Uses of Class
org.apache.sysds.runtime.controlprogram.caching.MatrixObject
-
-
Uses of MatrixObject in org.apache.sysds.api.mlcontext
Methods in org.apache.sysds.api.mlcontext that return MatrixObject Modifier and Type Method Description static 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.static MatrixObjectMLContextConversionUtil. dataFrameToMatrixObject(org.apache.spark.sql.Dataset<org.apache.spark.sql.Row> dataFrame)Convert aDataFrameto aMatrixObject.static MatrixObjectMLContextConversionUtil. dataFrameToMatrixObject(org.apache.spark.sql.Dataset<org.apache.spark.sql.Row> dataFrame, MatrixMetadata matrixMetadata)Convert aDataFrameto aMatrixObject.static MatrixObjectMLContextConversionUtil. doubleMatrixToMatrixObject(String variableName, double[][] doubleMatrix)Convert a two-dimensional double array to aMatrixObject.static MatrixObjectMLContextConversionUtil. doubleMatrixToMatrixObject(String variableName, double[][] doubleMatrix, MatrixMetadata matrixMetadata)Convert a two-dimensional double array to aMatrixObject.MatrixObjectMLResults. getMatrixObject(String outputName)Obtain an output as aMatrixObjectstatic MatrixObjectMLContextConversionUtil. javaRDDStringCSVToMatrixObject(org.apache.spark.api.java.JavaRDD<String> javaRDD)Convert aJavaRDD<String>in CSV format to aMatrixObjectstatic MatrixObjectMLContextConversionUtil. javaRDDStringCSVToMatrixObject(org.apache.spark.api.java.JavaRDD<String> javaRDD, MatrixMetadata matrixMetadata)Convert aJavaRDD<String>in CSV format to aMatrixObjectstatic MatrixObjectMLContextConversionUtil. javaRDDStringIJVToMatrixObject(org.apache.spark.api.java.JavaRDD<String> javaRDD, MatrixMetadata matrixMetadata)Convert aJavaRDD<String>in IJV format to aMatrixObject.static MatrixObjectMLContextConversionUtil. matrixBlockToMatrixObject(String variableName, MatrixBlock matrixBlock, MatrixMetadata matrixMetadata)Convert aMatrixBlockto aMatrixObject.static MatrixObjectMLContextConversionUtil. rddStringCSVToMatrixObject(org.apache.spark.rdd.RDD<String> rdd)Convert aRDD<String>in CSV format to aMatrixObjectstatic MatrixObjectMLContextConversionUtil. rddStringCSVToMatrixObject(org.apache.spark.rdd.RDD<String> rdd, MatrixMetadata matrixMetadata)Convert aRDD<String>in CSV format to aMatrixObjectstatic MatrixObjectMLContextConversionUtil. rddStringIJVToMatrixObject(org.apache.spark.rdd.RDD<String> rdd, MatrixMetadata matrixMetadata)Convert aRDD<String>in IJV format to aMatrixObject.MatrixObjectMatrix. toMatrixObject()Obtain the matrix as a SystemDS MatrixObject.static MatrixObjectMLContextConversionUtil. urlToMatrixObject(URL url, MatrixMetadata matrixMetadata)Convert a matrix at a URL to aMatrixObject.Methods in org.apache.sysds.api.mlcontext with parameters of type MatrixObject Modifier and Type Method Description static double[][]MLContextConversionUtil. matrixObjectTo2DDoubleArray(MatrixObject matrixObject)Convert aMatrixObjectto a two-dimensional double array.static org.apache.spark.api.java.JavaPairRDD<MatrixIndexes,MatrixBlock>MLContextConversionUtil. matrixObjectToBinaryBlocks(MatrixObject matrixObject, SparkExecutionContext sparkExecutionContext)Convert aMatrixObjectto aJavaPairRDD<MatrixIndexes, MatrixBlock>.static org.apache.spark.sql.Dataset<org.apache.spark.sql.Row>MLContextConversionUtil. matrixObjectToDataFrame(MatrixObject matrixObject, SparkExecutionContext sparkExecutionContext, boolean isVectorDF)Convert aMatrixObjectto aDataFrame.static org.apache.spark.api.java.JavaRDD<String>MLContextConversionUtil. matrixObjectToJavaRDDStringCSV(MatrixObject matrixObject)Convert aMatrixObjectto aJavaRDD<String>in CSV format.static org.apache.spark.api.java.JavaRDD<String>MLContextConversionUtil. matrixObjectToJavaRDDStringIJV(MatrixObject matrixObject)Convert aMatrixObjectto aJavaRDD<String>in IJV