cavis/libnd4j/include/helpers/cuda/ConstantTadHelper.cu

113 lines
4.7 KiB
Plaintext
Raw Normal View History

2019-06-06 14:21:15 +02:00
/*******************************************************************************
* Copyright (c) 2015-2018 Skymind, Inc.
*
* This program and the accompanying materials are made available under the
* terms of the Apache License, Version 2.0 which is available at
* https://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.
*
* SPDX-License-Identifier: Apache-2.0
******************************************************************************/
//
// @author raver119@gmail.com
//
#include "../ConstantTadHelper.h"
#include <TAD.h>
#include <ConstantHelper.h>
#include <exceptions/cuda_exception.h>
#include <execution/LaunchContext.h>
#include <ShapeUtils.h>
namespace nd4j {
ConstantTadHelper::ConstantTadHelper() {
auto numDevices = ConstantHelper::getNumberOfDevices();
for (int e = 0; e < numDevices; e++) {
std::map<TadDescriptor, TadPack> pack;
_cache.emplace_back(pack);
}
}
ConstantTadHelper* ConstantTadHelper::getInstance() {
if (!_INSTANCE)
_INSTANCE = new ConstantTadHelper();
return _INSTANCE;
}
TadPack& ConstantTadHelper::tadForDimensions(Nd4jLong *originalShape, int dimension, const bool keepUnitiesInShape) {
return tadForDimensions(originalShape, &dimension, 1, keepUnitiesInShape);
}
TadPack& ConstantTadHelper::tadForDimensions(Nd4jLong *originalShape, const std::vector<int> &dimensions, const bool keepUnitiesInShape) {
return tadForDimensions(originalShape, const_cast<int *>(dimensions.data()), dimensions.size(), keepUnitiesInShape);
}
TadPack& ConstantTadHelper::tadForDimensions(Nd4jLong *originalShape, int* dimensions, int dimLength, const bool keepUnitiesInShape) {
TadDescriptor tadDescriptor(originalShape, dimensions, dimLength, keepUnitiesInShape);
return tadForDimensions(tadDescriptor);
}
TadPack& ConstantTadHelper::tadForDimensions(ShapeDescriptor &descriptor, std::vector<int> &dimensions, const bool keepUnitiesInShape) {
TadDescriptor tadDescriptor(descriptor, dimensions, keepUnitiesInShape);
return tadForDimensions(tadDescriptor);
}
TadPack& ConstantTadHelper::tadForDimensions(TadDescriptor &descriptor) {
const int deviceId = ConstantHelper::getCurrentDevice();
_mutex.lock();
if (_cache[deviceId].count(descriptor) == 0) {
const auto shapeInfo = descriptor.originalShape().toShapeInfo();
const int rank = shape::rank(shapeInfo);
const std::vector<int> dimsToExclude = ShapeUtils::evalDimsToExclude(rank, descriptor.axis());
const Nd4jLong numOfSubArrs = ShapeUtils::getNumOfSubArrs(shapeInfo, dimsToExclude);
const int subArrRank = (rank == dimsToExclude.size() || descriptor.areUnitiesinShape()) ? rank : rank - dimsToExclude.size();
auto sPtr = new Nd4jLong[shape::shapeInfoLength(subArrRank)];
auto oPtr = new Nd4jLong[numOfSubArrs];
Dev branch merge: dev_20190606 (#7904) * correct logsoftmax looss (#2) * Small SameDiff listener fix (#4) * Various fixes (#6) * #7839 Fix for asXMatrix and tests * #7866 EmbeddingSequenceLayer dtype fix + test * #7856 SameDiff save/load stream methods * #7859 RegressionEvaluation rank 4 fix + tests + axis configuration * EvaluationBinary 3d/4d * More evaluation 3d/4d tests * #7847 Evaluation empty checks * Small test ifx * #7848 Fix median edge case * Improve DL4J samediff layer tests * [WIP] FastText wrapper implemented (#8) * FastText implemented * Some fixes * Fix shapes for wordsNearest * Validation of input vectors * Fixes * Fixed test * Thread tagged * Some tweaks * setContextClassLoader for DeallocatorServiceThread * Numpy format tests (#1) * Various fixes (#11) * #7852 SameDiff gather fix * #7892 SameDiff placeholder to constant conversion * #7890 validate input rank for MLN/CG init methods * Fix broken permute shape calculation * Permute and gather fixes * Tests * #7850 LogSumExp fix + test * Handful of test fixes * Empty arrays with non-scalar shapes (#10) * minor rearrangements for lambdas * empty tensors with non-scalar shapes * numpy empty tensors