cavis/libnd4j/tests_cpu/layers_tests/DeclarableOpsTests15.cpp

1068 lines
50 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
******************************************************************************/
//
// Created by raver on 8/4/2018.
//
#include "testlayers.h"
#include <ops/declarable/CustomOperations.h>
#include <NDArray.h>
#include <ops/ops.h>
#include <GradCheck.h>
[WIP] multi-device support (#80) * fix pad javadoc and @see links. (#72) Signed-off-by: Robert Altena <Rob@Ra-ai.com> * [WIP] More fixes (#73) * special tests for ConstantTadHelper/ConstantShapeHelper Signed-off-by: raver119 <raver119@gmail.com> * release methods for data buffers Signed-off-by: raver119 <raver119@gmail.com> * delete temporary buffer Java side Signed-off-by: raver119 <raver119@gmail.com> * delete temporary buffer Java side Signed-off-by: raver119 <raver119@gmail.com> * delete temporary TadPack C++/Java side (#74) Signed-off-by: raver119 <raver119@gmail.com> * Zoo model TF import test updates (#75) * argLine fix, update compression_gru comment * updated comment for xception * undid but commented argLine change * updated xlnet comment * copyright headers * - new NDArray methods like()/ulike() (#77) - fix for depthwise_conv2d_bp + special test Signed-off-by: raver119 <raver119@gmail.com> * upsampling2d fix CUDA Signed-off-by: raver119 <raver119@gmail.com> * DL4J trace logging (#79) * MLN/CG trace logging for debugging Signed-off-by: AlexDBlack <blacka101@gmail.com> * Tiny tweak Signed-off-by: AlexDBlack <blacka101@gmail.com> * strided_slice_bp shape fn leak fix Signed-off-by: raver119 <raver119@gmail.com> * SameDiff fixes and naming (#78) * remove SDVariable inplace methods * import methods * npe fix in OpVal * removed SameDiff inplace ops from tests * Naming updates, moved to centralized methods in SameDiff, should use op_#:# for everything * quick fixes * javadoc * SDVariable eval with placeholders * use regex match * better matching * initial commit Signed-off-by: raver119 <raver119@gmail.com> * initial commit Signed-off-by: raver119 <raver119@gmail.com> * fix javadoc. (#76) * fix javadoc. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * replace most @see with @link s. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * 4 additional tests Signed-off-by: raver119 <raver119@gmail.com> * launch context reorganization Signed-off-by: raver119 <raver119@gmail.com> * LaunchContext reorganization Signed-off-by: raver119 <raver119@gmail.com> * per-device LaunchContext Signed-off-by: raver119 <raver119@gmail.com> * Various DL4J/ND4J fixes (#81) * #7954 Force refresh of UI when switching tabs on overview page Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8017 Concurrent modification exception (synchronize) fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8033 Don't initialize updater in middle of writing memory crash dump Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8208 Fix shape checks for ND4J int[] creator methods Signed-off-by: AlexDBlack <blacka101@gmail.com> * #6385 #7992 Keras import naming fixes + cleanup Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8016 Upsampling3D - add NDHWC format support Signed-off-by: AlexDBlack <blacka101@gmail.com> * ContextBuffers as separate entity Signed-off-by: raver119 <raver119@gmail.com> * Refactor NativeOps.h to export C functions * Actually export functions from NativeOps.h * Adapt the Java wrappers in ND4J generated with JavaCPP * Create C wrappers for some of the C++ classes currently used by ND4J * ContextBuffers as separate entity Signed-off-by: raver119 <raver119@gmail.com> * remove duplicate code in createBufferDetached. (#83) Signed-off-by: Robert Altena <Rob@Ra-ai.com> * Keras model import - updater lr fix (#84) * Keras model import - updater lr fix Signed-off-by: eraly <susan.eraly@gmail.com> * Keras model import - updater lr fix, cleanup Signed-off-by: eraly <susan.eraly@gmail.com> * ContextBuffers as separate entity Signed-off-by: raver119 <raver119@gmail.com> * ContextBuffers as separate entity Signed-off-by: raver119 <raver119@gmail.com> * Fix functions of OpaqueVariablesSet * thread-local buffers/affinity Signed-off-by: raver119 <raver119@gmail.com> * thread safety for LaunchContext Signed-off-by: raver119 <raver119@gmail.com> * more of thread safety Signed-off-by: raver119 <raver119@gmail.com> * one more multi threaded test Signed-off-by: raver119 <raver119@gmail.com> * SameDiff Convolution Config validation, better output methods (#82) * Conv Config validation & tests Signed-off-by: Ryan Nett <rnett@skymind.io> * stackOutputs utility method Signed-off-by: Ryan Nett <rnett@skymind.io> * use constructor for validation, support negative kernel sizes (infered from weights) Signed-off-by: Ryan Nett <rnett@skymind.io> * better output methods Signed-off-by: Ryan Nett <rnett@skymind.io> * move output to be with fit and evaluate Signed-off-by: Ryan Nett <rnett@skymind.io> * fixes Signed-off-by: Ryan Nett <rnett@skymind.io> * more fixes Signed-off-by: Ryan Nett <rnett@skymind.io> * refactor duplicate code from pad methods. (#86) * refactor duplicate code from pad methods. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * replace switch with if. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * Various ND4J/DL4J fixes and improvements (#87) * Reshape and reallocate - small fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * Reshape and reallocate - small fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * #6488 ElementWiseVertex broadcast support Signed-off-by: AlexDBlack <blacka101@gmail.com> * Constructors and broadcast supported it Transforms.max/min Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8054 ElementWiseVertex now supports broadcast inputs Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8057 Nd4j.create overload dtype fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7551 ND4J Shape validation fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] Numpy boolean import (#91) * numpy bool type Signed-off-by: raver119 <raver119@gmail.com> * numpy bool java side Signed-off-by: raver119 <raver119@gmail.com> * remove create method with unused parameter. (#89) * remove create method with unused parameter. * removed more unused methods. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * removing more unused code. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * last removal of unused code. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * remove createSparse methods. (#92) Signed-off-by: Robert Altena <Rob@Ra-ai.com> * Various ND4J/DL4J fixes (#90) * Deprecate Old*Op instances Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8063 #8054 Broadcast exceptions + cleanup inplace ops Signed-off-by: AlexDBlack <blacka101@gmail.com> * Small fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Remove bad test condition Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7993 Fix shape function issue in crop_and_resize op Signed-off-by: AlexDBlack <blacka101@gmail.com> * DL4J SameDiff lambda layer fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8029 Fix for pnorm backprop math Signed-off-by: AlexDBlack <blacka101@gmail.com> * #8038 Fix Op profiler NaN/Inf triggering + add tests (#93) Signed-off-by: AlexDBlack <blacka101@gmail.com> * createUninitializedDetached refactoring. (#94) * wip * update interface, add null implementations. * Breaking one test in a weird way. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * createUninitializedDetached refactored. Signed-off-by: Robert Altena <Rob@Ra-ai.com> * cuda build fix for issues introduced by recent refactoring Signed-off-by: raver119 <raver119@gmail.com> * [WIP] More of CUDA (#95) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Implementation of hashcode cuda helper. Working edition. * Fixed parallel test input arangements. * Fixed tests for hashcode op. * Fixed shape calculation for image:crop_and_resize op and test. * NativeOps tests. Initial test suite. * Added tests for indexReduce methods. * Added test on execBroadcast with NDArray as dimensions. * Added test on execBroadcastBool with NDArray as dimensions. * Added tests on execPairwiseTransform and execPairwiseTransofrmBool. * Added tests for execReduce with scalar results. * Added reduce tests for non-empty dims array. * Added tests for reduce3. * Added tests for execScalar. * Added tests for execSummaryStats. * - provide cpu/cuda code for batch_to_space - testing it Signed-off-by: Yurii <yurii@skymind.io> * - remove old test for batch_to_space (had wrong format and numbers were not checked) Signed-off-by: Yurii <yurii@skymind.io> * Fixed complilation errors with test. * Added test for execTransformFloat. * Added test for execTransformSame. * Added test for execTransformBool. * Added test for execTransformStrict. * Added tests for execScalar/execScalarBool with TADs. * Added test for flatten. * - provide cpu/cuda code for space_to_Batch operaion Signed-off-by: Yurii <yurii@skymind.io> * Added test for concat. * comment unnecessary stuff in s_t_b Signed-off-by: Yurii <yurii@skymind.io> * Added test for specialConcat. * Added tests for memcpy/set routines. * Fixed pullRow cuda test. * Added pullRow test. * Added average test. * - correct typo in NDArray::applyPairwiseTransform(nd4j::pairwise::BoolOps op...) Signed-off-by: Yurii <yurii@skymind.io> * - debugging and fixing cuda tests in JavaInteropTests file Signed-off-by: Yurii <yurii@skymind.io> * - correct some tests Signed-off-by: Yurii <yurii@skymind.io> * Added test for shuffle. * Fixed ops declarations. * Restored omp and added shuffle test. * Added convertTypes test. * Added tests for execRandom. Eliminated usage of RandomBuffer with NativeOps. * Added sort tests. * Added tests for execCustomOp. * - further debuging and fixing tests terminated with crash Signed-off-by: Yurii <yurii@skymind.io> * Added tests for calculateOutputShapes. * Addded Benchmarks test. * Commented benchmark tests. * change assertion Signed-off-by: raver119 <raver119@gmail.com> * Added tests for apply_sgd op. Added cpu helper for that op. * Implement cuda helper for aplly_sgd op. Fixed tests for NativeOps. * Added test for assign broadcastable. * Added tests for assign_bp op. * Added tests for axpy op. * - assign/execScalar/execTransformAny signature change - minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Fixed axpy op. * meh Signed-off-by: raver119 <raver119@gmail.com> * - fix tests for nativeOps::concat Signed-off-by: Yurii <yurii@skymind.io> * sequential transform/scalar Signed-off-by: raver119 <raver119@gmail.com> * allow nested parallelism Signed-off-by: raver119 <raver119@gmail.com> * assign_bp leak fix Signed-off-by: raver119 <raver119@gmail.com> * block setRNG fix Signed-off-by: raver119 <raver119@gmail.com> * enable parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * enable nested parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * Added cuda implementation for row_count helper. * Added implementation for tnse gains op helper. * - take into account possible situations when input arrays are empty in reduce_ cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implemented tsne/edge_forces op cuda-based helper. Parallelized cpu-based helper for edge_forces. * Added kernel for tsne/symmetrized op heleper. * Implementation of tsne/symmetrized op cuda helper. Working edition. * Eliminated waste printfs. * Added test for broadcastgradientargs op. * host-only fallback for empty reduce float Signed-off-by: raver119 <raver119@gmail.com> * - some tests fixes Signed-off-by: Yurii <yurii@skymind.io> * - correct the rest of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * - further correction of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * Added test for Cbow op. Also added cuda implementation for cbow helpers. * - improve code of stack operation for scalar case Signed-off-by: Yurii <yurii@skymind.io> * - provide cuda kernel for gatherND operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cbow helpers with cuda kernels. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * - further correction of cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implementatation of cbow op helper with cuda kernels. Working edition. * Skip random testing for cudablas case. * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for ELU and ELU_BP ops. * Added tests for eq_scalar, gt_scalar, gte_scalar and lte_scalar ops. * Added tests for neq_scalar. * Added test for noop. * - further work on clipbynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * - get rid of concat op call, use instead direct concat helper call Signed-off-by: Yurii <yurii@skymind.io> * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for lrelu and lrelu_bp. * Added tests for selu and selu_bp. * Fixed lrelu derivative helpers. * - some corrections in lstm Signed-off-by: Yurii <yurii@skymind.io> * operator * result shape fix Signed-off-by: raver119 <raver119@gmail.com> * - correct typo in lstmCell Signed-off-by: Yurii <yurii@skymind.io> * few tests fixed Signed-off-by: raver119 <raver119@gmail.com> * CUDA inverse broadcast bool fix Signed-off-by: raver119 <raver119@gmail.com> * disable MMAP test for CUDA Signed-off-by: raver119 <raver119@gmail.com> * BooleanOp syncToDevice Signed-off-by: raver119 <raver119@gmail.com> * meh Signed-off-by: raver119 <raver119@gmail.com> * additional data types for im2col/col2im Signed-off-by: raver119 <raver119@gmail.com> * Added test for firas_sparse op. * one more RandomBuffer test excluded Signed-off-by: raver119 <raver119@gmail.com> * Added tests for flatten op. * Added test for Floor op. * bunch of tests fixed Signed-off-by: raver119 <raver119@gmail.com> * mmulDot tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Implemented floordiv_bp op and tests. * Fixed scalar case with cuda implementation for bds. * - work on cuda kernel for clip_by_norm backprop op is completed Signed-off-by: Yurii <yurii@skymind.io> * Eliminate cbow crach. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Eliminated abortion with batched nlp test. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Fixed shared flag initializing. * disabled bunch of cpu workspaces tests Signed-off-by: raver119 <raver119@gmail.com> * scalar operators fix: missing registerSpecialUse call Signed-off-by: raver119 <raver119@gmail.com> * Fixed logdet for cuda and tests. * - correct clipBynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * Fixed crop_and_resize shape datatype. * - correct some mmul tests Signed-off-by: Yurii <yurii@skymind.io> * build fix Signed-off-by: raver119 <raver119@gmail.com> * exclude two methods for JNI Signed-off-by: raver119 <raver119@gmail.com> * exclude two methods for JNI Signed-off-by: raver119 <raver119@gmail.com> * exclude two methods for JNI (#97) Signed-off-by: raver119 <raver119@gmail.com> * temporary stack fix Signed-off-by: raver119 <raver119@gmail.com> * round robin affinity test Signed-off-by: raver119 <raver119@gmail.com> * get rid of legacy CudaContext methods Signed-off-by: raver119 <raver119@gmail.com> * get rid of legacy ContextPool classes/methods Signed-off-by: raver119 <raver119@gmail.com> * one legacy test removed Signed-off-by: raver119 <raver119@gmail.com> * few more fields rearranged Signed-off-by: raver119 <raver119@gmail.com> * OpaqueLaunchContext Signed-off-by: raver119 <raver119@gmail.com> * OpaqueLaunchContext++ Signed-off-by: raver119 <raver119@gmail.com> * more of OpaqueLaunchContext methods Signed-off-by: raver119 <raver119@gmail.com> * LaunchContext -> CudaContext Signed-off-by: raver119 <raver119@gmail.com> * AffinityManger changes Signed-off-by: raver119 <raver119@gmail.com> * AffinityManger changes Signed-off-by: raver119 <raver119@gmail.com> * cusolver handles Signed-off-by: raver119 <raver119@gmail.com> * typo Signed-off-by: raver119 <raver119@gmail.com> * cusolver method Signed-off-by: raver119 <raver119@gmail.com> * cusolver handle propagated Signed-off-by: raver119 <raver119@gmail.com> * blas/solver handles Signed-off-by: raver119 <raver119@gmail.com> * one more test Signed-off-by: raver119 <raver119@gmail.com> * legacy concat implementations replaced with new CustomOp Signed-off-by: raver119 <raver119@gmail.com> * one more test Signed-off-by: raver119 <raver119@gmail.com> * concat now uses way more blocks Signed-off-by: raver119 <raver119@gmail.com> * print Signed-off-by: raver119 <raver119@gmail.com> * no more triple template mmul Signed-off-by: raver119 <raver119@gmail.com> * bunch of kernels have dtypes reconsidered Signed-off-by: raver119 <raver119@gmail.com> * bunch of kernels have dtypes reconsidered Signed-off-by: raver119 <raver119@gmail.com> * bitonic sort reorganized Signed-off-by: raver119 <raver119@gmail.com> * bunch of cpu stuff removed from cuda scope Signed-off-by: raver119 <raver119@gmail.com> * bunch of cpu stuff removed from cuda scope Signed-off-by: raver119 <raver119@gmail.com> * type conversions moved to generic impl Signed-off-by: raver119 <raver119@gmail.com> * cpu data types pass Signed-off-by: raver119 <raver119@gmail.com> * non_max_suppression Signed-off-by: raver119 <raver119@gmail.com> * sortByValue fix Signed-off-by: raver119 <raver119@gmail.com> * ignore all mixed datatype tests for mmul Signed-off-by: raver119 <raver119@gmail.com> * special handling of OpProfiler exceptions Signed-off-by: raver119 <raver119@gmail.com> * - one failing concat test in cpp - Nd4j.tile now uses op internally Signed-off-by: raver119 <raver119@gmail.com> * get back dtype exception for legacy arrays deserialization Signed-off-by: raver119 <raver119@gmail.com>