format.static List<String>MLContextConversionUtil. matrixObjectToListStringCSV(MatrixObject matrixObject)Convert aMatrixObjectto aList<String>in CSV format.static List<String>MLContextConversionUtil. matrixObjectToListStringIJV(MatrixObject matrixObject)Convert aMatrixObjectto aList<String>in IJV format.static org.apache.spark.rdd.RDD<String>MLContextConversionUtil. matrixObjectToRDDStringCSV(MatrixObject matrixObject)Convert aMatrixObjectto aRDD<String>in CSV format.static org.apache.spark.rdd.RDD<String>MLContextConversionUtil. matrixObjectToRDDStringIJV(MatrixObject matrixObject)Convert aMatrixObjectto aRDD<String>in IJV format.Constructors in org.apache.sysds.api.mlcontext with parameters of type MatrixObject Constructor Description Matrix(MatrixObject matrixObject, SparkExecutionContext sparkExecutionContext) -
Uses of MatrixObject in org.apache.sysds.hops.rewrite
Methods in org.apache.sysds.hops.rewrite with parameters of type MatrixObject Modifier and Type Method Description static voidHopRewriteUtils. copyLineNumbers(MatrixObject mo, Hop tread)static DataOpHopRewriteUtils. createTransientRead(String name, MatrixObject mo) -
Uses of MatrixObject in org.apache.sysds.runtime.codegen
Methods in org.apache.sysds.runtime.codegen that return MatrixObject Modifier and Type Method Description MatrixObjectSpoofCUDACellwise. execute(ExecutionContext ec, ArrayList<MatrixObject> inputs, ArrayList<ScalarObject> scalarObjects, String outputName)MatrixObjectSpoofCUDAOperator. execute(ExecutionContext ec, ArrayList<MatrixObject> inputs, ArrayList<ScalarObject> scalarObjects, String outputName)MatrixObjectSpoofCUDARowwise. execute(ExecutionContext ec, ArrayList<MatrixObject> inputs, ArrayList<ScalarObject> scalarObjects, String outputName)Methods in org.apache.sysds.runtime.codegen with parameters of type MatrixObject Modifier and Type Method Description default voidSpoofCUDAOperator. packDataForTransfer(ExecutionContext ec, ArrayList<MatrixObject> inputs, ArrayList<ScalarObject> scalarObjects, MatrixObject out_obj, int num_inputs, int ID, long grix, boolean tB1, jcuda.Pointer[] ptr)default voidSpoofCUDAOperator. prepareMatrixPointers(ByteBuffer buf, ExecutionContext ec, MatrixObject mo, boolean tB1) -
Uses of MatrixObject in org.apache.sysds.runtime.controlprogram.caching
Constructors in org.apache.sysds.runtime.controlprogram.caching with parameters of type MatrixObject Constructor Description MatrixObject(MatrixObject mo)Copy constructor that copies meta data but NO data. -
Uses of MatrixObject in org.apache.sysds.runtime.controlprogram.context
Methods in org.apache.sysds.runtime.controlprogram.context that return MatrixObject Modifier and Type Method Description MatrixObjectExecutionContext. allocateGPUMatrixObject(String varName, long numRows, long numCols)Allocates theGPUObjectfor a given LOPS Variable (eg.static MatrixObjectExecutionContext. createMatrixObject(MatrixBlock mb)static MatrixObjectExecutionContext. createMatrixObject(DataCharacteristics dc)MatrixObjectExecutionContext. getMatrixInputForGPUInstruction(String varName, String opcode)MatrixObjectExecutionContext. getMatrixObject(String varname)MatrixObjectExecutionContext. getMatrixObject(CPOperand input)Methods in org.apache.sysds.runtime.controlprogram.context that return types with arguments of type MatrixObject Modifier and Type Method Description Pair<MatrixObject,Boolean>ExecutionContext. getDenseMatrixOutputForGPUInstruction(String varName, long numRows, long numCols)Allocates a dense matrix on the GPU (for output)Pair<MatrixObject,Boolean>ExecutionContext. getDenseMatrixOutputForGPUInstruction(String varName, long numRows, long numCols, boolean initialize)Pair<MatrixObject,Boolean>ExecutionContext. getSparseMatrixOutputForGPUInstruction(String varName, long numRows, long