatrixBlockDictionary.MMDictSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) static voidDictLibMatrixMult.MMDictsWithScaling(IDictionary left, IDictionary right, IColIndex leftRows, IColIndex rightColumns, MatrixBlock result, int[] counts) Matrix multiply with scaling (left side transposed)static double[]MatrixBlockDictionary.sliceColumns(MatrixBlock mb, IntArrayList selectedColumns) static voidDictLibMatrixMult.TSMMDictionaryWithScaling(IDictionary dict, int[] counts, IColIndex rows, IColIndex cols, MatrixBlock ret) Perform the full tsmm with the dictionary (allocating into the entire output matrix.)voidADictionary.TSMMToUpperTriangle(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidDictionary.TSMMToUpperTriangle(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) static voidDictLibMatrixMult.TSMMToUpperTriangle(IDictionary left, IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) Does two matrix multiplications in one go but only add to the upper triangle.voidIDictionary.TSMMToUpperTriangle(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) Matrix multiplication but allocate output in upper triangle and twice if on diagonal, note this is leftvoidMatrixBlockDictionary.TSMMToUpperTriangle(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidADictionary.TSMMToUpperTriangleDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidDictionary.TSMMToUpperTriangleDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidIDictionary.TSMMToUpperTriangleDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) Matrix multiplication but allocate output in upper triangle and twice if on diagonal, note this is rightvoidMatrixBlockDictionary.TSMMToUpperTriangleDense(double[] left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidADictionary.TSMMToUpperTriangleDenseScaling(double[] left, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) voidDictionary.TSMMToUpperTriangleDenseScaling(double[] left, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) voidIDictionary.TSMMToUpperTriangleDenseScaling(double[] left, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) Matrix multiplication but allocate output in upper triangle and twice if on diagonal, note this is rightvoidMatrixBlockDictionary.TSMMToUpperTriangleDenseScaling(double[] left, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) voidADictionary.TSMMToUpperTriangleScaling(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) voidDictionary.TSMMToUpperTriangleScaling(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) static voidDictLibMatrixMult.TSMMToUpperTriangleScaling(IDictionary left, IDictionary right, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) Does two matrix multiplications in one go but only add to the upper triangle with scaling.voidIDictionary.TSMMToUpperTriangleScaling(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) Matrix multiplication but allocate output in upper triangle and twice if on diagonal, note this is leftvoidMatrixBlockDictionary.TSMMToUpperTriangleScaling(IDictionary right, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) voidADictionary.TSMMToUpperTriangleSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidDictionary.TSMMToUpperTriangleSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidIDictionary.TSMMToUpperTriangleSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) Matrix multiplication but allocate output in upper triangle and twice if on diagonal, note this is rightvoidMatrixBlockDictionary.TSMMToUpperTriangleSparse(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, MatrixBlock result) voidADictionary.TSMMToUpperTriangleSparseScaling(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) voidDictionary.TSMMToUpperTriangleSparseScaling(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) voidIDictionary.TSMMToUpperTriangleSparseScaling(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) Matrix multiplication but allocate output in upper triangle and twice if on diagonal, note this is rightvoidMatrixBlockDictionary.TSMMToUpperTriangleSparseScaling(SparseBlock left, IColIndex rowsLeft, IColIndex colsRight, int[] scale, MatrixBlock result) voidADictionary.TSMMWithScaling(int[] counts, IColIndex rows, IColIndex cols, MatrixBlock ret) voidDictionary.TSMMWithScaling(int[] counts, IColIndex rows, IColIndex cols, MatrixBlock ret) voidIDictionary.TSMMWithScaling(int[] counts, IColIndex rows, IColIndex cols, MatrixBlock ret) Transpose self matrix multiplication with a scaling factor on each pair of values.voidMatrixBlockDictionary.TSMMWithScaling(int[] counts, IColIndex rows, IColIndex cols, MatrixBlock ret) -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.colgroup.functional
Methods in org.apache.sysds.runtime.compress.colgroup.functional with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic double[]LinearRegression.regressMatrixBlock(MatrixBlock rawBlock, IColIndex colIndexes, boolean transposed) -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.colgroup.mapping
Methods in org.apache.sysds.runtime.compress.colgroup.mapping with parameters of type MatrixBlockModifier and TypeMethodDescriptionfinal voidAMapToData.preAggregateDense(MatrixBlock m, double[] preAV, int rl, int ru, int cl, int cu) Pre aggregate a dense matrix m into pre, subject to only including a row segment and column segment. -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.colgroup.scheme
Methods in org.apache.sysds.runtime.compress.colgroup.scheme with parameters of type MatrixBlockModifier and TypeMethodDescriptionfinal AColGroupACLAScheme.encode(MatrixBlock data) final AColGroupACLAScheme.encode(MatrixBlock data, IColIndex columns) CompressionScheme.encode(MatrixBlock mb) Encode the given matrix block, it is assumed that the given MatrixBlock already fit the current scheme.CompressionScheme.encode(MatrixBlock mb, int k) Encode the given matrix block, it is assumed that the given MatrixBlock already fit the current scheme.ICLAScheme.encode(MatrixBlock data) Encode the given matrix block into the scheme provided in the instance.ICLAScheme.encode(MatrixBlock data, IColIndex columns) Encode a given matrix block into the scheme provided in the instance but overwrite what columns to use.final AColGroupACLAScheme.encodeT(MatrixBlock data) final AColGroupACLAScheme.encodeT(MatrixBlock data, IColIndex columns) ICLAScheme.encodeT(MatrixBlock data) Encode the given matrix block into the scheme provided in the instance, the input data is transposed The method is unsafe in the sense that if the encoding scheme does not fit, there is no guarantee that an error is thrown.ICLAScheme.encodeT(MatrixBlock data, IColIndex columns) Encode a given matrix block into the scheme provided in the instance but overwrite what columns to use.final ICLASchemeACLAScheme.update(MatrixBlock data) final ICLASchemeACLAScheme.update(MatrixBlock data, IColIndex columns) CompressionScheme.update(MatrixBlock mb) Update the encodings contained to also enable compression of the given mb.CompressionScheme.update(MatrixBlock mb, int k) Update the encodings contained to also enable compression of the given mb.ICLAScheme.update(MatrixBlock data) Update the encoding scheme to enable compression of the given data.ICLAScheme.update(MatrixBlock data, IColIndex columns) Update the encoding scheme to enable compression of the given data.final Pair<ICLAScheme,AColGroup> ACLAScheme.updateAndEncode(MatrixBlock data) final Pair<ICLAScheme,AColGroup> ACLAScheme.updateAndEncode(MatrixBlock data, IColIndex columns) CompressionScheme.updateAndEncode(MatrixBlock mb) CompressionScheme.updateAndEncode(MatrixBlock mb, int k) ICLAScheme.updateAndEncode(MatrixBlock data) Update and encode the given block in a single pass.ICLAScheme.updateAndEncode(MatrixBlock data, IColIndex columns) Try to update and encode in a single pass over the data.final Pair<ICLAScheme,AColGroup> ACLAScheme.updateAndEncodeT(MatrixBlock data) final Pair<ICLAScheme,AColGroup> ACLAScheme.updateAndEncodeT(MatrixBlock data, IColIndex columns) ICLAScheme.updateAndEncodeT(MatrixBlock data) Update and encode the given block in a single pass.ICLAScheme.updateAndEncodeT(MatrixBlock data, IColIndex columns) Try to update and encode in a single pass over the data.final ICLASchemeACLAScheme.updateT(MatrixBlock data) final ICLASchemeACLAScheme.updateT(MatrixBlock data, IColIndex columns) ICLAScheme.updateT(MatrixBlock data) Update the encoding scheme to enable compression of the given data.ICLAScheme.updateT(MatrixBlock data, IColIndex columns) Update the encoding scheme to enable compression of the given data. -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.cost
Methods in org.apache.sysds.runtime.compress.cost with parameters of type MatrixBlockModifier and TypeMethodDescriptionabstract doubleACostEstimate.getCost(MatrixBlock mb) Get the cost of a matrix block.doubleComputationCostEstimator.getCost(MatrixBlock mb) doubleDistinctCostEstimator.getCost(MatrixBlock mb) doubleHybridCostEstimator.getCost(MatrixBlock mb) doubleMemoryCostEstimator.getCost(MatrixBlock mb) -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.estim
Methods in org.apache.sysds.runtime.compress.estim with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic AComEstComEstFactory.createEstimator(MatrixBlock data, CompressionSettings cs, int k) Create an estimator for the input data with the given settings and parallelization degree.static AComEstComEstFactory.createEstimator(MatrixBlock data, CompressionSettings cs, int sampleSize, int k) Create an estimator for the input data with the given settings and parallelization degree.Constructors in org.apache.sysds.runtime.compress.estim with parameters of type MatrixBlockModifierConstructorDescriptionComEstExact(MatrixBlock data, CompressionSettings compSettings) ComEstSample(MatrixBlock data, CompressionSettings cs, int sampleSize, int k) CompressedSizeEstimatorSample, samples from the input data and estimates the size of the compressed matrix.ComEstSample(MatrixBlock sample, CompressionSettings cs, MatrixBlock full, int k) -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.estim.encoding
Methods in org.apache.sysds.runtime.compress.estim.encoding with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic IEncodeEncodingFactory.createFromMatrixBlock(MatrixBlock m, boolean transposed, int rowCol) Create encoding of a single specific column inside the matrix input.static IEncodeEncodingFactory.createFromMatrixBlock(MatrixBlock m, boolean transposed, int rowCol, double[] scaleFactors) Create encoding of a single specific column inside the matrix input.static IEncodeEncodingFactory.createFromMatrixBlock(MatrixBlock m, boolean transposed, IColIndex rowCols) Encode a list of columns together from the input matrix, as if it is cocoded.static IEncodeEncodingFactory.createFromMatrixBlock(MatrixBlock m, boolean transposed, IColIndex rowCols, double[] scaleFactors) static IEncodeEncodingFactory.createFromMatrixBlockDelta(MatrixBlock m, boolean transposed, IColIndex rowCols) Encode a full delta representation of the matrix input taking all rows into account.static IEncodeEncodingFactory.createFromMatrixBlockDelta(MatrixBlock m, boolean transposed, IColIndex rowCols, int sampleSize) Encode a delta representation of the matrix input taking the first "sampleSize" rows into account. -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.io
Fields in org.apache.sysds.runtime.compress.io declared as MatrixBlockMethods in org.apache.sysds.runtime.compress.io that return MatrixBlockModifier and TypeMethodDescriptionCompressUnwrap.call(CompressedWriteBlock arg0) CompressedWriteBlock.get()static MatrixBlockReaderCompressed.readCompressedMatrixFromHDFS(String fname, long rlen, long clen, int blen) ReaderCompressed.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderCompressed.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) Methods in org.apache.sysds.runtime.compress.io that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> Methods in org.apache.sysds.runtime.compress.io with parameters of type MatrixBlockModifier and TypeMethodDescriptionCompressWrap.call(MatrixBlock arg0) static voidWriterCompressed.writeCompressedMatrixToHDFS(MatrixBlock src, String fname) static voidWriterCompressed.writeCompressedMatrixToHDFS(MatrixBlock src, String fname, int blen) static voidWriterCompressed.writeCompressedMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) voidWriterCompressed.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) Method parameters in org.apache.sysds.runtime.compress.io with type arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic voidWriterCompressed.writeRDDToHDFS(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> src, String path) static voidWriterCompressed.writeRDDToHDFS(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> src, String path, int blen, DataCharacteristics mc) Constructors in org.apache.sysds.runtime.compress.io with parameters of type MatrixBlock -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.lib
Methods in org.apache.sysds.runtime.compress.lib that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockCLALibAggTernaryOp.agg(MatrixBlock m1, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret, AggregateTernaryOperator op, boolean inCP) static MatrixBlockCLALibCompAgg.aggregateUnary(CompressedMatrixBlock inputMatrix, MatrixBlock result, AggregateUnaryOperator op, int blen, MatrixIndexes indexesIn, boolean inCP) static MatrixBlockCLALibBinaryCellOp.binaryOperationsLeft(BinaryOperator op, CompressedMatrixBlock m1, MatrixBlock that) static MatrixBlockCLALibBinaryCellOp.binaryOperationsRight(BinaryOperator op, CompressedMatrixBlock m1, MatrixBlock that) static MatrixBlockCLALibCBind.cbind(MatrixBlock left, MatrixBlock[] right, int k) static MatrixBlockCLALibCBind.cbind(MatrixBlock left, MatrixBlock right, int k) static MatrixBlockCLALibStack.combine(Map<MatrixIndexes, MatrixBlock> m, Map<Integer, List<IDictionary>> d, int k) Combine the map of index matrix blocks into a single MatrixBlock.static MatrixBlockCLALibStack.combine(Map<MatrixIndexes, MatrixBlock> m, Map<Integer, List<IDictionary>> d, int rlen, int clen, int blen, int k) Alternative read, when dimensions are known.static MatrixBlockCLALibDecompress.decompress(CompressedMatrixBlock cmb, int k) static MatrixBlockCLALibLeftMultBy.leftMultByMatrix(CompressedMatrixBlock right, MatrixBlock left, MatrixBlock ret, int k) Left multiplication with two CompressedMatrixBlock following the equation: ret = left %*% rightstatic MatrixBlockCLALibLeftMultBy.leftMultByMatrixTransposed(CompressedMatrixBlock right, CompressedMatrixBlock left, MatrixBlock ret, int k) Left multiplication with two CompressedMatrixBlock following the equation:static MatrixBlockCLALibLeftMultBy.leftMultByMatrixTransposed(CompressedMatrixBlock right, MatrixBlock left, MatrixBlock ret, int k) Left multiplication with a CompressedMatrixBlock on the right following the equation:static MatrixBlockCLALibTSMM.leftMultByTransposeSelf(CompressedMatrixBlock cmb, int k) static MatrixBlockCLALibTSMM.leftMultByTransposeSelf(CompressedMatrixBlock cmb, MatrixBlock ret, int k) Self left Matrix multiplication (tsmm) t(x) %*% xstatic MatrixBlockCLALibSelectionMult.leftSelection(CompressedMatrixBlock right, MatrixBlock left, MatrixBlock ret, int k) Left selection where the left matrix is sparse with a max 1 non zero per row and that non zero is a 1.static MatrixBlockCLALibMatrixMult.matrixMultiply(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) static MatrixBlockCLALibMatrixMult.matrixMultiply(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k, boolean transposeLeft, boolean transposeRight) static MatrixBlockCLALibMerge.merge(MatrixBlock a, MatrixBlock b, boolean appendOnly, boolean par, boolean deep) Merge b into a, if possible otherwisestatic MatrixBlockCLALibMMChain.mmChain(CompressedMatrixBlock x, MatrixBlock v, MatrixBlock w, MatrixBlock out, MapMultChain.ChainType ctype, int k) Support compressed MM chain operation to fuse the following cases :static MatrixBlockCLALibReorg.reorg(CompressedMatrixBlock cmb, ReorgOperator op, MatrixBlock ret, int startRow, int startColumn, int length) static MatrixBlockCLALibReplace.replace(CompressedMatrixBlock in, MatrixBlock out, double pattern, double replacement, int k) static MatrixBlockCLALibReshape.reshape(CompressedMatrixBlock in, int rows, int cols, boolean rowwise) static MatrixBlockCLALibReshape.reshape(CompressedMatrixBlock in, int rows, int cols, boolean rowwise, int k) static MatrixBlockCLALibRexpand.rexpand(int seqHeight, MatrixBlock A) static MatrixBlockCLALibRexpand.rexpand(int seqHeight, MatrixBlock A, int nColOut) static MatrixBlockCLALibRexpand.rexpand(int seqHeight, MatrixBlock A, int nColOut, int k) static MatrixBlockCLALibRexpand.rexpand(CompressedMatrixBlock in, MatrixBlock ret, double max, boolean rows, boolean cast, boolean ignore, int k) static MatrixBlockCLALibRightMultBy.rightMultByMatrix(CompressedMatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) static MatrixBlockCLALibRightMultBy.rightMultByMatrix(CompressedMatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k, boolean allowOverlap) static MatrixBlockCLALibRemoveEmpty.rmempty(CompressedMatrixBlock in, MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) CP rmempty operation (single input, single output matrix)static MatrixBlockCLALibScalar.scalarOperations(ScalarOperator sop, CompressedMatrixBlock m1, MatrixValue result) static MatrixBlockCLALibSlice.slice(CompressedMatrixBlock cmb, int rl, int ru, int cl, int cu, boolean deep) static MatrixBlockCLALibSlice.sliceRowsCompressed(CompressedMatrixBlock cmb, int rl, int ru) static MatrixBlockCLALibSort.sort(CompressedMatrixBlock mb, SortIndex fn) Sort (order) a compressed matrix in place of theorderbuilt-in, while keeping the result compressed.static MatrixBlockCLALibSort.sort(CompressedMatrixBlock mb, MatrixValue weights, MatrixBlock result, int k) Compute the sorted value/weight table used by the quantile/median/IQM operations (thesort/ qsort lop), exploiting compression to sort the few distinct values instead of all rows.static MatrixBlockCLALibTernaryOp.ternaryOperations(TernaryOperator op, MatrixBlock m1, MatrixBlock m2, MatrixBlock m3) static MatrixBlockCLALibUnary.unaryOperations(CompressedMatrixBlock m, UnaryOperator op, MatrixValue result) Methods in org.apache.sysds.runtime.compress.lib that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic org.apache.commons.lang3.tuple.Pair<MatrixBlock,FrameBlock> CLALibBinCompress.binCompress(CacheBlock<?> X, MatrixBlock d, int k) static List<MatrixBlock>CLALibSlice.sliceBlocks(CompressedMatrixBlock cmb, int blen, int k) Slice blocks of compressed matrices from the compressed representation.Methods in org.apache.sysds.runtime.compress.lib with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockCLALibAggTernaryOp.agg(MatrixBlock m1, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret, AggregateTernaryOperator op, boolean inCP) static MatrixBlockCLALibCompAgg.aggregateUnary(CompressedMatrixBlock inputMatrix, MatrixBlock result, AggregateUnaryOperator op, int blen, MatrixIndexes indexesIn, boolean inCP) static MatrixBlockCLALibBinaryCellOp.binaryOperationsLeft(BinaryOperator op, CompressedMatrixBlock m1, MatrixBlock that) static MatrixBlockCLALibBinaryCellOp.binaryOperationsRight(BinaryOperator op, CompressedMatrixBlock m1, MatrixBlock that) static org.apache.commons.lang3.tuple.Pair<MatrixBlock,FrameBlock> CLALibBinCompress.binCompress(CacheBlock<?