2019-08-14 15:52:34 +02:00
#include <array>
2019-06-06 14:21:15 +02:00
using namespace nd4j;
class DeclarableOpsTests15 : public testing::Test {
public:
DeclarableOpsTests15() {
printf("\n");
fflush(stdout);
}
};
TEST_F(DeclarableOpsTests15, Test_NormalizeMoments_1) {
auto d = NDArrayFactory::create<double>('c', {10, 10});
auto w = NDArrayFactory::create<double>(10);
auto x = NDArrayFactory::create<double>('c', {10});
auto y = NDArrayFactory::create<double>('c', {10});
auto z0 = NDArrayFactory::create<double>('c', {10});
auto z1 = NDArrayFactory::create<double>('c', {10});
nd4j::ops::normalize_moments op;
auto result = op.execute({&w, &x, &y}, {&z0, &z1}, {1e-4}, {}, {});
ASSERT_EQ(Status::OK(), result);
}
TEST_F(DeclarableOpsTests15, Test_Add_1) {
auto x = NDArrayFactory::create<int>('c', {5}, {1, 1, 1, 1, 1});
auto y = NDArrayFactory::create<int>('c', {5}, {1, 1, 1, 1, 1});
auto e = NDArrayFactory::create<int>('c', {5}, {2, 2, 2, 2, 2});
nd4j::ops::add op;
auto result = op.execute({&x, &y}, {&x}, {}, {}, {});
ASSERT_EQ(Status::OK(), result);
ASSERT_EQ(e, x);
}
TEST_F(DeclarableOpsTests15, Test_Half_assign_1) {
auto x = NDArrayFactory::create<float16>('c', {2, 5});
int y = 1;
x.assign(y);
ASSERT_EQ(10, x.sumNumber().e<int>(0));
}
TEST_F(DeclarableOpsTests15, test_avgpooling_edge_1) {
[WIP] More of CUDA (#95) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Implementation of hashcode cuda helper. Working edition. * Fixed parallel test input arangements. * Fixed tests for hashcode op. * Fixed shape calculation for image:crop_and_resize op and test. * NativeOps tests. Initial test suite. * Added tests for indexReduce methods. * Added test on execBroadcast with NDArray as dimensions. * Added test on execBroadcastBool with NDArray as dimensions. * Added tests on execPairwiseTransform and execPairwiseTransofrmBool. * Added tests for execReduce with scalar results. * Added reduce tests for non-empty dims array. * Added tests for reduce3. * Added tests for execScalar. * Added tests for execSummaryStats. * - provide cpu/cuda code for batch_to_space - testing it Signed-off-by: Yurii <yurii@skymind.io> * - remove old test for batch_to_space (had wrong format and numbers were not checked) Signed-off-by: Yurii <yurii@skymind.io> * Fixed complilation errors with test. * Added test for execTransformFloat. * Added test for execTransformSame. * Added test for execTransformBool. * Added test for execTransformStrict. * Added tests for execScalar/execScalarBool with TADs. * Added test for flatten. * - provide cpu/cuda code for space_to_Batch operaion Signed-off-by: Yurii <yurii@skymind.io> * Added test for concat. * comment unnecessary stuff in s_t_b Signed-off-by: Yurii <yurii@skymind.io> * Added test for specialConcat. * Added tests for memcpy/set routines. * Fixed pullRow cuda test. * Added pullRow test. * Added average test. * - correct typo in NDArray::applyPairwiseTransform(nd4j::pairwise::BoolOps op...) Signed-off-by: Yurii <yurii@skymind.io> * - debugging and fixing cuda tests in JavaInteropTests file Signed-off-by: Yurii <yurii@skymind.io> * - correct some tests Signed-off-by: Yurii <yurii@skymind.io> * Added test for shuffle. * Fixed ops declarations. * Restored omp and added shuffle test. * Added convertTypes test. * Added tests for execRandom. Eliminated usage of RandomBuffer with NativeOps. * Added sort tests. * Added tests for execCustomOp. * - further debuging and fixing tests terminated with crash Signed-off-by: Yurii <yurii@skymind.io> * Added tests for calculateOutputShapes. * Addded Benchmarks test. * Commented benchmark tests. * change assertion Signed-off-by: raver119 <raver119@gmail.com> * Added tests for apply_sgd op. Added cpu helper for that op. * Implement cuda helper for aplly_sgd op. Fixed tests for NativeOps. * Added test for assign broadcastable. * Added tests for assign_bp op. * Added tests for axpy op. * - assign/execScalar/execTransformAny signature change - minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Fixed axpy op. * meh Signed-off-by: raver119 <raver119@gmail.com> * - fix tests for nativeOps::concat Signed-off-by: Yurii <yurii@skymind.io> * sequential transform/scalar Signed-off-by: raver119 <raver119@gmail.com> * allow nested parallelism Signed-off-by: raver119 <raver119@gmail.com> * assign_bp leak fix Signed-off-by: raver119 <raver119@gmail.com> * block setRNG fix Signed-off-by: raver119 <raver119@gmail.com> * enable parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * enable nested parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * Added cuda implementation for row_count helper. * Added implementation for tnse gains op helper. * - take into account possible situations when input arrays are empty in reduce_ cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implemented tsne/edge_forces op cuda-based helper. Parallelized cpu-based helper for edge_forces. * Added kernel for tsne/symmetrized op heleper. * Implementation of tsne/symmetrized op cuda helper. Working edition. * Eliminated waste printfs. * Added test for broadcastgradientargs op. * host-only fallback for empty reduce float Signed-off-by: raver119 <raver119@gmail.com> * - some tests fixes Signed-off-by: Yurii <yurii@skymind.io> * - correct the rest of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * - further correction of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * Added test for Cbow op. Also added cuda implementation for cbow helpers. * - improve code of stack operation for scalar case Signed-off-by: Yurii <yurii@skymind.io> * - provide cuda kernel for gatherND operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cbow helpers with cuda kernels. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * - further correction of cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implementatation of cbow op helper with cuda kernels. Working edition. * Skip random testing for cudablas case. * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for ELU and ELU_BP ops. * Added tests for eq_scalar, gt_scalar, gte_scalar and lte_scalar ops. * Added tests for neq_scalar. * Added test for noop. * - further work on clipbynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * - get rid of concat op call, use instead direct concat helper call Signed-off-by: Yurii <yurii@skymind.io> * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for lrelu and lrelu_bp. * Added tests for selu and selu_bp. * Fixed lrelu derivative helpers. * - some corrections in lstm Signed-off-by: Yurii <yurii@skymind.io> * operator * result shape fix Signed-off-by: raver119 <raver119@gmail.com> * - correct typo in lstmCell Signed-off-by: Yurii <yurii@skymind.io> * few tests fixed Signed-off-by: raver119 <raver119@gmail.com> * CUDA inverse broadcast bool fix Signed-off-by: raver119 <raver119@gmail.com> * disable MMAP test for CUDA Signed-off-by: raver119 <raver119@gmail.com> * BooleanOp syncToDevice Signed-off-by: raver119 <raver119@gmail.com> * meh Signed-off-by: raver119 <raver119@gmail.com> * additional data types for im2col/col2im Signed-off-by: raver119 <raver119@gmail.com> * Added test for firas_sparse op. * one more RandomBuffer test excluded Signed-off-by: raver119 <raver119@gmail.com> * Added tests for flatten op. * Added test for Floor op. * bunch of tests fixed Signed-off-by: raver119 <raver119@gmail.com> * mmulDot tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Implemented floordiv_bp op and tests. * Fixed scalar case with cuda implementation for bds. * - work on cuda kernel for clip_by_norm backprop op is completed Signed-off-by: Yurii <yurii@skymind.io> * Eliminate cbow crach. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Eliminated abortion with batched nlp test. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Fixed shared flag initializing. * disabled bunch of cpu workspaces tests Signed-off-by: raver119 <raver119@gmail.com> * scalar operators fix: missing registerSpecialUse call Signed-off-by: raver119 <raver119@gmail.com> * Fixed logdet for cuda and tests. * - correct clipBynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * Fixed crop_and_resize shape datatype. * - correct some mmul tests Signed-off-by: Yurii <yurii@skymind.io>
2019-08-02 19:01:03 +02:00
int inOutH = 5;// 35;
int inOutW = 5;// 35;
int inOutC = 10;// 192;
2019-06-06 14:21:15 +02:00
auto x = NDArrayFactory::create<double>('c', {1, inOutH, inOutW, inOutC});
x.linspace(1.0);
nd4j::ops::avgpool2d op;
auto result = op.execute({&x}, {}, {3,3, 1,1, 0,0, 1,1, 1, 0, 1});
ASSERT_EQ(Status::OK(), result->status());
auto z = result->at(0);
int totalPadHeight = (inOutH - 1) * 1 + 3 - inOutH;
int padTop = totalPadHeight / 2;
int padBottom = totalPadHeight - totalPadHeight / 2;
int k = 3;
auto m = NDArrayFactory::create<double>('c', {1, inOutH, inOutW, inOutC});
auto c = NDArrayFactory::create<double>('c', {1, inOutH, inOutW, inOutC});
for (int h = 0; h < inOutH; h++) {
for (int w = 0; w < inOutW; w++) {
int hFrom = h - padTop;
int wFrom = w - padBottom;
int hTo = hFrom + k;
int wTo = wFrom + k;
hFrom = nd4j::math::nd4j_max<int>(0, hFrom);
wFrom = nd4j::math::nd4j_max<int>(0, wFrom);
hTo = nd4j::math::nd4j_min<int>(inOutH, hTo);
wTo = nd4j::math::nd4j_min<int>(inOutW, wTo);
int idxOut[4];
int idxIn[4];
for (int ch = 0; ch < inOutC; ch++) {
idxOut[1] = h;
idxOut[2] = w;
idxOut[3] = ch;
idxIn[3] = ch;
for (int kh = hFrom; kh < hTo; kh++) {
for (int kw = wFrom; kw < wTo; kw++) {
idxIn[1] = kh;
idxIn[2] = kw;
auto inVal = x.e<double>(0, kh, kw, ch);
m.p(0, h, w, ch, inVal + m.e<double>(0, h, w, ch));
c.p(0, h, w, ch, 1 + c.e<int>(0, h, w, ch));
}
}
}
}
}
m /= c;
ASSERT_EQ(m, *z);
delete result;
}
TEST_F(DeclarableOpsTests15, Test_standarize_1) {
auto x = NDArrayFactory::create<float>('c', {5}, {1.f, 1.f, 1.f, 1.f, 1.f});
auto e = NDArrayFactory::create<float>('c', {5}, {0.f, 0f, 0.f, 0.f, 0.f});
2019-06-06 14:21:15 +02:00
nd4j::ops::standardize op;
auto result = op.execute({&x}, {&x}, {}, {0}, {});
ASSERT_EQ(Status::OK(), result);
ASSERT_EQ(e, x);
}
TEST_F(DeclarableOpsTests15, Test_standarize_bp_1) {
auto x = NDArrayFactory::create<float>('c', {5}, {1.f, 1.f, 1.f, 1.f, 1.f});
auto eps = NDArrayFactory::create<float>('c', {5}, {0.f, 0.f, 0.f, 0.f, 0.f});
2019-06-06 14:21:15 +02:00
nd4j::ops::standardize_bp op;
auto result = op.execute({&x, &eps}, {}, {0}, {});
ASSERT_EQ(Status::OK(), result->status());
delete result;
}
2019-09-30 17:24:12 +02:00
TEST_F(DeclarableOpsTests15, Test_AdjustContrast_1) {
auto x = NDArrayFactory::create<double>('c', {4,4,3});
NDArray factor = NDArrayFactory::create<double>(2.);
auto e = NDArrayFactory::create<double>('c', {4,4,3}, {-21.5, -20.5, -19.5, -15.5, -14.5, -13.5, -9.5, -8.5, -7.5, -3.5, -2.5, -1.5,
2.5, 3.5, 4.5, 8.5, 9.5, 10.5, 14.5, 15.5, 16.5, 20.5, 21.5, 22.5,
26.5, 27.5, 28.5, 32.5, 33.5, 34.5, 38.5, 39.5, 40.5, 44.5, 45.5, 46.5,
50.5, 51.5, 52.5, 56.5, 57.5, 58.5, 62.5, 63.5, 64.5, 68.5, 69.5, 70.5});
2019-09-30 17:24:12 +02:00
x.linspace(1.);
nd4j::ops::adjust_contrast op;
auto result = op.execute({&x, &factor}, {}, {}, {});
2019-09-30 17:24:12 +02:00
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
2019-09-30 17:24:12 +02:00
ASSERT_TRUE(e.equalsTo(out));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_AdjustContrast_2) {
auto x = NDArrayFactory::create<float>('c', {1, 4,4,3});
auto e = NDArrayFactory::create<float>('c', {1, 4,4,3}, {
-21.5f, -20.5f, -19.5f, -15.5f, -14.5f, -13.5f, -9.5f, -8.5f, -7.5f, -3.5f, -2.5f, -1.5f,
2.5f, 3.5f, 4.5f, 8.5f, 9.5f, 10.5f, 14.5f, 15.5f, 16.5f, 20.5f, 21.5f, 22.5f,
26.5f, 27.5f, 28.5f, 32.5f, 33.5f, 34.5f, 38.5f, 39.5f, 40.5f, 44.5f, 45.5f, 46.5f,
50.5f, 51.5f, 52.5f, 56.5f, 57.5f, 58.5f, 62.5f, 63.5f, 64.5f, 68.5f, 69.5f, 70.5f
2019-09-30 17:24:12 +02:00
});
x.linspace(1.);
nd4j::ops::adjust_contrast op;
auto result = op.execute({&x}, {2.}, {}, {});
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
// out->printIndexedBuffer("Adjusted Constrast");
ASSERT_TRUE(e.equalsTo(out));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_AdjustContrast_3) {
auto x = NDArrayFactory::create<float>('c', {1, 4,4,3});
auto e = NDArrayFactory::create<float>('c', {1, 4,4,3}, {
-21.5f, -20.5f, -19.5f, -15.5f, -14.5f, -13.5f, -9.5f, -8.5f, -7.5f, -3.5f, -2.5f, -1.5f,
2.5f, 3.5f, 4.5f, 8.5f, 9.5f, 10.5f, 14.5f, 15.5f, 16.5f, 20.5f, 21.5f, 22.5f,
26.5f, 27.5f, 28.5f, 32.5f, 33.5f, 34.5f, 38.5f, 39.5f, 40.5f, 44.5f, 45.5f, 46.5f,
50.5f, 51.5f, 52.5f, 56.5f, 57.5f, 58.5f, 62.5f, 63.5f, 64.5f, 68.5f, 69.5f, 70.5f
});
x.linspace(1.);
nd4j::ops::adjust_contrast_v2 op;
auto result = op.execute({&x}, {2.}, {}, {});
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
// out->printIndexedBuffer("Adjusted Constrast");
ASSERT_TRUE(e.equalsTo(out));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_AdjustContrast_4) {
auto x = NDArrayFactory::create<double>('c', {4, 4, 3});
auto e = NDArrayFactory::create<double>('c', {4, 4, 3}, {
-21.5, -20.5, -19.5, -15.5, -14.5, -13.5, -9.5, -8.5, -7.5, -3.5, -2.5, -1.5,
2.5, 3.5, 4.5, 8.5, 9.5, 10.5, 14.5, 15.5, 16.5, 20.5, 21.5, 22.5,
26.5, 27.5, 28.5, 32.5, 33.5, 34.5, 38.5, 39.5, 40.5, 44.5, 45.5, 46.5,
50.5, 51.5, 52.5, 56.5, 57.5, 58.5, 62.5, 63.5, 64.5, 68.5, 69.5, 70.5
});
x.linspace(1.);
nd4j::ops::adjust_contrast_v2 op;
auto result = op.execute({&x}, {2.}, {}, {});
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
// out->printIndexedBuffer("Adjusted Constrast");
ASSERT_TRUE(e.equalsTo(out));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_AdjustContrast_5) {
auto x = NDArrayFactory::create<double>('c', {1, 3, 4});
auto e = NDArrayFactory::create<double>('c', {1, 3, 4}, {
-3., -2., -1., 0., 5., 6., 7., 8., 13., 14., 15., 16.