with non-scalar shapes * few more empty tweaks * Small fixes * conv3d signature update * micro fix in batchnorm mkldnn * Import fixes * Fix * MKL-DNN update * Small fill fix * fill with empty input + test * Fixes * Small error improvement * Fix * one special test * couple of fixes for lstm * Rewrite TFGraphMapper.getNDArrayFromTensor to be maintainable and less error prone * Fixes * FP16 * Unsigned * BFloat16 * Fill op - empty tweaks * - couple of fixes for empty arrays construction - stack updated * strided slice fix * one transform test * provide method for reducing shapeInfo in case of input array is empty * Fixed reduceAlongDimensions to use empty input properly. * couple of broadcast tests * couple of tests broadcast tests + tweak to make them pass * add check of non-empty to methods producing sub-arrays * Fixed reshapeC with zeros in shape. * complete empty check in reduce_... legacy ops * Concat and cumsum/prod * Tweak to empty shape inference on import * add empty check to the rest of reduce legacy ops * one more test * correct typo in evalReduceShapeInfoEmpty * Added tests for reduce_* ops to tests with zero shapes. * few more tests for empty reductions * Fixed strided_slice op with empty case and tests. * one more empty reduction test * Fixed strided_slice test. * add empty check to NDArray::reshapei * infOrMax * empty min/max with infinity tests * made unstack working correctly with empty arrays * few IndexReduce tests + tweaks for empty shapes * add test for empty concat * few tests fixed * Validation fix for reductions on empty shapes * Reverse fix * Reduction shape calc fixes * SameDiff.generateOutputVariable: don't use shape function to determine number of outputs * Range fix * - NDArray constructor updated for scalars/empty arrays - few tests fixed * More fixes * Empty creator fixes * concat fix * concat fix * TF import tests: allow 'both all NaN' and 'both all inf' to pass * Slice, zero fraction, and reshape fixes * transpose, gather * Zero fraction * scalar cast fix * Empty reduction axis support * few more tests fixed * Fixed input checks conforming with TF for concat op and tests. * few tests fixed * matmul scalar shape fix * Fixed checkout for data type and scalarity with concat to allow non-empty scalars with vector concats. * broadcast bool fix * few more tests * few more tests * correct evalReduceShapeInfoEmpty * argmax/argmin + tests * one more empty edge case + one more test * argmax/argmin/realdiv_bp tweaks * empty reshape test + fix * Helper fixes * Small fixes * Gather test fix * Gather test fix * Small fixes * reduce scalar zero values * scalar mean workaround * Remove debug code * along dim mean workaround * one more test * - equalsTo() tweak for empty arrays - one more test * broadcast tweaks
2019-06-15 13:34:34 +02:00
if (numOfSubArrs > 0)
shape::calcSubArrShapeAndOffsets(shapeInfo, numOfSubArrs, dimsToExclude.size(), dimsToExclude.data(), sPtr, oPtr, descriptor.areUnitiesinShape());
2019-06-06 14:21:15 +02:00
Nd4jPointer soPtr;
auto res = cudaMalloc(reinterpret_cast<void**>(&soPtr), numOfSubArrs * sizeof(Nd4jLong));
if (res != 0)
throw cuda_exception::build("Memory allocation for tadOffsets failed", res);
res = cudaMemcpy(soPtr, oPtr, numOfSubArrs * sizeof(Nd4jLong), cudaMemcpyHostToDevice);
if (res != 0)
throw cuda_exception::build("tadOffsets copy failed", res);
auto ssPtr = ConstantHelper::getInstance()->replicatePointer(sPtr, shape::shapeInfoByteLength(subArrRank));
ConstantDataBuffer shapesBuffer(sPtr, ssPtr, shape::shapeInfoLength(subArrRank) * sizeof(Nd4jLong), DataType::INT64);
ConstantDataBuffer offsetsBuffer(oPtr, soPtr, numOfSubArrs * sizeof(Nd4jLong), DataType::INT64);
TadPack t(shapesBuffer, offsetsBuffer, numOfSubArrs);
_cache[deviceId][descriptor] = t;
TadPack &r = _cache[deviceId][descriptor];
_mutex.unlock();
delete[] shapeInfo;
return r;
} else {
TadPack &r = _cache[deviceId][descriptor];
_mutex.unlock();
return r;
}
}
nd4j::ConstantTadHelper* nd4j::ConstantTadHelper::_INSTANCE = 0;
}