numCols, long nnz)Allocates a sparse matrix in CSR format on the GPU.Pair<MatrixObject,Boolean>ExecutionContext. getSparseMatrixOutputForGPUInstruction(String varName, long numRows, long numCols, long nnz, boolean initialize)Methods in org.apache.sysds.runtime.controlprogram.context with parameters of type MatrixObject Modifier and Type Method Description PartitionedBroadcast<MatrixBlock>SparkExecutionContext. getBroadcastForMatrixObject(MatrixObject mo)longExecutionContext. getGPUDensePointerAddress(MatrixObject obj)CSRPointerExecutionContext. getGPUSparsePointerAddress(MatrixObject obj)org.apache.spark.api.java.JavaPairRDD<?,?>SparkExecutionContext. getRDDHandleForMatrixObject(MatrixObject mo, Types.FileFormat fmt)org.apache.spark.api.java.JavaPairRDD<?,?>SparkExecutionContext. getRDDHandleForMatrixObject(MatrixObject mo, Types.FileFormat fmt, int numParts, boolean inclEmpty)voidSparkExecutionContext. setBroadcastHandle(MatrixObject mo) -
Uses of MatrixObject in org.apache.sysds.runtime.controlprogram.federated
Methods in org.apache.sysds.runtime.controlprogram.federated that return MatrixObject Modifier and Type Method Description MatrixObjectMatrixLineagePair. getMO()Methods in org.apache.sysds.runtime.controlprogram.federated with parameters of type MatrixObject Modifier and Type Method Description static voidFederationUtils. checkFedMapType(MatrixObject mo)static MatrixLineagePairMatrixLineagePair. of(MatrixObject mo, LineageItem li)Constructors in org.apache.sysds.runtime.controlprogram.federated with parameters of type MatrixObject Constructor Description MatrixLineagePair(MatrixObject mo, LineageItem li) -
Uses of MatrixObject in org.apache.sysds.runtime.controlprogram.paramserv
Methods in org.apache.sysds.runtime.controlprogram.paramserv that return MatrixObject Modifier and Type Method Description static MatrixObjectParamservUtils. createShallowCopy(MatrixObject mo)MatrixObjectPSWorker. getFeatures()MatrixObjectPSWorker. getLabels()static MatrixObjectParamservUtils. newMatrixObject(MatrixBlock mb)static MatrixObjectParamservUtils. newMatrixObject(MatrixBlock mb, boolean cleanup)static MatrixObjectParamservUtils. sliceMatrix(MatrixObject mo, long rl, long rh)Slice the matrixMethods in org.apache.sysds.runtime.controlprogram.paramserv with parameters of type MatrixObject Modifier and Type Method Description static HEParamServerHEParamServer. create(ListObject model, String aggFunc, Statement.PSUpdateType updateType, Statement.PSFrequency freq, ExecutionContext ec, int workerNum, String valFunc, int numBatchesPerEpoch, MatrixObject valFeatures, MatrixObject valLabels, int nbatches)static LocalParamServerLocalParamServer. create(ListObject model, String aggFunc, Statement.PSUpdateType updateType, Statement.PSFrequency freq, ExecutionContext ec, int workerNum, String valFunc, int numBatchesPerEpoch, MatrixObject valFeatures, MatrixObject valLabels, int nbatches, boolean modelAvg)static MatrixObjectParamservUtils. createShallowCopy(MatrixObject mo)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)voidPSWorker. setFeatures(MatrixObject features)voidPSWorker. setLabels(MatrixObject labels)static MatrixObjectParamservUtils. sliceMatrix(MatrixObject mo, long rl, long rh)Slice the matrix -
Uses of MatrixObject in org.apache.sysds.runtime.controlprogram.paramserv.dp
Fields in org.apache.sysds.runtime.controlprogram.paramserv.dp with type parameters of type MatrixObject Modifier and Type Field Description List<MatrixObject>DataPartitionFederatedScheme.Result. _pFeaturesList<MatrixObject>DataPartitionFederatedScheme.Result. _pLabelsList<MatrixObject>DataPartitionLocalScheme.Result. pFeaturesList<MatrixObject>DataPartitionLocalScheme.Result. pLabelsConstructor parameters in org.apache.sysds.runtime.controlprogram.paramserv.dp with type arguments of type MatrixObject Constructor Description Result(List<MatrixObject> pFeatures, List<MatrixObject> pLabels, int workerNum, DataPartitionFederatedScheme.BalanceMetrics balanceMetrics, List<Double> weightingFactors)Result(List<MatrixObject> pFeatures, List<MatrixObject> pLabels) -