> X, MatrixBlock d, int k) static MatrixBlockCLALibCBind.cbind(MatrixBlock left, MatrixBlock[] right, int k) static MatrixBlockCLALibCBind.cbind(MatrixBlock left, MatrixBlock right, int k) static voidCLALibDecompress.decompressDense(MatrixBlock ret, List<AColGroup> groups, double[] constV, double eps, int k, boolean overlapping) static voidCLALibDecompress.decompressTo(CompressedMatrixBlock cmb, MatrixBlock ret, int rowOffset, int colOffset, int k, boolean countNNz) static voidCLALibDecompress.decompressTo(CompressedMatrixBlock cmb, MatrixBlock ret, int rowOffset, int colOffset, int k, boolean countNNz, boolean reset) static booleanCLALibSelectionMult.isSelectionMatrix(MatrixBlock mb) Analyze if the given matrix is a selection matrix if on the left side of a matrix multiplication.static MatrixBlockCLALibLeftMultBy.leftMultByMatrix(CompressedMatrixBlock right, MatrixBlock left, MatrixBlock ret, int k) Left multiplication with two CompressedMatrixBlock following the equation: ret = left %*% rightstatic MatrixBlockCLALibLeftMultBy.leftMultByMatrixTransposed(CompressedMatrixBlock right, CompressedMatrixBlock left, MatrixBlock ret, int k) Left multiplication with two CompressedMatrixBlock following the equation:static MatrixBlockCLALibLeftMultBy.leftMultByMatrixTransposed(CompressedMatrixBlock right, MatrixBlock left, MatrixBlock ret, int k) Left multiplication with a CompressedMatrixBlock on the right following the equation:static MatrixBlockCLALibTSMM.leftMultByTransposeSelf(CompressedMatrixBlock cmb, MatrixBlock ret, int k) Self left Matrix multiplication (tsmm) t(x) %*% xstatic MatrixBlockCLALibSelectionMult.leftSelection(CompressedMatrixBlock right, MatrixBlock left, MatrixBlock ret, int k) Left selection where the left matrix is sparse with a max 1 non zero per row and that non zero is a 1.static MatrixBlockCLALibMatrixMult.matrixMultiply(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) static MatrixBlockCLALibMatrixMult.matrixMultiply(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k, boolean transposeLeft, boolean transposeRight) static MatrixBlockCLALibMerge.merge(MatrixBlock a, MatrixBlock b, boolean appendOnly, boolean par, boolean deep) Merge b into a, if possible otherwisestatic MatrixBlockCLALibMMChain.mmChain(CompressedMatrixBlock x, MatrixBlock v, MatrixBlock w, MatrixBlock out, MapMultChain.ChainType ctype, int k) Support compressed MM chain operation to fuse the following cases :static MatrixBlockCLALibReorg.reorg(CompressedMatrixBlock cmb, ReorgOperator op, MatrixBlock ret, int startRow, int startColumn, int length) static MatrixBlockCLALibReplace.replace(CompressedMatrixBlock in, MatrixBlock out, double pattern, double replacement, int k) static MatrixBlockCLALibRexpand.rexpand(int seqHeight, MatrixBlock A) static MatrixBlockCLALibRexpand.rexpand(int seqHeight, MatrixBlock A, int nColOut) static MatrixBlockCLALibRexpand.rexpand(int seqHeight, MatrixBlock A, int nColOut, int k) static MatrixBlockCLALibRexpand.rexpand(CompressedMatrixBlock in, MatrixBlock ret, double max, boolean rows, boolean cast, boolean ignore, int k) static MatrixBlockCLALibRightMultBy.rightMultByMatrix(CompressedMatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) static MatrixBlockCLALibRightMultBy.rightMultByMatrix(CompressedMatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k, boolean allowOverlap) static MatrixBlockCLALibRemoveEmpty.rmempty(CompressedMatrixBlock in, MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) CP rmempty operation (single input, single output matrix)static double[]CLALibLeftMultBy.rowSum(MatrixBlock mb, int rl, int ru, int cl, int cu) static MatrixBlockCLALibSort.sort(CompressedMatrixBlock mb, MatrixValue weights, MatrixBlock result, int k) Compute the sorted value/weight table used by the quantile/median/IQM operations (thesort/ qsort lop), exploiting compression to sort the few distinct values instead of all rows.static MatrixBlockCLALibTernaryOp.ternaryOperations(TernaryOperator op, MatrixBlock m1, MatrixBlock m2, MatrixBlock m3) Method parameters in org.apache.sysds.runtime.compress.lib with type arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockCLALibStack.combine(Map<MatrixIndexes, MatrixBlock> m, Map<Integer, List<IDictionary>> d, int k) Combine the map of index matrix blocks into a single MatrixBlock.static MatrixBlockCLALibStack.combine(Map<MatrixIndexes, MatrixBlock> m, Map<Integer, List<IDictionary>> d, int rlen, int clen, int blen, int k) Alternative read, when dimensions are known. -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.plan
Methods in org.apache.sysds.runtime.compress.plan with parameters of type MatrixBlockModifier and TypeMethodDescriptionIPlanEncode.encode(MatrixBlock in) Encode a given matrix block subject to the plan given, while erroring out in case of breaking compressionNaivePlanEncode.encode(MatrixBlock in) voidIPlanEncode.expandPlan(MatrixBlock in) Given a block expand the plan to enable encoding of the given block.voidNaivePlanEncode.expandPlan(MatrixBlock in) voidNaivePlanEncode.expandPlanMultiThread(MatrixBlock in) voidNaivePlanEncode.expandPlanSingleThread(MatrixBlock in) -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.readers
Methods in org.apache.sysds.runtime.compress.readers with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic ReaderColumnSelectionReaderColumnSelection.createDeltaReader(MatrixBlock rawBlock, IColIndex colIndices, boolean transposed) Create a reader of the matrix block that computes delta values (current row - previous row) on-the-fly.static ReaderColumnSelectionReaderColumnSelection.createDeltaReader(MatrixBlock rawBlock, IColIndex colIndices, boolean transposed, int rl, int ru) Create a reader of the matrix block that computes delta values (current row - previous row) on-the-fly.static ReaderColumnSelectionReaderColumnSelection.createQuantizedReader(MatrixBlock rawBlock, IColIndex colIndices, boolean transposed, double[] scaleFactors) Create a reader of the matrix block that directly reads quantized values using scale factors.static ReaderColumnSelectionReaderColumnSelection.createQuantizedReader(MatrixBlock rawBlock, IColIndex colIndices, boolean transposed, int rl, int ru, double[] scaleFactors) Create a reader of the matrix block that directly reads quantized values using scale factors.static ReaderColumnSelectionReaderColumnSelection.createReader(MatrixBlock rawBlock, IColIndex colIndices, boolean transposed) Create a reader of the matrix block that is able to iterate though all the rows and return as dense double arrays.static ReaderColumnSelectionReaderColumnSelection.createReader(MatrixBlock rawBlock, IColIndex colIndices, boolean transposed, int rl, int ru) Create a reader of the matrix block that is able to iterate though all the rows and return as dense double arrays. -
Uses of MatrixBlock in org.apache.sysds.runtime.compress.utils
Methods in org.apache.sysds.runtime.compress.utils that return MatrixBlockModifier and TypeMethodDescriptionCompressRDDClean.call(MatrixBlock mb) static MatrixBlockUtil.extractValues(double[] v, IColIndex colIndexes) static MatrixBlockUtil.matrixBlockFromDenseArray(double[] values, int nCol) static MatrixBlockUtil.matrixBlockFromDenseArray(double[] values, int nCol, boolean check) Methods in org.apache.sysds.runtime.compress.utils with parameters of type MatrixBlock -
Uses of MatrixBlock in org.apache.sysds.runtime.controlprogram.caching
Methods in org.apache.sysds.runtime.controlprogram.caching that return MatrixBlockModifier and TypeMethodDescriptionMatrixObject.readMatrixPartition(IndexRange pred) NOTE: for reading matrix partitions, we could cache (in its real sense) the read block with soft references (no need for eviction, as partitioning only applied for read-only matrices).Methods in org.apache.sysds.runtime.controlprogram.caching with parameters of type MatrixBlockConstructors in org.apache.sysds.runtime.controlprogram.caching with parameters of type MatrixBlockModifierConstructorDescriptionMatrixObject(Types.ValueType vt, String file, MetaData mtd, MatrixBlock data) Constructor that takes the value type, HDFS filename and associated metadata and a MatrixBlock used for creation after serialization -
Uses of MatrixBlock in org.apache.sysds.runtime.controlprogram.context
Methods in org.apache.sysds.runtime.controlprogram.context that return MatrixBlockModifier and TypeMethodDescriptionMatrixObjectFuture.acquireRead()ExecutionContext.getMatrixInput(String varName) Pins a matrix variable into memory and returns the internal matrix block.ExecutionContext.getMatrixInput(CPOperand input) static MatrixBlockSparkExecutionContext.toMatrixBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> rdd, int rlen, int clen, int blen, long nnz) Utility method for creating a single matrix block out of a binary block RDD.static MatrixBlockSparkExecutionContext.toMatrixBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixCell> rdd, int rlen, int clen, long nnz) Utility method for creating a single matrix block out of a binary cell RDD.static MatrixBlockSparkExecutionContext.toMatrixBlock(RDDObject rdd, int rlen, int clen, int blen, long nnz) This method is a generic abstraction for calls from the buffer pool.static MatrixBlockSparkExecutionContext.toMatrixBlock(RDDObject rdd, int rlen, int clen, long nnz) Methods in org.apache.sysds.runtime.controlprogram.context that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionorg.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkExecutionContext.getBinaryMatrixBlockRDDHandleForVariable(String varname) Spark instructions should call this for all matrix inputs except broadcast variables.org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkExecutionContext.getBinaryMatrixBlockRDDHandleForVariable(String varname, int numParts, boolean inclEmpty) SparkExecutionContext.getBroadcastForMatrixObject(MatrixObject mo) SparkExecutionContext.getBroadcastForVariable(String varname) ExecutionContext.getMatrixInputs(CPOperand[] inputs) ExecutionContext.getMatrixInputs(CPOperand[] inputs, boolean includeList) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkExecutionContext.toMatrixJavaPairRDD(org.apache.spark.api.java.JavaSparkContext sc, MatrixBlock src, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkExecutionContext.toMatrixJavaPairRDD(org.apache.spark.api.java.JavaSparkContext sc, MatrixBlock src, int blen, int numParts, boolean inclEmpty) static PartitionedBlock<MatrixBlock>SparkExecutionContext.toPartitionedMatrixBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> rdd, int rlen, int clen, int blen, long nnz) Methods in org.apache.sysds.runtime.controlprogram.context with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixObjectExecutionContext.createMatrixObject(MatrixBlock mb) voidExecutionContext.setMatrixOutput(String varName, MatrixBlock outputData) voidExecutionContext.setMatrixOutput(String varName, MatrixBlock outputData, MatrixObject.UpdateType flag) voidExecutionContext.setMatrixOutputAndLineage(String varName, MatrixBlock outputData, LineageItem li) voidExecutionContext.setMatrixOutputAndLineage(CPOperand var, MatrixBlock outputData, LineageItem li) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkExecutionContext.toMatrixJavaPairRDD(org.apache.spark.api.java.JavaSparkContext sc, MatrixBlock src, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkExecutionContext.toMatrixJavaPairRDD(org.apache.spark.api.java.JavaSparkContext sc, MatrixBlock src, int blen, int numParts, boolean inclEmpty) Method parameters in org.apache.sysds.runtime.controlprogram.context with type arguments of type MatrixBlockModifier and TypeMethodDescriptionvoidExecutionContext.setMatrixOutput(String varName, Future<MatrixBlock> fmb) voidExecutionContext.setMatrixOutputAndLineage(String varName, Future<MatrixBlock> fmb, LineageItem li) static MatrixBlockSparkExecutionContext.toMatrixBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> rdd, int rlen, int clen, int blen, long nnz) Utility method for creating a single matrix block out of a binary block RDD.static PartitionedBlock<MatrixBlock>SparkExecutionContext.toPartitionedMatrixBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> rdd, int rlen, int clen, int blen, long nnz) Constructor parameters in org.apache.sysds.runtime.controlprogram.context with type arguments of type MatrixBlockModifierConstructorDescriptionMatrixObjectFuture(Types.ValueType vt, String file, Future<MatrixBlock> fmb) MatrixObjectFuture(MatrixObject mo, Future<MatrixBlock> fmb) -
Uses of MatrixBlock in org.apache.sysds.runtime.controlprogram.federated
Methods in org.apache.sysds.runtime.controlprogram.federated that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockFederationUtils.aggAdd(Future<FederatedResponse>[] ffr) static MatrixBlockFederationUtils.aggMatrix(AggregateUnaryOperator aop, Future<FederatedResponse>[] ffr, Future<FederatedResponse>[] meanFfr, FederationMap map) static MatrixBlockFederationUtils.aggMatrix(AggregateUnaryOperator aop, Future<FederatedResponse>[] ffr, FederationMap map) static MatrixBlockFederationUtils.aggMean(Future<FederatedResponse>[] ffr, FederationMap map) static MatrixBlockFederationUtils.aggMinMax(Future<FederatedResponse>[] ffr, boolean isMin, boolean isScalar, Optional<FTypes.FType> fedType) static MatrixBlockFederationUtils.aggMinMaxIndex(Future<FederatedResponse>[] ffr, boolean isMin, FederationMap map) static MatrixBlockFederationUtils.aggProd(Future<FederatedResponse>[] ffr, FederationMap fedMap, AggregateUnaryOperator aop) static MatrixBlockFederationUtils.aggregateResponses(List<org.apache.commons.lang3.tuple.Pair<FederatedRange, Future<FederatedResponse>>> readResponses) Aggregate partially aggregated data from federated workers by adding values with the same index in different federated locations.static MatrixBlockFederationUtils.aggVar(Future<FederatedResponse>[] ffr, Future<FederatedResponse>[] meanFfr, FederationMap map, boolean isRowAggregate, boolean isScalar) static MatrixBlockFederationUtils.bind(Future<FederatedResponse>[] ffr, boolean cbind) static MatrixBlockFederationUtils.bindResponses(List<org.apache.commons.lang3.tuple.Pair<FederatedRange, Future<FederatedResponse>>> readResponses, long[] dims) Bind data from federated workers based on non-overlapping federated ranges.static MatrixBlock[]FederationUtils.getResults(Future<FederatedResponse>[] ffr) -
Uses of MatrixBlock in org.apache.sysds.runtime.controlprogram.paramserv
Methods in org.apache.sysds.runtime.controlprogram.paramserv that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockParamservUtils.cbindMatrix(MatrixBlock left, MatrixBlock right) static MatrixBlockParamservUtils.generatePermutation(int numEntries, long seed) Generate the permutationstatic MatrixBlockParamservUtils.generateReplicationMatrix(int nsamples, int nrows, long seed) Generates a matrix which when left multiplied with the input matrix will replicate n data rowsstatic MatrixBlockParamservUtils.generateSubsampleMatrix(int nsamples, int nrows, long seed) Generates a matrix which when left multiplied with the input matrix will subsamplestatic MatrixBlockParamservUtils.sliceMatrixBlock(MatrixBlock mb, long rl, long rh) Slice the matrix block and return a matrix block (used in spark)Methods in org.apache.sysds.runtime.controlprogram.paramserv that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic org.apache.spark.api.java.JavaPairRDD<Long,scala.Tuple2<MatrixBlock, MatrixBlock>> SparkParamservUtils.assembleTrainingData(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> featuresRDD, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> labelsRDD) Assemble the matrix of features and labels according to the rowIDstatic org.apache.spark.api.java.JavaPairRDD<Long,scala.Tuple2<MatrixBlock, MatrixBlock>> SparkParamservUtils.assembleTrainingData(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> featuresRDD, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> labelsRDD) Assemble the matrix of features and labels according to the rowIDstatic org.apache.spark.api.java.JavaPairRDD<Integer,scala.Tuple2<MatrixBlock, MatrixBlock>> SparkParamservUtils.doPartitionOnSpark(SparkExecutionContext sec, MatrixObject features, MatrixObject labels, Statement.PSScheme scheme, int workerNum) static org.apache.spark.api.java.JavaPairRDD<Integer,scala.Tuple2<MatrixBlock, MatrixBlock>> SparkParamservUtils.doPartitionOnSpark(SparkExecutionContext sec, MatrixObject features, MatrixObject labels, Statement.PSScheme scheme, int workerNum) Methods in org.apache.sysds.runtime.controlprogram.paramserv with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockParamservUtils.cbindMatrix(MatrixBlock left, MatrixBlock right) static MatrixObjectParamservUtils.newMatrixObject(MatrixBlock mb) static MatrixObjectParamservUtils.newMatrixObject(MatrixBlock mb, boolean cleanup) static MatrixBlockParamservUtils.sliceMatrixBlock(MatrixBlock mb, long rl, long rh) Slice the matrix block and return a matrix block (used in spark)Method parameters in org.apache.sysds.runtime.controlprogram.paramserv with type arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic org.apache.spark.api.java.JavaPairRDD<Long,scala.Tuple2<MatrixBlock, MatrixBlock>> SparkParamservUtils.assembleTrainingData(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> featuresRDD, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> labelsRDD) Assemble the matrix of features and labels according to the rowIDvoidSparkPSWorker.call(scala.Tuple2<Integer, scala.Tuple2<MatrixBlock, MatrixBlock>> input) voidSparkPSWorker.call(scala.Tuple2<Integer, scala.Tuple2<MatrixBlock, MatrixBlock>> input) -
Uses of MatrixBlock in org.apache.sysds.runtime.controlprogram.paramserv.dp
