/* * Licensed to the Apache Software Foundation (ASF) under one * or more contributor license agreements. See the NOTICE file * distributed with this work for additional information * regarding copyright ownership. The ASF licenses this file * to you under the Apache License, Version 2.0 (the * "License"); you may not use this file except in compliance * with the License. You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, * software distributed under the License is distributed on an * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY * KIND, either express or implied. See the License for the * specific language governing permissions and limitations * under the License. */ package org.apache.sysml.runtime.instructions.cp; import org.apache.sysml.lops.Ternary; import org.apache.sysml.parser.Expression.DataType; import org.apache.sysml.parser.Expression.ValueType; import org.apache.sysml.runtime.DMLRuntimeException; import org.apache.sysml.runtime.controlprogram.context.ExecutionContext; import org.apache.sysml.runtime.instructions.Instruction; import org.apache.sysml.runtime.instructions.InstructionUtils; import org.apache.sysml.runtime.matrix.data.CTableMap; import org.apache.sysml.runtime.matrix.data.MatrixBlock; import org.apache.sysml.runtime.matrix.operators.Operator; import org.apache.sysml.runtime.matrix.operators.SimpleOperator; import org.apache.sysml.runtime.util.DataConverter; public class TernaryCPInstruction extends ComputationCPInstruction { private String _outDim1; private String _outDim2; private boolean _dim1Literal; private boolean _dim2Literal; private boolean _isExpand; private boolean _ignoreZeros; public TernaryCPInstruction(Operator op, CPOperand in1, CPOperand in2, CPOperand in3, CPOperand out, String outputDim1, boolean dim1Literal,String outputDim2, boolean dim2Literal, boolean isExpand, boolean ignoreZeros, String opcode, String istr ) { super(op, in1, in2, in3, out, opcode, istr); _outDim1 = outputDim1; _dim1Literal = dim1Literal; _outDim2 = outputDim2; _dim2Literal = dim2Literal; _isExpand = isExpand; _ignoreZeros = ignoreZeros; } public static TernaryCPInstruction parseInstruction(String inst) throws DMLRuntimeException { String[] parts = InstructionUtils.getInstructionPartsWithValueType(inst); InstructionUtils.checkNumFields ( parts, 7 ); String opcode = parts[0]; //handle opcode if ( !(opcode.equalsIgnoreCase("ctable") || opcode.equalsIgnoreCase("ctableexpand")) ) { throw new DMLRuntimeException("Unexpected opcode in TertiaryCPInstruction: " + inst); } boolean isExpand = opcode.equalsIgnoreCase("ctableexpand"); //handle operands CPOperand in1 = new CPOperand(parts[1]); CPOperand in2 = new CPOperand(parts[2]); CPOperand in3 = new CPOperand(parts[3]); //handle known dimension information String[] dim1Fields = parts[4].split(Instruction.LITERAL_PREFIX); String[] dim2Fields = parts[5].split(Instruction.LITERAL_PREFIX); CPOperand out = new CPOperand(parts[6]); boolean ignoreZeros = Boolean.parseBoolean(parts[7]); // ctable does not require any operator, so we simply pass-in a dummy operator with null functionobject return new TernaryCPInstruction(new SimpleOperator(null), in1, in2, in3, out, dim1Fields[0], Boolean.parseBoolean(dim1Fields[1]), dim2Fields[0], Boolean.parseBoolean(dim2Fields[1]), isExpand, ignoreZeros, opcode, inst); } private Ternary.OperationTypes findCtableOperation() { DataType dt1 = input1.getDataType(); DataType dt2 = input2.getDataType(); DataType dt3 = input3.getDataType(); return Ternary.findCtableOperationByInputDataTypes(dt1, dt2, dt3); } @Override public void processInstruction(ExecutionContext ec) throws DMLRuntimeException { MatrixBlock matBlock1 = ec.getMatrixInput(input1.getName()); MatrixBlock matBlock2=null, wtBlock=null; double