});
x.linspace(1.);
nd4j::ops::adjust_contrast_v2 op;
auto result = op.execute({&x}, {2.}, {}, {});
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
// out->printIndexedBuffer("Adjusted Constrast");
ASSERT_TRUE(e.equalsTo(out));
delete result;
}
/*
* public void testAdjustContrast1() {
INDArray in = Nd4j.createFromArray(new float[]{0.7788f,0.8012f,0.7244f,0.2309f,0.7271f,0.1804f,
0.5056f,0.8925f,0.5461f,0.9234f,0.0856f,0.7938f,0.6591f,0.5555f,0.1596f,0.3087f,0.1548f,0.4695f,
0.9939f,0.6113f,0.6765f,0.1800f,0.6750f,0.2246f,0.0509f,0.4601f,0.8284f,0.2354f,0.9752f,0.8361f,
0.2585f,0.4189f,0.7028f,0.7679f,0.5373f,0.7234f,0.2690f,0.0062f,0.0327f,0.0644f,0.8428f,0.7494f,
0.0755f,0.6245f,0.3491f,0.5793f,0.5730f,0.1822f,0.6420f,0.9143f,0.3019f,
0.3574f,0.1704f,0.8395f,0.5468f,0.0744f,0.9011f,0.6574f,0.4124f,0.2445f,0.4248f,0.5219f,
0.6952f,0.4900f,0.2158f,0.9549f,0.1386f,0.1544f,0.5365f,0.0134f,0.4163f,0.1456f,0.4109f,
0.2484f, 0.3330f,0.2974f,0.6636f,0.3808f,0.8664f, 0.1896f, 0.7530f, 0.7215f, 0.6612f, 0.7270f,
0.5704f,0.2666f,0.7453f,0.0444f,0.3024f,0.4850f,0.7982f,0.0965f,0.7843f,0.5075f,
0.0844f,0.8370f,0.6103f,0.4604f,0.6087f, 0.8594f, 0.4599f, 0.6714f, 0.2744f, 0.1981f, 0.4143f,
0.7821f,0.3505f,0.5040f,0.1180f,0.8307f,0.1817f,0.8442f,0.5074f,0.4471f,0.5105f,0.6666f,
0.2576f,0.2341f,0.6801f,0.2652f,0.5394f,0.4690f,0.6146f,0.1210f,0.2576f,0.0769f,0.4643f,
0.1628f,0.2026f,0.3774f,0.0506f,0.3462f,0.5720f,0.0838f,0.4228f,0.0588f,0.5362f,0.4756f,
0.2530f,0.1778f,0.0751f,0.8977f,0.3648f,0.3065f,0.4739f,0.7014f,0.4473f,0.5171f,0.1744f,
0.3487f,0.7759f,0.9491f,0.2072f,0.2182f,0.6520f,0.3092f,0.9545f,0.1881f,0.9579f,0.1785f,
0.9636f,0.4830f,0.6569f,0.3353f,0.9997f,0.5869f,0.5747f,0.0238f,0.2943f,0.5248f,0.5879f,
.7266f,0.1965f,0.9167f,0.9726f,0.9206f,0.0519f,0.2997f,0.0039f,0.7652f,0.5498f,
0.3794f,0.3791f,0.3528f,0.2873f,0.8082f,0.4732f,0.4399f,0.6606f,0.5991f,0.0034f,0.4874f
}).reshape(8,8,3,1);
INDArray out = Nd4j.create(DataType.FLOAT, in.shape());
INDArray[] res = Nd4j.exec(new AdjustContrast(in, 2.0, out));
assertArrayEquals(out.shape(), in.shape());
//assertEquals(expected, out);
}
* */
TEST_F(DeclarableOpsTests15, Test_AdjustContrast_6) {
auto x = NDArrayFactory::create<float>('c', {8,8, 3, 1}, {0.7788f,0.8012f,0.7244f,0.2309f,0.7271f,0.1804f,
0.5056f,0.8925f,0.5461f,0.9234f,0.0856f,0.7938f,0.6591f,0.5555f,0.1596f,0.3087f,0.1548f,0.4695f,
0.9939f,0.6113f,0.6765f,0.1800f,0.6750f,0.2246f,0.0509f,0.4601f,0.8284f,0.2354f,0.9752f,0.8361f,
0.2585f,0.4189f,0.7028f,0.7679f,0.5373f,0.7234f,0.2690f,0.0062f,0.0327f,0.0644f,0.8428f,0.7494f,
0.0755f,0.6245f,0.3491f,0.5793f,0.5730f,0.1822f,0.6420f,0.9143f,0.3019f,
0.3574f,0.1704f,0.8395f,0.5468f,0.0744f,0.9011f,0.6574f,0.4124f,0.2445f,0.4248f,0.5219f,
0.6952f,0.4900f,0.2158f,0.9549f,0.1386f,0.1544f,0.5365f,0.0134f,0.4163f,0.1456f,0.4109f,
0.2484f, 0.3330f,0.2974f,0.6636f,0.3808f,0.8664f, 0.1896f, 0.7530f, 0.7215f, 0.6612f, 0.7270f,
0.5704f,0.2666f,0.7453f,0.0444f,0.3024f,0.4850f,0.7982f,0.0965f,0.7843f,0.5075f,
0.0844f,0.8370f,0.6103f,0.4604f,0.6087f, 0.8594f, 0.4599f, 0.6714f, 0.2744f, 0.1981f, 0.4143f,
0.7821f,0.3505f,0.5040f,0.1180f,0.8307f,0.1817f,0.8442f,0.5074f,0.4471f,0.5105f,0.6666f,
0.2576f,0.2341f,0.6801f,0.2652f,0.5394f,0.4690f,0.6146f,0.1210f,0.2576f,0.0769f,0.4643f,
0.1628f,0.2026f,0.3774f,0.0506f,0.3462f,0.5720f,0.0838f,0.4228f,0.0588f,0.5362f,0.4756f,
0.2530f,0.1778f,0.0751f,0.8977f,0.3648f,0.3065f,0.4739f,0.7014f,0.4473f,0.5171f,0.1744f,
0.3487f,0.7759f,0.9491f,0.2072f,0.2182f,0.6520f,0.3092f,0.9545f,0.1881f,0.9579f,0.1785f,
0.9636f,0.4830f,0.6569f,0.3353f,0.9997f,0.5869f,0.5747f,0.0238f,0.2943f,0.5248f,0.5879f,
.7266f,0.1965f,0.9167f,0.9726f,0.9206f,0.0519f,0.2997f,0.0039f,0.7652f,0.5498f,
0.3794f,0.3791f,0.3528f,0.2873f,0.8082f,0.4732f,0.4399f,0.6606f,0.5991f,0.0034f,0.4874f});
auto e = NDArrayFactory::create<float>('c', {8, 8, 3, 1}, {
1.0218375f, 1.0666375f, 0.9130375f,
-0.07396251f, 0.91843754f, -0.17496246f,
0.47543746f, 1.2492375f, 0.55643755f,
1.3110375f, -0.36456245f, 1.0518374f,
0.7824375f, 0.57523745f, -0.21656245f,
0.0816375f, -0.2261625f, 0.40323752f,
1.4520376f, 0.6868375f, 0.81723756f,
-0.17576247f, 0.81423753f, -0.08656245f,
-0.36249164f, 0.45590833f, 1.1925083f,
0.00650835f, 1.4861084f, 1.2079083f,
0.05270836f, 0.37350836f, 0.94130826f,
1.0715083f, 0.6103083f, 0.9825083f,
0.07370833f, -0.4518917f, -0.39889166f,
-0.3354917f, 1.2213084f, 1.0345083f,
-0.3132917f, 0.78470826f, 0.23390833f,
0.6943083f, 0.68170834f, -0.09989169f,
0.8352709f, 1.3798709f, 0.15507084f,
0.26607084f, -0.10792917f, 1.2302709f,
0.6448709f, -0.29992914f, 1.3534708f,
0.86607087f, 0.37607086f, 0.04027084f,
0.40087086f, 0.59507084f, 0.9416709f,
0.53127086f, -0.01712915f, 1.4610709f,
-0.17152917f, -0.13992918f, 0.6242708f,
-0.42192918f, 0.38387084f, -0.15752912f,
0.3311833f, 0.00618333f, 0.17538333f,
0.10418332f, 0.8365834f, 0.27098334f,
1.2421833f, -0.1114167f, 1.0153834f,
0.9523833f, 0.8317833f, 0.9633833f,
0.6501833f, 0.04258335f, 0.9999833f,
-0.40181667f, 0.11418331f, 0.47938335f,
1.1057833f, -0.29761666f, 1.0779834f,
0.5243833f, -0.32181668f, 1.1833833f,
0.73157084f, 0.4317708f, 0.7283708f,
1.2297708f, 0.4307708f, 0.85377085f,
0.05977082f, -0.09282917f, 0.33957082f,
1.0751709f, 0.2119708f, 0.51897085f,
-0.25302917f, 1.1723708f, -0.12562919f,
1.1993709f, 0.5257708f, 0.40517086f,
0.53197086f, 0.8441708f, 0.02617085f,
-0.0208292f, 0.8711709f, 0.04137081f,
0.74936247f, 0.6085625f, 0.8997625f,
-0.08743751f, 0.18576252f, -0.17563748f,
0.5991625f, -0.0038375f, 0.07576251f,
0.42536253f, -0.22823751f, 0.36296248f,
0.81456256f, -0.16183749f, 0.5161625f,
-0.21183747f, 0.7429625f, 0.6217625f,
0.17656249f, 0.02616251f, -0.17923748f,
1.4659625f, 0.40016252f, 0.28356248f,
0.4195791f, 0.8745791f, 0.36637908f,
0.50597906f, -0.17942089f, 0.16917908f,
1.0235791f, 1.3699791f, -0.11382091f,
-0.0918209f, 0.7757791f, 0.09017909f,
1.3807791f, -0.15202093f, 1.3875791f,
-0.1712209f, 1.3989791f, 0.43777913f,
0.7855791f, 0.1423791f, 1.4711791f,
0.6455791f, 0.6211791f, -0.48062086f,
0.10189578f, 0.5628958f, 0.68909574f,
0.96649575f, -0.09370419f, 1.3466958f,
1.4584957f, 1.3544958f, -0.3829042f,
0.11269578f, -0.47890422f, 1.0436958f,
0.6128957f, 0.27209583f, 0.2714958f,
0.21889582f, 0.08789578f, 1.1296958f,
0.4596958f, 0.39309582f, 0.8344958f,
0.71149576f, -0.4799042f, 0.4880958f
});
nd4j::ops::adjust_contrast op;
auto result = op.execute({&x}, {2.}, {}, {});
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
// out->printBuffer("Adjusted Constrast6");
// e.printBuffer("Adjusted Expected 6");
// ASSERT_TRUE(e.equalsTo(out));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_AdjustContrast_7) {
auto x = NDArrayFactory::create<double>('c', {8,8, 3, 1}, {0.7788f,0.8012f,0.7244f,0.2309f,0.7271f,0.1804f,
0.5056f,0.8925f,0.5461f,0.9234f,0.0856f,0.7938f,0.6591f,0.5555f,0.1596f,0.3087f,0.1548f,0.4695f,
0.9939f,0.6113f,0.6765f,0.1800f,0.6750f,0.2246f,0.0509f,0.4601f,0.8284f,0.2354f,0.9752f,0.8361f,
0.2585f,0.4189f,0.7028f,0.7679f,0.5373f,0.7234f,0.2690f,0.0062f,0.0327f,0.0644f,0.8428f,0.7494f,
0.0755f,0.6245f,0.3491f,0.5793f,0.5730f,0.1822f,0.6420f,0.9143f,0.3019f,
0.3574f,0.1704f,0.8395f,0.5468f,0.0744f,0.9011f,0.6574f,0.4124f,0.2445f,0.4248f,0.5219f,
0.6952f,0.4900f,0.2158f,0.9549f,0.1386f,0.1544f,0.5365f,0.0134f,0.4163f,0.1456f,0.4109f,
0.2484f, 0.3330f,0.2974f,0.6636f,0.3808f,0.8664f, 0.1896f, 0.7530f, 0.7215f, 0.6612f, 0.7270f,
0.5704f,0.2666f,0.7453f,0.0444f,0.3024f,0.4850f,0.7982f,0.0965f,0.7843f,0.5075f,
0.0844f,0.8370f,0.6103f,0.4604f,0.6087f, 0.8594f, 0.4599f, 0.6714f, 0.2744f, 0.1981f, 0.4143f,
0.7821f,0.3505f,0.5040f,0.1180f,0.8307f,0.1817f,0.8442f,0.5074f,0.4471f,0.5105f,0.6666f,
0.2576f,0.2341f,0.6801f,0.2652f,0.5394f,0.4690f,0.6146f,0.1210f,0.2576f,0.0769f,0.4643f,
0.1628f,0.2026f,0.3774f,0.0506f,0.3462f,0.5720f,0.0838f,0.4228f,0.0588f,0.5362f,0.4756f,
0.2530f,0.1778f,0.0751f,0.8977f,0.3648f,0.3065f,0.4739f,0.7014f,0.4473f,0.5171f,0.1744f,
0.3487f,0.7759f,0.9491f,0.2072f,0.2182f,0.6520f,0.3092f,0.9545f,0.1881f,0.9579f,0.1785f,
0.9636f,0.4830f,0.6569f,0.3353f,0.9997f,0.5869f,0.5747f,0.0238f,0.2943f,0.5248f,0.5879f,
.7266f,0.1965f,0.9167f,0.9726f,0.9206f,0.0519f,0.2997f,0.0039f,0.7652f,0.5498f,
0.3794f,0.3791f,0.3528f,0.2873f,0.8082f,0.4732f,0.4399f,0.6606f,0.5991f,0.0034f,0.4874f});
auto e = NDArrayFactory::create<double>('c', {8, 8, 3, 1}, {
1.0218375, 1.0666375 , 0.9130375 ,
-0.07396251, 0.91843754, -0.17496246,
0.47543746, 1.2492375 , 0.55643755,
1.3110375 , -0.36456245, 1.0518374 ,
0.7824375 , 0.57523745, -0.21656245,
0.0816375 , -0.2261625 , 0.40323752,
1.4520376 , 0.6868375 , 0.81723756,
-0.17576247, 0.81423753, -0.08656245,
-0.36249164, 0.45590833, 1.1925083 ,
0.00650835, 1.4861084 , 1.2079083 ,
0.05270836, 0.37350836, 0.94130826,
1.0715083 , 0.6103083 , 0.9825083 ,
0.07370833, -0.4518917 , -0.39889166,
-0.3354917 , 1.2213084 , 1.0345083 ,
-0.3132917 , 0.78470826, 0.23390833,
0.6943083 , 0.68170834, -0.09989169,
0.8352709 , 1.3798709 , 0.15507084,
0.26607084, -0.10792917, 1.2302709 ,
0.6448709 , -0.29992914, 1.3534708 ,
0.86607087, 0.37607086, 0.04027084,
0.40087086, 0.59507084, 0.9416709 ,
0.53127086, -0.01712915, 1.4610709 ,
-0.17152917, -0.13992918, 0.6242708 ,
-0.42192918, 0.38387084, -0.15752912,
0.3311833 , 0.00618333, 0.17538333,
0.10418332, 0.8365834 , 0.27098334,
1.2421833 , -0.1114167 , 1.0153834 ,
0.9523833 , 0.8317833 , 0.9633833 ,
0.6501833 , 0.04258335, 0.9999833 ,
-0.40181667, 0.11418331, 0.47938335,
1.1057833 , -0.29761666, 1.0779834 ,
0.5243833 , -0.32181668, 1.1833833 ,
0.73157084, 0.4317708 , 0.7283708 ,
1.2297708 , 0.4307708 , 0.85377085,
0.05977082, -0.09282917, 0.33957082,
1.0751709 , 0.2119708 , 0.51897085,
-0.25302917, 1.1723708 , -0.12562919,
1.1993709 , 0.5257708 , 0.40517086,
0.53197086, 0.8441708 , 0.02617085,
-0.0208292 , 0.8711709 , 0.04137081,
0.74936247, 0.6085625 , 0.8997625 ,
-0.08743751, 0.18576252, -0.17563748,
0.5991625 , -0.0038375 , 0.07576251,
0.42536253, -0.22823751, 0.36296248,
0.81456256, -0.16183749, 0.5161625 ,
-0.21183747, 0.7429625 , 0.6217625 ,
0.17656249, 0.02616251, -0.17923748,
1.4659625 , 0.40016252, 0.28356248,
0.4195791 , 0.8745791 , 0.36637908,
0.50597906, -0.17942089, 0.16917908,
1.0235791 , 1.3699791 , -0.11382091,
-0.0918209 , 0.7757791 , 0.09017909,
1.3807791 , -0.15202093, 1.3875791 ,
-0.1712209 , 1.3989791 , 0.43777913,
0.7855791 , 0.1423791 , 1.4711791 ,
0.6455791 , 0.6211791 , -0.48062086,
0.10189578, 0.5628958 , 0.68909574,
0.96649575, -0.09370419, 1.3466958 ,
1.4584957 , 1.3544958 , -0.3829042 ,
0.11269578, -0.47890422, 1.0436958 ,
0.6128957 , 0.27209583, 0.2714958 ,
0.21889582, 0.08789578, 1.1296958 ,
0.4596958 , 0.39309582, 0.8344958 ,
0.71149576, -0.4799042, 0.4880958
});
// x.linspace(1.);
nd4j::ops::adjust_contrast_v2 op;
auto result = op.execute({&x}, {2.}, {}, {});
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
// out->printBuffer("Adjusted Constrast7");
// e.printBuffer("Adjusted expected 7");
auto diff = e - *out;
// diff.printBuffer("Adjusted subtract 7");
ASSERT_TRUE(e.equalsTo(out));
delete result;
}
2019-10-02 18:05:26 +02:00
TEST_F(DeclarableOpsTests15, Test_BitCast_1) {
2019-10-02 14:04:28 +02:00
auto x = NDArrayFactory::create<float>('c', {2, 2, 2});
auto e = NDArrayFactory::create<double>('c', {2, 2}, {2., 512., 8192., 131072.032 });
x.linspace(1.);
nd4j::ops::bitcast op;
auto result = op.execute({&x}, {}, {nd4j::DataType::DOUBLE}, {});
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
// out->printIndexedBuffer("Casted result");
ASSERT_TRUE(e.equalsTo(out));
delete result;
}
2019-10-02 18:05:26 +02:00
TEST_F(DeclarableOpsTests15, Test_BitCast_2) {
2019-10-02 14:04:28 +02:00
auto x = NDArrayFactory::create<float>('c', {2, 4});
auto e = NDArrayFactory::create<float16>('c', {2, 4, 2}, {0.f, 1.875f, 0.f, 2.f, 0.f, 2.125f, 0.f, 2.25f,
0.f, 2.312f, 0.f, 2.375f, 0.f, 2.438f, 0.f, 2.5f});
2019-10-02 14:04:28 +02:00
x.linspace(1.);
nd4j::ops::bitcast op;
auto result = op.execute({&x}, {}, {nd4j::DataType::HALF}, {});
ASSERT_EQ(Status::OK(), result->status());
auto out = result->at(0);
ASSERT_TRUE(e.equalsTo(out));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_BitCast_3) {
auto x = NDArrayFactory::create<float>('c', {1, 4});
x.linspace(1.);
nd4j::ops::bitcast op;
try {
auto result = op.execute({&x}, {}, {nd4j::DataType::INT64}, {});
ASSERT_NE(Status::OK(), result->status());
delete result;
} catch (std::exception& e) {
nd4j_printf("Error should be here `%s'. It's OK.\n", e.what());
}
}
TEST_F(DeclarableOpsTests15, Test_BitCast_4) {
auto x = NDArrayFactory::create<float>('c', {1, 4});
auto e = NDArrayFactory::create<Nd4jLong>('c', {1, 2}, {1234567890LL, 2468013579LL});
x.linspace(1.);
nd4j::ops::bitcast op;
try {
auto result = op.execute({&x}, {&e}, {}, {nd4j::DataType::INT64}, {});
ASSERT_NE(Status::OK(), result);
} catch(std::exception& e) {
nd4j_printf("Error `%s' should be here. It's OK.\n",e.what());
}
}
TEST_F(DeclarableOpsTests15, Test_BitCast_5) {
auto x = NDArrayFactory::create<float16>('c', {4, 4}, {
0.4922f, 0.2969f, 0.6172f, 0.8906f,
0.9297f, 0.0859f, 0.2344f, 0.3828f,
0.5781f, 0.7969f, 0.0391f, 0.1719f,
0.8359f, 0.9297f, 0.3438f, 0.0938f});
auto e = NDArrayFactory::create<Nd4jLong>('c', {4}, {4260467851820808160LL, 3900173902914993008LL, 3566895990128523424LL,
3314989625590692528LL});
nd4j::ops::bitcast op;
auto result = op.execute({&x}, {}, {nd4j::DataType::INT64}, {});
ASSERT_EQ(Status::OK(), result->status());
auto res = result->at(0);
// res->printIndexedBuffer("BITCAST5");
ASSERT_TRUE(e.equalsTo(res));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_BitCast_6) {
auto x = NDArrayFactory::create<float16>('c', {4, 4}, {
1.f, 2.f, 3.f, 4.f,
5.f, 6.f, 7.f, 8.f,
9.f, 10.f, 11.f, 12.f,
13.f, 14.f, 15.f, 16.f});
auto e = NDArrayFactory::create<Nd4jLong>('c', {4}, {4899988963420290048LL, 5188224837230806272LL, 5332342774136064128LL,
5476460161268730496LL});
nd4j::ops::bitcast op;
auto result = op.execute({&x}, {}, {nd4j::DataType::INT64}, {});
ASSERT_EQ(Status::OK(), result->status());
auto res = result->at(0);
// res->printIndexedBuffer("BITCAST6");
ASSERT_TRUE(e.equalsTo(res));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_BitCast_7) {
auto x = NDArrayFactory::create<float16>('c', {4, 4}, {
1.1f, 2.2f, 3.3f, 4.4f,
5.1f, 6.2f, 7.3f, 8.4f,
9.1f, 10.2f, 11.3f, 12.4f,
13.f, 14.2f, 15.3f, 16.4f});
auto e = NDArrayFactory::create<Nd4jLong>('c', {4}, {
4928700072476425318LL, 5202580391758873882LL, 5346698272827918477LL, 5483778673873668736LL});
nd4j::ops::bitcast op;
auto result = op.execute({&x}, {}, {nd4j::DataType::INT64}, {});
ASSERT_EQ(Status::OK(), result->status());
auto res = result->at(0);
// res->printIndexedBuffer("BITCAST7");
ASSERT_TRUE(e.equalsTo(res));
delete result;
}
TEST_F(DeclarableOpsTests15, Test_depthwise_bp_1) {
auto in = NDArrayFactory::create<float>('c', {4, 8, 64, 64});
auto w = NDArrayFactory::create<float>('c', {2, 2, 8, 2});
auto b = NDArrayFactory::create<float>('c', {1, 16});
auto grad = NDArrayFactory::create<float>('c', {4, 16, 64, 64});
auto gradI = in.like();
auto gradW = w.like();
auto gradB = b.like();
nd4j:ops::depthwise_conv2d_bp op;
auto status = op.execute({&in, &w, &b, &grad}, {&gradI, &gradW, &gradB}, {}, {2, 2, 1, 1, 0, 0, 1, 1, 1, 0}, {});
ASSERT_EQ(Status::OK(), status);
}
2019-06-06 14:21:15 +02:00
TEST_F(DeclarableOpsTests15, test_matmul_bp_1) {
auto a = NDArrayFactory::create<double>('c', {1, 3});
auto b = NDArrayFactory::create<double>('c', {1, 4});
auto gI = NDArrayFactory::create<double>('c', {3, 4});
auto gA = NDArrayFactory::create<double>('c', {1, 3});
auto gB = NDArrayFactory::create<double>('c', {1, 4});
nd4j::ops::matmul_bp op;
auto status = op.execute({&a, &b, &gI}, {&gA, &gB}, {}, {1, 0, 0}, {});
ASSERT_EQ(Status::OK(), status);
}
TEST_F(DeclarableOpsTests15, test_non_decreasing_1) {
auto x = NDArrayFactory::create<double>(1.0);
auto z = NDArrayFactory::create<bool>(false);
auto e = NDArrayFactory::create<bool>(true);
nd4j::ops::is_non_decreasing op;
Context ctx(1);
ctx.setInputArray(0, &x);
ctx.setOutputArray(0, &z);
auto status = op.execute(&ctx);
ASSERT_EQ(Status::OK(), status);
ASSERT_EQ(e, z);
}