Methods in org.apache.sysds.runtime.controlprogram.paramserv.dp that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockDRRLocalScheme.removeEmpty(MatrixBlock mb, int k, int workerId) Methods in org.apache.sysds.runtime.controlprogram.paramserv.dp that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionscala.Tuple2<Integer,scala.Tuple2<MatrixBlock, MatrixBlock>> DataPartitionerSparkAggregator.call(scala.Tuple2<Integer, LinkedList<scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>>>> input) Row-wise combine the matrixscala.Tuple2<Integer,scala.Tuple2<MatrixBlock, MatrixBlock>> DataPartitionerSparkAggregator.call(scala.Tuple2<Integer, LinkedList<scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>>>> input) Row-wise combine the matrixIterator<scala.Tuple2<Integer,scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>>>> DataPartitionerSparkMapper.call(scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>> input) Do data partitioningIterator<scala.Tuple2<Integer,scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>>>> DataPartitionerSparkMapper.call(scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>> input) Do data partitioningstatic List<MatrixBlock>DCLocalScheme.partition(int k, MatrixBlock mb) static List<MatrixBlock>ORLocalScheme.partition(int k, MatrixBlock mb, List<MatrixBlock> permutations) Methods in org.apache.sysds.runtime.controlprogram.paramserv.dp with parameters of type MatrixBlockModifier and TypeMethodDescriptionabstract DataPartitionLocalScheme.ResultDataPartitionLocalScheme.doPartitioning(int workersNum, MatrixBlock features, MatrixBlock labels) abstract org.apache.sysds.runtime.controlprogram.paramserv.dp.DataPartitionSparkScheme.ResultDataPartitionSparkScheme.doPartitioning(int numWorkers, int rblkID, MatrixBlock features, MatrixBlock labels) DCLocalScheme.doPartitioning(int workersNum, MatrixBlock features, MatrixBlock labels) org.apache.sysds.runtime.controlprogram.paramserv.dp.DataPartitionSparkScheme.ResultDCSparkScheme.doPartitioning(int numWorkers, int rblkID, MatrixBlock features, MatrixBlock labels) DRLocalScheme.doPartitioning(int workersNum, MatrixBlock features, MatrixBlock labels) DRRLocalScheme.doPartitioning(int workersNum, MatrixBlock features, MatrixBlock labels) org.apache.sysds.runtime.controlprogram.paramserv.dp.DataPartitionSparkScheme.ResultDRRSparkScheme.doPartitioning(int numWorkers, int rblkID, MatrixBlock features, MatrixBlock labels) org.apache.sysds.runtime.controlprogram.paramserv.dp.DataPartitionSparkScheme.ResultDRSparkScheme.doPartitioning(int numWorkers, int rblkID, MatrixBlock features, MatrixBlock labels) LocalDataPartitioner.doPartitioning(int workersNum, MatrixBlock features, MatrixBlock labels) ORLocalScheme.doPartitioning(int workersNum, MatrixBlock features, MatrixBlock labels) org.apache.sysds.runtime.controlprogram.paramserv.dp.DataPartitionSparkScheme.ResultORSparkScheme.doPartitioning(int numWorkers, int rblkID, MatrixBlock features, MatrixBlock labels) org.apache.sysds.runtime.controlprogram.paramserv.dp.DataPartitionSparkScheme.ResultSparkDataPartitioner.doPartitioning(int numWorkers, MatrixBlock features, MatrixBlock labels, long rowID) static List<MatrixBlock>DCLocalScheme.partition(int k, MatrixBlock mb) static List<MatrixBlock>ORLocalScheme.partition(int k, MatrixBlock mb, List<MatrixBlock> permutations) static MatrixBlockDRRLocalScheme.removeEmpty(MatrixBlock mb, int k, int workerId) Method parameters in org.apache.sysds.runtime.controlprogram.paramserv.dp with type arguments of type MatrixBlockModifier and TypeMethodDescriptionscala.Tuple2<Integer,scala.Tuple2<MatrixBlock, MatrixBlock>> DataPartitionerSparkAggregator.call(scala.Tuple2<Integer, LinkedList<scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>>>> input) Row-wise combine the matrixscala.Tuple2<Integer,scala.Tuple2<MatrixBlock, MatrixBlock>> DataPartitionerSparkAggregator.call(scala.Tuple2<Integer, LinkedList<scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>>>> input) Row-wise combine the matrixIterator<scala.Tuple2<Integer,scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>>>> DataPartitionerSparkMapper.call(scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>> input) Do data partitioningIterator<scala.Tuple2<Integer,scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>>>> DataPartitionerSparkMapper.call(scala.Tuple2<Long, scala.Tuple2<MatrixBlock, MatrixBlock>> input) Do data partitioningstatic List<MatrixBlock>ORLocalScheme.partition(int k, MatrixBlock mb, List<MatrixBlock> permutations) -
Uses of MatrixBlock in org.apache.sysds.runtime.controlprogram.parfor
Methods in org.apache.sysds.runtime.controlprogram.parfor that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockDataPartitioner.createReuseMatrixBlock(ParForProgramBlock.PDataPartitionFormat dpf, int rows, int cols) Methods in org.apache.sysds.runtime.controlprogram.parfor that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionscala.Tuple2<MatrixIndexes,MatrixBlock> ResultMergeRemoteSparkWCompare.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<Iterable<MatrixBlock>, MatrixBlock>> arg) Method parameters in org.apache.sysds.runtime.controlprogram.parfor with type arguments of type MatrixBlockModifier and TypeMethodDescriptionDataPartitionerRemoteSparkMapper.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) scala.Tuple2<MatrixIndexes,MatrixBlock> ResultMergeRemoteSparkWCompare.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<Iterable<MatrixBlock>, MatrixBlock>> arg) scala.Tuple2<MatrixIndexes,MatrixBlock> ResultMergeRemoteSparkWCompare.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<Iterable<MatrixBlock>, MatrixBlock>> arg) -
Uses of MatrixBlock in org.apache.sysds.runtime.controlprogram.parfor.util
Fields in org.apache.sysds.runtime.controlprogram.parfor.util declared as MatrixBlockConstructors in org.apache.sysds.runtime.controlprogram.parfor.util with parameters of type MatrixBlock -
Uses of MatrixBlock in org.apache.sysds.runtime.data
Methods in org.apache.sysds.runtime.data with parameters of type MatrixBlockModifier and TypeMethodDescriptionvoidBasicTensorBlock.set(MatrixBlock other) voidTensorBlock.set(MatrixBlock other) -
Uses of MatrixBlock in org.apache.sysds.runtime.einsum
Methods in org.apache.sysds.runtime.einsum that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockEOpNodeFuse.compute(EOpNodeFuse.EinsumRewriteType rewriteType, List<MatrixBlock> ABsInput, List<MatrixBlock> mbBAs, List<MatrixBlock> mbBs, List<MatrixBlock> mbAs, List<MatrixBlock> mbAZs, Double scalar, int numThreads) abstract MatrixBlockEOpNode.computeEOpNode(List<MatrixBlock> inputs, int numOfThreads, org.apache.commons.logging.Log LOG) EOpNodeBinary.computeEOpNode(List<MatrixBlock> inputs, int numThreads, org.apache.commons.logging.Log LOG) EOpNodeData.computeEOpNode(List<MatrixBlock> inputs, int numOfThreads, org.apache.commons.logging.Log LOG) EOpNodeFuse.computeEOpNode(List<MatrixBlock> inputs, int numThreads, org.apache.commons.logging.Log LOG) EOpNodeUnary.computeEOpNode(List<MatrixBlock> inputs, int numOfThreads, org.apache.commons.logging.Log LOG) Method parameters in org.apache.sysds.runtime.einsum with type arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockEOpNodeFuse.compute(EOpNodeFuse.EinsumRewriteType rewriteType, List<MatrixBlock> ABsInput, List<MatrixBlock> mbBAs, List<MatrixBlock> mbBs, List<MatrixBlock> mbAs, List<MatrixBlock> mbAZs, Double scalar, int numThreads) abstract MatrixBlockEOpNode.computeEOpNode(List<MatrixBlock> inputs, int numOfThreads, org.apache.commons.logging.Log LOG) EOpNodeBinary.computeEOpNode(List<MatrixBlock> inputs, int numThreads, org.apache.commons.logging.Log LOG) EOpNodeData.computeEOpNode(List<MatrixBlock> inputs, int numOfThreads, org.apache.commons.logging.Log LOG) EOpNodeFuse.computeEOpNode(List<MatrixBlock> inputs, int numThreads, org.apache.commons.logging.Log LOG) EOpNodeUnary.computeEOpNode(List<MatrixBlock> inputs, int numOfThreads, org.apache.commons.logging.Log LOG) static EinsumContextEinsumContext.getEinsumContext(String eqStr, List<MatrixBlock> inputs) -
Uses of MatrixBlock in org.apache.sysds.runtime.frame.data
Methods in org.apache.sysds.runtime.frame.data with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic FrameBlockFrameBlock.convertToFrameBlock(MatrixBlock mb, Types.ValueType[] schema, int k) FrameBlock.invalidByLength(MatrixBlock feaLen) This method validates the frame data against an attribute length constrain if data value in any cell is greater than the specified threshold of that attribute the output frame will store a null on that cell position, thus removing the length-violating values.FrameBlock.removeEmptyOperations(boolean rows, boolean emptyReturn, MatrixBlock select) -
Uses of MatrixBlock in org.apache.sysds.runtime.frame.data.lib
Methods in org.apache.sysds.runtime.frame.data.lib that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockMatrixBlockFromFrame.convertToMatrixBlock(FrameBlock frame, int k) Converts a frame block with arbitrary schema into a matrix block.static MatrixBlockMatrixBlockFromFrame.convertToMatrixBlock(FrameBlock frame, MatrixBlock ret, int k) Converts a frame block with arbitrary schema into a matrix block.Methods in org.apache.sysds.runtime.frame.data.lib with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic FrameBlockFrameFromMatrixBlock.convertToFrameBlock(MatrixBlock mb, int k) static FrameBlockFrameFromMatrixBlock.convertToFrameBlock(MatrixBlock mb, Types.ValueType[] schema, int k) static FrameBlockFrameFromMatrixBlock.convertToFrameBlock(MatrixBlock mb, Types.ValueType vt, int k) static MatrixBlockMatrixBlockFromFrame.convertToMatrixBlock(FrameBlock frame, MatrixBlock ret, int k) Converts a frame block with arbitrary schema into a matrix block.static FrameBlockFrameLibRemoveEmpty.removeEmpty(FrameBlock fb, boolean rows, boolean emptyReturn, MatrixBlock select) -
Uses of MatrixBlock in org.apache.sysds.runtime.functionobjects
Methods in org.apache.sysds.runtime.functionobjects with parameters of type MatrixBlockModifier and TypeMethodDescriptionvoidCTable.execute(double v1, double v2, double w, boolean ignoreZeros, CTableMap resultMap, MatrixBlock resultBlock) voidCTable.execute(double v1, double v2, double w, boolean ignoreZeros, MatrixBlock ctableResult) voidCTable.execute(MatrixBlock in1, MatrixBlock in2, MatrixBlock w, CTableMap resultMap, int k) CTable.getPartialCTableTask(MatrixBlock in1, MatrixBlock in2, MatrixBlock w, int startInd, int blockSize, ArrayList<CTableMap> pmaps) -
Uses of MatrixBlock in org.apache.sysds.runtime.instructions.cp
Methods in org.apache.sysds.runtime.instructions.cp with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic voidEinsumCPInstruction.ensureMatrixBlockColumnVector(MatrixBlock mb) static voidEinsumCPInstruction.ensureMatrixBlockRowVector(MatrixBlock mb) -
Uses of MatrixBlock in org.apache.sysds.runtime.instructions.fed
Methods in org.apache.sysds.runtime.instructions.fed that return MatrixBlockModifier and TypeMethodDescription<T> MatrixBlockQuantilePickFEDInstruction.getEquiHeightBins(ExecutionContext ec, int colID, double[] quantiles) -
Uses of MatrixBlock in org.apache.sysds.runtime.instructions.ooc
Methods in org.apache.sysds.runtime.instructions.ooc that return MatrixBlockMethods in org.apache.sysds.runtime.instructions.ooc with parameters of type MatrixBlockModifier and TypeMethodDescriptionbooleanOOCInstruction.OOCMatrixBlockTracker.putAndIncrementCount(Long idx, MatrixBlock block) Adds or updates a block for the given index and updates its internal count.voidOOCInstruction.OOCMatrixBlockTracker.putAndInitCount(Long idx, MatrixBlock block) -
Uses of MatrixBlock in org.apache.sysds.runtime.instructions.spark
Methods in org.apache.sysds.runtime.instructions.spark that return MatrixBlockModifier and TypeMethodDescriptionAggregateUnarySPInstruction.RDDUAggFunction2.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) BinUaggChainSPInstruction.RDDBinUaggChainFunction.call(MatrixBlock arg0) CompressionSPInstruction.CompressionFunction.call(MatrixBlock arg0) CompressionSPInstruction.CompressionWorkloadFunction.call(MatrixBlock arg0) CtableSPInstruction.MapJoinSignature1.call(MatrixBlock v1) CtableSPInstruction.MapJoinSignature2.call(scala.Tuple2<MatrixBlock, MatrixBlock> v1) CtableSPInstruction.MapJoinSignature3.call(scala.Tuple2<scala.Tuple2<MatrixBlock, MatrixBlock>, MatrixBlock> v1) DeCompressionSPInstruction.DeCompressionFunction.call(MatrixBlock arg0) ParameterizedBuiltinSPInstruction.RDDMapGroupedAggFunction2.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) ParameterizedBuiltinSPInstruction.RDDReplaceFunction.call(MatrixBlock arg0) static MatrixBlockMatrixIndexingSPInstruction.inmemoryIndexing(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in1, DataCharacteristics mcIn, DataCharacteristics mcOut, IndexRange ixrange) Methods in org.apache.sysds.runtime.instructions.spark that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionscala.Tuple2<MatrixIndexes,MatrixBlock> AppendGAlignedSPInstruction.ShiftColumnIndex.call(scala.Tuple2<MatrixIndexes, MatrixBlock> kv) scala.Tuple2<MatrixIndexes,MatrixBlock> AppendGSPInstruction.MergeWithShiftedBlocks.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<Iterable<MatrixBlock>, Iterable<MatrixBlock>>> kv) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> AppendGSPInstruction.ShiftMatrix.call(scala.Tuple2<MatrixIndexes, MatrixBlock> kv) scala.Tuple2<MatrixIndexes,MatrixBlock> BuiltinNarySPInstruction.PadBlocksFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDMapGroupedAggFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDRemoveEmptyFunction.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<MatrixBlock, MatrixBlock>> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDRemoveEmptyFunctionInMem.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDRExpandFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) scala.Tuple2<Long,MatrixBlock> ParameterizedBuiltinSPInstruction.RDDTransformApplyFunction2.call(scala.Tuple2<Long, FrameBlock> in) scala.Tuple2<MatrixIndexes,MatrixBlock> ParameterizedBuiltinSPInstruction.RDDTransformDecodeExpandFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> in) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> SpoofSPInstruction.ReplicateRightFactorFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Methods in org.apache.sysds.runtime.instructions.spark with parameters of type MatrixBlockModifier and TypeMethodDescriptionBinUaggChainSPInstruction.RDDBinUaggChainFunction.call(MatrixBlock arg0) CompressionSPInstruction.CompressionFunction.call(MatrixBlock arg0) CompressionSPInstruction.CompressionWorkloadFunction.call(MatrixBlock arg0) CompressionSPInstruction.SizeFunction.call(MatrixBlock arg0) CtableSPInstruction.MapJoinSignature1.call(MatrixBlock v1) DeCompressionSPInstruction.DeCompressionFunction.call(MatrixBlock arg0) ParameterizedBuiltinSPInstruction.RDDContainsFunction.call(MatrixBlock arg0) ParameterizedBuiltinSPInstruction.RDDReplaceFunction.call(MatrixBlock arg0) Method parameters in org.apache.sysds.runtime.instructions.spark with type arguments of type MatrixBlockModifier and TypeMethodDescriptionAggregateUnarySPInstruction.RDDUAggFunction2.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) scala.Tuple2<MatrixIndexes,MatrixBlock> AppendGAlignedSPInstruction.ShiftColumnIndex.call(scala.Tuple2<MatrixIndexes, MatrixBlock> kv) scala.Tuple2<MatrixIndexes,MatrixBlock> AppendGSPInstruction.MergeWithShiftedBlocks.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<Iterable<MatrixBlock>, Iterable<MatrixBlock>>> kv) scala.Tuple2<MatrixIndexes,MatrixBlock> AppendGSPInstruction.MergeWithShiftedBlocks.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<Iterable<MatrixBlock>, Iterable<MatrixBlock>>> kv) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> AppendGSPInstruction.ShiftMatrix.call(scala.Tuple2<MatrixIndexes, MatrixBlock> kv) scala.Tuple2<MatrixIndexes,MatrixBlock> BuiltinNarySPInstruction.PadBlocksFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) CtableSPInstruction.MapJoinSignature2.call(scala.Tuple2<MatrixBlock, MatrixBlock> v1) CtableSPInstruction.MapJoinSignature2.call(scala.Tuple2<MatrixBlock, MatrixBlock> v1) CtableSPInstruction.MapJoinSignature3.call(scala.Tuple2<scala.Tuple2<MatrixBlock, MatrixBlock>, MatrixBlock> v1) CtableSPInstruction.MapJoinSignature3.call(scala.Tuple2<scala.Tuple2<MatrixBlock, MatrixBlock>, MatrixBlock> v1) CtableSPInstruction.MapJoinSignature3.call(scala.Tuple2<scala.Tuple2<MatrixBlock, MatrixBlock>, MatrixBlock> v1) ParameterizedBuiltinSPInstruction.RDDContainsVectFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> input) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDMapGroupedAggFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) ParameterizedBuiltinSPInstruction.RDDMapGroupedAggFunction2.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDRemoveEmptyFunction.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<MatrixBlock, MatrixBlock>> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDRemoveEmptyFunction.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<MatrixBlock, MatrixBlock>> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDRemoveEmptyFunctionInMem.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ParameterizedBuiltinSPInstruction.RDDRExpandFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) scala.Tuple2<MatrixIndexes,MatrixBlock> ParameterizedBuiltinSPInstruction.RDDTransformDecodeExpandFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> in) scala.Tuple2<Long,FrameBlock> ParameterizedBuiltinSPInstruction.RDDTransformDecodeFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> in) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> SpoofSPInstruction.ReplicateRightFactorFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) static MatrixBlockMatrixIndexingSPInstruction.inmemoryIndexing(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in1, DataCharacteristics mcIn, DataCharacteristics mcOut, IndexRange ixrange) Constructor parameters in org.apache.sysds.runtime.instructions.spark with type arguments of type MatrixBlockModifierConstructorDescriptionRDDContainsVectFunction(PartitionedBroadcast<MatrixBlock> bc, int blocksize) RDDMapGroupedAggFunction(PartitionedBroadcast<MatrixBlock> pbm, Operator op, int ngroups, int blen) RDDMapGroupedAggFunction2(PartitionedBroadcast<MatrixBlock> pbm, Operator op, int ngroups) RDDRemoveEmptyFunctionInMem(boolean rmRows, long len, long blen, PartitionedBroadcast<MatrixBlock> off) -
Uses of MatrixBlock in org.apache.sysds.runtime.instructions.spark.data
Methods in org.apache.sysds.runtime.instructions.spark.data that return MatrixBlockModifier and TypeMethodDescriptionCorrMatrixBlock.getCorrection()CorrMatrixBlock.getValue()RowMatrixBlock.getValue()Methods in org.apache.sysds.runtime.instructions.spark.data with parameters of type MatrixBlockConstructors in org.apache.sysds.runtime.instructions.spark.data with parameters of type MatrixBlockModifierConstructorDescriptionCorrMatrixBlock(MatrixBlock value) CorrMatrixBlock(MatrixBlock value, MatrixBlock corr) RowMatrixBlock(int len, int row, MatrixBlock value) -
Uses of MatrixBlock in org.apache.sysds.runtime.instructions.spark.functions