Uses of MatrixObject in org.apache.sysds.runtime.controlprogram.paramserv.homomorphicEncryption
Methods in org.apache.sysds.runtime.controlprogram.paramserv.homomorphicEncryption that return MatrixObject Modifier and Type Method Description MatrixObjectSEALServer. average(CiphertextMatrix encrypted_sum, PlaintextMatrix[] partial_plaintexts)averages the partial decryptionsMethods in org.apache.sysds.runtime.controlprogram.paramserv.homomorphicEncryption with parameters of type MatrixObject Modifier and Type Method Description CiphertextMatrixSEALClient. encrypt(MatrixObject plaintext)encrypts one block of data with public key stored statically and returns it setPublicKey() must have been called before calling this -
Uses of MatrixObject in org.apache.sysds.runtime.controlprogram.parfor
Methods in org.apache.sysds.runtime.controlprogram.parfor that return MatrixObject Modifier and Type Method Description MatrixObjectDataPartitioner. createPartitionedMatrixObject(MatrixObject in, String fnameNew)MatrixObjectDataPartitioner. createPartitionedMatrixObject(MatrixObject in, String fnameNew, boolean force)MatrixObjectDataPartitioner. createPartitionedMatrixObject(MatrixObject in, MatrixObject out, boolean force)Creates a partitioned matrix object based on the given input matrix object, according to the specified split format.MatrixObjectResultMergeLocalAutomatic. executeParallelMerge(int par)MatrixObjectResultMergeLocalFile. executeParallelMerge(int par)MatrixObjectResultMergeLocalMemory. executeParallelMerge(int par)MatrixObjectResultMergeRemoteSpark. executeParallelMerge(int par)MatrixObjectResultMergeRemoteSparkWCompare. executeParallelMerge(int par)MatrixObjectResultMergeLocalAutomatic. executeSerialMerge()MatrixObjectResultMergeLocalFile. executeSerialMerge()MatrixObjectResultMergeLocalMemory. executeSerialMerge()MatrixObjectResultMergeRemoteSpark. executeSerialMerge()MatrixObjectResultMergeRemoteSparkWCompare. executeSerialMerge()Methods in org.apache.sysds.runtime.controlprogram.parfor with parameters of type MatrixObject Modifier and Type Method Description MatrixObjectDataPartitioner. createPartitionedMatrixObject(MatrixObject in, String fnameNew)MatrixObjectDataPartitioner. createPartitionedMatrixObject(MatrixObject in, String fnameNew, boolean force)MatrixObjectDataPartitioner. createPartitionedMatrixObject(MatrixObject in, MatrixObject out, boolean force)Creates a partitioned matrix object based on the given input matrix object, according to the specified split format.static RemoteParForJobReturnRemoteDPParForSpark. runJob(long pfid, String itervar, String matrixvar, String program, HashMap<String,byte[]> clsMap, String resultFile, MatrixObject input, ExecutionContext ec, ParForProgramBlock.PartitionFormat dpf, Types.FileFormat fmt, boolean tSparseCol, boolean enableCPCaching, int numReducers)Constructors in org.apache.sysds.runtime.controlprogram.parfor with parameters of type MatrixObject Constructor Description ResultMergeLocalAutomatic(MatrixObject out, MatrixObject[] in, String outputFilename, boolean accum)ResultMergeLocalFile(MatrixObject out, MatrixObject[] in, String outputFilename, boolean accum)ResultMergeLocalMemory(MatrixObject out, MatrixObject[] in, String outputFilename, boolean accum)ResultMergeMatrix(MatrixObject out, MatrixObject[] in, String outputFilename, boolean accum)ResultMergeRemoteSpark(MatrixObject out, MatrixObject[] in, String outputFilename, boolean accum, ExecutionContext ec, int numMappers, int numReducers) -
