cavis/libnd4j/include/ops/declarable/helpers/cpu/batchnorm.cpp

147 lines
6.1 KiB
C++
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 Yurii Shyrma, created on 25.02.2018
//
#include<ops/declarable/helpers/batchnorm.h>
#include <helpers/ShapeUtils.h>
#include <OmpLaunchHelper.h>
namespace nd4j {
namespace ops {
namespace helpers {
//////////////////////////////////////////////////////////////////////////
template <typename T>
static void batchnorm_(const NDArray* input, const NDArray* mean, const NDArray* variance, const NDArray* gamma, const NDArray* beta, NDArray* output, const std::vector<int>& axes, const double epsilon) {
NDArray sigmaInvGam(mean); // do not copy mean's buffer, take only its shapeInfo
T eps = epsilon;
if(gamma != nullptr) {
auto lambda = LAMBDA_TT(x, y, eps) {return x / nd4j::math::nd4j_sqrt<T, T>(y + eps);};
const_cast<NDArray*>(gamma)->applyPairwiseLambda<T>(variance, lambda, &sigmaInvGam);
}
else {
auto lambda = LAMBDA_T(x, eps) { return 1. / nd4j::math::nd4j_sqrt<T, T>(x + eps); };
const_cast<NDArray*>(variance)->applyLambda<T>(lambda, &sigmaInvGam);
}
// auto sigmaInvGam = (*variance + epsilon).transform(transform::RSqrt); // sigmaInvGam = 1 / sqrt(variance + epsilon)
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
// if(gamma != nullptr) sigmaInvGam *= *gamma;
2019-06-06 14:21:15 +02:00
const T* sigmaBuff = sigmaInvGam.bufferAsT<T>();
const T* meanBuff = mean->bufferAsT<T>();
const T* inBuff = input->bufferAsT<T>();
T* outBuff = output->bufferAsT<T>();
const Nd4jLong lenBig = input->lengthOf();
const Nd4jLong lenSmall = mean->lengthOf();
const Nd4jLong* inShapeInfo = input->getShapeInfo();
const Nd4jLong* meanShapeInfo = mean->getShapeInfo();
uint inShapeInfoCast[MAX_RANK];
uint meanShapeInfoCast[MAX_RANK];
bool canCastIn = nd4j::DataTypeUtils::castShapeInfo(inShapeInfo, inShapeInfoCast);
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
bool canCastMean = nd4j::DataTypeUtils::castShapeInfo(meanShapeInfo, meanShapeInfoCast);
2019-06-06 14:21:15 +02:00
const Nd4jLong step = lenBig / lenSmall;
std::vector<int> dimsToExclude = ShapeUtils::evalDimsToExclude(input->rankOf(), axes);
OmpLaunchHelper info(lenBig, lenSmall);
if(beta != nullptr) {
const T* betaBuff = beta->bufferAsT<T>();
PRAGMA_OMP_PARALLEL_THREADS(info._numThreads)
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
{
2019-06-06 14:21:15 +02:00
const auto threadNum = omp_get_thread_num();
Nd4jLong* inOffsets = new Nd4jLong[step];
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
Nd4jLong* memBuff = new Nd4jLong[2 * inShapeInfo[0]];
for (int j = 0; j < lenSmall; ++j) {
2019-06-06 14:21:15 +02:00
const bool isOwner = j < info._numThreads ? threadNum == j : threadNum == j % info._numThreads;
if (!isOwner) continue;
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
2019-06-06 14:21:15 +02:00
const Nd4jLong start = j * step;
const Nd4jLong end = start + step;
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
2019-06-06 14:21:15 +02:00
// calculate offset for mean, variance, gamma, beta (all of them have the same shape)
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
auto offsetSmall = shape::indexOffset(j, meanShapeInfo, meanShapeInfoCast, lenSmall, canCastMean);
2019-06-06 14:21:15 +02:00
// calculate offset for input and output (all of them have the same shape)
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
shape::outerArrayOffsets(inOffsets, j, inShapeInfo, meanShapeInfo, memBuff, dimsToExclude.data());
2019-06-06 14:21:15 +02:00
PRAGMA_OMP_SIMD
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
for (Nd4jLong i = 0; i < step; ++i) {
2019-06-06 14:21:15 +02:00
auto offsetBig = inOffsets[i];
outBuff[offsetBig] = (inBuff[offsetBig] - meanBuff[offsetSmall]) * sigmaBuff[offsetSmall] + betaBuff[offsetSmall];
}
}
delete []inOffsets;
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
delete []memBuff;
}
2019-06-06 14:21:15 +02:00
}
else {
PRAGMA_OMP_PARALLEL_THREADS(info._numThreads)
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
{
2019-06-06 14:21:15 +02:00
const auto threadNum = omp_get_thread_num();
Nd4jLong* inOffsets = new Nd4jLong[step];
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
Nd4jLong* memBuff = new Nd4jLong[2 * inShapeInfo[0]];
for (int j = 0; j < lenSmall; ++j) {
2019-06-06 14:21:15 +02:00
const bool isOwner = j < info._numThreads ? threadNum == j : threadNum == j % info._numThreads;
if (!isOwner) continue;
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
2019-06-06 14:21:15 +02:00
const Nd4jLong start = j * step;
const Nd4jLong end = start + step;
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
2019-06-06 14:21:15 +02:00
// calculate offset for mean, variance, gamma, beta (all of them have the same shape)
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