Methods in org.apache.sysds.runtime.instructions.spark.functions that return MatrixBlockModifier and TypeMethodDescriptionAggregateDropCorrectionFunction.call(MatrixBlock arg0) ComputeBinaryBlockNnzFunction.call(MatrixBlock arg0) CopyMatrixBlockFunction.call(MatrixBlock arg0) CreateSparseBlockFunction.call(MatrixBlock arg0) MapInputSignature.call(MatrixBlock v1) MapJoinSignature.call(scala.Tuple2<MatrixBlock[], MatrixBlock> v1) MatrixMatrixBinaryOpFunction.call(scala.Tuple2<MatrixBlock, MatrixBlock> arg0) MatrixScalarUnaryFunction.call(MatrixBlock arg0) Methods in org.apache.sysds.runtime.instructions.spark.functions that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionLazyIterableIterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> CopyMatrixBlockPairFunction.call(Iterator<scala.Tuple2<MatrixIndexes, MatrixBlock>> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ExtractBlockForBinaryReblock.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) LazyIterableIterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> MatrixVectorBinaryOpPartitionFunction.call(Iterator<scala.Tuple2<MatrixIndexes, MatrixBlock>> arg0) scala.Tuple2<MatrixIndexes,MatrixBlock> ReorgMapFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ReplicateBlockFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ReplicateVectorFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Methods in org.apache.sysds.runtime.instructions.spark.functions with parameters of type MatrixBlockModifier and TypeMethodDescriptionAggregateDropCorrectionFunction.call(MatrixBlock arg0) ComputeBinaryBlockNnzFunction.call(MatrixBlock arg0) CopyMatrixBlockFunction.call(MatrixBlock arg0) CreateSparseBlockFunction.call(MatrixBlock arg0) FilterNonEmptyBlocksFunction2.call(MatrixBlock arg0) MapInputSignature.call(MatrixBlock v1) MatrixScalarUnaryFunction.call(MatrixBlock arg0) Method parameters in org.apache.sysds.runtime.instructions.spark.functions with type arguments of type MatrixBlockModifier and TypeMethodDescriptionConvertMatrixBlockToIJVLines.call(scala.Tuple2<MatrixIndexes, MatrixBlock> kv) LazyIterableIterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> CopyMatrixBlockPairFunction.call(Iterator<scala.Tuple2<MatrixIndexes, MatrixBlock>> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ExtractBlockForBinaryReblock.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,WeightedCell>> ExtractGroup.ExtractGroupBroadcast.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg) Iterator<scala.Tuple2<MatrixIndexes,WeightedCell>> ExtractGroup.ExtractGroupJoin.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<MatrixBlock, MatrixBlock>> arg) Iterator<scala.Tuple2<MatrixIndexes,WeightedCell>> ExtractGroup.ExtractGroupJoin.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<MatrixBlock, MatrixBlock>> arg) Iterator<scala.Tuple2<MatrixIndexes,WeightedCell>> ExtractGroupNWeights.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<scala.Tuple2<MatrixBlock, MatrixBlock>, MatrixBlock>> arg) Iterator<scala.Tuple2<MatrixIndexes,WeightedCell>> ExtractGroupNWeights.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<scala.Tuple2<MatrixBlock, MatrixBlock>, MatrixBlock>> arg) Iterator<scala.Tuple2<MatrixIndexes,WeightedCell>> ExtractGroupNWeights.call(scala.Tuple2<MatrixIndexes, scala.Tuple2<scala.Tuple2<MatrixBlock, MatrixBlock>, MatrixBlock>> arg) FilterDiagMatrixBlocksFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) FilterNonEmptyBlocksFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) IsBlockInList.call(scala.Tuple2<MatrixIndexes, MatrixBlock> kv) IsBlockInRange.call(scala.Tuple2<MatrixIndexes, MatrixBlock> kv) MapJoinSignature.call(scala.Tuple2<MatrixBlock[], MatrixBlock> v1) MatrixMatrixBinaryOpFunction.call(scala.Tuple2<MatrixBlock, MatrixBlock> arg0) MatrixMatrixBinaryOpFunction.call(scala.Tuple2<MatrixBlock, MatrixBlock> arg0) LazyIterableIterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> MatrixVectorBinaryOpPartitionFunction.call(Iterator<scala.Tuple2<MatrixIndexes, MatrixBlock>> arg0) org.apache.sysds.runtime.instructions.spark.functions.OuterVectorBinaryOpFunction.OuterVectorBinaryOpIteratorOuterVectorBinaryOpFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) RecomputeNnzFunction.call(Iterator<MatrixBlock> iter) scala.Tuple2<MatrixIndexes,MatrixBlock> ReorgMapFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ReplicateBlockFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> ReplicateVectorFunction.call(scala.Tuple2<MatrixIndexes, MatrixBlock> arg0) Constructor parameters in org.apache.sysds.runtime.instructions.spark.functions with type arguments of type MatrixBlockModifierConstructorDescriptionExtractGroupBroadcast(PartitionedBroadcast<MatrixBlock> pbm, long blen, long ngroups, Operator op) MatrixVectorBinaryOpPartitionFunction(BinaryOperator op, PartitionedBroadcast<MatrixBlock> binput, BinaryM.VectorType vtype) -
Uses of MatrixBlock in org.apache.sysds.runtime.instructions.spark.utils
Methods in org.apache.sysds.runtime.instructions.spark.utils that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockRDDAggregateUtils.aggStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, AggregateOperator aop) Single block aggregation over pair rdds with corrections for numerical stability.static MatrixBlockRDDAggregateUtils.aggStable(org.apache.spark.api.java.JavaRDD<MatrixBlock> in, AggregateOperator aop) Single block aggregation over rdds with corrections for numerical stability.static MatrixBlockRDDAggregateUtils.sumStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) static MatrixBlockRDDAggregateUtils.sumStable(org.apache.spark.api.java.JavaRDD<MatrixBlock> in) Methods in org.apache.sysds.runtime.instructions.spark.utils that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.aggByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, AggregateOperator aop) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.aggByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, AggregateOperator aop, boolean deepCopyCombiner) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.aggByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, AggregateOperator aop, int numPartitions, boolean deepCopyCombiner) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.binaryBlockToBinaryBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, DataCharacteristics mcIn, DataCharacteristics mcOut) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> FrameRDDConverterUtils.binaryBlockToMatrixBlock(org.apache.spark.api.java.JavaPairRDD<Long, FrameBlock> input, DataCharacteristics mcIn, DataCharacteristics mcOut) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.binaryCellToBinaryBlock(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixCell> input, DataCharacteristics mcOut, boolean outputEmptyBlocks) Iterator<scala.Tuple2<MatrixIndexes,MatrixBlock>> RDDConverterUtils.BinaryCellToBinaryBlockFunction.call(Iterator<scala.Tuple2<MatrixIndexes, MatrixCell>> arg0) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtilsExt.coordinateMatrixToBinaryBlock(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.mllib.linalg.distributed.CoordinateMatrix input, DataCharacteristics mcIn, boolean outputEmptyBlocks) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtilsExt.coordinateMatrixToBinaryBlock(org.apache.spark.SparkContext sc, org.apache.spark.mllib.linalg.distributed.CoordinateMatrix input, DataCharacteristics mcIn, boolean outputEmptyBlocks) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkUtils.copyBinaryBlockMatrix(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) Creates a partitioning-preserving deep copy of the input matrix RDD, where the indexes and values are copied.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkUtils.copyBinaryBlockMatrix(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, boolean deep) Creates a partitioning-preserving copy of the input matrix RDD.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.csvToBinaryBlock(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.api.java.JavaPairRDD<org.apache.hadoop.io.LongWritable, org.apache.hadoop.io.Text> input, DataCharacteristics mc, boolean hasHeader, String delim, boolean fill, double fillValue, Set<String> naStrings) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.csvToBinaryBlock(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.api.java.JavaRDD<String> input, DataCharacteristics mcOut, boolean hasHeader, String delim, boolean fill, double fillValue, Set<String> naStrings) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.dataFrameToBinaryBlock(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.sql.Dataset<org.apache.spark.sql.Row> df, DataCharacteristics mc, boolean containsID, boolean isVector) static List<scala.Tuple2<MatrixIndexes,MatrixBlock>> SparkUtils.fromIndexedMatrixBlock(List<IndexedMatrixValue> in) static scala.Tuple2<MatrixIndexes,MatrixBlock> SparkUtils.fromIndexedMatrixBlock(IndexedMatrixValue in) static List<Pair<MatrixIndexes,MatrixBlock>> SparkUtils.fromIndexedMatrixBlockToPair(List<IndexedMatrixValue> in) static Pair<MatrixIndexes,MatrixBlock> SparkUtils.fromIndexedMatrixBlockToPair(IndexedMatrixValue in) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkUtils.getEmptyBlockRDD(org.apache.spark.api.java.JavaSparkContext sc, DataCharacteristics mc) Creates an RDD of empty blocks according to the given matrix characteristics.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.libsvmToBinaryBlock(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.api.java.JavaPairRDD<org.apache.hadoop.io.LongWritable, org.apache.hadoop.io.Text> input, DataCharacteristics mc, String delim, String indexDelim) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.matrixBlockToAlignedMatrixBlock(org.apache.spark.api.java.JavaPairRDD<Long, MatrixBlock> input, DataCharacteristics mcIn, DataCharacteristics mcOut) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.mergeByKey(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) Merges disjoint data of all blocks per key.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.mergeByKey(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, boolean deepCopyCombiner) Merges disjoint data of all blocks per key.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.mergeByKey(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, int numPartitions, boolean deepCopyCombiner) Merges disjoint data of all blocks per key.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.mergeRowsByKey(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, RowMatrixBlock> in) Merges disjoint data of all blocks per key.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortByVal(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, long rlen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortByVal(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in2, long rlen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortByVals(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, long rlen, long clen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortDataByIx(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> data, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> ixmap, long rlen, long clen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortDataByVal(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> val, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> data, boolean asc, long rlen, long clen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortDataByValMemSort(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> val, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> data, boolean asc, long rlen, long clen, int blen, SparkExecutionContext sec, ReorgOperator r_op) This function collects and sorts value column in memory and then broadcasts it.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortDataByVals(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> val, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> data, boolean asc, long rlen, long clen, long clen2, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortIndexesByVal(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> val, boolean asc, long rlen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortIndexesByVals(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, boolean asc, long rlen, long clen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.sumByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.sumByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, boolean deepCopyCombiner) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.sumByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, int numPartitions, boolean deepCopyCombiner) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.textCellToBinaryBlock(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.api.java.JavaPairRDD<org.apache.hadoop.io.LongWritable, org.apache.hadoop.io.Text> input, DataCharacteristics mcOut, boolean outputEmptyBlocks, FileFormatPropertiesMM mmProps) Methods in org.apache.sysds.runtime.instructions.spark.utils with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic voidRDDConverterUtilsExt.copyRowBlocks(MatrixBlock mb, int rowIndex, MatrixBlock ret, int numRowsPerBlock, int rlen, int clen) static voidRDDConverterUtilsExt.copyRowBlocks(MatrixBlock mb, int rowIndex, MatrixBlock ret, long numRowsPerBlock, long rlen, long clen) static voidRDDConverterUtilsExt.copyRowBlocks(MatrixBlock mb, long rowIndex, MatrixBlock ret, int numRowsPerBlock, int rlen, int clen) static voidRDDConverterUtilsExt.copyRowBlocks(MatrixBlock mb, long rowIndex, MatrixBlock ret, long numRowsPerBlock, long rlen, long clen) static voidRDDConverterUtilsExt.postProcessAfterCopying(MatrixBlock ret) static IndexedMatrixValueSparkUtils.toIndexedMatrixBlock(MatrixIndexes ix, MatrixBlock mb) Method parameters in org.apache.sysds.runtime.instructions.spark.utils with type arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.aggByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, AggregateOperator aop) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.aggByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, AggregateOperator aop, boolean deepCopyCombiner) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.aggByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, AggregateOperator aop, int numPartitions, boolean deepCopyCombiner) static MatrixBlockRDDAggregateUtils.aggStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, AggregateOperator aop) Single block aggregation over pair rdds with corrections for numerical stability.static MatrixBlockRDDAggregateUtils.aggStable(org.apache.spark.api.java.JavaRDD<MatrixBlock> in, AggregateOperator aop) Single block aggregation over rdds with corrections for numerical stability.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.binaryBlockToBinaryBlock(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, DataCharacteristics mcIn, DataCharacteristics mcOut) static org.apache.spark.api.java.JavaRDD<String>RDDConverterUtils.binaryBlockToCsv(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, DataCharacteristics mcIn, FileFormatPropertiesCSV props, boolean strict) static org.apache.spark.sql.Dataset<org.apache.spark.sql.Row>RDDConverterUtils.binaryBlockToDataFrame(org.apache.spark.sql.SparkSession sparkSession, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, DataCharacteristics mc, boolean toVector) static org.apache.spark.sql.Dataset<org.apache.spark.sql.Row>RDDConverterUtils.binaryBlockToDataFrame(org.apache.spark.sql.SQLContext sqlContext, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, DataCharacteristics mc, boolean toVector) Deprecated.static org.apache.spark.api.java.JavaRDD<org.apache.spark.ml.feature.LabeledPoint>RDDConverterUtils.binaryBlockToLabeledPoints(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) Converter from binary block rdd to rdd of labeled points.static org.apache.spark.api.java.JavaRDD<String>RDDConverterUtils.binaryBlockToLibsvm(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, DataCharacteristics mcIn, FileFormatPropertiesLIBSVM props, boolean strict) static org.apache.spark.api.java.JavaRDD<String>RDDConverterUtils.binaryBlockToTextCell(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, DataCharacteristics mc) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkUtils.copyBinaryBlockMatrix(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) Creates a partitioning-preserving deep copy of the input matrix RDD, where the indexes and values are copied.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> SparkUtils.copyBinaryBlockMatrix(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, boolean deep) Creates a partitioning-preserving copy of the input matrix RDD.static longSparkUtils.getNonZeros(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> input) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDConverterUtils.matrixBlockToAlignedMatrixBlock(org.apache.spark.api.java.JavaPairRDD<Long, MatrixBlock> input, DataCharacteristics mcIn, DataCharacteristics mcOut) static org.apache.spark.api.java.JavaPairRDD<org.apache.hadoop.io.LongWritable,FrameBlock> FrameRDDConverterUtils.matrixBlockToBinaryBlock(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> input, DataCharacteristics mcIn) static org.apache.spark.api.java.JavaPairRDD<Long,FrameBlock> FrameRDDConverterUtils.matrixBlockToBinaryBlockLongIndex(org.apache.spark.api.java.JavaSparkContext sc, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> input, DataCharacteristics dcIn) static doubleRDDAggregateUtils.max(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.mergeByKey(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) Merges disjoint data of all blocks per key.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.mergeByKey(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, boolean deepCopyCombiner) Merges disjoint data of all blocks per key.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.mergeByKey(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, int numPartitions, boolean deepCopyCombiner) Merges disjoint data of all blocks per key.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortByVal(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, long rlen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortByVal(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in2, long rlen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortByVals(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, long rlen, long clen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortDataByIx(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> data, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> ixmap, long rlen, long clen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortDataByVal(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> val, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> data, boolean asc, long rlen, long clen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortDataByValMemSort(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> val, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> data, boolean asc, long rlen, long clen, int blen, SparkExecutionContext sec, ReorgOperator r_op) This function collects and sorts value column in memory and then broadcasts it.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortDataByVals(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> val, org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> data, boolean asc, long rlen, long clen, long clen2, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortIndexesByVal(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> val, boolean asc, long rlen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDSortUtils.sortIndexesByVals(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, boolean asc, long rlen, long clen, int blen) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.sumByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.sumByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, boolean deepCopyCombiner) static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock> RDDAggregateUtils.sumByKeyStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in, int numPartitions, boolean deepCopyCombiner) static MatrixBlockRDDAggregateUtils.sumStable(org.apache.spark.api.java.JavaPairRDD<MatrixIndexes, MatrixBlock> in) static MatrixBlockRDDAggregateUtils.sumStable(org.apache.spark.api.java.JavaRDD<MatrixBlock> in) static IndexedMatrixValueSparkUtils.toIndexedMatrixBlock(scala.Tuple2<MatrixIndexes, MatrixBlock> in) -
Uses of MatrixBlock in org.apache.sysds.runtime.io