Merge master to upstream (#7945) * Shugeo strided slice zeros (#14) * Modified strided_slice op to properly work with empty-like shapes. * Fixed test for reduce_mean with empty-like input. * [WIP] Last merge (#15) * 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 * [WIP] Fixing outstanding issues for NLP (#9) * Avoid using not-inited objects * Test fixed. * Redundant method avoided for models like FastText * KMeans++ implementation * KMeans++ implementation * Disable parallel execution * KMeans++ * Tests * Dev branch merge (#16) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Fix some issues on master (#17) * Fix DataVec test issue * Fix issue with dl4j SameDiff output layer * Dtype fix for lambda layers * #7912 BertIterator dtype fix (use float32 not global default) * [WIP] Next set of CUDA stuff (#7) New CUDA implementations and improvements * bad file * Dev branch master merge (#23) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Compatibility of deserialization (#18) Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com> * SameDiff: add activation gradient checking support for debugging (#19) * SameDiff gradient checker: first pass on activation gradient checks * Fixes + tests for activation gradient checking * Javadoc * [WIP] Some nd4j data type corrections (#20) * Adjust data type * Set correct Data type. * Size of proper data type. * fix averaged cpu load (#22) * SameDiff ops, TF import and fixes (#24) * CheckNumerics tests + fixes + misc fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fake quant Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * FakeQuantWithMinMaxArgs Signed-off-by: AlexDBlack <blacka101@gmail.com> * CheckNumerics fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix libnd4j ALL_INTS and ALL_FLOATS declaration (uint and bfloat types) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Small fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Javadoc Signed-off-by: AlexDBlack <blacka101@gmail.com> * Exception tweak Signed-off-by: AlexDBlack <blacka101@gmail.com> * fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix for out of scope stack allocated var use Signed-off-by: AlexDBlack <blacka101@gmail.com> * Ignores Signed-off-by: AlexDBlack <blacka101@gmail.com> * Ignore for known failing test (already logged issue) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Merge upstream to fork (#25) * Add thousand-separator commas to TotalParams (#7915) * Add thousand-separator commas to TotalParams The number of parameters can be quite large, and it would help the reading of the summary printout to have the TotalParams column & values at the bottom have thousand-separator-commas in them. * Add thousand-separator commas to MultiLayerNetwork Corresponding change to MultiLayerNetwork Signed-off-by: Jxtps Jxtps <jxtps435@gmail.com> * Update contributing and issue/PR templates (#7934) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix link to AdaDelta paper (#7942) Fix link to AdaDelta paper hosted on matthewzeiler.com Signed-off-by: Jxtps * Fixes, and ignores for known/logged failing issues (#7943) Signed-off-by: AlexDBlack <blacka101@gmail.com> * SameDiff + DL4J/SameDiff: Multiple fixes (#28) * #7919 HDF5 attribute buffer length fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7909 Arbiter constructor exception ux improvements Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7925 RNN output layer length checks Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7939 Add listener for validating inputs are not incorrectly modified Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7939 Integrate NonInplaceValidationListener into tests * #7844 DL4J SameDiff fixes for variable minibatch size * DL4J SameDiff fixes - ensure gradient for input placeholder is available Signed-off-by: AlexDBlack <blacka101@gmail.com> * Tweaks to ExternalErrorsFunction - use placeholders, make more robust * Another fix * More fixes * More SameDiff/DL4J fixes * Scope out scalar array creation in BaseScalarOp * Remove debug code Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] Final dev branch merge (#29) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Compatibility of deserialization (#18) Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com> * SameDiff: add activation gradient checking support for debugging (#19) * SameDiff gradient checker: first pass on activation gradient checks * Fixes + tests for activation gradient checking * Javadoc * [WIP] Some nd4j data type corrections (#20) * Adjust data type * Set correct Data type. * Size of proper data type. * fix averaged cpu load (#22) * [WIP] Multiple dataset iterators (#27) * Splitting dataset into arbitrary number * Fixes * Multiple split of iterator * Test * Test * Some fixes * signature change * one more tweak Signed-off-by: raver119 <raver119@gmail.com> * one more test for sequential use of DataSetIteratorSplitter Signed-off-by: raver119 <raver119@gmail.com> * Fixes * Fixes * one more test for Alexander Signed-off-by: raver119 <raver119@gmail.com> * Some fixes * Some fixes * one more test for Alexander Signed-off-by: raver119 <raver119@gmail.com> * minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Some fixes * Some fixes * couple of assertions tweaked Signed-off-by: raver119 <raver119@gmail.com> * MDS splitter test :/ Signed-off-by: raver119 <raver119@gmail.com> * Minor refactoring * Multi dataset * Some fixes * More tests * Small number of test fixes/improvements (failures on CI) (#31) Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] More CUDA stuff (#26) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * LRN BP CUDA Signed-off-by: raver119 <raver119@gmail.com> * less memory Signed-off-by: raver119 <raver119@gmail.com> * Fixed bug with crop_and_resize op helper. * get rid of unnecessary index-calculation dunction Signed-off-by: Yurii <yurii@skymind.io> * Fixed sort with nth_element cuda-based helper. * Refactored nth_element. * Refactored nth_element op and tests. * Modified usage of dim array with sortTad routine. * Refactored main routine of helper for non_max_image_suppression op. * non_max_image_suppression op helper with cuda kernel implementation. Initial revision. * fix vol2col cuda kernel * meh Signed-off-by: raver119 <raver119@gmail.com> * topK concept Signed-off-by: raver119 <raver119@gmail.com> * unsorted topK with scanWitdh of 1 Signed-off-by: raver119 <raver119@gmail.com> * correct vol2col tests * sorted/unsorted topK Signed-off-by: raver119 <raver119@gmail.com> * implementation and fixing col2im/col2vol * Corrected usage flags with input/output with reverse op. * dup is const now Signed-off-by: raver119 <raver119@gmail.com> * percentile op Signed-off-by: raver119 <raver119@gmail.com> * group tests for mapool2d Signed-off-by: Yurii <yurii@skymind.io> * special test for george Signed-off-by: raver119 <raver119@gmail.com> * less threads for sortTad Signed-off-by: raver119 <raver119@gmail.com> * provide conv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * remove auther in sort tad kernel code Signed-off-by: Yurii <yurii@skymind.io> * provide depthwise_conv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * - max_pooling_with_argmax - null check for special use Signed-off-by: raver119 <raver119@gmail.com> * dts cuda Signed-off-by: raver119 <raver119@gmail.com> * provide sconv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * std cuda Signed-off-by: raver119 <raver119@gmail.com> * Refactored non_max_suppression op to conform TF implementation. * Improved suppression helper. * provide pooling3d for cuda Signed-off-by: Yurii <yurii@skymind.io> * minor lstm rearrangements Signed-off-by: raver119 <raver119@gmail.com> * more of minor lstm rearrangements Signed-off-by: raver119 <raver119@gmail.com> * (bi)dynamic_rnn Signed-off-by: raver119 <raver119@gmail.com> * templates init order Signed-off-by: raver119 <raver119@gmail.com> * Refactored non_max_suppression op. * Added cuda kernel for non_max_suppression. * CPU sort by key/value Signed-off-by: raver119 <raver119@gmail.com> * CPU sort TAD by key/value Signed-off-by: raver119 <raver119@gmail.com> * CPU sort TAD by key/value tests Signed-off-by: raver119 <raver119@gmail.com> * Eliminate compiler error with cuda implementation. * - repaired gradCheck in cuda - provide conv2d_bp for cuda Signed-off-by: Yurii <yurii@skymind.io> * missed signature Signed-off-by: raver119 <raver119@gmail.com> * provide depthwise_conv2d_bp for cuda Signed-off-by: Yurii <yurii@skymind.io> * Implementation of lup helper with cuda kernel. Initial commit. * further work on backprops for convolutions Signed-off-by: Yurii <yurii@skymind.io> * CUDA linear sort by key/val Signed-off-by: raver119 <raver119@gmail.com> * CUDA tad sort by key/val Signed-off-by: raver119 <raver119@gmail.com> * start providing of backprop for pooling2d/3d Signed-off-by: Yurii <yurii@skymind.io> * Added atomicAdd for bool datatype. * dynamic partition concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic partition concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic partition scalar CUDA Signed-off-by: raver119 <raver119@gmail.com> * important comment Signed-off-by: raver119 <raver119@gmail.com> * fix pooling2d/3d backprop helpers Signed-off-by: Yurii <yurii@skymind.io> * Added non-linear test with dynamic_partition. * Improved test for dynamic_partition. * dynamic_partition TAD concept Signed-off-by: raver119 <raver119@gmail.com> * - dynamic_partition TAD CUDA impl - dynamic_partition TAD CPU fix Signed-off-by: raver119 <raver119@gmail.com> * - rewrite cpu code for usampling2d/3d - write cuda code for usampling2d/3d Signed-off-by: Yurii <yurii@skymind.io> * dynamic_stitch CUDA vector case Signed-off-by: raver119 <raver119@gmail.com> * dynamic_stitch CUDA TAD case concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic_stitch CUDA TAD case impl Signed-off-by: raver119 <raver119@gmail.com> * Added tests for dynamic_stitch 3D-4D cases. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * Fixed type check for dynamic stitch. * min/max bp Signed-off-by: raver119 <raver119@gmail.com> * rewrite code for upsampling2d/3d cpu Signed-off-by: Yurii <yurii@skymind.io> * reduce min/max/norm_max bp Signed-off-by: raver119 <raver119@gmail.com> * lup implementation. Additional enhancements. * provide code for upsamling2d/3d backprop Signed-off-by: Yurii <yurii@skymind.io> * weightedCrossEntropyWithLogits Signed-off-by: raver119 <raver119@gmail.com> * Fixed template math atomicMul for 64bit ints. * Refactored dynamic_partition_bp op. * inverseBroadcast fix Signed-off-by: raver119 <raver119@gmail.com> * DynamicPartitionBP test datatype fixed. * - nd4j_atomicMul Windows fix - cpu/NDArrayLambda.hpp excluded from CUDA Signed-off-by: raver119 <raver119@gmail.com>
2019-06-27 17:37:04 +02:00
TEST_F(DeclarableOpsTests15, test_check_numeric_1) {
auto x = NDArrayFactory::create<float>('c', {3},{1.f, 2.f, 3.f});
auto y = NDArrayFactory::string("shouldn't ever trigger");
nd4j::ops::check_numerics op;
auto result = op.execute({&x, &y}, {}, {});
ASSERT_EQ(Status::OK(), result->status());
auto z = result->at(0);
ASSERT_EQ(x, *z);
delete result;
}
TEST_F(DeclarableOpsTests15, test_check_numeric_2) {
auto x = NDArrayFactory::create<float>('c', {3},{1.f, 2.f, std::numeric_limits<float>::infinity()});
auto y = NDArrayFactory::string("should trigger");
auto z = NDArrayFactory::create<float>('c', {3} );
nd4j::ops::check_numerics op;
try {
auto status = op.execute({&x, &y}, {&z}, {}, {}, {});
ASSERT_TRUE(false);
} catch (std::invalid_argument &e) {
//
}
}
TEST_F(DeclarableOpsTests15, test_check_numeric_3) {
auto x = NDArrayFactory::create<float>('c', {3},{1.f, 2.f, std::numeric_limits<float>::quiet_NaN()});
auto y = NDArrayFactory::string("should trigger");
auto z = NDArrayFactory::create<float>('c', {3} );
nd4j::ops::check_numerics op;
try {
auto status = op.execute({&x, &y}, {&z}, {}, {}, {});
ASSERT_TRUE(false);
} catch (std::invalid_argument &e) {
//
}
}
2019-06-06 14:21:15 +02:00
TEST_F(DeclarableOpsTests15, Test_layer_norm_1) {
auto x = NDArrayFactory::create<float>('c', {1, 5}, {1.f, 2.f, 3.f, 4.f, 5.f});
auto g = NDArrayFactory::create<float>('c', {5}, {1.f, 2.f, 3.f, 4.f, 5.f});
auto b = NDArrayFactory::create<float>('c', {5}, {1.f, 2.f, 3.f, 4.f, 5.f});
2019-06-06 14:21:15 +02:00
nd4j::ops::layer_norm op;
auto result = op.execute({&x, &g, &b}, {}, {0}, {false});
2019-06-06 14:21:15 +02:00
ASSERT_EQ(Status::OK(), result->status());
delete result;
}
TEST_F(DeclarableOpsTests15, Test_layer_norm_bp_1) {
auto x = NDArrayFactory::create<float>('c', {1, 5}, {1.f, 2.f, 3.f, 4.f, 5.f});
auto g = NDArrayFactory::create<float>('c', {5}, {1.f, 2.f, 3.f, 4.f, 5.f});
auto b = NDArrayFactory::create<float>('c', {5}, {1.f, 2.f, 3.f, 4.f, 5.f});
auto eps = NDArrayFactory::create<float>('c', {1, 5}, {0.f, 0.f, 0.f, 0.f, 0.f});
2019-06-06 14:21:15 +02:00
nd4j::ops::layer_norm_bp op;
auto result = op.execute({&x, &g, &b, &eps}, {}, {0}, {false});
2019-06-06 14:21:15 +02:00
ASSERT_EQ(Status::OK(), result->status());
delete result;
}
Merge master to upstream (#7945) * Shugeo strided slice zeros (#14) * Modified strided_slice op to properly work with empty-like shapes. * Fixed test for reduce_mean with empty-like input. * [WIP] Last merge (#15) * 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 * [WIP] Fixing outstanding issues for NLP (#9) * Avoid using not-inited objects * Test fixed. * Redundant method avoided for models like FastText * KMeans++ implementation * KMeans++ implementation * Disable parallel execution * KMeans++ * Tests * Dev branch merge (#16) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Fix some issues on master (#17) * Fix DataVec test issue * Fix issue with dl4j SameDiff output layer * Dtype fix for lambda layers * #7912 BertIterator dtype fix (use float32 not global default) * [WIP] Next set of CUDA stuff (#7) New CUDA implementations and improvements * bad file * Dev branch master merge (#23) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Compatibility of deserialization (#18) Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com> * SameDiff: add activation gradient checking support for debugging (#19) * SameDiff gradient checker: first pass on activation gradient checks * Fixes + tests for activation gradient checking * Javadoc * [WIP] Some nd4j data type corrections (#20) * Adjust data type * Set correct Data type. * Size of proper data type. * fix averaged cpu load (#22) * SameDiff ops, TF import and fixes (#24) * CheckNumerics tests + fixes + misc fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fake quant Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * FakeQuantWithMinMaxArgs Signed-off-by: AlexDBlack <blacka101@gmail.com> * CheckNumerics fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix libnd4j ALL_INTS and ALL_FLOATS declaration (uint and bfloat types) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Small fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Javadoc Signed-off-by: AlexDBlack <blacka101@gmail.com> * Exception tweak Signed-off-by: AlexDBlack <blacka101@gmail.com> * fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix for out of scope stack allocated var use Signed-off-by: AlexDBlack <blacka101@gmail.com> * Ignores Signed-off-by: AlexDBlack <blacka101@gmail.com> * Ignore for known failing test (already logged issue) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Merge upstream to fork (#25) * Add thousand-separator commas to TotalParams (#7915) * Add thousand-separator commas to TotalParams The number of parameters can be quite large, and it would help the reading of the summary printout to have the TotalParams column & values at the bottom have thousand-separator-commas in them. * Add thousand-separator commas to MultiLayerNetwork Corresponding change to MultiLayerNetwork Signed-off-by: Jxtps Jxtps <jxtps435@gmail.com> * Update contributing and issue/PR templates (#7934) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix link to AdaDelta paper (#7942) Fix link to AdaDelta paper hosted on matthewzeiler.com Signed-off-by: Jxtps * Fixes, and ignores for known/logged failing issues (#7943) Signed-off-by: AlexDBlack <blacka101@gmail.com> * SameDiff + DL4J/SameDiff: Multiple fixes (#28) * #7919 HDF5 attribute buffer length fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7909 Arbiter constructor exception ux improvements Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7925 RNN output layer length checks Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7939 Add listener for validating inputs are not incorrectly modified Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7939 Integrate NonInplaceValidationListener into tests * #7844 DL4J SameDiff fixes for variable minibatch size * DL4J SameDiff fixes - ensure gradient for input placeholder is available Signed-off-by: AlexDBlack <blacka101@gmail.com> * Tweaks to ExternalErrorsFunction - use placeholders, make more robust * Another fix * More fixes * More SameDiff/DL4J fixes * Scope out scalar array creation in BaseScalarOp * Remove debug code Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] Final dev branch merge (#29) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Compatibility of deserialization (#18) Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com> * SameDiff: add activation gradient checking support for