Uses of MatrixObject in org.apache.sysds.runtime.controlprogram.parfor.opt
Methods in org.apache.sysds.runtime.controlprogram.parfor.opt with parameters of type MatrixObject Modifier and Type Method Description static booleanOptimizerRuleBased. allowsBinaryCellPartitions(MatrixObject mo, ParForProgramBlock.PartitionFormat dpf) -
Uses of MatrixObject in org.apache.sysds.runtime.instructions.cp
Constructors in org.apache.sysds.runtime.instructions.cp with parameters of type MatrixObject Constructor Description TriggerBroadcastTask(ExecutionContext ec, MatrixObject mo)TriggerRemoteOperationsTask(MatrixObject mo) -
Uses of MatrixObject in org.apache.sysds.runtime.instructions.fed
Methods in org.apache.sysds.runtime.instructions.fed with parameters of type MatrixObject Modifier and Type Method Description static voidMultiReturnParameterizedBuiltinFEDInstruction. encodeFederatedFrames(FederationMap fedMapping, MultiColumnEncoder globalencoder, MatrixObject transformedMat)voidUnaryMatrixFEDInstruction. processCumulativeInstruction(ExecutionContext ec, MatrixObject mo1) -
Uses of MatrixObject in org.apache.sysds.runtime.instructions.gpu.context
Methods in org.apache.sysds.runtime.instructions.gpu.context that return MatrixObject Modifier and Type Method Description MatrixObjectGPUObject. getMatrixObject()Methods in org.apache.sysds.runtime.instructions.gpu.context with parameters of type MatrixObject Modifier and Type Method Description GPUObjectGPUContext. createGPUObject(MatrixObject mo)Instantiates a newGPUObjectinitialized with the givenMatrixObject.GPUObjectGPUContext. shallowCopyGPUObject(GPUObject source, MatrixObject mo)Shallow copy the given sourceGPUObjectto a newGPUObjectand assign that to the givenMatrixObject.Constructors in org.apache.sysds.runtime.instructions.gpu.context with parameters of type MatrixObject Constructor Description GPUObject(GPUContext gCtx, GPUObject that, MatrixObject mat) -
Uses of MatrixObject in org.apache.sysds.runtime.instructions.spark.utils
Methods in org.apache.sysds.runtime.instructions.spark.utils with parameters of type MatrixObject Modifier and Type Method Description static longSparkUtils. getNonZeros(MatrixObject mo)static voidSparkUtils. postprocessUltraSparseOutput(MatrixObject mo, DataCharacteristics mcOut) -
Uses of MatrixObject in org.apache.sysds.runtime.lineage
Methods in org.apache.sysds.runtime.lineage with parameters of type MatrixObject Modifier and Type Method Description static FederatedResponseLineageItemUtils. setUDFResponse(FederatedUDF udf, MatrixObject mo) -
Uses of MatrixObject in org.apache.sysds.runtime.matrix.data
Methods in org.apache.sysds.runtime.matrix.data that return MatrixObject Modifier and Type Method Description static MatrixObjectLibMatrixCUDA. getDenseMatrixOutputForGPUInstruction(ExecutionContext ec, String instName, String name, long numRows, long numCols)Helper method to get the output block (allocated on the GPU) Also records performance information intoStatisticsstatic MatrixObjectLibMatrixCUDA. getDenseMatrixOutputForGPUInstruction(ExecutionContext ec, String instName, String name, long numRows, long numCols, boolean initialize)static MatrixObjectLibMatrixCuMatMult. matmult(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject left, MatrixObject right, String outputName, boolean isLeftTransposed, boolean isRightTransposed)Matrix multiply on GPU Examines sparsity and shapes and routes call to appropriate method from cuBLAS or cuSparse C = op(A) x op(B) The user is expected to call ec.releaseMatrixOutputForGPUInstruction(outputName);Methods in org.apache.sysds.runtime.matrix.data with parameters of type MatrixObject Modifier and Type Method Description static voidLibMatrixCUDA. abs(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "abs" operation on a