Methods in org.apache.sysds.runtime.io that return MatrixBlockModifier and TypeMethodDescriptionReaderCOGParallel.TileProcessor.call()static MatrixBlockReaderHDF5.computeHDF5Size(List<org.apache.hadoop.fs.Path> files, org.apache.hadoop.fs.FileSystem fs, String datasetName, long estnnz) static MatrixBlock[]MatrixWriter.createMatrixBlocksForReuse(long rlen, long clen, int blen, boolean sparse, long nonZeros) BinaryBlockSerialization.MatrixBlockDeserializer.deserialize(MatrixBlock mb) static MatrixBlockMatrixWriter.getMatrixBlockForReuse(MatrixBlock[] blocks, int rows, int cols, int blen) ReaderTextCSVParallel.readMatrixAsStream(OOCStream<IndexedMatrixValue> outStream, String fname, long rlen, long clen, int blen, long estnnz) abstract MatrixBlockMatrixReader.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderBinaryBlock.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderBinaryBlockParallel.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderCOG.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderCOGParallel.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderDelta.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderDeltaParallel.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderHDF5.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderHDF5Parallel.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderTextCell.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderTextCSV.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderTextCSVParallel.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderTextLIBSVM.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) ReaderTextLIBSVMParallel.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) abstract MatrixBlockMatrixReader.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderBinaryBlock.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderCOG.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderCOGParallel.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderDelta.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderHDF5.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderHDF5Parallel.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderTextCell.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderTextCSV.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderTextCSVParallel.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderTextLIBSVM.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) ReaderTextLIBSVMParallel.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) Methods in org.apache.sysds.runtime.io that return types with arguments of type MatrixBlockModifier and TypeMethodDescriptionorg.apache.hadoop.io.serializer.Deserializer<MatrixBlock>BinaryBlockSerialization.getDeserializer(Class arg0) org.apache.hadoop.io.serializer.Serializer<MatrixBlock>BinaryBlockSerialization.getSerializer(Class arg0) Methods in org.apache.sysds.runtime.io with parameters of type MatrixBlockModifier and TypeMethodDescriptionBinaryBlockSerialization.MatrixBlockDeserializer.deserialize(MatrixBlock mb) static MatrixBlockMatrixWriter.getMatrixBlockForReuse(MatrixBlock[] blocks, int rows, int cols, int blen) static longReaderHDF5.readMatrixFromHDF5(H5ByteReader byteReader, String datasetName, MatrixBlock dest, int rl, long ru, long clen, int blen) voidBinaryBlockSerialization.MatrixBlockSerializer.serialize(MatrixBlock mb) voidWriterHDF5Parallel.writeHDF5MatrixToHDFS(org.apache.hadoop.fs.Path path, org.apache.hadoop.mapred.JobConf job, org.apache.hadoop.fs.FileSystem fs, MatrixBlock src) voidMatrixWriter.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz) abstract voidMatrixWriter.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) final voidWriterBinaryBlock.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) voidWriterDelta.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) voidWriterHDF5.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) final voidWriterMatrixMarket.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) final voidWriterTextCell.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) final voidWriterTextCSV.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) final voidWriterTextLIBSVM.writeMatrixToHDFS(MatrixBlock src, String fname, long rlen, long clen, int blen, long nnz, boolean diag) Constructors in org.apache.sysds.runtime.io with parameters of type MatrixBlockModifierConstructorDescriptionReadTask(org.apache.hadoop.mapred.InputSplit split, org.apache.hadoop.mapred.TextInputFormat informat, org.apache.hadoop.mapred.JobConf job, MatrixBlock dest, long rlen, long clen, boolean mm, FileFormatPropertiesMM mmProps) TileProcessor(int clen, byte[] tileData, int tileRow, int tileCol, int tileWidth, int tileLength, int bands, int[] bitsPerSample, SampleFormatDataTypes[] sampleFormat, COGHeader cogHeader, MatrixBlock dest, int planarConfiguration) TileProcessor(int clen, byte[] tileData, int tileRow, int tileCol, int tileWidth, int tileLength, int bands, int[] bitsPerSample, SampleFormatDataTypes[] sampleFormat, COGHeader cogHeader, MatrixBlock dest, int planarConfiguration, int band) -
Uses of MatrixBlock in org.apache.sysds.runtime.iogen
Methods in org.apache.sysds.runtime.iogen that return MatrixBlockMethods in org.apache.sysds.runtime.iogen with parameters of type MatrixBlockModifier and TypeMethodDescriptionvoidSampleProperties.setSampleMatrix(MatrixBlock sampleMatrix) Constructors in org.apache.sysds.runtime.iogen with parameters of type MatrixBlockModifierConstructorDescriptionFormatIdentifyer(String raw, MatrixBlock matrix) GenerateReaderMatrix(String sampleRaw, MatrixBlock sampleMatrix, boolean parallel) ReaderMapping(int nlines, int nrows, int ncols, RawIndex[] sampleRawIndexes, MatrixBlock matrix) ReaderMapping(String raw, MatrixBlock matrix) ReaderMappingIndex(int nlines, int nrows, int ncols, RawIndex[] sampleRawIndexes, MatrixBlock matrix) ReaderMappingIndex(String raw, MatrixBlock matrix) SampleProperties(String sampleRaw, MatrixBlock sampleMatrix) -
Uses of MatrixBlock in org.apache.sysds.runtime.iogen.template
Methods in org.apache.sysds.runtime.iogen.template that return MatrixBlockModifier and TypeMethodDescriptionMatrixGenerateReader.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) MatrixGenerateReaderParallel.readMatrixFromHDFS(String fname, long rlen, long clen, int blen, long estnnz) MatrixGenerateReader.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) MatrixGenerateReaderParallel.readMatrixFromInputStream(InputStream is, long rlen, long clen, int blen, long estnnz) -
Uses of MatrixBlock in org.apache.sysds.runtime.lineage
Methods in org.apache.sysds.runtime.lineage that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockLineageCache.getMatrix(LineageItem key) LineageCacheEntry.getMBValue()Methods in org.apache.sysds.runtime.lineage with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic voidLineageCache.putValueAsyncOp(LineageItem instLI, Data data, MatrixBlock mb, long starttime) voidLineageCacheEntry.setValue(MatrixBlock val) voidLineageCacheEntry.setValue(MatrixBlock val, long computetime) Constructors in org.apache.sysds.runtime.lineage with parameters of type MatrixBlockModifierConstructorDescriptionLineageCacheEntry(LineageItem key, Types.DataType dt, MatrixBlock Mval, ScalarObject Sval, long computetime) -
Uses of MatrixBlock in org.apache.sysds.runtime.matrix.data
Fields in org.apache.sysds.runtime.matrix.data declared as MatrixBlockModifier and TypeFieldDescriptionDnnParameters.biasDnnParameters.input1DnnParameters.input2DnnParameters.input3DnnParameters.input4DnnParameters.input5DnnParameters.input6DnnParameters.input7DnnParameters.input8DnnParameters.input9DnnParameters.outputDnnParameters.output2DnnParameters.output3DnnParameters.output4DnnParameters.output5Methods in org.apache.sysds.runtime.matrix.data that return MatrixBlockModifier and TypeMethodDescriptionMatrixBlock.aggregateBinaryOperations(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, AggregateBinaryOperator op) final MatrixBlockMatrixBlock.aggregateBinaryOperations(MatrixBlock m1, MatrixBlock m2, AggregateBinaryOperator op) static MatrixBlockLibMatrixAgg.aggregateTernary(MatrixBlock in1, MatrixBlock in2, MatrixBlock in3, MatrixBlock ret, AggregateTernaryOperator op) static MatrixBlockLibMatrixAgg.aggregateTernary(MatrixBlock in1, MatrixBlock in2, MatrixBlock in3, MatrixBlock ret, AggregateTernaryOperator op, int k) static MatrixBlockMatrixBlock.aggregateTernaryOperations(MatrixBlock m1, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret, AggregateTernaryOperator op, boolean inCP) static MatrixBlockLibMatrixAgg.aggregateUnaryMatrix(AggregateUnaryOperator op, MatrixBlock in, MatrixValue result, int blen, MatrixIndexes indexesIn, boolean inCP) final MatrixBlockMatrixBlock.aggregateUnaryOperations(AggregateUnaryOperator op) MatrixBlock.aggregateUnaryOperations(AggregateUnaryOperator op, MatrixValue result, int blen, MatrixIndexes indexesIn, boolean inCP) final MatrixBlockMatrixBlock.allocateBlock()final MatrixBlockMatrixBlock.allocateDenseBlock()static MatrixBlockLibMatrixAppend.append(MatrixBlock a, MatrixBlock[] that, MatrixBlock result, boolean cbind) static MatrixBlockMatrixBlock.append(List<MatrixBlock> that, MatrixBlock ret, boolean cbind, int k) Append that list of matrixblocks together.final MatrixBlockMatrixBlock.append(MatrixBlock that) Append that matrix to this matrix, while allocating a new matrix.MatrixBlock.append(MatrixBlock[] that, MatrixBlock result, boolean cbind) Append that list of matrixes to this matrix.final MatrixBlockMatrixBlock.append(MatrixBlock that, boolean cbind) Append that matrix to this matrix, while allocating a new matrix.final MatrixBlockMatrixBlock.append(MatrixBlock that, MatrixBlock ret) Append that matrix to this matrix.final MatrixBlockMatrixBlock.append(MatrixBlock that, MatrixBlock ret, boolean cbind) Append that matrix to this matrix.final MatrixBlockMatrixBlock.binaryOperations(BinaryOperator op, MatrixValue thatValue) MatrixBlock.binaryOperations(BinaryOperator op, MatrixValue thatValue, MatrixValue result) MatrixBlock.binaryOperationsInPlace(BinaryOperator op, MatrixValue thatValue) static MatrixBlockLibMatrixBincell.bincellOp(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, BinaryOperator op) static MatrixBlockLibMatrixBincell.bincellOpInPlace(MatrixBlock m1ret, MatrixBlock m2, BinaryOperator op) NOTE: operations in place always require m1 and m2 to be of equal dimensions defaults to right side operations, updating the m1 matrix with like: m1ret op m2static MatrixBlockLibMatrixBincell.bincellOpInPlaceLeft(MatrixBlock m1ret, MatrixBlock m2, BinaryOperator op) Left side operations, updating the m1 matrix like: m2 op m1retstatic MatrixBlockLibMatrixBincell.bincellOpInPlaceRight(MatrixBlock m1ret, MatrixBlock m2, BinaryOperator op) Right side operations, updating the m1 matrix like: m1ret op m2static MatrixBlockLibMatrixBincell.bincellOpScalar(MatrixBlock m1, MatrixBlock ret, ScalarOperator op, int k) MatrixBlock.chainMatrixMultOperations(MatrixBlock v, MatrixBlock w, MatrixBlock out, MapMultChain.ChainType ctype) MatrixBlock.chainMatrixMultOperations(MatrixBlock v, MatrixBlock w, MatrixBlock out, MapMultChain.ChainType ctype, int k) final MatrixBlockMatrixBlock.colMax()Wrapper method for reduceall-colMin of a matrix.final MatrixBlockMatrixBlock.colMax(int k) final MatrixBlockMatrixBlock.colMin()Wrapper method for reduceall-colMin of a matrix.final MatrixBlockMatrixBlock.colMin(int k) MatrixBlock.colSum()Wrapper method for single threaded reduceall-colSum of a matrix.static MatrixBlockLibCommonsMath.computeSolve(MatrixBlock in1, MatrixBlock in2) Function to solve a given system of equations.MatrixBlock.copyShallow(MatrixBlock that) static MatrixBlockLibMatrixCountDistinct.countDistinctValuesFromSketch(CountDistinctOperator op, CorrMatrixBlock corrBlkIn) final MatrixBlockMatrixBlock.ctableSeqOperations(MatrixValue thatMatrix, double thatScalar, MatrixBlock resultBlock) D = ctable(seq,A,w) this <- seq; thatMatrix <- A; thatScalar <- w; result <- D (i1,j1,v1) from input1 (this) (i1,j1,v2) from input2 (that) (w) from scalar_input3 (scalarThat2)MatrixBlock.ctableSeqOperations(MatrixValue thatMatrix, double thatScalar, MatrixBlock ret, boolean updateClen) D = ctable(seq,A,w)static MatrixBlockLibMatrixAgg.cumaggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, UnaryOperator uop) static MatrixBlockLibMatrixAgg.cumaggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, UnaryOperator uop, double[] agg) static MatrixBlockLibMatrixAgg.cumaggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, UnaryOperator uop, int k) static MatrixBlockLibMatrixReorg.diag(MatrixBlock in, MatrixBlock out) static MatrixBlockLibMatrixMult.emptyMatrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret) static MatrixBlockLibMatrixCountDistinct.estimateDistinctValues(MatrixBlock in, CountDistinctOperator op) Public method to count the number of distinct values inside a matrix.MatrixBlock.extractTriangular(MatrixBlock ret, boolean lower, boolean diag, boolean values) static MatrixBlock[]LibMatrixFourier.fft(MatrixBlock re, int threads) Function to perform FFT for a given matrix.static MatrixBlock[]LibMatrixFourier.fft(MatrixBlock re, MatrixBlock im, int threads) Function to perform FFT for two given matrices.static MatrixBlock[]LibMatrixFourier.fft_linearized(MatrixBlock re, int threads) Function to perform FFT for each row of a given matrix.static MatrixBlock[]LibMatrixFourier.fft_linearized(MatrixBlock re, MatrixBlock im, int threads) Function to perform FFT for each row of two given matrices.MatrixBlock.fill(double v) static MatrixBlockLibMatrixReorg.fusedSeqRexpand(int seqHeight, MatrixBlock A, double w) The DML code to activate this function:static MatrixBlockLibMatrixReorg.fusedSeqRexpand(int seqHeight, MatrixBlock A, double w, MatrixBlock ret, boolean updateClen) The DML code to activate this function:static MatrixBlockLibMatrixReorg.fusedSeqRexpand(int seqHeight, MatrixBlock A, double w, MatrixBlock ret, boolean updateClen, int k) The DML code to activate this function:static MatrixBlockLibMatrixSketch.getUniqueValues(MatrixBlock blkIn, Types.Direction dir) final MatrixBlockMatrixBlock.groupedAggOperations(MatrixValue tgt, MatrixValue wghts, MatrixValue ret, int ngroups, Operator op) Invocation from CP instructions.MatrixBlock.groupedAggOperations(MatrixValue tgt, MatrixValue wghts, MatrixValue ret, int ngroups, Operator op, int k) static MatrixBlock[]LibMatrixFourier.ifft(MatrixBlock re, int threads) Function to perform IFFT for a given matrix.static MatrixBlock[]LibMatrixFourier.ifft(MatrixBlock re, MatrixBlock im, int threads) Function to perform IFFT for two given matrices.static MatrixBlock[]LibMatrixFourier.ifft_linearized(MatrixBlock re, int threads) Function to perform IFFT for each row of a given matrix.static MatrixBlock[]LibMatrixFourier.ifft_linearized(MatrixBlock re, MatrixBlock im, int threads) Function to perform IFFT for each row of two given matrices.MatrixBlock.leftIndexingOperations(ScalarObject scalar, int rl, int cl, MatrixBlock ret, MatrixObject.UpdateType update) Explicitly allow left indexing for scalars.MatrixBlock.leftIndexingOperations(MatrixBlock rhsMatrix, int rl, int ru, int cl, int cu, MatrixBlock ret, MatrixObject.UpdateType update) final MatrixBlockMatrixBlock.leftIndexingOperations(MatrixBlock rhsMatrix, IndexRange ixrange, MatrixBlock ret, MatrixObject.UpdateType update) static MatrixBlockOperationsOnMatrixValues.matMult(MatrixBlock value1, MatrixBlock value2, MatrixBlock valueOut, AggregateBinaryOperator op) static MatrixBlockOperationsOnMatrixValues.matMult(MatrixBlock value1, MatrixBlock value2, AggregateBinaryOperator op) static MatrixBlockOperationsOnMatrixValues.matMult(MatrixIndexes indexes1, MatrixBlock value1, MatrixIndexes indexes2, MatrixBlock value2, MatrixIndexes indexesOut, MatrixBlock valueOut, AggregateBinaryOperator op) static MatrixBlockLibCommonsMath.matrixMatrixOperations(MatrixBlock in1, MatrixBlock in2, String opcode) static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2) Performs a matrix multiplication All variants use a IKJ access pattern, and internally use dense output.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, int k) Performs a matrix multiplication All variants use a IKJ access pattern, and internally use dense output.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret) Performs a matrix multiplication and stores the result in the output matrix.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, boolean fixedRet) This method allows one to disabling exam sparsity.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, boolean fixedRet, int k) Performs a matrix multiplication and stores the result in the output matrix.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) Performs a multi-threaded matrix multiplication and stores the result in the output matrix.static MatrixBlockLibMatrixNative.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) Performs matrix multiplication using native library if BLAS is available or else falls back to Java BLAS.static MatrixBlockLibMatrixMult.matrixMultNonNative(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) static MatrixBlockLibMatrixMult.matrixMultTransposeSelf(MatrixBlock m1, MatrixBlock ret, boolean leftTranspose) MatrixBlock.max(int k) Wrapper method for reduceall-max of a matrix.MatrixBlock.merge(MatrixBlock that, boolean appendOnly) MatrixBlock.merge(MatrixBlock that, boolean appendOnly, boolean par) MatrixBlock.merge(MatrixBlock that, boolean appendOnly, boolean par, boolean deep) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in, String opcode, int threads) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in, String opcode, int threads, int num_iterations, double tol) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in, String opcode, int threads, long seed) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in1, MatrixBlock in2, String opcode) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in1, MatrixBlock in2, String opcode, int threads, long seed) static MatrixBlockMatrixBlock.naryOperations(Operator op, MatrixBlock[] matrices, ScalarObject[] scalars, MatrixBlock ret) static MatrixBlockLibMatrixReorg.permute(MatrixBlock in, int[] inDims, int[] perm) static MatrixBlockLibMatrixReorg.permute(MatrixBlock in, int[] inDims, int[] perm, int k) final MatrixBlockMatrixBlock.pickValues(MatrixValue quantiles, MatrixValue ret) Pick the quantiles out of this matrix.MatrixBlock.pickValues(MatrixValue quantiles, MatrixValue ret, boolean average) static MatrixBlockLibMatrixAgg.prepareAggregateUnaryOutput(MatrixBlock in, AggregateUnaryOperator op, MatrixValue result, int blen) static MatrixBlockLibMatrixMult.prepMatrixMultTransposeSelfInput(MatrixBlock m1, boolean leftTranspose, boolean par) final MatrixBlockMatrixBlock.quaternaryOperations(QuaternaryOperator qop, MatrixBlock um, MatrixBlock vm, MatrixBlock wm, MatrixBlock out) MatrixBlock.quaternaryOperations(QuaternaryOperator qop, MatrixBlock U, MatrixBlock V, MatrixBlock wm, MatrixBlock out, int k) static MatrixBlockMatrixBlock.randOperations(int rows, int cols, double sparsity) static MatrixBlockMatrixBlock.randOperations(int rows, int cols, double sparsity, double min, double max, String pdf, long seed) Function to generate the random matrix with specified dimensions (block sizes are not specified).static MatrixBlockMatrixBlock.randOperations(int rows, int cols, double sparsity, double min, double max, String pdf, long seed, int k) Function to generate the random matrix with specified dimensions (block sizes are not specified).static MatrixBlockMatrixBlock.randOperations(RandomMatrixGenerator rgen, long seed) Function to generate the random matrix with specified dimensions and block dimensions.static MatrixBlockMatrixBlock.randOperations(RandomMatrixGenerator rgen, long seed, int k) Function to generate the random matrix with specified dimensions and block dimensions.MatrixBlock.randOperationsInPlace(RandomMatrixGenerator rgen, org.apache.commons.math3.random.Well1024a bigrand, long bSeed) Function to generate a matrix of random numbers.MatrixBlock.randOperationsInPlace(RandomMatrixGenerator rgen, org.apache.commons.math3.random.Well1024a bigrand, long bSeed, int k) Function to generate a matrix of random numbers.final MatrixBlockMatrixBlock.rbind(MatrixBlock that) final MatrixBlockMatrixBlock.removeEmptyOperations(MatrixBlock ret, boolean rows, boolean emptyReturn) MatrixBlock.removeEmptyOperations(MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) static MatrixBlockLibMatrixReorg.removeEmptyRows(MatrixBlock in, MatrixBlock ret, boolean emptyReturn, boolean selectNull, boolean[] flags, int rlen2) Remove selected