debugging (#19) * SameDiff gradient checker: first pass on activation gradient checks * Fixes + tests for activation gradient checking * Javadoc * [WIP] Some nd4j data type corrections (#20) * Adjust data type * Set correct Data type. * Size of proper data type. * fix averaged cpu load (#22) * [WIP] Multiple dataset iterators (#27) * Splitting dataset into arbitrary number * Fixes * Multiple split of iterator * Test * Test * Some fixes * signature change * one more tweak Signed-off-by: raver119 <raver119@gmail.com> * one more test for sequential use of DataSetIteratorSplitter Signed-off-by: raver119 <raver119@gmail.com> * Fixes * Fixes * one more test for Alexander Signed-off-by: raver119 <raver119@gmail.com> * Some fixes * Some fixes * one more test for Alexander Signed-off-by: raver119 <raver119@gmail.com> * minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Some fixes * Some fixes * couple of assertions tweaked Signed-off-by: raver119 <raver119@gmail.com> * MDS splitter test :/ Signed-off-by: raver119 <raver119@gmail.com> * Minor refactoring * Multi dataset * Some fixes * More tests * Small number of test fixes/improvements (failures on CI) (#31) Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] More CUDA stuff (#26) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * LRN BP CUDA Signed-off-by: raver119 <raver119@gmail.com> * less memory Signed-off-by: raver119 <raver119@gmail.com> * Fixed bug with crop_and_resize op helper. * get rid of unnecessary index-calculation dunction Signed-off-by: Yurii <yurii@skymind.io> * Fixed sort with nth_element cuda-based helper. * Refactored nth_element. * Refactored nth_element op and tests. * Modified usage of dim array with sortTad routine. * Refactored main routine of helper for non_max_image_suppression op. * non_max_image_suppression op helper with cuda kernel implementation. Initial revision. * fix vol2col cuda kernel * meh Signed-off-by: raver119 <raver119@gmail.com> * topK concept Signed-off-by: raver119 <raver119@gmail.com> * unsorted topK with scanWitdh of 1 Signed-off-by: raver119 <raver119@gmail.com> * correct vol2col tests * sorted/unsorted topK Signed-off-by: raver119 <raver119@gmail.com> * implementation and fixing col2im/col2vol * Corrected usage flags with input/output with reverse op. * dup is const now Signed-off-by: raver119 <raver119@gmail.com> * percentile op Signed-off-by: raver119 <raver119@gmail.com> * group tests for mapool2d Signed-off-by: Yurii <yurii@skymind.io> * special test for george Signed-off-by: raver119 <raver119@gmail.com> * less threads for sortTad Signed-off-by: raver119 <raver119@gmail.com> * provide conv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * remove auther in sort tad kernel code Signed-off-by: Yurii <yurii@skymind.io> * provide depthwise_conv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * - max_pooling_with_argmax - null check for special use Signed-off-by: raver119 <raver119@gmail.com> * dts cuda Signed-off-by: raver119 <raver119@gmail.com> * provide sconv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * std cuda Signed-off-by: raver119 <raver119@gmail.com> * Refactored non_max_suppression op to conform TF implementation. * Improved suppression helper. * provide pooling3d for cuda Signed-off-by: Yurii <yurii@skymind.io> * minor lstm rearrangements Signed-off-by: raver119 <raver119@gmail.com> * more of minor lstm rearrangements Signed-off-by: raver119 <raver119@gmail.com> * (bi)dynamic_rnn Signed-off-by: raver119 <raver119@gmail.com> * templates init order Signed-off-by: raver119 <raver119@gmail.com> * Refactored non_max_suppression op. * Added cuda kernel for non_max_suppression. * CPU sort by key/value Signed-off-by: raver119 <raver119@gmail.com> * CPU sort TAD by key/value Signed-off-by: raver119 <raver119@gmail.com> * CPU sort TAD by key/value tests Signed-off-by: raver119 <raver119@gmail.com> * Eliminate compiler error with cuda implementation. * - repaired gradCheck in cuda - provide conv2d_bp for cuda Signed-off-by: Yurii <yurii@skymind.io> * missed signature Signed-off-by: raver119 <raver119@gmail.com> * provide depthwise_conv2d_bp for cuda Signed-off-by: Yurii <yurii@skymind.io> * Implementation of lup helper with cuda kernel. Initial commit. * further work on backprops for convolutions Signed-off-by: Yurii <yurii@skymind.io> * CUDA linear sort by key/val Signed-off-by: raver119 <raver119@gmail.com> * CUDA tad sort by key/val Signed-off-by: raver119 <raver119@gmail.com> * start providing of backprop for pooling2d/3d Signed-off-by: Yurii <yurii@skymind.io> * Added atomicAdd for bool datatype. * dynamic partition concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic partition concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic partition scalar CUDA Signed-off-by: raver119 <raver119@gmail.com> * important comment Signed-off-by: raver119 <raver119@gmail.com> * fix pooling2d/3d backprop helpers Signed-off-by: Yurii <yurii@skymind.io> * Added non-linear test with dynamic_partition. * Improved test for dynamic_partition. * dynamic_partition TAD concept Signed-off-by: raver119 <raver119@gmail.com> * - dynamic_partition TAD CUDA impl - dynamic_partition TAD CPU fix Signed-off-by: raver119 <raver119@gmail.com> * - rewrite cpu code for usampling2d/3d - write cuda code for usampling2d/3d Signed-off-by: Yurii <yurii@skymind.io> * dynamic_stitch CUDA vector case Signed-off-by: raver119 <raver119@gmail.com> * dynamic_stitch CUDA TAD case concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic_stitch CUDA TAD case impl Signed-off-by: raver119 <raver119@gmail.com> * Added tests for dynamic_stitch 3D-4D cases. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * Fixed type check for dynamic stitch. * min/max bp Signed-off-by: raver119 <raver119@gmail.com> * rewrite code for upsampling2d/3d cpu Signed-off-by: Yurii <yurii@skymind.io> * reduce min/max/norm_max bp Signed-off-by: raver119 <raver119@gmail.com> * lup implementation. Additional enhancements. * provide code for upsamling2d/3d backprop Signed-off-by: Yurii <yurii@skymind.io> * weightedCrossEntropyWithLogits Signed-off-by: raver119 <raver119@gmail.com> * Fixed template math atomicMul for 64bit ints. * Refactored dynamic_partition_bp op. * inverseBroadcast fix Signed-off-by: raver119 <raver119@gmail.com> * DynamicPartitionBP test datatype fixed. * - nd4j_atomicMul Windows fix - cpu/NDArrayLambda.hpp excluded from CUDA Signed-off-by: raver119 <raver119@gmail.com>
2019-06-27 17:37:04 +02:00
//////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, Test_layer_norm_bp_2) {
NDArray x('c', {3, 4, 8, 8}, nd4j::DataType::FLOAT32);
NDArray gain('c', {4}, {-0.1, 0.1, -0.2, 0.2}, nd4j::DataType::FLOAT32);
NDArray bias('c', {4}, {-0.05, 0.05, -1.05, 1.05}, nd4j::DataType::FLOAT32);
NDArray gradO('c', {3, 4, 8, 8}, nd4j::DataType::FLOAT32);
NDArray gradI('c', {3, 4, 8, 8}, nd4j::DataType::FLOAT32);
NDArray gradG('c', {4}, nd4j::DataType::FLOAT32);
NDArray gradB('c', {4}, nd4j::DataType::FLOAT32);
x.linspace(-20, 0.5);
gradO.linspace(-4, 0.05);
nd4j::ops::layer_norm_bp op;
auto status = op.execute({&x, &gain, &bias, &gradO}, {&gradI, &gradG, &gradB}, {}, {1,2,3}, {true});
ASSERT_EQ(Status::OK(), status);
}
TEST_F(DeclarableOpsTests15, test_hashCode_1) {
auto x = NDArrayFactory::create<int>('c', {10});
auto y = NDArrayFactory::create<int>('c', {10});
x.linspace(1.);
y.linspace(2.);
nd4j::ops::hashcode op;
[WIP] More of CUDA (#95) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Implementation of hashcode cuda helper. Working edition. * Fixed parallel test input arangements. * Fixed tests for hashcode op. * Fixed shape calculation for image:crop_and_resize op and test. * NativeOps tests. Initial test suite. * Added tests for indexReduce methods. * Added test on execBroadcast with NDArray as dimensions. * Added test on execBroadcastBool with NDArray as dimensions. * Added tests on execPairwiseTransform and execPairwiseTransofrmBool. * Added tests for execReduce with scalar results. * Added reduce tests for non-empty dims array. * Added tests for reduce3. * Added tests for execScalar. * Added tests for execSummaryStats. * - provide cpu/cuda code for batch_to_space - testing it Signed-off-by: Yurii <yurii@skymind.io> * - remove old test for batch_to_space (had wrong format and numbers were not checked) Signed-off-by: Yurii <yurii@skymind.io> * Fixed complilation errors with test. * Added test for execTransformFloat. * Added test for execTransformSame. * Added test for execTransformBool. * Added test for execTransformStrict. * Added tests for execScalar/execScalarBool with TADs. * Added test for flatten. * - provide cpu/cuda code for space_to_Batch operaion Signed-off-by: Yurii <yurii@skymind.io> * Added test for concat. * comment unnecessary stuff in s_t_b Signed-off-by: Yurii <yurii@skymind.io> * Added test for specialConcat. * Added tests for memcpy/set routines. * Fixed pullRow cuda test. * Added pullRow test. * Added average test. * - correct typo in NDArray::applyPairwiseTransform(nd4j::pairwise::BoolOps op...) Signed-off-by: Yurii <yurii@skymind.io> * - debugging and fixing cuda tests in JavaInteropTests file Signed-off-by: Yurii <yurii@skymind.io> * - correct some tests Signed-off-by: Yurii <yurii@skymind.io> * Added test for shuffle. * Fixed ops declarations. * Restored omp and added shuffle test. * Added convertTypes test. * Added tests for execRandom. Eliminated usage of RandomBuffer with NativeOps. * Added sort tests. * Added tests for execCustomOp. * - further debuging and fixing tests terminated with crash Signed-off-by: Yurii <yurii@skymind.io> * Added tests for calculateOutputShapes. * Addded Benchmarks test. * Commented benchmark tests. * change assertion Signed-off-by: raver119 <raver119@gmail.com> * Added tests for apply_sgd op. Added cpu helper for that op. * Implement cuda helper for aplly_sgd op. Fixed tests for NativeOps. * Added test for assign broadcastable. * Added tests for assign_bp op. * Added tests for axpy op. * - assign/execScalar/execTransformAny signature change - minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Fixed axpy op. * meh Signed-off-by: raver119 <raver119@gmail.com> * - fix tests for nativeOps::concat Signed-off-by: Yurii <yurii@skymind.io> * sequential transform/scalar Signed-off-by: raver119 <raver119@gmail.com> * allow nested parallelism Signed-off-by: raver119 <raver119@gmail.com> * assign_bp leak fix Signed-off-by: raver119 <raver119@gmail.com> * block setRNG fix Signed-off-by: raver119 <raver119@gmail.com> * enable parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * enable nested parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * Added cuda implementation for row_count helper. * Added implementation for tnse gains op helper. * - take into account possible situations when input arrays are empty in reduce_ cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implemented tsne/edge_forces op cuda-based helper. Parallelized cpu-based helper for edge_forces. * Added kernel for tsne/symmetrized op heleper. * Implementation of tsne/symmetrized op cuda helper. Working edition. * Eliminated waste printfs. * Added test for broadcastgradientargs op. * host-only fallback for empty reduce float Signed-off-by: raver119 <raver119@gmail.com> * - some tests fixes Signed-off-by: Yurii <yurii@skymind.io> * - correct the rest of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * - further correction of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * Added test for Cbow op. Also added cuda implementation for cbow helpers. * - improve code of stack operation for scalar case Signed-off-by: Yurii <yurii@skymind.io> * - provide cuda kernel for gatherND operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cbow helpers with cuda kernels. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * - further correction of cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implementatation of cbow op helper with cuda kernels. Working edition. * Skip random testing for cudablas case. * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for ELU and ELU_BP ops. * Added tests for eq_scalar, gt_scalar, gte_scalar and lte_scalar ops. * Added tests for neq_scalar. * Added test for noop. * - further work on clipbynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * - get rid of concat op call, use instead direct concat helper call Signed-off-by: Yurii <yurii@skymind.io> * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for lrelu and lrelu_bp. * Added tests for selu and selu_bp. * Fixed lrelu derivative helpers. * - some corrections in lstm Signed-off-by: Yurii <yurii@skymind.io> * operator * result shape fix Signed-off-by: raver119 <raver119@gmail.com> * - correct typo in lstmCell Signed-off-by: Yurii <yurii@skymind.io> * few tests fixed Signed-off-by: raver119 <raver119@gmail.com> * CUDA inverse broadcast bool fix Signed-off-by: raver119 <raver119@gmail.com> * disable MMAP test for CUDA Signed-off-by: raver119 <raver119@gmail.com> * BooleanOp syncToDevice Signed-off-by: raver119 <raver119@gmail.com> * meh Signed-off-by: raver119 <raver119@gmail.com> * additional data types for im2col/col2im Signed-off-by: raver119 <raver119@gmail.com> * Added test for firas_sparse op. * one more RandomBuffer test excluded Signed-off-by: raver119 <raver119@gmail.com> * Added tests for flatten op. * Added test for Floor op. * bunch of tests fixed Signed-off-by: raver119 <raver119@gmail.com> * mmulDot tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Implemented floordiv_bp op and tests. * Fixed scalar case with cuda implementation for bds. * - work on cuda kernel for clip_by_norm backprop op is completed Signed-off-by: Yurii <yurii@skymind.io> * Eliminate cbow crach. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Eliminated abortion with batched nlp test. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Fixed shared flag initializing. * disabled bunch of cpu workspaces tests Signed-off-by: raver119 <raver119@gmail.com> * scalar operators fix: missing registerSpecialUse call Signed-off-by: raver119 <raver119@gmail.com> * Fixed logdet for cuda and tests. * - correct clipBynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * Fixed crop_and_resize shape datatype. * - correct some mmul tests Signed-off-by: Yurii <yurii@skymind.io>
2019-08-02 19:01:03 +02:00
auto resultA0 = op.execute({&x}, {}, {}, {}, false, nd4j::DataType::INT64);
auto resultA1 = op.execute({&x}, {}, {}, {}, false, nd4j::DataType::INT64);
auto resultB0 = op.execute({&y}, {}, {}, {}, false, nd4j::DataType::INT64);
// resultA0->at(0)->printIndexedBuffer("A0");
// resultA1->at(0)->printIndexedBuffer("A1");
// resultB0->at(0)->printIndexedBuffer("B0");
ASSERT_EQ(*resultA0->at(0), *resultA1->at(0));
ASSERT_NE(*resultA0->at(0), *resultB0->at(0));
delete resultA0;
delete resultA1;
delete resultB0;
}
TEST_F(DeclarableOpsTests15, test_hashCode_2) {
auto x = NDArrayFactory::create<int>('c', {1027});
auto y = NDArrayFactory::create<int>('c', {1027});
x.linspace(1.);
y.linspace(2.);
nd4j::ops::hashcode op;