matrix on the GPUstatic voidLibMatrixCUDA. acos(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "acos" operation on a matrix on the GPUstatic voidLibMatrixCUDA. asin(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "asin" operation on a matrix on the GPUstatic voidLibMatrixCUDA. atan(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "atan" operation on a matrix on the GPUstatic voidLibMatrixCUDA. axpy(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, MatrixObject in2, String outputName, double constant)Performs daxpy operationstatic voidLibMatrixCuDNN. batchNormalizationBackward(GPUContext gCtx, String instName, MatrixObject image, MatrixObject dout, MatrixObject scale, MatrixObject dX, MatrixObject dScale, MatrixObject dBias, double epsilon, MatrixObject resultSaveMean, MatrixObject resultSaveInvVariance)This method computes the backpropagation errors for image, scale and bias of batch normalization layerstatic voidLibMatrixCuDNN. batchNormalizationForwardInference(GPUContext gCtx, String instName, MatrixObject image, MatrixObject scale, MatrixObject bias, MatrixObject runningMean, MatrixObject runningVar, MatrixObject ret, double epsilon)Performs the forward BatchNormalization layer computation for inferencestatic voidLibMatrixCuDNN. batchNormalizationForwardTraining(GPUContext gCtx, String instName, MatrixObject image, MatrixObject scale, MatrixObject bias, MatrixObject runningMean, MatrixObject runningVar, MatrixObject ret, MatrixObject retRunningMean, MatrixObject retRunningVar, double epsilon, double exponentialAverageFactor, MatrixObject resultSaveMean, MatrixObject resultSaveInvVariance)Performs the forward BatchNormalization layer computation for trainingstatic voidLibMatrixCUDA. biasAdd(GPUContext gCtx, String instName, MatrixObject input, MatrixObject bias, MatrixObject outputBlock)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 voidLibMatrixCUDA. biasMultiply(GPUContext gCtx, String instName, MatrixObject input, MatrixObject bias, MatrixObject outputBlock)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 voidLibMatrixCUDA. cbind(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, MatrixObject in2, String outputName)static voidLibMatrixCUDA. ceil(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "ceil" operation on a matrix on the GPUstatic voidLibMatrixCUDA. channelSums(GPUContext gCtx, String instName, MatrixObject input, MatrixObject outputBlock, long C, long HW)Perform channel_sums operations: out = rowSums(matrix(colSums(A), rows=C, cols=HW))static voidLibMatrixCuDNN. conv2d(GPUContext gCtx, String instName, MatrixObject image, MatrixObject filter, MatrixObject outputBlock, int N, int C, int H, int W, int K, int R, int S, int pad_h, int pad_w, int stride_h, int stride_w, int P, int Q, double intermediateMemoryBudget)Performs a 2D convolutionstatic voidLibMatrixCuDNN. conv2dBackwardData(GPUContext gCtx, String instName, MatrixObject filter, MatrixObject dout, MatrixObject output, int N, int C, int H, int W, int K, int R, int S, int pad_h, int pad_w, int stride_h, int stride_w, int P, int Q, double intermediateMemoryBudget)This method computes the backpropogation errors for previous layer of convolution operationstatic voidLibMatrixCuDNN. conv2dBackwardFilter(GPUContext gCtx, String instName, MatrixObject image, MatrixObject dout, MatrixObject outputBlock, int N, int C, int H, int W, int K, int R, int S, int pad_h, int pad_w, int stride_h, int stride_w, int P, int Q, double intermediateMemoryBudget)This method computes the backpropogation errors for filter of convolution operationstatic voidLibMatrixCuDNN. conv2dBiasAdd(GPUContext gCtx, String instName, MatrixObject image, MatrixObject bias, MatrixObject filter, MatrixObject output, int N, int C, int H, int W, int K, int R, int S, int pad_h, int