rows, based on the boolean array given.static MatrixBlockLibMatrixReorg.reorg(MatrixBlock in, MatrixBlock out, ReorgOperator op) static MatrixBlockLibMatrixReorg.reorgInPlace(MatrixBlock in, ReorgOperator op) MatrixBlock.reorgOperations(ReorgOperator op, MatrixValue ret, int startRow, int startColumn, int length) static MatrixBlockLibMatrixReplace.replaceOperations(MatrixBlock in, MatrixBlock ret, double pattern, double replacement) static MatrixBlockLibMatrixReplace.replaceOperations(MatrixBlock in, MatrixBlock ret, double pattern, double replacement, int k) final MatrixBlockMatrixBlock.replaceOperations(MatrixValue result, double pattern, double replacement) MatrixBlock.replaceOperations(MatrixValue result, double pattern, double replacement, int k) static MatrixBlockLibMatrixReorg.reshape(MatrixBlock in, int rows, int cols, boolean rowwise) CP reshape operation (single input, single output matrix) NOTE: In contrast to R, the rowwise parameter specifies both the read and write order, with row-wise being the default, while R uses always a column-wise read, rowwise specifying the write order and column-wise being the default.static MatrixBlockLibMatrixReorg.reshape(MatrixBlock in, MatrixBlock out, int rows, int cols, boolean rowwise) CP reshape operation (single input, single output matrix) NOTE: In contrast to R, the rowwise parameter specifies both the read and write order, with row-wise being the default, while R uses always a column-wise read, rowwise specifying the write order and column-wise being the default.static MatrixBlockLibMatrixReorg.reshape(MatrixBlock in, MatrixBlock out, int rows, int cols, boolean rowwise, int k) CP reshape operation (single input, single output matrix) NOTE: In contrast to R, the rowwise parameter specifies both the read and write order, with row-wise being the default, while R uses always a column-wise read, rowwise specifying the write order and column-wise being the default.MatrixBlock.reshape(int rows, int cols, boolean byRow) static MatrixBlockLibMatrixReorg.rev(MatrixBlock in, MatrixBlock out) static MatrixBlockLibMatrixReorg.rev(MatrixBlock in, MatrixBlock out, int k) static MatrixBlockLibMatrixReorg.rexpand(MatrixBlock in, MatrixBlock ret, double max, boolean rows, boolean cast, boolean ignore, int k) CP rexpand operation (single input, single output), the classic example of this operation is one hot encoding of a column to multiple columns.static MatrixBlockLibMatrixReorg.rexpand(MatrixBlock in, MatrixBlock ret, int max, boolean rows, boolean cast, boolean ignore, int k) CP rexpand operation (single input, single output), the classic example of this operation is one hot encoding of a column to multiple columns.MatrixBlock.rexpandOperations(MatrixBlock ret, double max, boolean rows, boolean cast, boolean ignore, int k) static MatrixBlockLibMatrixReorg.rmempty(MatrixBlock in, MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) CP rmempty operation (single input, single output matrix)static MatrixBlockLibMatrixReorg.rmemptyEarlyAbort(MatrixBlock in, MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) Handle the early-termination cases of removeEmpty that do not require scanning for empty rows/columns.static MatrixBlockLibMatrixReorg.rmemptyUnsafe(MatrixBlock in, MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) static MatrixBlockLibMatrixReorg.roll(MatrixBlock in, MatrixBlock out, int shift) static MatrixBlockLibMatrixReorg.roll(MatrixBlock input, MatrixBlock output, int shift, int numThreads) final MatrixBlockMatrixBlock.rowSum()Wrapper method for single threaded reduceall-rowSum of a matrix.final MatrixBlockMatrixBlock.rowSum(int k) Wrapper method for multi threaded reduceall-rowSum of a matrix.static MatrixBlockMatrixBlock.sampleOperations(long range, int size, boolean replace, long seed) MatrixBlock.scalarOperations(ScalarOperator op, MatrixValue result) static MatrixBlockMatrixBlock.seqOperations(double from, double to, double incr) Method to generate a sequence according to the given parameters.MatrixBlock.seqOperationsInPlace(double from, double to, double incr) final MatrixBlockMatrixBlock.slice(int rl, int ru) final MatrixBlockMatrixBlock.slice(int rl, int ru, boolean deep) final MatrixBlockMatrixBlock.slice(int rl, int ru, int cl, int cu) final MatrixBlockMatrixBlock.slice(int rl, int ru, int cl, int cu, boolean deep) MatrixBlock.slice(int rl, int ru, int cl, int cu, boolean deep, MatrixBlock ret) final MatrixBlockMatrixBlock.slice(int rl, int ru, int cl, int cu, MatrixBlock ret) final MatrixBlockMatrixBlock.slice(IndexRange ixrange, MatrixBlock ret) static MatrixBlockLibMatrixReorg.sort(MatrixBlock in, MatrixBlock out, int[] by, boolean desc, boolean ixret) static MatrixBlockLibMatrixReorg.sort(MatrixBlock in, MatrixBlock out, int[] by, boolean desc, boolean ixret, int k) final MatrixBlockMatrixBlock.sortOperations()final MatrixBlockMatrixBlock.sortOperations(MatrixValue weights) final MatrixBlockMatrixBlock.sortOperations(MatrixValue weights, MatrixBlock result) MatrixBlock.sortOperations(MatrixValue weights, MatrixBlock result, int k) final MatrixBlockMatrixBlock.sparseToDense()MatrixBlock.sparseToDense(int k) static MatrixBlock[]LibMatrixSTFT.stft(MatrixBlock re, int windowSize, int overlap, int threads) Function to perform STFT on a given matrices with windowSize and overlap.static MatrixBlock[]LibMatrixSTFT.stft(MatrixBlock re, MatrixBlock im, int windowSize, int overlap, int threads) Function to perform STFT on two given matrices with windowSize and overlap.MatrixBlock.sum(int k) Wrapper method for reduceall-sum of a matrix parallelfinal MatrixBlockMatrixBlock.ternaryOperations(TernaryOperator op, MatrixBlock m2, MatrixBlock m3) MatrixBlock.ternaryOperations(TernaryOperator op, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret) CTableMap.toMatrixBlock(int rlen, int clen) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, int k) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, int k, boolean allowCSR) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, MatrixBlock out) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, MatrixBlock out, int k) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, MatrixBlock out, int k, boolean allowCSR) final MatrixBlockMatrixBlock.transpose()Transpose this MatrixBlockMatrixBlock.transpose(int k) Transpose this MatrixBlock leveraging parallelzation degree kstatic MatrixBlockLibMatrixReorg.transposeInPlace(MatrixBlock in, int k) MatrixBlock.transposeSelfMatrixMultOperations(MatrixBlock out, MMTSJ.MMTSJType tstype) MatrixBlock.transposeSelfMatrixMultOperations(MatrixBlock out, MMTSJ.MMTSJType tstype, int k) MatrixBlock.uaggouterchainOperations(MatrixBlock mbLeft, MatrixBlock mbRight, MatrixBlock mbOut, BinaryOperator bOp, AggregateUnaryOperator uaggOp) static MatrixBlockLibCommonsMath.unaryOperations(MatrixBlock inj, String opcode) final MatrixBlockMatrixBlock.unaryOperations(UnaryOperator op) MatrixBlock.unaryOperations(UnaryOperator op, MatrixValue result) static MatrixBlockLibMatrixBincell.uncellOp(MatrixBlock m1, MatrixBlock ret, UnaryOperator op) MatrixBlock.unionOperations(MatrixBlock m1, MatrixBlock m2) MatrixBlock.zeroOutOperations(MatrixValue result, IndexRange range) Methods in org.apache.sysds.runtime.matrix.data that return types with arguments of type MatrixBlockMethods in org.apache.sysds.runtime.matrix.data with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic voidLibMatrixAgg.aggregateBinaryMatrix(MatrixBlock in, MatrixBlock aggVal, MatrixBlock aggCorr, boolean deep) Core incremental matrix aggregate (ak+) as used in mapmult, tsmm, cpmm, etc.static voidLibMatrixAgg.aggregateBinaryMatrix(MatrixBlock in, MatrixBlock aggVal, AggregateOperator aop) Core incremental matrix aggregate (ak+) as used for uack+ and acrk+.MatrixBlock.aggregateBinaryOperations(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, AggregateBinaryOperator op) final MatrixBlockMatrixBlock.aggregateBinaryOperations(MatrixBlock m1, MatrixBlock m2, AggregateBinaryOperator op) static CmCovObjectLibMatrixAgg.aggregateCmCov(MatrixBlock in1, MatrixBlock in2, MatrixBlock in3, ValueFunction fn) Single threaded Covariance and Central Moment operations CM = Central Moment COV = Covariancestatic CmCovObjectLibMatrixAgg.aggregateCmCov(MatrixBlock in1, MatrixBlock in2, MatrixBlock in3, ValueFunction fn, int k) Multi threaded Covariance and Central Moment operations CM = Central Moment COV = Covariancestatic voidLibMatrixOuterAgg.aggregateMatrix(MatrixBlock in1Val, MatrixBlock outVal, double[] bv, int[] bvi, BinaryOperator bOp, AggregateUnaryOperator uaggOp) static MatrixBlockLibMatrixAgg.aggregateTernary(MatrixBlock in1, MatrixBlock in2, MatrixBlock in3, MatrixBlock ret, AggregateTernaryOperator op) static MatrixBlockLibMatrixAgg.aggregateTernary(MatrixBlock in1, MatrixBlock in2, MatrixBlock in3, MatrixBlock ret, AggregateTernaryOperator op, int k) static MatrixBlockMatrixBlock.aggregateTernaryOperations(MatrixBlock m1, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret, AggregateTernaryOperator op, boolean inCP) static voidLibMatrixAggUnarySpecialization.aggregateUnary(MatrixBlock mb, AggregateUnaryOperator op, MatrixBlock result, int blen, MatrixIndexes indexesIn) static voidLibMatrixAgg.aggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, AggregateUnaryOperator uaop) static voidLibMatrixAgg.aggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, AggregateUnaryOperator uaop, boolean allowReformatToSparse) static voidLibMatrixAgg.aggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, AggregateUnaryOperator uaop, int k) static MatrixBlockLibMatrixAgg.aggregateUnaryMatrix(AggregateUnaryOperator op, MatrixBlock in, MatrixValue result, int blen, MatrixIndexes indexesIn, boolean inCP) static MatrixBlockLibMatrixAppend.append(MatrixBlock a, MatrixBlock[] that, MatrixBlock result, boolean cbind) static MatrixBlockMatrixBlock.append(List<MatrixBlock> that, MatrixBlock ret, boolean cbind, int k) Append that list of matrixblocks together.final MatrixBlockMatrixBlock.append(MatrixBlock that) Append that matrix to this matrix, while allocating a new matrix.MatrixBlock.append(MatrixBlock[] that, MatrixBlock result, boolean cbind) Append that list of matrixes to this matrix.final MatrixBlockMatrixBlock.append(MatrixBlock that, boolean cbind) Append that matrix to this matrix, while allocating a new matrix.final MatrixBlockMatrixBlock.append(MatrixBlock that, MatrixBlock ret) Append that matrix to this matrix.final MatrixBlockMatrixBlock.append(MatrixBlock that, MatrixBlock ret, boolean cbind) Append that matrix to this matrix.voidMatrixBlock.appendRowToSparse(SparseBlock dest, MatrixBlock src, int i, int rowoffset, int coloffset, boolean deep) voidMatrixBlock.appendToSparse(MatrixBlock that, int rowoffset, int coloffset) static voidLibMatrixDNN.batchNorm2D(MatrixBlock image, MatrixBlock scale, MatrixBlock bias, MatrixBlock runningMean, MatrixBlock runningVar, String phase, double epsilon, double mu, MatrixBlock ret, MatrixBlock retRunningMean, MatrixBlock retRunningVar, MatrixBlock resultSaveMean, MatrixBlock resultSaveInvVariance) static voidLibMatrixDNN.batchNorm2DBackward(MatrixBlock image, MatrixBlock dout, MatrixBlock scale, double epsilon, MatrixBlock resultSaveMean, MatrixBlock resultSaveInvVariance, MatrixBlock dX, MatrixBlock dScale, MatrixBlock dBias) static voidLibMatrixDNN.biasAdd(MatrixBlock input, MatrixBlock bias, MatrixBlock outputBlock, int numThreads) Performs the operation corresponding to the DML script: ones = matrix(1, rows=1, cols=Hout*Wout) output = input + matrix(bias %*% ones, rows=1, cols=F*Hout*Wout) This operation is often followed by conv2d and hence we have introduced bias_add(input, bias) built-in functionstatic voidLibMatrixDNN.biasMultiply(MatrixBlock input, MatrixBlock bias, MatrixBlock outputBlock, int numThreads) Performs the operation corresponding to the DML script: ones = matrix(1, rows=1, cols=Hout*Wout) output = input * matrix(bias %*% ones, rows=1, cols=F*Hout*Wout) This operation is often followed by conv2d and hence we have introduced bias_multiply(input, bias) built-in functionstatic MatrixBlockLibMatrixBincell.bincellOp(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, BinaryOperator op) static MatrixBlockLibMatrixBincell.bincellOpInPlace(MatrixBlock m1ret, MatrixBlock m2, BinaryOperator op) NOTE: operations in place always require m1 and m2 to be of equal dimensions defaults to right side operations, updating the m1 matrix with like: m1ret op m2static MatrixBlockLibMatrixBincell.bincellOpInPlaceLeft(MatrixBlock m1ret, MatrixBlock m2, BinaryOperator op) Left side operations, updating the m1 matrix like: m2 op m1retstatic MatrixBlockLibMatrixBincell.bincellOpInPlaceRight(MatrixBlock m1ret, MatrixBlock m2, BinaryOperator op) Right side operations, updating the m1 matrix like: m1ret op m2static MatrixBlockLibMatrixBincell.bincellOpScalar(MatrixBlock m1, MatrixBlock ret, ScalarOperator op, int k) MatrixBlock.chainMatrixMultOperations(MatrixBlock v, MatrixBlock w, MatrixBlock out, MapMultChain.ChainType ctype) MatrixBlock.chainMatrixMultOperations(MatrixBlock v, MatrixBlock w, MatrixBlock out, MapMultChain.ChainType ctype, int k) static voidLibMatrixDNN.channelSums(MatrixBlock input, MatrixBlock outputBlock, int C, int HW) Perform channel sum operationstatic voidMatrixBlock.checkCMOperations(MatrixBlock mb, CMOperator op) static voidLibMatrixAppend.checkDimensionsForAppend(MatrixBlock[] in, boolean cbind, int rlen, int clen) static voidLibMatrixReorg.checkRexpand(MatrixBlock in, boolean ignore) Quick check if the input is valid for rexpand, this check does not guarantee that the input is valid for rexpandMatrixBlock.cmOperations(CMOperator op, MatrixBlock weights) static voidLibMatrixDNNIm2Col.col2imOverSingleImage(int outputN, MatrixBlock input, DnnParameters params) static MatrixBlockLibCommonsMath.computeSolve(MatrixBlock in1, MatrixBlock in2) Function to solve a given system of equations.MatrixBlock.containsVector(MatrixBlock pattern, boolean earlyAbort) static voidLibMatrixDNN.conv2d(MatrixBlock input, MatrixBlock filter, MatrixBlock outputBlock, DnnParameters params) This method performs convolution (i.e.static voidLibMatrixNative.conv2d(MatrixBlock input, MatrixBlock filter, MatrixBlock outputBlock, DnnParameters params) This method performs convolution (i.e.static voidLibMatrixDNN.conv2dBackwardData(MatrixBlock filter, MatrixBlock dout, MatrixBlock outputBlock, DnnParameters params) This method computes the backpropogation errors for previous layer of convolution operationstatic voidLibMatrixNative.conv2dBackwardData(MatrixBlock filter, MatrixBlock dout, MatrixBlock outputBlock, DnnParameters params) This method computes the backpropagation errors for previous layer of convolution operationstatic voidLibMatrixDNN.conv2dBackwardFilter(MatrixBlock input, MatrixBlock dout, MatrixBlock outputBlock, DnnParameters params) This method computes the backpropogation errors for filter of convolution operationstatic voidLibMatrixNative.conv2dBackwardFilter(MatrixBlock input, MatrixBlock dout, MatrixBlock outputBlock, DnnParameters params) This method computes the backpropogation errors for filter of convolution operationvoidBinaryBlockToTextCellConverter.convert(MatrixIndexes k1, MatrixBlock v1) Before calling convert, please make sure to setBlockSize(blen, blen);voidMatrixBlock.copy(int rl, int ru, int cl, int cu, MatrixBlock src, boolean awareDestNZ) In-place copy of matrix src into the index range of the existing current matrix.static voidLibMatrixReorg.copyDenseMtx(MatrixBlock in, MatrixBlock out, int inIdx, int outIdx, int copyLen, boolean isAllocated, boolean copyTotalNonZeros) static voidLibMatrixReorg.copyMtx(MatrixBlock in, MatrixBlock out, int inStart, int outStart, int copyLen, boolean isAllocated, boolean copyTotalNonZeros) MatrixBlock.copyShallow(MatrixBlock that) static longLibMatrixMult.copyUpperToLowerTriangle(MatrixBlock ret) static longLibMatrixMult.copyUpperToLowerTriangleDense(MatrixBlock ret) Used for all version of TSMM where the result is known to be symmetric.static longLibMatrixMult.copyUpperToLowerTriangleSparse(MatrixBlock ret) static Future<int[]>LibMatrixReorg.countNNZColumns(MatrixBlock in, int k, ExecutorService pool) LibMatrixReorg.countNNZColumnsFuture(MatrixBlock in, int k, ExecutorService pool) static int[]LibMatrixReorg.countNnzPerColumn(MatrixBlock in) static int[]LibMatrixReorg.countNnzPerColumn(MatrixBlock in, int rl, int ru) MatrixBlock.covOperations(COVOperator op, MatrixBlock that) MatrixBlock.covOperations(COVOperator op, MatrixBlock that, MatrixBlock weights) static CorrMatrixBlockLibMatrixCountDistinct.createSketch(CountDistinctOperator op, MatrixBlock blkIn) voidMatrixBlock.ctableOperations(Operator op, double scalarThat, double scalarThat2, CTableMap resultMap, MatrixBlock resultBlock) D = ctable(A,v2,w) this <- A; scalar_that <- v2; scalar_that2 <- w; result <- D (i1,j1,v1) from input1 (this) (v2) from sclar_input2 (scalarThat) (w) from scalar_input3 (scalarThat2)voidMatrixBlock.ctableOperations(Operator op, double scalarThat, MatrixValue that2Val, CTableMap resultMap, MatrixBlock resultBlock) D = ctable(A,v2,W) this <- A; scalarThat <- v2; that2 <- W; result <- D (i1,j1,v1) from input1 (this) (v2) from sclar_input2 (scalarThat) (i3,j3,w) from input3 (that2)voidMatrixBlock.ctableOperations(Operator op, MatrixIndexes ix1, double scalarThat, boolean left, int blen, CTableMap resultMap, MatrixBlock resultBlock) Specific ctable case of ctable(seq(...),X), where X is the only matrix input.voidMatrixBlock.ctableOperations(Operator op, MatrixValue thatVal, double scalarThat2, boolean ignoreZeros, CTableMap resultMap, MatrixBlock resultBlock) D = ctable(A,B,w) this <- A; that <- B; scalar_that2 <- w; result <- D (i1,j1,v1) from input1 (this) (i1,j1,v2) from input2 (that) (w) from scalar_input3 (scalarThat2) NOTE: This method supports both vectors and matrices.voidMatrixBlock.ctableOperations(Operator op, MatrixValue thatVal, MatrixValue that2Val, CTableMap resultMap, MatrixBlock resultBlock) voidMatrixCell.ctableOperations(Operator op, double scalarThat, double scalarThat2, CTableMap resultMap, MatrixBlock resultBlock) voidMatrixCell.ctableOperations(Operator op, double scalarThat, MatrixValue that2, CTableMap resultMap, MatrixBlock resultBlock) voidMatrixCell.ctableOperations(Operator op, MatrixIndexes ix1, double scalarThat, boolean left, int blen, CTableMap resultMap, MatrixBlock resultBlock) voidMatrixCell.ctableOperations(Operator op, MatrixValue that, double scalarThat2, boolean ignoreZeros, CTableMap ctableResult, MatrixBlock ctableResultBlock) voidMatrixCell.ctableOperations(Operator op, MatrixValue that, MatrixValue that2, CTableMap resultMap, MatrixBlock resultBlock) abstract voidMatrixValue.ctableOperations(Operator op, double scalar_that, double scalar_that2, CTableMap resultMap, MatrixBlock resultBlock) abstract voidMatrixValue.ctableOperations(Operator op, double scalarThat, MatrixValue that2, CTableMap ctableResult, MatrixBlock ctableResultBlock) abstract voidMatrixValue.ctableOperations(Operator op, MatrixIndexes ix1, double scalar_that, boolean left, int blen, CTableMap resultMap, MatrixBlock resultBlock) abstract voidMatrixValue.ctableOperations(Operator op, MatrixValue that, double scalar_that2, boolean ignoreZeros, CTableMap resultMap, MatrixBlock resultBlock) abstract voidMatrixValue.ctableOperations(Operator op, MatrixValue that, MatrixValue that2, CTableMap resultMap, MatrixBlock resultBlock) final MatrixBlockMatrixBlock.ctableSeqOperations(MatrixValue