[WIP] More of CUDA (#95) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Implementation of hashcode cuda helper. Working edition. * Fixed parallel test input arangements. * Fixed tests for hashcode op. * Fixed shape calculation for image:crop_and_resize op and test. * NativeOps tests. Initial test suite. * Added tests for indexReduce methods. * Added test on execBroadcast with NDArray as dimensions. * Added test on execBroadcastBool with NDArray as dimensions. * Added tests on execPairwiseTransform and execPairwiseTransofrmBool. * Added tests for execReduce with scalar results. * Added reduce tests for non-empty dims array. * Added tests for reduce3. * Added tests for execScalar. * Added tests for execSummaryStats. * - provide cpu/cuda code for batch_to_space - testing it Signed-off-by: Yurii <yurii@skymind.io> * - remove old test for batch_to_space (had wrong format and numbers were not checked) Signed-off-by: Yurii <yurii@skymind.io> * Fixed complilation errors with test. * Added test for execTransformFloat. * Added test for execTransformSame. * Added test for execTransformBool. * Added test for execTransformStrict. * Added tests for execScalar/execScalarBool with TADs. * Added test for flatten. * - provide cpu/cuda code for space_to_Batch operaion Signed-off-by: Yurii <yurii@skymind.io> * Added test for concat. * comment unnecessary stuff in s_t_b Signed-off-by: Yurii <yurii@skymind.io> * Added test for specialConcat. * Added tests for memcpy/set routines. * Fixed pullRow cuda test. * Added pullRow test. * Added average test. * - correct typo in NDArray::applyPairwiseTransform(nd4j::pairwise::BoolOps op...) Signed-off-by: Yurii <yurii@skymind.io> * - debugging and fixing cuda tests in JavaInteropTests file Signed-off-by: Yurii <yurii@skymind.io> * - correct some tests Signed-off-by: Yurii <yurii@skymind.io> * Added test for shuffle. * Fixed ops declarations. * Restored omp and added shuffle test. * Added convertTypes test. * Added tests for execRandom. Eliminated usage of RandomBuffer with NativeOps. * Added sort tests. * Added tests for execCustomOp. * - further debuging and fixing tests terminated with crash Signed-off-by: Yurii <yurii@skymind.io> * Added tests for calculateOutputShapes. * Addded Benchmarks test. * Commented benchmark tests. * change assertion Signed-off-by: raver119 <raver119@gmail.com> * Added tests for apply_sgd op. Added cpu helper for that op. * Implement cuda helper for aplly_sgd op. Fixed tests for NativeOps. * Added test for assign broadcastable. * Added tests for assign_bp op. * Added tests for axpy op. * - assign/execScalar/execTransformAny signature change - minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Fixed axpy op. * meh Signed-off-by: raver119 <raver119@gmail.com> * - fix tests for nativeOps::concat Signed-off-by: Yurii <yurii@skymind.io> * sequential transform/scalar Signed-off-by: raver119 <raver119@gmail.com> * allow nested parallelism Signed-off-by: raver119 <raver119@gmail.com> * assign_bp leak fix Signed-off-by: raver119 <raver119@gmail.com> * block setRNG fix Signed-off-by: raver119 <raver119@gmail.com> * enable parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * enable nested parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * Added cuda implementation for row_count helper. * Added implementation for tnse gains op helper. * - take into account possible situations when input arrays are empty in reduce_ cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implemented tsne/edge_forces op cuda-based helper. Parallelized cpu-based helper for edge_forces. * Added kernel for tsne/symmetrized op heleper. * Implementation of tsne/symmetrized op cuda helper. Working edition. * Eliminated waste printfs. * Added test for broadcastgradientargs op. * host-only fallback for empty reduce float Signed-off-by: raver119 <raver119@gmail.com> * - some tests fixes Signed-off-by: Yurii <yurii@skymind.io> * - correct the rest of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * - further correction of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * Added test for Cbow op. Also added cuda implementation for cbow helpers. * - improve code of stack operation for scalar case Signed-off-by: Yurii <yurii@skymind.io> * - provide cuda kernel for gatherND operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cbow helpers with cuda kernels. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * - further correction of cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implementatation of cbow op helper with cuda kernels. Working edition. * Skip random testing for cudablas case. * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for ELU and ELU_BP ops. * Added tests for eq_scalar, gt_scalar, gte_scalar and lte_scalar ops. * Added tests for neq_scalar. * Added test for noop. * - further work on clipbynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * - get rid of concat op call, use instead direct concat helper call Signed-off-by: Yurii <yurii@skymind.io> * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for lrelu and lrelu_bp. * Added tests for selu and selu_bp. * Fixed lrelu derivative helpers. * - some corrections in lstm Signed-off-by: Yurii <yurii@skymind.io> * operator * result shape fix Signed-off-by: raver119 <raver119@gmail.com> * - correct typo in lstmCell Signed-off-by: Yurii <yurii@skymind.io> * few tests fixed Signed-off-by: raver119 <raver119@gmail.com> * CUDA inverse broadcast bool fix Signed-off-by: raver119 <raver119@gmail.com> * disable MMAP test for CUDA Signed-off-by: raver119 <raver119@gmail.com> * BooleanOp syncToDevice Signed-off-by: raver119 <raver119@gmail.com> * meh Signed-off-by: raver119 <raver119@gmail.com> * additional data types for im2col/col2im Signed-off-by: raver119 <raver119@gmail.com> * Added test for firas_sparse op. * one more RandomBuffer test excluded Signed-off-by: raver119 <raver119@gmail.com> * Added tests for flatten op. * Added test for Floor op. * bunch of tests fixed Signed-off-by: raver119 <raver119@gmail.com> * mmulDot tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Implemented floordiv_bp op and tests. * Fixed scalar case with cuda implementation for bds. * - work on cuda kernel for clip_by_norm backprop op is completed Signed-off-by: Yurii <yurii@skymind.io> * Eliminate cbow crach. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Eliminated abortion with batched nlp test. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Fixed shared flag initializing. * disabled bunch of cpu workspaces tests Signed-off-by: raver119 <raver119@gmail.com> * scalar operators fix: missing registerSpecialUse call Signed-off-by: raver119 <raver119@gmail.com> * Fixed logdet for cuda and tests. * - correct clipBynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * Fixed crop_and_resize shape datatype. * - correct some mmul tests Signed-off-by: Yurii <yurii@skymind.io>
2019-08-02 19:01:03 +02:00
auto resultA0 = op.execute({&x}, {}, {}, {}, false, nd4j::DataType::INT64);
auto resultA1 = op.execute({&x}, {}, {}, {}, false, nd4j::DataType::INT64);
auto resultB0 = op.execute({&y}, {}, {}, {}, false, nd4j::DataType::INT64);
// resultA0->at(0)->printIndexedBuffer("A0");
// resultA1->at(0)->printIndexedBuffer("A1");
// resultB0->at(0)->printIndexedBuffer("B0");
ASSERT_EQ(*resultA0->at(0), *resultA1->at(0));
ASSERT_NE(*resultA0->at(0), *resultB0->at(0));
delete resultA0;
delete resultA1;
delete resultB0;
}
TEST_F(DeclarableOpsTests15, test_reshape_to_scalar_1) {
auto array = NDArrayFactory::create<float>(119.f);
auto e = NDArrayFactory::create<float>('c', {1, 1}, {119.f});
nd4j::ops::reshape op;
auto result = op.execute({&array}, {}, {1, 1});
ASSERT_EQ(Status::OK(), result->status());
auto z = result->at(0);
ASSERT_EQ(e, *z);
delete result;
}
TEST_F(DeclarableOpsTests15, test_reshape_to_scalar_2) {
auto array = NDArrayFactory::create<float>(119.f);
auto e = NDArrayFactory::create<float>('c', {1, 1}, {119.f});
auto z = NDArrayFactory::create<float>('c', {1, 1});
nd4j::ops::reshape op;
auto result = op.execute({&array}, {&z}, {}, {1, 1}, {});
ASSERT_EQ(Status::OK(), result);
ASSERT_EQ(e, z);
}
TEST_F(DeclarableOpsTests15, test_rank_1) {
auto array = NDArrayFactory::create<float>('c', {4, 64});
auto e = NDArrayFactory::create<int>('c', {}, {2});
auto z = NDArrayFactory::create<int>('c', {});
nd4j::ops::rank op;
auto result = op.execute({&array}, {&z}, {}, {}, {});
ASSERT_EQ(Status::OK(), result);
ASSERT_EQ(e, z);
}
TEST_F(DeclarableOpsTests15, test_rank_2) {
auto array = NDArrayFactory::create<float>('c', {4, 64});
auto e = NDArrayFactory::create<int>('c', {}, {2});
nd4j::ops::rank op;
auto result = op.execute({&array}, {}, {});
ASSERT_EQ(Status::OK(), result->status());
auto z = result->at(0);
ASSERT_EQ(e, *z);
delete result;
}
Merge master to upstream (#7945) * Shugeo strided slice zeros (#14) * Modified strided_slice op to properly work with empty-like shapes. * Fixed test for reduce_mean with empty-like input. * [WIP] Last merge (#15) * 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 * [WIP] Fixing outstanding issues for NLP (#9) * Avoid using not-inited objects * Test fixed. * Redundant method avoided for models like FastText * KMeans++ implementation * KMeans++ implementation * Disable parallel execution * KMeans++ * Tests * Dev branch merge (#16) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Fix some issues on master (#17) * Fix DataVec test issue * Fix issue with dl4j SameDiff output layer * Dtype fix for lambda layers * #7912 BertIterator dtype fix (use float32 not global default) * [WIP] Next set of CUDA stuff (#7) New CUDA implementations and improvements * bad file * Dev branch master merge (#23) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Compatibility of deserialization (#18) Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com> * SameDiff: add activation gradient checking support for debugging (#19) * SameDiff gradient checker: first pass on activation gradient checks * Fixes + tests for activation gradient checking * Javadoc * [WIP] Some nd4j data type corrections (#20) * Adjust data type * Set correct Data type. * Size of proper data type. * fix averaged cpu load (#22) * SameDiff ops, TF import and fixes (#24) * CheckNumerics tests + fixes + misc fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fake quant Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * FakeQuantWithMinMaxArgs Signed-off-by: AlexDBlack <blacka101@gmail.com> * CheckNumerics fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix libnd4j ALL_INTS and ALL_FLOATS declaration (uint and bfloat types) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Small fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Javadoc Signed-off-by: AlexDBlack <blacka101@gmail.com> * Exception tweak Signed-off-by: AlexDBlack <blacka101@gmail.com> * fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix for out of scope stack allocated var use Signed-off-by: AlexDBlack <blacka101@gmail.com> * Ignores Signed-off-by: AlexDBlack <blacka101@gmail.com> * Ignore for known failing test (already logged issue) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Merge upstream to fork (#25) * Add thousand-separator commas to TotalParams (#7915) * Add thousand-separator commas to TotalParams The number of parameters can be quite large, and it would help the reading of the summary printout to have the TotalParams column & values at the bottom have thousand-separator-commas in them. * Add thousand-separator commas to MultiLayerNetwork Corresponding change to MultiLayerNetwork Signed-off-by: Jxtps Jxtps <jxtps435@gmail.com> * Update contributing and issue/PR templates (#7934) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix link to AdaDelta paper (#7942) Fix link to AdaDelta paper hosted on matthewzeiler.com Signed-off-by: Jxtps * Fixes, and ignores for known/logged failing issues (#7943) Signed-off-by: AlexDBlack <blacka101@gmail.com> * SameDiff + DL4J/SameDiff: Multiple fixes (#28) * #7919 HDF5 attribute buffer length fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7909 Arbiter constructor exception ux improvements Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7925 RNN output layer length checks Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7939 Add listener for validating inputs are not incorrectly modified Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7939 Integrate NonInplaceValidationListener into tests * #7844 DL4J SameDiff fixes for variable minibatch size * DL4J SameDiff fixes - ensure gradient for input placeholder is available Signed-off-by: AlexDBlack <blacka101@gmail.com> * Tweaks to ExternalErrorsFunction - use placeholders, make more robust * Another fix * More fixes * More SameDiff/DL4J fixes * Scope out scalar array creation in BaseScalarOp * Remove debug code Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] Final dev branch merge (#29) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Compatibility of deserialization (#18) Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com> * SameDiff: add activation gradient checking support for debugging (#19) * SameDiff gradient checker: first pass on activation gradient checks * Fixes + tests for activation gradient checking * Javadoc * [WIP] Some nd4j data type corrections (#20) * Adjust data type * Set correct Data type. * Size of proper data type. * fix averaged cpu load (#22) * [WIP] Multiple dataset iterators (#27) * Splitting dataset into arbitrary number * Fixes * Multiple split of iterator * Test * Test * Some fixes * signature change * one more tweak Signed-off-by: raver119 <raver119@gmail.com> * one more test for sequential use of DataSetIteratorSplitter Signed-off-by: raver119 <raver119@gmail.com> * Fixes * Fixes * one more test for Alexander Signed-off-by: raver119 <raver119@gmail.com> * Some fixes * Some fixes * one more test for Alexander Signed-off-by: raver119 <raver119@gmail.com> * minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Some fixes * Some fixes * couple of assertions tweaked Signed-off-by: raver119 <raver119@gmail.com> * MDS splitter test :/ Signed-off-by: raver119 <raver119@gmail.com> * Minor refactoring * Multi dataset * Some fixes * More tests * Small number of test fixes/improvements (failures on CI) (#31) Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] More CUDA stuff (#26) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * LRN BP CUDA Signed-off-by: raver119 <raver119@gmail.com> * less memory Signed-off-by: raver119 <raver119@gmail.com> * Fixed bug with crop_and_resize op helper. * get rid of unnecessary index-calculation dunction Signed-off-by: Yurii <yurii@skymind.io> * Fixed sort with nth_element cuda-based helper. * Refactored nth_element. * Refactored nth_element op and tests. * Modified usage of dim array with sortTad routine. * Refactored main routine of helper for non_max_image_suppression op. * non_max_image_suppression op helper with cuda kernel implementation. Initial revision. * fix vol2col cuda kernel * meh Signed-off-by: raver119 <raver119@gmail.com> * topK concept Signed-off-by: raver119 <raver119@gmail.com> * unsorted topK with scanWitdh of 1 Signed-off-by: raver119 <raver119@gmail.com> * correct vol2col tests * sorted/unsorted topK Signed-off-by: raver119 <raver119@gmail.com> * implementation and fixing col2im/col2vol * Corrected usage flags with input/output with reverse op. * dup is const now Signed-off-by: raver119 <raver119@gmail.com> * percentile op Signed-off-by: raver119 <raver119@gmail.com> * group tests for mapool2d Signed-off-by: Yurii <yurii@skymind.io> * special test for george Signed-off-by: raver119 <raver119@gmail.com> * less threads for sortTad Signed-off-by: raver119 <raver119@gmail.com> * provide conv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * remove auther in sort tad kernel code Signed-off-by: Yurii <yurii@skymind.io> * provide depthwise_conv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * - max_pooling_with_argmax - null check for special use Signed-off-by: raver119 <raver119@gmail.com> * dts cuda Signed-off-by: raver119 <raver119@gmail.com> * provide sconv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * std cuda Signed-off-by: raver119 <raver119@gmail.com> * Refactored non_max_suppression op to conform TF implementation. * Improved suppression helper. * provide pooling3d for cuda Signed-off-by: Yurii <yurii@skymind.io> * minor lstm rearrangements Signed-off-by: raver119 <raver119@gmail.com> * more of minor lstm rearrangements Signed-off-by: raver119 <raver119@gmail.com> * (bi)dynamic_rnn Signed-off-by: raver119 <raver119@gmail.com> * templates init order Signed-off-by: raver119 <raver119@gmail.com> * Refactored non_max_suppression op. * Added cuda kernel for non_max_suppression. * CPU sort by key/value Signed-off-by: raver119 <raver119@gmail.com> * CPU sort TAD by key/value Signed-off-by: raver119 <raver119@gmail.com> * CPU sort TAD by key/value tests Signed-off-by: raver119 <raver119@gmail.com> * Eliminate compiler error with cuda implementation. * - repaired gradCheck in cuda - provide conv2d_bp for cuda Signed-off-by: Yurii <yurii@skymind.io> * missed signature Signed-off-by: raver119 <raver119@gmail.com> * provide depthwise_conv2d_bp for cuda Signed-off-by: Yurii <yurii@skymind.io> * Implementation of lup helper with cuda kernel. Initial commit. * further work on backprops for convolutions Signed-off-by: Yurii <yurii@skymind.io> * CUDA linear sort by key/val Signed-off-by: raver119 <raver119@gmail.com> * CUDA tad sort by key/val Signed-off-by: raver119 <raver119@gmail.com> * start providing of backprop for pooling2d/3d Signed-off-by: Yurii <yurii@skymind.io> * Added atomicAdd for bool datatype. * dynamic partition concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic partition concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic partition scalar CUDA Signed-off-by: raver119 <raver119@gmail.com> * important comment Signed-off-by: raver119 <raver119@gmail.com> * fix pooling2d/3d backprop helpers Signed-off-by: Yurii <yurii@skymind.io> * Added non-linear test with dynamic_partition. * Improved test for dynamic_partition. * dynamic_partition TAD concept Signed-off-by: raver119 <raver119@gmail.com> * - dynamic_partition TAD CUDA impl - dynamic_partition TAD CPU fix Signed-off-by: raver119 <raver119@gmail.com> * - rewrite cpu code for usampling2d/3d - write cuda code for usampling2d/3d Signed-off-by: Yurii <yurii@skymind.io> * dynamic_stitch CUDA vector case Signed-off-by: raver119 <raver119@gmail.com> * dynamic_stitch CUDA TAD case concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic_stitch CUDA TAD case impl Signed-off-by: raver119 <raver119@gmail.com> * Added tests for dynamic_stitch 3D-4D cases. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * Fixed type check for dynamic stitch. * min/max bp Signed-off-by: raver119 <raver119@gmail.com> * rewrite code for upsampling2d/3d cpu Signed-off-by: Yurii <yurii@skymind.io> * reduce min/max/norm_max bp Signed-off-by: raver119 <raver119@gmail.com> * lup implementation. Additional enhancements. * provide code for upsamling2d/3d backprop Signed-off-by: Yurii <yurii@skymind.io> * weightedCrossEntropyWithLogits Signed-off-by: raver119 <raver119@gmail.com> * Fixed template math atomicMul for 64bit ints. * Refactored dynamic_partition_bp op. * inverseBroadcast fix Signed-off-by: raver119 <raver119@gmail.com> * DynamicPartitionBP test datatype fixed. * - nd4j_atomicMul Windows fix - cpu/NDArrayLambda.hpp excluded from CUDA Signed-off-by: raver119 <raver119@gmail.com>