pad_w, int stride_h, int stride_w, int P, int Q, double intermediateMemoryBudget)Does a 2D convolution followed by a bias_addstatic voidLibMatrixCUDA. cos(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "cos" operation on a matrix on the GPUstatic voidLibMatrixCUDA. cosh(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "cosh" operation on a matrix on the GPUstatic voidLibMatrixCUDA. cumulativeScan(ExecutionContext ec, GPUContext gCtx, String instName, String kernelFunction, MatrixObject in, String outputName)Cumulative scanstatic voidLibMatrixCUDA. cumulativeSumProduct(ExecutionContext ec, GPUContext gCtx, String instName, String kernelFunction, MatrixObject in, String outputName)Cumulative sum-product kernel cascade invokationstatic voidLibMatrixCUDA. exp(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "exp" operation on a matrix on the GPUstatic voidLibMatrixCUDA. floor(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "floor" operation on a matrix on the GPUstatic jcuda.PointerLibMatrixCUDA. getDensePointer(GPUContext gCtx, MatrixObject input, String instName)Convenience method to get jcudaDenseMatrixPtr.static jcuda.PointerLibMatrixCuDNN. getDensePointerForCuDNN(GPUContext gCtx, MatrixObject image, String instName, int numRows, int numCols)Convenience method to get jcudaDenseMatrixPtr.static longLibMatrixCUDA. getNnz(GPUContext gCtx, String instName, MatrixObject mo, boolean recomputeDenseNNZ)Note: if the matrix is in dense format, it explicitly re-computes the number of nonzeros.static booleanLibMatrixCUDA. isInSparseFormat(GPUContext gCtx, MatrixObject mo)static voidLibMatrixCUDA. log(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "log" operation on a matrix on the GPUstatic MatrixObjectLibMatrixCuMatMult. matmult(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject left, MatrixObject right, String outputName, boolean isLeftTransposed, boolean isRightTransposed)Matrix multiply on GPU Examines sparsity and shapes and routes call to appropriate method from cuBLAS or cuSparse C = op(A) x op(B) The user is expected to call ec.releaseMatrixOutputForGPUInstruction(outputName);static voidLibMatrixCUDA. matmultTSMM(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject left, String outputName, boolean isLeftTransposed)Performs tsmm, A %*% A' or A' %*% A, on GPU by exploiting cublasDsyrk(...)static voidLibMatrixCUDA. matrixMatrixArithmetic(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, MatrixObject in2, String outputName, boolean isLeftTransposed, boolean isRightTransposed, BinaryOperator op)Performs elementwise arithmetic operation specified by op of two input matrices in1 and in2static voidLibMatrixCUDA. matrixMatrixRelational(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, MatrixObject in2, String outputName, BinaryOperator op)Performs elementwise operation relational specified by op of two input matrices in1 and in2static voidLibMatrixCUDA. matrixScalarArithmetic(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in, String outputName, boolean isInputTransposed, ScalarOperator op)Entry point to perform elementwise matrix-scalar arithmetic operation specified by opstatic voidLibMatrixCUDA. matrixScalarOp(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in, String outputName, boolean isInputTransposed, ScalarOperator op)Utility to do matrix-scalar operation kernelstatic voidLibMatrixCUDA. matrixScalarRelational(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in, String outputName, ScalarOperator op)Entry point to perform elementwise matrix-scalar relational operation specified by opstatic voidLibMatrixCuDNN. pooling(GPUContext gCtx, String instName, MatrixObject image, MatrixObject outputBlock, int N, int C, int H, int W, int K, int R, int S, int pad_h, int pad_w, int