thatMatrix, double thatScalar, MatrixBlock resultBlock) D = ctable(seq,A,w) this <- seq; thatMatrix <- A; thatScalar <- w; result <- D (i1,j1,v1) from input1 (this) (i1,j1,v2) from input2 (that) (w) from scalar_input3 (scalarThat2)MatrixBlock.ctableSeqOperations(MatrixValue thatMatrix, double thatScalar, MatrixBlock ret, boolean updateClen) D = ctable(seq,A,w)static MatrixBlockLibMatrixAgg.cumaggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, UnaryOperator uop) static MatrixBlockLibMatrixAgg.cumaggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, UnaryOperator uop, double[] agg) static MatrixBlockLibMatrixAgg.cumaggregateUnaryMatrix(MatrixBlock in, MatrixBlock out, UnaryOperator uop, int k) static voidLibMatrixDenseToSparse.denseToSparse(MatrixBlock r, boolean allowCSR) Convert the given matrix block to a sparse allocation.static voidLibMatrixDenseToSparse.denseToSparse(MatrixBlock r, boolean allowCSR, int k) static MatrixBlockLibMatrixReorg.diag(MatrixBlock in, MatrixBlock out) static MatrixBlockLibMatrixMult.emptyMatrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret) static booleanLibMatrixEquals.equals(MatrixBlock a, MatrixBlock b) Analyze if the two matrix blocks are equivalent, this functions even if the underlying allocation and data structure varies.static booleanLibMatrixEquals.equals(MatrixBlock a, MatrixBlock b, double eps) Analyze if the two matrix blocks are equivalent, this functions even if the underlying allocation and data structure varies.final booleanMatrixBlock.equals(MatrixBlock arg0) Analyze if the matrixBlocks are equivalent, the comparsion supports if the differnet sides are differently allocated such as sparse and dense.static MatrixBlockLibMatrixCountDistinct.estimateDistinctValues(MatrixBlock in, CountDistinctOperator op) Public method to count the number of distinct values inside a matrix.MatrixBlock.extractTriangular(MatrixBlock ret, boolean lower, boolean diag, boolean values) static MatrixBlock[]LibMatrixFourier.fft(MatrixBlock re, int threads) Function to perform FFT for a given matrix.static MatrixBlock[]LibMatrixFourier.fft(MatrixBlock re, MatrixBlock im, int threads) Function to perform FFT for two given matrices.static MatrixBlock[]LibMatrixFourier.fft_linearized(MatrixBlock re, int threads) Function to perform FFT for each row of a given matrix.static MatrixBlock[]LibMatrixFourier.fft_linearized(MatrixBlock re, MatrixBlock im, int threads) Function to perform FFT for each row of two given matrices.static MatrixBlockLibMatrixReorg.fusedSeqRexpand(int seqHeight, MatrixBlock A, double w) The DML code to activate this function:static MatrixBlockLibMatrixReorg.fusedSeqRexpand(int seqHeight, MatrixBlock A, double w, MatrixBlock ret, boolean updateClen) The DML code to activate this function:static MatrixBlockLibMatrixReorg.fusedSeqRexpand(int seqHeight, MatrixBlock A, double w, MatrixBlock ret, boolean updateClen, int k) The DML code to activate this function:static voidLibMatrixDatagen.generateRandomMatrix(MatrixBlock out, RandomMatrixGenerator rgen, org.apache.commons.math3.random.Well1024a bigrand, long bSeed) Function to generate a matrix of random numbers.static voidLibMatrixDatagen.generateRandomMatrix(MatrixBlock out, RandomMatrixGenerator rgen, org.apache.commons.math3.random.Well1024a bigrand, long bSeed, int k) Function to generate a matrix of random numbers.static voidLibMatrixDatagen.generateSample(MatrixBlock out, long range, int size, boolean replace, long seed) Generates a sample of sizesizefrom a range of values [1,range].static voidLibMatrixDatagen.generateSequence(MatrixBlock out, double from, double to, double incr) Method to generate a sequence according to the given parameters.static voidLibMatrixDatagen.genRandomNumbers(boolean invokedFromCP, int rl, int ru, int cl, int cu, MatrixBlock out, RandomMatrixGenerator rgen, long bSeed, long[] seeds) LibMatrixBincell.getBinaryAccessType(MatrixBlock m1, MatrixBlock m2) LibMatrixBincell.getBinaryAccessTypeExtended(MatrixBlock m1, MatrixBlock m2) static MatrixBlockLibMatrixSketch.getUniqueValues(MatrixBlock blkIn, Types.Direction dir) LibMatrixDNNRotate180.Rotate180Worker.getWorker(MatrixBlock in, MatrixBlock out, DnnParameters params, boolean zeroOutSparseOutput, boolean trans) static voidLibMatrixAgg.groupedAggregate(MatrixBlock groups, MatrixBlock target, MatrixBlock weights, MatrixBlock result, int numGroups, Operator op) static voidLibMatrixAgg.groupedAggregate(MatrixBlock groups, MatrixBlock target, MatrixBlock weights, MatrixBlock result, int numGroups, Operator op, int k) static MatrixBlock[]LibMatrixFourier.ifft(MatrixBlock re, int threads) Function to perform IFFT for a given matrix.static MatrixBlock[]LibMatrixFourier.ifft(MatrixBlock re, MatrixBlock im, int threads) Function to perform IFFT for two given matrices.static MatrixBlock[]LibMatrixFourier.ifft_linearized(MatrixBlock re, int threads) Function to perform IFFT for each row of a given matrix.static MatrixBlock[]LibMatrixFourier.ifft_linearized(MatrixBlock re, MatrixBlock im, int threads) Function to perform IFFT for each row of two given matrices.static voidLibMatrixDNNIm2Col.im2col(MatrixBlock in, MatrixBlock out, int r, int C, int R, int S, int H, int W, int P, int Q, int stride_h, int stride_w, int pad_h, int pad_w, boolean trans) static voidLibMatrixDNNIm2Col.im2col(MatrixBlock in, MatrixBlock out, int r, DnnParameters params, boolean trans) static voidLibMatrixDNNIm2Col.im2colSparse(MatrixBlock in, MatrixBlock out, int r, int C, int R, int S, int H, int W, int P, int Q, int stride_h, int stride_w, int pad_h, int pad_w, boolean trans) static booleanLibMatrixMult.isSparseOutputMatrixMult(MatrixBlock m1, MatrixBlock m2) static booleanLibMatrixMult.isSparseOutputTSMM(MatrixBlock m1) static booleanLibMatrixMult.isSparseOutputTSMM(MatrixBlock m1, boolean ultraSparse) static booleanLibMatrixMult.isUltraSparseMatrixMult(MatrixBlock m1, MatrixBlock m2, boolean m1Perm) static voidLibMatrixBincell.isValidDimensionsBinary(MatrixBlock m1, MatrixBlock m2) MatrixBlock.leftIndexingOperations(ScalarObject scalar, int rl, int cl, MatrixBlock ret, MatrixObject.UpdateType update) Explicitly allow left indexing for scalars.MatrixBlock.leftIndexingOperations(MatrixBlock rhsMatrix, int rl, int ru, int cl, int cu, MatrixBlock ret, MatrixObject.UpdateType update) final MatrixBlockMatrixBlock.leftIndexingOperations(MatrixBlock rhsMatrix, IndexRange ixrange, MatrixBlock ret, MatrixObject.UpdateType update) static voidLibMatrixDNNLSTM.lstmTile(int n, int d, int T, int m, int start, int end, MatrixBlock x, MatrixBlock w, MatrixBlock bias, MatrixBlock out0, MatrixBlock c0, boolean return_sequences, MatrixBlock out, MatrixBlock cout, MatrixBlock cache_out, MatrixBlock cache_c, MatrixBlock cache_ifog) static MatrixBlockOperationsOnMatrixValues.matMult(MatrixBlock value1, MatrixBlock value2, MatrixBlock valueOut, AggregateBinaryOperator op) static MatrixBlockOperationsOnMatrixValues.matMult(MatrixBlock value1, MatrixBlock value2, AggregateBinaryOperator op) static MatrixBlockOperationsOnMatrixValues.matMult(MatrixIndexes indexes1, MatrixBlock value1, MatrixIndexes indexes2, MatrixBlock value2, MatrixIndexes indexesOut, MatrixBlock valueOut, AggregateBinaryOperator op) static MatrixBlockLibCommonsMath.matrixMatrixOperations(MatrixBlock in1, MatrixBlock in2, String opcode) static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2) Performs a matrix multiplication All variants use a IKJ access pattern, and internally use dense output.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, int k) Performs a matrix multiplication All variants use a IKJ access pattern, and internally use dense output.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret) Performs a matrix multiplication and stores the result in the output matrix.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, boolean fixedRet) This method allows one to disabling exam sparsity.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, boolean fixedRet, int k) Performs a matrix multiplication and stores the result in the output matrix.static MatrixBlockLibMatrixMult.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) Performs a multi-threaded matrix multiplication and stores the result in the output matrix.static MatrixBlockLibMatrixNative.matrixMult(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) Performs matrix multiplication using native library if BLAS is available or else falls back to Java BLAS.static voidLibMatrixMult.matrixMultChain(MatrixBlock mX, MatrixBlock mV, MatrixBlock mW, MatrixBlock ret, MapMultChain.ChainType ct) Performs a matrix multiplication chain operation of type t(X)%*%(X%*%v) or t(X)%*%(w*(X%*%v)).static voidLibMatrixMult.matrixMultChain(MatrixBlock mX, MatrixBlock mV, MatrixBlock mW, MatrixBlock ret, MapMultChain.ChainType ct, int k) Performs a parallel matrix multiplication chain operation of type t(X)%*%(X%*%v) or t(X)%*%(w*(X%*%v)).static MatrixBlockLibMatrixMult.matrixMultNonNative(MatrixBlock m1, MatrixBlock m2, MatrixBlock ret, int k) static voidLibMatrixMult.matrixMultPermute(MatrixBlock pm1, MatrixBlock m2, MatrixBlock ret1, MatrixBlock ret2) static voidLibMatrixMult.matrixMultPermute(MatrixBlock pm1, MatrixBlock m2, MatrixBlock ret1, MatrixBlock ret2, int k) static MatrixBlockLibMatrixMult.matrixMultTransposeSelf(MatrixBlock m1, MatrixBlock ret, boolean leftTranspose) static voidLibMatrixMult.matrixMultTransposeSelf(MatrixBlock m1, MatrixBlock ret, boolean leftTranspose, boolean copyToLowerTriangle) static voidLibMatrixMult.matrixMultTransposeSelf(MatrixBlock m1, MatrixBlock ret, boolean leftTranspose, int k) TSMM with optional transposed left side or not (Transposed self matrix multiplication)static voidLibMatrixMult.matrixMultWCeMM(MatrixBlock mW, MatrixBlock mU, MatrixBlock mV, double eps, MatrixBlock ret, WeightedCrossEntropy.WCeMMType wt) static voidLibMatrixMult.matrixMultWCeMM(MatrixBlock mW, MatrixBlock mU, MatrixBlock mV, double eps, MatrixBlock ret, WeightedCrossEntropy.WCeMMType wt, int k) static voidLibMatrixMult.matrixMultWDivMM(MatrixBlock mW, MatrixBlock mU, MatrixBlock mV, MatrixBlock mX, MatrixBlock ret, WeightedDivMM.WDivMMType wt) NOTE: This operation has limited NaN support, which is acceptable because all our sparse-safe operations have only limited NaN support.static voidLibMatrixMult.matrixMultWDivMM(MatrixBlock mW, MatrixBlock mU, MatrixBlock mV, MatrixBlock mX, MatrixBlock ret, WeightedDivMM.WDivMMType wt, int k) NOTE: This operation has limited NaN support, which is acceptable because all our sparse-safe operations have only limited NaN support.static voidLibMatrixMult.matrixMultWSigmoid(MatrixBlock mW, MatrixBlock mU, MatrixBlock mV, MatrixBlock ret, WeightedSigmoid.WSigmoidType wt) static voidLibMatrixMult.matrixMultWSigmoid(MatrixBlock mW, MatrixBlock mU, MatrixBlock mV, MatrixBlock ret, WeightedSigmoid.WSigmoidType wt, int k) static voidLibMatrixMult.matrixMultWSLoss(MatrixBlock mX, MatrixBlock mU, MatrixBlock mV, MatrixBlock mW, MatrixBlock ret, WeightedSquaredLoss.WeightsType wt) static voidLibMatrixMult.matrixMultWSLoss(MatrixBlock mX, MatrixBlock mU, MatrixBlock mV, MatrixBlock mW, MatrixBlock ret, WeightedSquaredLoss.WeightsType wt, int k) static voidLibMatrixMult.matrixMultWuMM(MatrixBlock mW, MatrixBlock mU, MatrixBlock mV, MatrixBlock ret, WeightedUnaryMM.WUMMType wt, ValueFunction fn) static voidLibMatrixMult.matrixMultWuMM(MatrixBlock mW, MatrixBlock mU, MatrixBlock mV, MatrixBlock ret, WeightedUnaryMM.WUMMType wt, ValueFunction fn, int k) MatrixBlock.merge(MatrixBlock that, boolean appendOnly) MatrixBlock.merge(MatrixBlock that, boolean appendOnly, boolean par) MatrixBlock.merge(MatrixBlock that, boolean appendOnly, boolean par, boolean deep) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in, String opcode, int threads) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in, String opcode, int threads, int num_iterations, double tol) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in, String opcode, int threads, long seed) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in1, MatrixBlock in2, String opcode) static MatrixBlock[]LibCommonsMath.multiReturnOperations(MatrixBlock in1, MatrixBlock in2, String opcode, int threads, long seed) static MatrixBlockMatrixBlock.naryOperations(Operator op, MatrixBlock[] matrices, ScalarObject[] scalars, MatrixBlock ret) static voidOperationsOnMatrixValues.performMapGroupedAggregate(Operator op, IndexedMatrixValue inTarget, MatrixBlock groups, int ngroups, int blen, ArrayList<IndexedMatrixValue> outlist) static ListOperationsOnMatrixValues.performSlice(IndexRange ixrange, int blen, int iix, int jix, MatrixBlock in) static MatrixBlockLibMatrixReorg.permute(MatrixBlock in, int[] inDims, int[] perm) static MatrixBlockLibMatrixReorg.permute(MatrixBlock in, int[] inDims, int[] perm, int k) static voidLibMatrixDNN.pooling(MatrixBlock input, MatrixBlock output, DnnParameters params, LibMatrixDNN.PoolingType poolType) static voidLibMatrixDNN.poolingBackward(MatrixBlock input, MatrixBlock dout, MatrixBlock outputBlock, DnnParameters params, boolean performReluBackward, LibMatrixDNN.PoolingType poolType) This method computes the backpropogation errors for previous layer of pooling operationstatic voidLibMatrixDNNIm2Col.preallocateSparseOutput(MatrixBlock in, MatrixBlock out) static MatrixBlockLibMatrixAgg.prepareAggregateUnaryOutput(MatrixBlock in, AggregateUnaryOperator op, MatrixValue result, int blen) static MatrixBlockLibMatrixMult.prepMatrixMultTransposeSelfInput(MatrixBlock m1, boolean leftTranspose, boolean par) voidMatrixBlock.putInto(MatrixBlock target, int rowOffset, int colOffset, boolean sparseCopyShallow) Method for copying this matrix into a target matrix.final MatrixBlockMatrixBlock.quaternaryOperations(QuaternaryOperator qop, MatrixBlock um, MatrixBlock vm, MatrixBlock wm, MatrixBlock out) MatrixBlock.quaternaryOperations(QuaternaryOperator qop, MatrixBlock U, MatrixBlock V, MatrixBlock wm, MatrixBlock out, int k) final MatrixBlockMatrixBlock.rbind(MatrixBlock that) static voidLibMatrixAgg.recomputeIndexes(MatrixBlock out, AggregateUnaryOperator op, int blen, MatrixIndexes ix) Recompute outputs (e.g., maxindex or minindex) according to block indexes from MR.static voidLibMatrixDNN.reluBackward(MatrixBlock input, MatrixBlock dout, MatrixBlock outputBlock, int numThreads) This method computes the backpropagation errors for previous layer of relu operationfinal MatrixBlockMatrixBlock.removeEmptyOperations(MatrixBlock ret, boolean rows, boolean emptyReturn) MatrixBlock.removeEmptyOperations(MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) static MatrixBlockLibMatrixReorg.removeEmptyRows(MatrixBlock in, MatrixBlock ret, boolean emptyReturn, boolean selectNull, boolean[] flags, int rlen2) Remove selected rows, based on the boolean array given.static MatrixBlockLibMatrixReorg.reorg(MatrixBlock in, MatrixBlock out, ReorgOperator op) static MatrixBlockLibMatrixReorg.reorgInPlace(MatrixBlock in, ReorgOperator op) static MatrixBlockLibMatrixReplace.replaceOperations(MatrixBlock in, MatrixBlock ret, double pattern, double replacement) static MatrixBlockLibMatrixReplace.replaceOperations(MatrixBlock in, MatrixBlock ret, double pattern, double replacement, int k) static BinaryOperatorLibMatrixBincell.replaceOpWithSparseSafeIfApplicable(MatrixBlock m1, MatrixBlock m2, BinaryOperator op) static voidLibMatrixOuterAgg.resetOutputMatrix(MatrixIndexes in1Ix, MatrixBlock in1Val, MatrixIndexes outIx, MatrixBlock outVal, AggregateUnaryOperator uaggOp) ReSet output matrixstatic MatrixBlockLibMatrixReorg.reshape(MatrixBlock in, int rows, int cols, boolean rowwise) CP reshape operation (single input, single output matrix) NOTE: In contrast to R, the rowwise parameter specifies both the read and write order, with row-wise being the default, while R uses always a column-wise read, rowwise specifying the write order and column-wise being the default.static MatrixBlockLibMatrixReorg.reshape(MatrixBlock in, MatrixBlock out, int rows, int cols, boolean rowwise) CP reshape operation (single input, single output matrix) NOTE: In contrast to R, the rowwise parameter specifies both the read and write order, with row-wise being the default, while R uses always a column-wise read, rowwise specifying the write order and column-wise being the default.static MatrixBlockLibMatrixReorg.reshape(MatrixBlock in, MatrixBlock out, int rows, int cols, boolean rowwise, int k) CP reshape operation (single input, single output matrix) NOTE: In contrast to R, the rowwise parameter specifies both the read and write order, with row-wise being the default, while R uses always a column-wise read, rowwise specifying the write order and column-wise being the default.static MatrixBlockLibMatrixReorg.rev(MatrixBlock in, MatrixBlock out) static MatrixBlockLibMatrixReorg.rev(MatrixBlock in, MatrixBlock out, int k) static MatrixBlockLibMatrixReorg.rexpand(MatrixBlock in, MatrixBlock ret, double max, boolean rows, boolean cast, boolean ignore, int k) CP rexpand operation (single input, single output), the classic example of this operation is one hot encoding of a column to multiple columns.static MatrixBlockLibMatrixReorg.rexpand(MatrixBlock in, MatrixBlock ret, int max, boolean rows, boolean cast, boolean ignore, int k) CP rexpand operation (single input, single output), the classic example of this operation is one hot encoding of a column to multiple columns.MatrixBlock.rexpandOperations(MatrixBlock ret, double max, boolean rows, boolean cast, boolean ignore, int k) static MatrixBlockLibMatrixReorg.rmempty(MatrixBlock in, MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) CP rmempty operation (single input, single output matrix)static MatrixBlockLibMatrixReorg.rmemptyEarlyAbort(MatrixBlock in, MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) Handle the early-termination cases of removeEmpty that do not require scanning for empty rows/columns.static MatrixBlockLibMatrixReorg.rmemptyUnsafe(MatrixBlock in, MatrixBlock ret, boolean rows, boolean emptyReturn, MatrixBlock select) static MatrixBlockLibMatrixReorg.roll(MatrixBlock in, MatrixBlock out, int shift) static MatrixBlockLibMatrixReorg.roll(MatrixBlock input, MatrixBlock output, int shift, int numThreads) static booleanLibMatrixAgg.satisfiesMultiThreadingConstraints(MatrixBlock in, int k) static booleanLibMatrixAgg.satisfiesMultiThreadingConstraints(MatrixBlock in, MatrixBlock out, AggregateUnaryOperator uaop, int k) static booleanLibMatrixMult.satisfiesMultiThreadingConstraints(MatrixBlock m1, boolean checkMem, boolean checkFLOPs, long FPfactor, int k) static booleanLibMatrixMult.satisfiesMultiThreadingConstraints(MatrixBlock m1, int k) static booleanLibMatrixMult.satisfiesMultiThreadingConstraints(MatrixBlock m1, MatrixBlock m2, boolean checkMem, boolean checkFLOPs, long FPfactor, int k) MatrixBlock.slice(int rl, int ru, int cl, int cu, boolean deep, MatrixBlock ret) final MatrixBlockMatrixBlock.slice(int rl, int ru, int cl, int cu, MatrixBlock ret) final MatrixBlockMatrixBlock.slice(IndexRange ixrange, MatrixBlock ret) static MatrixBlockLibMatrixReorg.sort(MatrixBlock in, MatrixBlock out, int[] by, boolean desc, boolean ixret) static MatrixBlockLibMatrixReorg.sort(MatrixBlock in, MatrixBlock out, int[] by, boolean desc, boolean ixret, int k) final MatrixBlockMatrixBlock.sortOperations(MatrixValue weights, MatrixBlock result) MatrixBlock.sortOperations(MatrixValue weights, MatrixBlock result, int k) static voidLibMatrixSparseToDense.sparseToDense(MatrixBlock r, int k) Convert the given matrix block