public class LibMatrixCuDNNInputRowFetcher extends LibMatrixCUDA implements AutoCloseable
cudaSupportFunctions, CUDNN_DATA_TYPE, customKernelSuffix, maxNumElementsOfCuDNNTensor, sizeOfDataType| Constructor and Description |
|---|
LibMatrixCuDNNInputRowFetcher(GPUContext gCtx,
String instName,
org.apache.sysml.runtime.controlprogram.caching.MatrixObject image)
Initialize the input fetcher
|
| Modifier and Type | Method and Description |
|---|---|
void |
close()
Deallocates temporary pointer
|
jcuda.Pointer |
getNthRow(int n)
Copy the nth row and return the dense pointer
|
abs, acos, asin, atan, axpy, biasAdd, biasMultiply, cbind, ceil, channelSums, computeNNZ, cos, cosh, dataTypePointerTo, denseTranspose, deviceCopy, double2float, exp, float2double, floor, getCublasHandle, getCudaKernels, getCusparseHandle, getDenseMatrixOutputForGPUInstruction, getDensePointer, getNnz, getSparsePointer, isInSparseFormat, log, matmultTSMM, matrixMatrixArithmetic, matrixMatrixRelational, matrixScalarArithmetic, matrixScalarOp, matrixScalarRelational, one, rbind, reluBackward, resetFloatingPointPrecision, round, sigmoid, sign, sin, sinh, sliceDenseDense, sliceOperations, sliceSparseDense, solve, sqrt, tan, tanh, toInt, transpose, unaryAggregate, zeropublic LibMatrixCuDNNInputRowFetcher(GPUContext gCtx, String instName, org.apache.sysml.runtime.controlprogram.caching.MatrixObject image)
gCtx - current gpu contextinstName - name of the instructionimage - input matrix object.public jcuda.Pointer getNthRow(int n)
n - zero-based row indexpublic void close()
close in interface AutoCloseableCopyright © 2018 The Apache Software Foundation. All rights reserved.