stride_h, int stride_w, int P, int Q, LibMatrixDNN.PoolingType poolingType, double intermediateMemoryBudget)performs maxpooling on GPU by exploiting cudnnPoolingForward(...)static voidLibMatrixCuDNN. poolingBackward(GPUContext gCtx, String instName, MatrixObject image, MatrixObject dout, MatrixObject maxpoolOutput, MatrixObject outputBlock, int N, int C, int H, int W, int K, int R, int S, int pad_h, int pad_w, int stride_h, int stride_w, int P, int Q, LibMatrixDNN.PoolingType poolingType, double intermediateMemoryBudget)Performs maxpoolingBackward on GPU by exploiting cudnnPoolingBackward(...) This method computes the backpropogation errors for previous layer of maxpooling operationstatic voidLibMatrixCUDA. rbind(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, MatrixObject in2, String outputName)static voidLibMatrixCuDNN. relu(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in, String outputName)Performs the relu operation on the GPU.static voidLibMatrixCUDA. reluBackward(GPUContext gCtx, String instName, MatrixObject input, MatrixObject dout, MatrixObject outputBlock)This method computes the backpropagation errors for previous layer of relu operationstatic voidLibMatrixCUDA. round(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "round" operation on a matrix on the GPUstatic voidLibMatrixCUDA. sigmoid(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "sigmoid" operation on a matrix on the GPUstatic voidLibMatrixCUDA. sign(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "sign" operation on a matrix on the GPUstatic voidLibMatrixCUDA. sin(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "sin" operation on a matrix on the GPUstatic voidLibMatrixCUDA. sinh(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "sinh" operation on a matrix on the GPUstatic voidLibMatrixCUDA. sliceOperations(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, IndexRange ixrange, String outputName)Method to perform rightIndex operation for a given lower and upper bounds in row and column dimensions.static voidLibMatrixCuDNN. softmax(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "softmax" operation on a matrix on the GPUstatic voidLibMatrixCUDA. solve(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, MatrixObject in2, String outputName)Implements the "solve" function for systemds Ax = B (A is of size m*n, B is of size m*1, x is of size n*1)static voidLibMatrixCUDA. sqrt(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "sqrt" operation on a matrix on the GPUstatic voidLibMatrixCUDA. tan(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "tan" operation on a matrix on the GPUstatic voidLibMatrixCUDA. tanh(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String outputName)Performs an "tanh" operation on a matrix on the GPUstatic voidLibMatrixCUDA. transpose(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in, String outputName)Transposes the input matrix using cublasDgeamstatic voidLibMatrixCUDA. unaryAggregate(ExecutionContext ec, GPUContext gCtx, String instName, MatrixObject in1, String output, AggregateUnaryOperator op)Entry point to perform Unary aggregate operations on the GPU.Constructors in org.apache.sysds.runtime.matrix.data with parameters of type MatrixObject Constructor Description LibMatrixCuDNNInputRowFetcher(GPUContext gCtx, String instName, MatrixObject image)Initialize the input fetcher -
Uses of MatrixObject in org.apache.sysds.runtime.util
Methods in org.apache.sysds.runtime.util with parameters of type MatrixObject Modifier and Type Method Description static ListObjectAutoDiff. getBackward(MatrixObject mo, ArrayList<Data> lineage, ExecutionContext adec)static List<Data>AutoDiff. parseNComputeAutoDiffFromLineage(MatrixObject mo, String mainTrace, ArrayList<String> names, ExecutionContext ec)
-