to a Dense allocation.static MatrixBlock[]LibMatrixSTFT.stft(MatrixBlock re, int windowSize, int overlap, int threads) Function to perform STFT on a given matrices with windowSize and overlap.static MatrixBlock[]LibMatrixSTFT.stft(MatrixBlock re, MatrixBlock im, int windowSize, int overlap, int threads) Function to perform STFT on two given matrices with windowSize and overlap.static voidLibMatrixTercell.tercellOp(MatrixBlock m1, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret, TernaryOperator op) final MatrixBlockMatrixBlock.ternaryOperations(TernaryOperator op, MatrixBlock m2, MatrixBlock m3) MatrixBlock.ternaryOperations(TernaryOperator op, MatrixBlock m2, MatrixBlock m3, MatrixBlock ret) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, int k) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, int k, boolean allowCSR) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, MatrixBlock out) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, MatrixBlock out, int k) static MatrixBlockLibMatrixReorg.transpose(MatrixBlock in, MatrixBlock out, int k, boolean allowCSR) static MatrixBlockLibMatrixReorg.transposeInPlace(MatrixBlock in, int k) static voidLibMatrixReorg.transposeInPlaceDenseBrenner(MatrixBlock in, int k) Transposes a dense matrix in-place using following cycles based on Brenner's method.static booleanLibMatrixReorg.transposeInPlaceTensor(MatrixBlock in, int[] shape, int[] perm) Performs prime in-place tensor transposition for arbitrary permutations.MatrixBlock.transposeSelfMatrixMultOperations(MatrixBlock out, MMTSJ.MMTSJType tstype) MatrixBlock.transposeSelfMatrixMultOperations(MatrixBlock out, MMTSJ.MMTSJType tstype, int k) static voidLibMatrixNative.tsmm(MatrixBlock m1, MatrixBlock ret, boolean leftTrans, int k) MatrixBlock.uaggouterchainOperations(MatrixBlock mbLeft, MatrixBlock mbRight, MatrixBlock mbOut, BinaryOperator bOp, AggregateUnaryOperator uaggOp) static MatrixBlockLibCommonsMath.unaryOperations(MatrixBlock inj, String opcode) static MatrixBlockLibMatrixBincell.uncellOp(MatrixBlock m1, MatrixBlock ret, UnaryOperator op) MatrixBlock.unionOperations(MatrixBlock m1, MatrixBlock m2) Method parameters in org.apache.sysds.runtime.matrix.data with type arguments of type MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockMatrixBlock.append(List<MatrixBlock> that, MatrixBlock ret, boolean cbind, int k) Append that list of matrixblocks together.Constructors in org.apache.sysds.runtime.matrix.data with parameters of type MatrixBlockModifierConstructorDescriptionDnnParameters(int N, int D, int T, int M, MatrixBlock x, MatrixBlock w, MatrixBlock bias, MatrixBlock out0, MatrixBlock c0, boolean return_sequences, int numThreads) DnnParameters(int n, int d, int t, int m, MatrixBlock x, MatrixBlock w, MatrixBlock bias, MatrixBlock out0, MatrixBlock c0, MatrixBlock cache_out, MatrixBlock cache_c, MatrixBlock cache_ifog, boolean return_sequences, MatrixBlock dout, MatrixBlock dc, MatrixBlock dx, MatrixBlock dw, MatrixBlock db, MatrixBlock dout0, MatrixBlock dc0, int numThreads) MatrixBlock(MatrixBlock that) MatrixBlock(MatrixBlock that, boolean sp) MatrixBlock(MatrixBlock that, SparseBlock.Type stype, boolean deep) -
Uses of MatrixBlock in org.apache.sysds.runtime.matrix.data.sketch
Methods in org.apache.sysds.runtime.matrix.data.sketch that return MatrixBlockModifier and TypeMethodDescriptionMatrixSketch.getValue(MatrixBlock blkIn) Get scalar distinct count from an input matrix block.MatrixSketch.getValueFromSketch(CorrMatrixBlock blkIn) Obtain matrix distinct count value from estimation Used for estimating distinct in rows or columns.Methods in org.apache.sysds.runtime.matrix.data.sketch with parameters of type MatrixBlockModifier and TypeMethodDescriptionMatrixSketch.create(MatrixBlock blkIn) Create an initial sketch of a given block.MatrixSketch.getValue(MatrixBlock blkIn) Get scalar distinct count from an input matrix block. -
Uses of MatrixBlock in org.apache.sysds.runtime.matrix.data.sketch.countdistinct
Methods in org.apache.sysds.runtime.matrix.data.sketch.countdistinct that return MatrixBlockModifier and TypeMethodDescriptionCountDistinctFunctionSketch.getValue(MatrixBlock blkIn) CountDistinctFunctionSketch.getValueFromSketch(CorrMatrixBlock blkIn) Methods in org.apache.sysds.runtime.matrix.data.sketch.countdistinct with parameters of type MatrixBlockModifier and TypeMethodDescriptionCountDistinctFunctionSketch.create(MatrixBlock blkIn) CountDistinctFunctionSketch.getValue(MatrixBlock blkIn) -
Uses of MatrixBlock in org.apache.sysds.runtime.matrix.data.sketch.countdistinctapprox
Methods in org.apache.sysds.runtime.matrix.data.sketch.countdistinctapprox that return MatrixBlockModifier and TypeMethodDescriptionKMVSketch.getValue(MatrixBlock blkIn) KMVSketch.getValueFromSketch(CorrMatrixBlock arg0) Methods in org.apache.sysds.runtime.matrix.data.sketch.countdistinctapprox with parameters of type MatrixBlockModifier and TypeMethodDescriptionKMVSketch.create(MatrixBlock blkIn) KMVSketch.getValue(MatrixBlock blkIn) -
Uses of MatrixBlock in org.apache.sysds.runtime.matrix.operators
Methods in org.apache.sysds.runtime.matrix.operators with parameters of type MatrixBlockModifier and TypeMethodDescriptionbooleanBinaryOperator.isIntroducingZerosLeft(MatrixBlock row) Check if the operation returns zeros if the input is contained in row.booleanBinaryOperator.isIntroducingZerosRight(MatrixBlock row) Check if the operation returns zeros if the input is contained in row.booleanBinaryOperator.isRowSafeLeft(MatrixBlock row) Check if the operation returns zeros if the input zero.booleanBinaryOperator.isRowSafeRight(MatrixBlock row) Check if the operation returns zeros if the input zero. -
Uses of MatrixBlock in org.apache.sysds.runtime.transform.decode
Methods in org.apache.sysds.runtime.transform.decode with parameters of type MatrixBlockModifier and TypeMethodDescriptionabstract FrameBlockDecoder.decode(MatrixBlock in, FrameBlock out) Block decode API converting a matrix block into a frame block.Decoder.decode(MatrixBlock in, FrameBlock out, int k) Block decode API converting a matrix block into a frame block in parallel.abstract voidDecoder.decode(MatrixBlock in, FrameBlock out, int rl, int ru) Block decode row blockDecoderBin.decode(MatrixBlock in, FrameBlock out) voidDecoderBin.decode(MatrixBlock in, FrameBlock out, int rl, int ru) DecoderComposite.decode(MatrixBlock in, FrameBlock out) voidDecoderComposite.decode(MatrixBlock in, FrameBlock out, int rl, int ru) DecoderDummycode.decode(MatrixBlock in, FrameBlock out) voidDecoderDummycode.decode(MatrixBlock in, FrameBlock out, int rl, int ru) DecoderPassThrough.decode(MatrixBlock in, FrameBlock out) voidDecoderPassThrough.decode(MatrixBlock in, FrameBlock out, int rl, int ru) DecoderRecode.decode(MatrixBlock in, FrameBlock out) voidDecoderRecode.decode(MatrixBlock in, FrameBlock out, int rl, int ru) -
Uses of MatrixBlock in org.apache.sysds.runtime.transform.encode
Methods in org.apache.sysds.runtime.transform.encode that return MatrixBlockModifier and TypeMethodDescriptionColumnEncoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol) Apply Functions are only used in Single Threaded or Multi-Threaded Dense context.ColumnEncoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol, int rowStart, int blk) ColumnEncoderComposite.apply(CacheBlock<?> in, MatrixBlock out, int outputCol, int rowStart, int blk) Encoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol) Apply the generated metadata to the FrameBlock and saved the result in out.EncoderMVImpute.apply(FrameBlock in, MatrixBlock out) EncoderOmit.apply(FrameBlock in, MatrixBlock out) abstract MatrixBlockLegacyEncoder.apply(FrameBlock in, MatrixBlock out) Encode input data blockwise according to existing transform meta data (transform apply).MultiColumnEncoder.apply(CacheBlock<?> in) MultiColumnEncoder.apply(CacheBlock<?> in, int k) MultiColumnEncoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol) MultiColumnEncoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol, int k, org.apache.sysds.runtime.transform.encode.MultiColumnEncoder.EncoderMeta encm, long nnz) static MatrixBlockCompressedEncode.encode(MultiColumnEncoder enc, FrameBlock in, int k) EncoderMVImpute.encode(FrameBlock in, MatrixBlock out) EncoderOmit.encode(FrameBlock in, MatrixBlock out) abstract MatrixBlockLegacyEncoder.encode(FrameBlock in, MatrixBlock out) Block encode: build and apply (transform encode).MultiColumnEncoder.encode(CacheBlock<?> in) MultiColumnEncoder.encode(CacheBlock<?> in, boolean compressedOut) MultiColumnEncoder.encode(CacheBlock<?> in, int k) MultiColumnEncoder.encode(CacheBlock<?> in, int k, boolean compressedOut) ColumnEncoder.getColMapping(FrameBlock meta) Obtain the column mapping of encoded frames based on the passed meta data frame.LegacyEncoder.getColMapping(FrameBlock meta, MatrixBlock out) Obtain the column mapping of encoded frames based on the passed meta data frame.MultiColumnEncoder.getColMapping(FrameBlock meta) Obtain the column mapping of encoded frames based on the passed meta data frame.Methods in org.apache.sysds.runtime.transform.encode with parameters of type MatrixBlockModifier and TypeMethodDescriptionColumnEncoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol) Apply Functions are only used in Single Threaded or Multi-Threaded Dense context.ColumnEncoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol, int rowStart, int blk) ColumnEncoderComposite.apply(CacheBlock<?> in, MatrixBlock out, int outputCol, int rowStart, int blk) Encoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol) Apply the generated metadata to the FrameBlock and saved the result in out.EncoderMVImpute.apply(FrameBlock in, MatrixBlock out) EncoderOmit.apply(FrameBlock in, MatrixBlock out) abstract MatrixBlockLegacyEncoder.apply(FrameBlock in, MatrixBlock out) Encode input data blockwise according to existing transform meta data (transform apply).MultiColumnEncoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol) MultiColumnEncoder.apply(CacheBlock<?> in, MatrixBlock out, int outputCol, int k, org.apache.sysds.runtime.transform.encode.MultiColumnEncoder.EncoderMeta encm, long nnz) voidColumnEncoderUDF.applyDense(CacheBlock<?> in, MatrixBlock out, int outputCol, int rowStart, int blk) voidColumnEncoderWordEmbedding.applyDense(CacheBlock<?> in, MatrixBlock out, int outputCol, int rowStart, int blk) static MultiColumnEncoderEncoderFactory.createEncoder(String spec, String[] colnames, int clen, FrameBlock meta, MatrixBlock embeddings) static MultiColumnEncoderEncoderFactory.createEncoder(String spec, String[] colnames, int clen, FrameBlock meta, MatrixBlock embeddings, int minCol, int maxCol) static MultiColumnEncoderEncoderFactory.createEncoder(String spec, String[] colnames, Types.ValueType[] schema, int clen, FrameBlock meta, MatrixBlock embeddings) static MultiColumnEncoderEncoderFactory.createEncoder(String spec, String[] colnames, Types.ValueType[] schema, FrameBlock meta, MatrixBlock embeddings) EncoderMVImpute.encode(FrameBlock in, MatrixBlock out) EncoderOmit.encode(FrameBlock in, MatrixBlock out) abstract MatrixBlockLegacyEncoder.encode(FrameBlock in, MatrixBlock out) Block encode: build and apply (transform encode).List<DependencyTask<?>>ColumnEncoder.getApplyTasks(CacheBlock<?> in, MatrixBlock out, int outputCol, int[] sparseRowPointerOffsets) List<DependencyTask<?>>ColumnEncoderComposite.getApplyTasks(CacheBlock<?> in, MatrixBlock out, int outputCol, int[] sparseRowPointerOffsets) LegacyEncoder.getColMapping(FrameBlock meta, MatrixBlock out) Obtain the column mapping of encoded frames based on the passed meta data frame.voidColumnEncoder.initEmbeddings(MatrixBlock embeddings) voidColumnEncoderComposite.initEmbeddings(MatrixBlock embeddings) voidColumnEncoderWordEmbedding.initEmbeddings(MatrixBlock embeddings) voidMultiColumnEncoder.initEmbeddings(MatrixBlock embeddings) Constructors in org.apache.sysds.runtime.transform.encode with parameters of type MatrixBlockModifierConstructorDescriptionFeatureHashSparseApplyTask(ColumnEncoderFeatureHash encoder, CacheBlock<?> input, MatrixBlock out, int outputCol) FeatureHashSparseApplyTask(ColumnEncoderFeatureHash encoder, CacheBlock<?> input, MatrixBlock out, int outputCol, int startRow, int blk) -
Uses of MatrixBlock in org.apache.sysds.runtime.util
Methods in org.apache.sysds.runtime.util that return MatrixBlockModifier and TypeMethodDescriptionstatic MatrixBlockPy4jConverterUtils.allocateDenseOrSparse(int rlen, int clen, boolean isSparse) static MatrixBlockPy4jConverterUtils.allocateDenseOrSparse(long rlen, long clen, boolean isSparse) static MatrixBlockPy4jConverterUtils.convertPy4JArrayToMB(byte[] data, int rlen, int clen) static MatrixBlockPy4jConverterUtils.convertPy4JArrayToMB(byte[] data, int rlen, int clen, boolean isSparse, Types.ValueType valueType) static MatrixBlockPy4jConverterUtils.convertPy4JArrayToMB(byte[] data, int rlen, int clen, Types.ValueType valueType) static MatrixBlockPy4jConverterUtils.convertSciPyCOOToMB(byte[] data, byte[] row, byte[] col, int rlen, int clen, int nnz) static MatrixBlockPy4jConverterUtils.convertSciPyCSRToMB(byte[] data, byte[] indices, byte[] indptr, int rlen, int clen, int nnz) static MatrixBlockDataConverter.convertToMatrixBlock(double[][] data) Creates a dense Matrix Block and copies the given double matrix into it.static MatrixBlockDataConverter.convertToMatrixBlock(double[] data, boolean columnVector) Creates a dense Matrix Block and copies the given double vector into it.static MatrixBlockDataConverter.convertToMatrixBlock(int[][] data) Converts an Integer matrix to an MatrixBlockstatic MatrixBlockDataConverter.convertToMatrixBlock(HashMap<MatrixIndexes, Double> map) static MatrixBlockDataConverter.convertToMatrixBlock(HashMap<MatrixIndexes, Double> map, int rlen, int clen) NOTE: this method also ensures the specified matrix dimensionsstatic MatrixBlockDataConverter.convertToMatrixBlock(org.apache.commons.math3.linear.RealMatrix rm) static MatrixBlockDataConverter.convertToMatrixBlock(FrameBlock frame) Converts a frame block with arbitrary schema into a matrix block.static MatrixBlockDataConverter.convertToMatrixBlock(CTableMap map) static MatrixBlockDataConverter.convertToMatrixBlock(CTableMap map, int rlen, int clen) NOTE: this method also ensures the specified matrix dimensionsstatic MatrixBlock[]DataConverter.convertToMatrixBlockPartitions(MatrixBlock mb, boolean colwise) static MatrixBlockLocalFileUtils.readMatrixBlockFromLocal(String fname) Reads a matrix block from local file system.static MatrixBlockLocalFileUtils.readMatrixBlockFromLocal(String fname, MatrixBlock reuse) Reads a matrix block from local file system.static MatrixBlockLocalFileUtils.readMatrixBlockFromStream(InputStream is) Reads a matrix block from an input stream, using a fused buffered reader with special support for matrix blocks.static MatrixBlockDataConverter.readMatrixFromHDFS(String dir, Types.FileFormat fmt, long rlen, long clen, int blen) static MatrixBlockDataConverter.readMatrixFromHDFS(String dir, Types.FileFormat fmt, long rlen, long clen, int blen, long expectedNnz) static MatrixBlockDataConverter.readMatrixFromHDFS(String dir, Types.FileFormat fmt, long rlen, long clen, int blen, long expectedNnz, FileFormatProperties formatProperties) static MatrixBlockDataConverter.readMatrixFromHDFS(ReadProperties prop) Core method for reading matrices in format textcell, matrixmarket, binarycell, or binaryblock from HDFS into main memory.Methods in org.apache.sysds.runtime.util with parameters of type MatrixBlockModifier and TypeMethodDescriptionstatic byte[]Py4jConverterUtils.convertMBtoPy4JDenseArr(MatrixBlock mb) static org.apache.commons.math3.linear.Array2DRowRealMatrixDataConverter.convertToArray2DRowRealMatrix(MatrixBlock mb) Helper method that converts SystemDS matrix variable (varname) into a Array2DRowRealMatrix format, which is useful in invoking Apache CommonsMath.static org.apache.commons.math3.linear.BlockRealMatrixDataConverter.convertToBlockRealMatrix(MatrixBlock mb) static boolean[]DataConverter.convertToBooleanVector(MatrixBlock mb) static DenseBlockDataConverter.convertToDenseBlock(MatrixBlock mb) static DenseBlockDataConverter.convertToDenseBlock(MatrixBlock mb, boolean deep) DataConverter.convertToDoubleList(MatrixBlock mb) static double[][]DataConverter.convertToDoubleMatrix(MatrixBlock mb) Creates a two-dimensional double matrix of the input matrix block.static double[]DataConverter.convertToDoubleVector(MatrixBlock mb) static double[]DataConverter.convertToDoubleVector(MatrixBlock mb, boolean deep) static double[]DataConverter.convertToDoubleVector(MatrixBlock mb, boolean deep, boolean allowNull) static FrameBlockDataConverter.convertToFrameBlock(MatrixBlock mb) Converts a matrix block into a frame block of value type double.static FrameBlockDataConverter.convertToFrameBlock(MatrixBlock mb, int k) Converts a matrix block into a frame block of value type double.static FrameBlockDataConverter.convertToFrameBlock(MatrixBlock mb, Types.ValueType vt) Converts a matrix block into a frame block of value type given.static FrameBlockDataConverter.convertToFrameBlock(MatrixBlock mb, Types.ValueType[] schema) Converts a matrix block into a frame block of with the given schemastatic FrameBlockDataConverter.convertToFrameBlock(MatrixBlock mb, Types.ValueType[] schema, int k) Converts a matrix block into a frame block of with the given schemastatic FrameBlockDataConverter.convertToFrameBlock(MatrixBlock mb, Types.ValueType vt, int k) Converts a matrix block into a frame block of a given value type.static int[]DataConverter.convertToIntVector(MatrixBlock mb) static long[]DataConverter.convertToLongVector(MatrixBlock mb) static MatrixBlock[]DataConverter.convertToMatrixBlockPartitions(MatrixBlock mb, boolean colwise) static TensorBlockDataConverter.convertToTensorBlock(MatrixBlock mb, Types.ValueType vt, boolean toBasicTensor) static int[]DataConverter.convertVectorToIndexList(MatrixBlock mb) static voidDataConverter.copyToDoubleVector(MatrixBlock mb, double[] dest, int destPos) static IndexedMatrixValueUtilFunctions.createIndexedMatrixBlock(MatrixBlock mb, DataCharacteristics mc, long ix) static voidDnnUtils.fillBias(MatrixBlock bias, double[] outputArray, int src_rl, int src_ru, int N, int K, int PQ) static booleanSortUtils.isSorted(MatrixBlock in) static MatrixBlockLocalFileUtils.readMatrixBlockFromLocal(String fname, MatrixBlock reuse) Reads a matrix block from local file system.static StringDataConverter.toString(MatrixBlock mb) static StringDataConverter.toString(MatrixBlock mb, boolean sparse, String separator, String lineseparator, int rowsToPrint, int colsToPrint, int decimal) Returns a string representation of a matrixstatic voidLocalFileUtils.writeMatrixBlockToLocal(String fname, MatrixBlock mb) Writes a matrix block to local file system.static voidDataConverter.writeMatrixToHDFS(MatrixBlock mat, String dir, Types.FileFormat fmt, DataCharacteristics dc) static voidDataConverter.writeMatrixToHDFS(MatrixBlock mat, String dir, Types.FileFormat fmt, DataCharacteristics dc, int replication, FileFormatProperties formatProperties) static voidDataConverter.writeMatrixToHDFS(MatrixBlock mat, String dir, Types.FileFormat fmt, DataCharacteristics dc, int replication, FileFormatProperties formatProperties, boolean diag) static longUnixPipeUtils.writeNumpyArrayInBatches(BufferedOutputStream out, int id, int batchSize, int numElem, Types.ValueType type, MatrixBlock mb) Symmetric with readNumpyArrayInBatches — writes data in batches with handshake.