2019-06-27 17:37:04 +02:00
TEST_F(DeclarableOpsTests15, test_lstmBlock_1) {
auto x0 = NDArrayFactory::create<Nd4jLong>(5);
auto x1 = NDArrayFactory::create<float>('c', {5, 1, 4}, {0.7787856f, 0.80119777f, 0.72437465f, 0.23089433f, 0.72714126f, 0.18039072f, 0.50563407f, 0.89252293f, 0.5461209f, 0.92336726f, 0.085571885f, 0.7937801f, 0.65908563f, 0.55552566f, 0.15962744f, 0.30874777f, 0.15476847f, 0.46954823f, 0.9938899f, 0.6112741f});
auto x2 = NDArrayFactory::create<float>('c', {1, 3}, {0.7717289f, 0.9280778f, 0.98455656f});
auto x3 = NDArrayFactory::create<float>('c', {1, 3}, {0.94414854f, 0.5956861f, 0.8668989f});
auto x4 = NDArrayFactory::create<float>('c', {7, 12}, {0.460692f, 0.042572856f, 0.08420354f, -0.09538093f, -0.11416581f, -0.53166187f, 0.40133476f, -0.24381405f, 0.30778718f, 0.52713746f, 0.16253126f, -0.034891903f, 0.011679292f, -0.19076681f, 0.14710993f, -0.3704369f, 0.51872355f, 0.13536876f, -0.5568739f, -0.08727971f, 0.07601875f, -0.074174374f, -0.5345982f, -0.3581748f, -0.28263924f, -0.25141674f, 0.43328637f, -0.50227314f, -0.26641843f, -0.38241976f, -0.19636461f, -0.04020852f, -0.27312332f, 0.5207915f, -0.37247592f, -0.4713087f, -0.25670746f, -0.14942765f, -0.015806139f, -0.22531253f, 0.5582536f, 0.3093416f, 0.3221351f, -0.0964683f, 0.14318448f, 0.42279094f, -0.46992f, -0.43399644f, -0.51704615f, -0.11854091f, 0.21697259f, -0.049382925f, 0.14059627f, 0.3912331f, -0.41345632f, 0.5067368f, -0.3420229f, 0.485789f, 0.044918716f, 0.26209074f, 0.12357575f, 0.21778125f, -0.53791714f, 0.18346387f, 0.054183125f, 0.5480431f, 0.03675288f, -0.26656917f, -0.018610716f, 0.19917983f, 0.5566165f, 0.43570566f, -0.35720813f, 0.31097364f, -0.47134516f, -0.289197f, 0.091138184f, 0.13300979f, -0.36592877f, -0.17540845f, 0.21732038f, 0.4393713f, 0.42800313f, 0.5006979f});
auto x5 = NDArrayFactory::create<float>('c', {1, 3});
auto x6 = NDArrayFactory::create<float>('c', {1, 3});
auto x7 = NDArrayFactory::create<float>('c', {1, 3});
auto x8 = NDArrayFactory::create<float>('c', {12});
nd4j::ops::lstmBlock op;
auto result = op.execute({&x0, &x1, &x2, &x3, &x4, &x5, &x6, &x7, &x8}, {2.0, 0.3}, {0, 0});
ASSERT_EQ(Status::OK(), result->status());
auto z = result->at(0);
[WIP] More of CUDA (#95) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Implementation of hashcode cuda helper. Working edition. * Fixed parallel test input arangements. * Fixed tests for hashcode op. * Fixed shape calculation for image:crop_and_resize op and test. * NativeOps tests. Initial test suite. * Added tests for indexReduce methods. * Added test on execBroadcast with NDArray as dimensions. * Added test on execBroadcastBool with NDArray as dimensions. * Added tests on execPairwiseTransform and execPairwiseTransofrmBool. * Added tests for execReduce with scalar results. * Added reduce tests for non-empty dims array. * Added tests for reduce3. * Added tests for execScalar. * Added tests for execSummaryStats. * - provide cpu/cuda code for batch_to_space - testing it Signed-off-by: Yurii <yurii@skymind.io> * - remove old test for batch_to_space (had wrong format and numbers were not checked) Signed-off-by: Yurii <yurii@skymind.io> * Fixed complilation errors with test. * Added test for execTransformFloat. * Added test for execTransformSame. * Added test for execTransformBool. * Added test for execTransformStrict. * Added tests for execScalar/execScalarBool with TADs. * Added test for flatten. * - provide cpu/cuda code for space_to_Batch operaion Signed-off-by: Yurii <yurii@skymind.io> * Added test for concat. * comment unnecessary stuff in s_t_b Signed-off-by: Yurii <yurii@skymind.io> * Added test for specialConcat. * Added tests for memcpy/set routines. * Fixed pullRow cuda test. * Added pullRow test. * Added average test. * - correct typo in NDArray::applyPairwiseTransform(nd4j::pairwise::BoolOps op...) Signed-off-by: Yurii <yurii@skymind.io> * - debugging and fixing cuda tests in JavaInteropTests file Signed-off-by: Yurii <yurii@skymind.io> * - correct some tests Signed-off-by: Yurii <yurii@skymind.io> * Added test for shuffle. * Fixed ops declarations. * Restored omp and added shuffle test. * Added convertTypes test. * Added tests for execRandom. Eliminated usage of RandomBuffer with NativeOps. * Added sort tests. * Added tests for execCustomOp. * - further debuging and fixing tests terminated with crash Signed-off-by: Yurii <yurii@skymind.io> * Added tests for calculateOutputShapes. * Addded Benchmarks test. * Commented benchmark tests. * change assertion Signed-off-by: raver119 <raver119@gmail.com> * Added tests for apply_sgd op. Added cpu helper for that op. * Implement cuda helper for aplly_sgd op. Fixed tests for NativeOps. * Added test for assign broadcastable. * Added tests for assign_bp op. * Added tests for axpy op. * - assign/execScalar/execTransformAny signature change - minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Fixed axpy op. * meh Signed-off-by: raver119 <raver119@gmail.com> * - fix tests for nativeOps::concat Signed-off-by: Yurii <yurii@skymind.io> * sequential transform/scalar Signed-off-by: raver119 <raver119@gmail.com> * allow nested parallelism Signed-off-by: raver119 <raver119@gmail.com> * assign_bp leak fix Signed-off-by: raver119 <raver119@gmail.com> * block setRNG fix Signed-off-by: raver119 <raver119@gmail.com> * enable parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * enable nested parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * Added cuda implementation for row_count helper. * Added implementation for tnse gains op helper. * - take into account possible situations when input arrays are empty in reduce_ cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implemented tsne/edge_forces op cuda-based helper. Parallelized cpu-based helper for edge_forces. * Added kernel for tsne/symmetrized op heleper. * Implementation of tsne/symmetrized op cuda helper. Working edition. * Eliminated waste printfs. * Added test for broadcastgradientargs op. * host-only fallback for empty reduce float Signed-off-by: raver119 <raver119@gmail.com> * - some tests fixes Signed-off-by: Yurii <yurii@skymind.io> * - correct the rest of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * - further correction of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * Added test for Cbow op. Also added cuda implementation for cbow helpers. * - improve code of stack operation for scalar case Signed-off-by: Yurii <yurii@skymind.io> * - provide cuda kernel for gatherND operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cbow helpers with cuda kernels. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * - further correction of cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implementatation of cbow op helper with cuda kernels. Working edition. * Skip random testing for cudablas case. * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for ELU and ELU_BP ops. * Added tests for eq_scalar, gt_scalar, gte_scalar and lte_scalar ops. * Added tests for neq_scalar. * Added test for noop. * - further work on clipbynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * - get rid of concat op call, use instead direct concat helper call Signed-off-by: Yurii <yurii@skymind.io> * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for lrelu and lrelu_bp. * Added tests for selu and selu_bp. * Fixed lrelu derivative helpers. * - some corrections in lstm Signed-off-by: Yurii <yurii@skymind.io> * operator * result shape fix Signed-off-by: raver119 <raver119@gmail.com> * - correct typo in lstmCell Signed-off-by: Yurii <yurii@skymind.io> * few tests fixed Signed-off-by: raver119 <raver119@gmail.com> * CUDA inverse broadcast bool fix Signed-off-by: raver119 <raver119@gmail.com> * disable MMAP test for CUDA Signed-off-by: raver119 <raver119@gmail.com> * BooleanOp syncToDevice Signed-off-by: raver119 <raver119@gmail.com> * meh Signed-off-by: raver119 <raver119@gmail.com> * additional data types for im2col/col2im Signed-off-by: raver119 <raver119@gmail.com> * Added test for firas_sparse op. * one more RandomBuffer test excluded Signed-off-by: raver119 <raver119@gmail.com> * Added tests for flatten op. * Added test for Floor op. * bunch of tests fixed Signed-off-by: raver119 <raver119@gmail.com> * mmulDot tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Implemented floordiv_bp op and tests. * Fixed scalar case with cuda implementation for bds. * - work on cuda kernel for clip_by_norm backprop op is completed Signed-off-by: Yurii <yurii@skymind.io> * Eliminate cbow crach. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Eliminated abortion with batched nlp test. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Fixed shared flag initializing. * disabled bunch of cpu workspaces tests Signed-off-by: raver119 <raver119@gmail.com> * scalar operators fix: missing registerSpecialUse call Signed-off-by: raver119 <raver119@gmail.com> * Fixed logdet for cuda and tests. * - correct clipBynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * Fixed crop_and_resize shape datatype. * - correct some mmul tests Signed-off-by: Yurii <yurii@skymind.io>
2019-08-02 19:01:03 +02:00
// z->printIndexedBuffer("Z");
Merge master to upstream (#7945) * Shugeo strided slice zeros (#14) * Modified strided_slice op to properly work with empty-like shapes. * Fixed test for reduce_mean with empty-like input. * [WIP] Last merge (#15) * 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 * [WIP] Fixing outstanding issues for NLP (#9) * Avoid using not-inited objects * Test fixed. * Redundant method avoided for models like FastText * KMeans++ implementation * KMeans++ implementation * Disable parallel execution * KMeans++ * Tests * Dev branch merge (#16) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Fix some issues on master (#17) * Fix DataVec test issue * Fix issue with dl4j SameDiff output layer * Dtype fix for lambda layers * #7912 BertIterator dtype fix (use float32 not global default) * [WIP] Next set of CUDA stuff (#7) New CUDA implementations and improvements * bad file * Dev branch master merge (#23) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Compatibility of deserialization (#18) Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com> * SameDiff: add activation gradient checking support for debugging (#19) * SameDiff gradient checker: first pass on activation gradient checks * Fixes + tests for activation gradient checking * Javadoc * [WIP] Some nd4j data type corrections (#20) * Adjust data type * Set correct Data type. * Size of proper data type. * fix averaged cpu load (#22) * SameDiff ops, TF import and fixes (#24) * CheckNumerics tests + fixes + misc fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fake quant Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fixes Signed-off-by: AlexDBlack <blacka101@gmail.com> * FakeQuantWithMinMaxArgs Signed-off-by: AlexDBlack <blacka101@gmail.com> * CheckNumerics fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix libnd4j ALL_INTS and ALL_FLOATS declaration (uint and bfloat types) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Small fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Javadoc Signed-off-by: AlexDBlack <blacka101@gmail.com> * Exception tweak Signed-off-by: AlexDBlack <blacka101@gmail.com> * fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix for out of scope stack allocated var use Signed-off-by: AlexDBlack <blacka101@gmail.com> * Ignores Signed-off-by: AlexDBlack <blacka101@gmail.com> * Ignore for known failing test (already logged issue) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Merge upstream to fork (#25) * Add thousand-separator commas to TotalParams (#7915) * Add thousand-separator commas to TotalParams The number of parameters can be quite large, and it would help the reading of the summary printout to have the TotalParams column & values at the bottom have thousand-separator-commas in them. * Add thousand-separator commas to MultiLayerNetwork Corresponding change to MultiLayerNetwork Signed-off-by: Jxtps Jxtps <jxtps435@gmail.com> * Update contributing and issue/PR templates (#7934) Signed-off-by: AlexDBlack <blacka101@gmail.com> * Fix link to AdaDelta paper (#7942) Fix link to AdaDelta paper hosted on matthewzeiler.com Signed-off-by: Jxtps * Fixes, and ignores for known/logged failing issues (#7943) Signed-off-by: AlexDBlack <blacka101@gmail.com> * SameDiff + DL4J/SameDiff: Multiple fixes (#28) * #7919 HDF5 attribute buffer length fix Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7909 Arbiter constructor exception ux improvements Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7925 RNN output layer length checks Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7939 Add listener for validating inputs are not incorrectly modified Signed-off-by: AlexDBlack <blacka101@gmail.com> * #7939 Integrate NonInplaceValidationListener into tests * #7844 DL4J SameDiff fixes for variable minibatch size * DL4J SameDiff fixes - ensure gradient for input placeholder is available Signed-off-by: AlexDBlack <blacka101@gmail.com> * Tweaks to ExternalErrorsFunction - use placeholders, make more robust * Another fix * More fixes * More SameDiff/DL4J fixes * Scope out scalar array creation in BaseScalarOp * Remove debug code Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] Final dev branch merge (#29) * SameDiff: convertDataType and gradient check util improvements (#12) * GradCheck util improvements * StopGradient constructor + test * SameDiff: Add datatype conversion * Javadoc and add DataType.isNumerical() * Small fix * Fix SameDiff TF import test cases intermediate naming (workaround for bad default) * TFGraphTestAllHelper: check intermediates in execution order * Add missing debug listener * [WIP] lstmBlock fix + other changes (#13) - fixes lstmBlock issue - changes NDArray method reshape(), permute(), transpose() by making them return instance instead of pointer - CheckNumerics op - fixes for ReduceBool IsInfOrNan & IsFinite * Small test fix * CheckNumerics op wrapper * Compatibility of deserialization (#18) Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com> * SameDiff: add activation gradient checking support for debugging (#19) * SameDiff gradient checker: first pass on activation gradient checks * Fixes + tests for activation gradient checking * Javadoc * [WIP] Some nd4j data type corrections (#20) * Adjust data type * Set correct Data type. * Size of proper data type. * fix averaged cpu load (#22) * [WIP] Multiple dataset iterators (#27) * Splitting dataset into arbitrary number * Fixes * Multiple split of iterator * Test * Test * Some fixes * signature change * one more tweak Signed-off-by: raver119 <raver119@gmail.com> * one more test for sequential use of DataSetIteratorSplitter Signed-off-by: raver119 <raver119@gmail.com> * Fixes * Fixes * one more test for Alexander Signed-off-by: raver119 <raver119@gmail.com> * Some fixes * Some fixes * one more test for Alexander Signed-off-by: raver119 <raver119@gmail.com> * minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Some fixes * Some fixes * couple of assertions tweaked Signed-off-by: raver119 <raver119@gmail.com> * MDS splitter test :/ Signed-off-by: raver119 <raver119@gmail.com> * Minor refactoring * Multi dataset * Some fixes * More tests * Small number of test fixes/improvements (failures on CI) (#31) Signed-off-by: AlexDBlack <blacka101@gmail.com> * [WIP] More CUDA stuff (#26) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * LRN BP CUDA Signed-off-by: raver119 <raver119@gmail.com> * less memory Signed-off-by: raver119 <raver119@gmail.com> * Fixed bug with crop_and_resize