cst1, cst2; CTableMap resultMap = new CTableMap(); MatrixBlock resultBlock = null; Ternary.OperationTypes ctableOp = findCtableOperation(); ctableOp = _isExpand ? Ternary.OperationTypes.CTABLE_EXPAND_SCALAR_WEIGHT : ctableOp; long outputDim1 = (_dim1Literal ? (long) Double.parseDouble(_outDim1) : (ec.getScalarInput(_outDim1, ValueType.DOUBLE, false)).getLongValue()); long outputDim2 = (_dim2Literal ? (long) Double.parseDouble(_outDim2) : (ec.getScalarInput(_outDim2, ValueType.DOUBLE, false)).getLongValue()); boolean outputDimsKnown = (outputDim1 != -1 && outputDim2 != -1); if ( outputDimsKnown ) { int inputRows = matBlock1.getNumRows(); int inputCols = matBlock1.getNumColumns(); boolean sparse = MatrixBlock.evalSparseFormatInMemory(outputDim1, outputDim2, inputRows*inputCols); //only create result block if dense; it is important not to aggregate on sparse result //blocks because it would implicitly turn the O(N) algorithm into O(N log N). if( !sparse ) resultBlock = new MatrixBlock((int)outputDim1, (int)outputDim2, false); } if( _isExpand ){ resultBlock = new MatrixBlock( matBlock1.getNumRows(), Integer.MAX_VALUE, true ); } switch(ctableOp) { case CTABLE_TRANSFORM: //(VECTOR) // F=ctable(A,B,W) matBlock2 = ec.getMatrixInput(input2.getName()); wtBlock = ec.getMatrixInput(input3.getName()); matBlock1.ternaryOperations((SimpleOperator)_optr, matBlock2, wtBlock, resultMap, resultBlock); break; case CTABLE_TRANSFORM_SCALAR_WEIGHT: //(VECTOR/MATRIX) // F = ctable(A,B) or F = ctable(A,B,1) matBlock2 = ec.getMatrixInput(input2.getName()); cst1 = ec.getScalarInput(input3.getName(), input3.getValueType(), input3.isLiteral()).getDoubleValue(); matBlock1.ternaryOperations((SimpleOperator)_optr, matBlock2, cst1, _ignoreZeros, resultMap, resultBlock); break; case CTABLE_EXPAND_SCALAR_WEIGHT: //(VECTOR) // F = ctable(seq,A) or F = ctable(seq,B,1) matBlock2 = ec.getMatrixInput(input2.getName()); cst1 = ec.getScalarInput(input3.getName(), input3.getValueType(), input3.isLiteral()).getDoubleValue(); // only resultBlock.rlen known, resultBlock.clen set in operation matBlock1.ternaryOperations((SimpleOperator)_optr, matBlock2, cst1, resultBlock); break; case CTABLE_TRANSFORM_HISTOGRAM: //(VECTOR) // F=ctable(A,1) or F = ctable(A,1,1) cst1 = ec.getScalarInput(input2.getName(), input2.getValueType(), input2.isLiteral()).getDoubleValue(); cst2 = ec.getScalarInput(input3.getName(), input3.getValueType(), input3.isLiteral()).getDoubleValue(); matBlock1.ternaryOperations((SimpleOperator)_optr, cst1, cst2, resultMap, resultBlock); break; case CTABLE_TRANSFORM_WEIGHTED_HISTOGRAM: //(VECTOR) // F=ctable(A,1,W) wtBlock = ec.getMatrixInput(input3.getName()); cst1 = ec.getScalarInput(input2.getName(), input2.getValueType(), input2.isLiteral()).getDoubleValue(); matBlock1.ternaryOperations((SimpleOperator)_optr, cst1, wtBlock, resultMap, resultBlock); break; default: throw new DMLRuntimeException("Encountered an invalid ctable operation ("+ctableOp+") while executing instruction: " + this.toString()); } if(input1.getDataType() == DataType.MATRIX) ec.releaseMatrixInput(input1.getName()); if(input2.getDataType() == DataType.MATRIX) ec.releaseMatrixInput(input2.getName()); if(input3.getDataType() == DataType.MATRIX) ec.releaseMatrixInput(input3.getName()); if ( resultBlock == null ){ //we need to respect potentially specified output dimensions here, because we might have //decided for hash-aggregation just to prevent inefficiency in case of sparse outputs. if( outputDimsKnown ) resultBlock = DataConverter.convertToMatrixBlock( resultMap, (int)outputDim1, (int)outputDim2 ); else resultBlock = DataConverter.convertToMatrixBlock( resultMap ); } else resultBlock.examSparsity(); ec.setMatrixOutput(output.getName(), resultBlock); } }