auto offsetSmall = shape::indexOffset(j, meanShapeInfo, meanShapeInfoCast, lenSmall, canCastMean);
2019-06-06 14:21:15 +02:00
// calculate offset for input and output (all of them have the same shape)
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
shape::outerArrayOffsets(inOffsets, j, inShapeInfo, meanShapeInfo, memBuff, dimsToExclude.data());
2019-06-06 14:21:15 +02:00
PRAGMA_OMP_SIMD
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
for (Nd4jLong i = 0; i < step; ++i) {
2019-06-06 14:21:15 +02:00
auto offsetBig = inOffsets[i];
outBuff[offsetBig] = (inBuff[offsetBig] - meanBuff[offsetSmall]) * sigmaBuff[offsetSmall];
}
}
delete []inOffsets;
[WIP] more CUDA stuff (#57) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Added gradcheck test for dynamic_partition_bp op. * - implementation of dilation op (cpu and cuda) Signed-off-by: Yurii <yurii@skymind.io> * Fixed broadcast_dynamic_shape 1D case and tests. * Fixed usage of default integer arguments. * Fixed dynamic_partition_bp op and tests. * Eliminated test with grad check for dynamic_partition_bp op. * start working on cuda svd - porting available corresponding api from cuSOLVER library Signed-off-by: Yurii <yurii@skymind.io> * provide prelu_bp Signed-off-by: Yurii <yurii@skymind.io> * - provide gruCell_bp (old version ??) Signed-off-by: Yurii <yurii@skymind.io> * - polishing cumsum_bp and cumprod_bp tests Signed-off-by: Yurii <yurii@skymind.io> * provide sparseSoftmaxCrossEntropyWithLogits and sparseSoftmaxCrossEntropyWithLogits_grad Signed-off-by: Yurii <yurii@skymind.io> * Fixed atomicMul with float input/output * implementation of cuda kernel for triu_bp operation Signed-off-by: Yurii <yurii@skymind.io> * Refactored lup helper to add parrallel computing. * cusolver libraries Signed-off-by: raver119 <raver119@gmail.com> * uncomment cuSolver APIs in svd.cu Signed-off-by: Yurii <yurii@skymind.io> * cusolver var Signed-off-by: raver119 <raver119@gmail.com> * - further work on cuSolver svd Signed-off-by: Yurii <yurii@skymind.io> * Implement usage of cuda solver to LUP decomposition. * - correct naames in lup functions Signed-off-by: Yurii <yurii@skymind.io> * correct svdQR cuda Signed-off-by: Yurii <yurii@skymind.io> * - provide transpositions of input matrices in case of c order in svdCudaQR Signed-off-by: Yurii <yurii@skymind.io> * Fixed implementation issues with LUP usign cuda solver. * Implementation of matrix_determinant helper with cuda kernels. Working revision. * Implemented log_matrix_determinant helper with cuda kernels. * - implementation of batched cuda svd Signed-off-by: Yurii <yurii@skymind.io> * Refactored cholesky helper and implementation of cuda solver cholesky batch. * - implementation of cuda kernel for tile bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cholesky and logdet with cuda kernels. * - implementation of cuda kernel for sru_bidirectional Signed-off-by: Yurii <yurii@skymind.io> * Fixed cholesky helper. * Cholesky op helper implementation. Working double-based cublas implementation. * bad import excluded Signed-off-by: raver119 <raver119@gmail.com> * Finished with cuda implementation of cholesky helper and tests. * - implementation of cuda kernel for sru_bidirectional_backprop operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse op helper with cuda kernels. The first revision. * - start working on gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * Implementation of matrix_inverse helper. * - further work on new gruCell_bp Signed-off-by: Yurii <yurii@skymind.io> * cuBLAS related fixes Signed-off-by: raver119 <raver119@gmail.com> * calculateOutputShapes() now passes device buffers as well Signed-off-by: raver119 <raver119@gmail.com> * special concat/average/accumulate init host pointers now Signed-off-by: raver119 <raver119@gmail.com> * few more tweaks Signed-off-by: raver119 <raver119@gmail.com> * additional CudaDataBufferFactory signatures certain for data types Signed-off-by: raver119 <raver119@gmail.com> * cuSolver host buffer Signed-off-by: raver119 <raver119@gmail.com> * buffer to buffer memcpy host ptr allocation Signed-off-by: raver119 <raver119@gmail.com>
2019-07-12 10:51:51 +02:00
delete []memBuff;
2019-06-06 14:21:15 +02:00
}
}
}
//////////////////////////////////////////////////////////////////////////
void batchnorm(const NDArray* input, const NDArray* mean, const NDArray* variance, const NDArray* gamma, const NDArray* beta, NDArray* output, const std::vector<int>& axes, const double epsilon) {
BUILD_SINGLE_SELECTOR(input->dataType(), batchnorm_, (input, mean, variance, gamma, beta, output, axes, epsilon), FLOAT_TYPES);
}
BUILD_SINGLE_TEMPLATE(template void batchnorm_, (const NDArray* input, const NDArray* mean, const NDArray* variance, const NDArray* gamma, const NDArray* beta, NDArray* output, const std::vector<int>& axes, const double epsilon), FLOAT_TYPES);
}
}
}