op helper. * get rid of unnecessary index-calculation dunction Signed-off-by: Yurii <yurii@skymind.io> * Fixed sort with nth_element cuda-based helper. * Refactored nth_element. * Refactored nth_element op and tests. * Modified usage of dim array with sortTad routine. * Refactored main routine of helper for non_max_image_suppression op. * non_max_image_suppression op helper with cuda kernel implementation. Initial revision. * fix vol2col cuda kernel * meh Signed-off-by: raver119 <raver119@gmail.com> * topK concept Signed-off-by: raver119 <raver119@gmail.com> * unsorted topK with scanWitdh of 1 Signed-off-by: raver119 <raver119@gmail.com> * correct vol2col tests * sorted/unsorted topK Signed-off-by: raver119 <raver119@gmail.com> * implementation and fixing col2im/col2vol * Corrected usage flags with input/output with reverse op. * dup is const now Signed-off-by: raver119 <raver119@gmail.com> * percentile op Signed-off-by: raver119 <raver119@gmail.com> * group tests for mapool2d Signed-off-by: Yurii <yurii@skymind.io> * special test for george Signed-off-by: raver119 <raver119@gmail.com> * less threads for sortTad Signed-off-by: raver119 <raver119@gmail.com> * provide conv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * remove auther in sort tad kernel code Signed-off-by: Yurii <yurii@skymind.io> * provide depthwise_conv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * - max_pooling_with_argmax - null check for special use Signed-off-by: raver119 <raver119@gmail.com> * dts cuda Signed-off-by: raver119 <raver119@gmail.com> * provide sconv2d for cuda Signed-off-by: Yurii <yurii@skymind.io> * std cuda Signed-off-by: raver119 <raver119@gmail.com> * Refactored non_max_suppression op to conform TF implementation. * Improved suppression helper. * provide pooling3d for cuda Signed-off-by: Yurii <yurii@skymind.io> * minor lstm rearrangements Signed-off-by: raver119 <raver119@gmail.com> * more of minor lstm rearrangements Signed-off-by: raver119 <raver119@gmail.com> * (bi)dynamic_rnn Signed-off-by: raver119 <raver119@gmail.com> * templates init order Signed-off-by: raver119 <raver119@gmail.com> * Refactored non_max_suppression op. * Added cuda kernel for non_max_suppression. * CPU sort by key/value Signed-off-by: raver119 <raver119@gmail.com> * CPU sort TAD by key/value Signed-off-by: raver119 <raver119@gmail.com> * CPU sort TAD by key/value tests Signed-off-by: raver119 <raver119@gmail.com> * Eliminate compiler error with cuda implementation. * - repaired gradCheck in cuda - provide conv2d_bp for cuda Signed-off-by: Yurii <yurii@skymind.io> * missed signature Signed-off-by: raver119 <raver119@gmail.com> * provide depthwise_conv2d_bp for cuda Signed-off-by: Yurii <yurii@skymind.io> * Implementation of lup helper with cuda kernel. Initial commit. * further work on backprops for convolutions Signed-off-by: Yurii <yurii@skymind.io> * CUDA linear sort by key/val Signed-off-by: raver119 <raver119@gmail.com> * CUDA tad sort by key/val Signed-off-by: raver119 <raver119@gmail.com> * start providing of backprop for pooling2d/3d Signed-off-by: Yurii <yurii@skymind.io> * Added atomicAdd for bool datatype. * dynamic partition concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic partition concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic partition scalar CUDA Signed-off-by: raver119 <raver119@gmail.com> * important comment Signed-off-by: raver119 <raver119@gmail.com> * fix pooling2d/3d backprop helpers Signed-off-by: Yurii <yurii@skymind.io> * Added non-linear test with dynamic_partition. * Improved test for dynamic_partition. * dynamic_partition TAD concept Signed-off-by: raver119 <raver119@gmail.com> * - dynamic_partition TAD CUDA impl - dynamic_partition TAD CPU fix Signed-off-by: raver119 <raver119@gmail.com> * - rewrite cpu code for usampling2d/3d - write cuda code for usampling2d/3d Signed-off-by: Yurii <yurii@skymind.io> * dynamic_stitch CUDA vector case Signed-off-by: raver119 <raver119@gmail.com> * dynamic_stitch CUDA TAD case concept Signed-off-by: raver119 <raver119@gmail.com> * dynamic_stitch CUDA TAD case impl Signed-off-by: raver119 <raver119@gmail.com> * Added tests for dynamic_stitch 3D-4D cases. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * Fixed type check for dynamic stitch. * min/max bp Signed-off-by: raver119 <raver119@gmail.com> * rewrite code for upsampling2d/3d cpu Signed-off-by: Yurii <yurii@skymind.io> * reduce min/max/norm_max bp Signed-off-by: raver119 <raver119@gmail.com> * lup implementation. Additional enhancements. * provide code for upsamling2d/3d backprop Signed-off-by: Yurii <yurii@skymind.io> * weightedCrossEntropyWithLogits Signed-off-by: raver119 <raver119@gmail.com> * Fixed template math atomicMul for 64bit ints. * Refactored dynamic_partition_bp op. * inverseBroadcast fix Signed-off-by: raver119 <raver119@gmail.com> * DynamicPartitionBP test datatype fixed. * - nd4j_atomicMul Windows fix - cpu/NDArrayLambda.hpp excluded from CUDA Signed-off-by: raver119 <raver119@gmail.com>
2019-06-27 17:37:04 +02:00
delete result;
}
TEST_F(DeclarableOpsTests15, test_lstmBlock_2) {
int seqLen = 8;
int bS = 16;
int nIn = 8;
auto x0 = NDArrayFactory::create<Nd4jLong>(5);
[WIP] More of CUDA (#95) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Implementation of hashcode cuda helper. Working edition. * Fixed parallel test input arangements. * Fixed tests for hashcode op. * Fixed shape calculation for image:crop_and_resize op and test. * NativeOps tests. Initial test suite. * Added tests for indexReduce methods. * Added test on execBroadcast with NDArray as dimensions. * Added test on execBroadcastBool with NDArray as dimensions. * Added tests on execPairwiseTransform and execPairwiseTransofrmBool. * Added tests for execReduce with scalar results. * Added reduce tests for non-empty dims array. * Added tests for reduce3. * Added tests for execScalar. * Added tests for execSummaryStats. * - provide cpu/cuda code for batch_to_space - testing it Signed-off-by: Yurii <yurii@skymind.io> * - remove old test for batch_to_space (had wrong format and numbers were not checked) Signed-off-by: Yurii <yurii@skymind.io> * Fixed complilation errors with test. * Added test for execTransformFloat. * Added test for execTransformSame. * Added test for execTransformBool. * Added test for execTransformStrict. * Added tests for execScalar/execScalarBool with TADs. * Added test for flatten. * - provide cpu/cuda code for space_to_Batch operaion Signed-off-by: Yurii <yurii@skymind.io> * Added test for concat. * comment unnecessary stuff in s_t_b Signed-off-by: Yurii <yurii@skymind.io> * Added test for specialConcat. * Added tests for memcpy/set routines. * Fixed pullRow cuda test. * Added pullRow test. * Added average test. * - correct typo in NDArray::applyPairwiseTransform(nd4j::pairwise::BoolOps op...) Signed-off-by: Yurii <yurii@skymind.io> * - debugging and fixing cuda tests in JavaInteropTests file Signed-off-by: Yurii <yurii@skymind.io> * - correct some tests Signed-off-by: Yurii <yurii@skymind.io> * Added test for shuffle. * Fixed ops declarations. * Restored omp and added shuffle test. * Added convertTypes test. * Added tests for execRandom. Eliminated usage of RandomBuffer with NativeOps. * Added sort tests. * Added tests for execCustomOp. * - further debuging and fixing tests terminated with crash Signed-off-by: Yurii <yurii@skymind.io> * Added tests for calculateOutputShapes. * Addded Benchmarks test. * Commented benchmark tests. * change assertion Signed-off-by: raver119 <raver119@gmail.com> * Added tests for apply_sgd op. Added cpu helper for that op. * Implement cuda helper for aplly_sgd op. Fixed tests for NativeOps. * Added test for assign broadcastable. * Added tests for assign_bp op. * Added tests for axpy op. * - assign/execScalar/execTransformAny signature change - minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Fixed axpy op. * meh Signed-off-by: raver119 <raver119@gmail.com> * - fix tests for nativeOps::concat Signed-off-by: Yurii <yurii@skymind.io> * sequential transform/scalar Signed-off-by: raver119 <raver119@gmail.com> * allow nested parallelism Signed-off-by: raver119 <raver119@gmail.com> * assign_bp leak fix Signed-off-by: raver119 <raver119@gmail.com> * block setRNG fix Signed-off-by: raver119 <raver119@gmail.com> * enable parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * enable nested parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * Added cuda implementation for row_count helper. * Added implementation for tnse gains op helper. * - take into account possible situations when input arrays are empty in reduce_ cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implemented tsne/edge_forces op cuda-based helper. Parallelized cpu-based helper for edge_forces. * Added kernel for tsne/symmetrized op heleper. * Implementation of tsne/symmetrized op cuda helper. Working edition. * Eliminated waste printfs. * Added test for broadcastgradientargs op. * host-only fallback for empty reduce float Signed-off-by: raver119 <raver119@gmail.com> * - some tests fixes Signed-off-by: Yurii <yurii@skymind.io> * - correct the rest of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * - further correction of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * Added test for Cbow op. Also added cuda implementation for cbow helpers. * - improve code of stack operation for scalar case Signed-off-by: Yurii <yurii@skymind.io> * - provide cuda kernel for gatherND operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cbow helpers with cuda kernels. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * - further correction of cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implementatation of cbow op helper with cuda kernels. Working edition. * Skip random testing for cudablas case. * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for ELU and ELU_BP ops. * Added tests for eq_scalar, gt_scalar, gte_scalar and lte_scalar ops. * Added tests for neq_scalar. * Added test for noop. * - further work on clipbynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * - get rid of concat op call, use instead direct concat helper call Signed-off-by: Yurii <yurii@skymind.io> * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for lrelu and lrelu_bp. * Added tests for selu and selu_bp. * Fixed lrelu derivative helpers. * - some corrections in lstm Signed-off-by: Yurii <yurii@skymind.io> * operator * result shape fix Signed-off-by: raver119 <raver119@gmail.com> * - correct typo in lstmCell Signed-off-by: Yurii <yurii@skymind.io> * few tests fixed Signed-off-by: raver119 <raver119@gmail.com> * CUDA inverse broadcast bool fix Signed-off-by: raver119 <raver119@gmail.com> * disable MMAP test for CUDA Signed-off-by: raver119 <raver119@gmail.com> * BooleanOp syncToDevice Signed-off-by: raver119 <raver119@gmail.com> * meh Signed-off-by: raver119 <raver119@gmail.com> * additional data types for im2col/col2im Signed-off-by: raver119 <raver119@gmail.com> * Added test for firas_sparse op. * one more RandomBuffer test excluded Signed-off-by: raver119 <raver119@gmail.com> * Added tests for flatten op. * Added test for Floor op. * bunch of tests fixed Signed-off-by: raver119 <raver119@gmail.com> * mmulDot tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Implemented floordiv_bp op and tests. * Fixed scalar case with cuda implementation for bds. * - work on cuda kernel for clip_by_norm backprop op is completed Signed-off-by: Yurii <yurii@skymind.io> * Eliminate cbow crach. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Eliminated abortion with batched nlp test. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Fixed shared flag initializing. * disabled bunch of cpu workspaces tests Signed-off-by: raver119 <raver119@gmail.com> * scalar operators fix: missing registerSpecialUse call Signed-off-by: raver119 <raver119@gmail.com> * Fixed logdet for cuda and tests. * - correct clipBynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * Fixed crop_and_resize shape datatype. * - correct some mmul tests Signed-off-by: Yurii <yurii@skymind.io>
2019-08-02 19:01:03 +02:00
auto x1 = NDArrayFactory::create<float>('f', {bS, nIn, seqLen});
auto x2 = NDArrayFactory::create<float>('f', {bS, nIn}); // nIn == nOut
auto x3 = NDArrayFactory::create<float>('f', {bS, nIn});
auto x4 = NDArrayFactory::create<float>('f', {2 * nIn, 4 * nIn});
auto x5 = NDArrayFactory::create<float>('f', {nIn});
auto x6 = NDArrayFactory::create<float>('f', {nIn});
auto x7 = NDArrayFactory::create<float>('f', {nIn});
auto x8 = NDArrayFactory::create<float>('f', {4 * nIn});
nd4j::ops::lstmBlock op;
auto result = op.execute({&x0, &x1, &x2, &x3, &x4, &x5, &x6, &x7, &x8}, {1.0, 0.0}, {0, 1});
ASSERT_EQ(Status::OK(), result->status());
auto z = result->at(0);
delete result;
}
[WIP] More of CUDA (#95) * initial commit Signed-off-by: raver119 <raver119@gmail.com> * Implementation of hashcode cuda helper. Working edition. * Fixed parallel test input arangements. * Fixed tests for hashcode op. * Fixed shape calculation for image:crop_and_resize op and test. * NativeOps tests. Initial test suite. * Added tests for indexReduce methods. * Added test on execBroadcast with NDArray as dimensions. * Added test on execBroadcastBool with NDArray as dimensions. * Added tests on execPairwiseTransform and execPairwiseTransofrmBool. * Added tests for execReduce with scalar results. * Added reduce tests for non-empty dims array. * Added tests for reduce3. * Added tests for execScalar. * Added tests for execSummaryStats. * - provide cpu/cuda code for batch_to_space - testing it Signed-off-by: Yurii <yurii@skymind.io> * - remove old test for batch_to_space (had wrong format and numbers were not checked) Signed-off-by: Yurii <yurii@skymind.io> * Fixed complilation errors with test. * Added test for execTransformFloat. * Added test for execTransformSame. * Added test for execTransformBool. * Added test for execTransformStrict. * Added tests for execScalar/execScalarBool with TADs. * Added test for flatten. * - provide cpu/cuda code for space_to_Batch operaion Signed-off-by: Yurii <yurii@skymind.io> * Added test for concat. * comment unnecessary stuff in s_t_b Signed-off-by: Yurii <yurii@skymind.io> * Added test for specialConcat. * Added tests for memcpy/set routines. * Fixed pullRow cuda test. * Added pullRow test. * Added average test. * - correct typo in NDArray::applyPairwiseTransform(nd4j::pairwise::BoolOps op...) Signed-off-by: Yurii <yurii@skymind.io> * - debugging and fixing cuda tests in JavaInteropTests file Signed-off-by: Yurii <yurii@skymind.io> * - correct some tests Signed-off-by: Yurii <yurii@skymind.io> * Added test for shuffle. * Fixed ops declarations. * Restored omp and added shuffle test. * Added convertTypes test. * Added tests for execRandom. Eliminated usage of RandomBuffer with NativeOps. * Added sort tests. * Added tests for execCustomOp. * - further debuging and fixing tests terminated with crash Signed-off-by: Yurii <yurii@skymind.io> * Added tests for calculateOutputShapes. * Addded Benchmarks test. * Commented benchmark tests. * change assertion Signed-off-by: raver119 <raver119@gmail.com> * Added tests for apply_sgd op. Added cpu helper for that op. * Implement cuda helper for aplly_sgd op. Fixed tests for NativeOps. * Added test for assign broadcastable. * Added tests for assign_bp op. * Added tests for axpy op. * - assign/execScalar/execTransformAny signature change - minor test fix Signed-off-by: raver119 <raver119@gmail.com> * Fixed axpy op. * meh Signed-off-by: raver119 <raver119@gmail.com> * - fix tests for nativeOps::concat Signed-off-by: Yurii <yurii@skymind.io> * sequential transform/scalar Signed-off-by: raver119 <raver119@gmail.com> * allow nested parallelism Signed-off-by: raver119 <raver119@gmail.com> * assign_bp leak fix Signed-off-by: raver119 <raver119@gmail.com> * block setRNG fix Signed-off-by: raver119 <raver119@gmail.com> * enable parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * enable nested parallelism by default Signed-off-by: raver119 <raver119@gmail.com> * Added cuda implementation for row_count helper. * Added implementation for tnse gains op helper. * - take into account possible situations when input arrays are empty in reduce_ cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implemented tsne/edge_forces op cuda-based helper. Parallelized cpu-based helper for edge_forces. * Added kernel for tsne/symmetrized op heleper. * Implementation of tsne/symmetrized op cuda helper. Working edition. * Eliminated waste printfs. * Added test for broadcastgradientargs op. * host-only fallback for empty reduce float Signed-off-by: raver119 <raver119@gmail.com> * - some tests fixes Signed-off-by: Yurii <yurii@skymind.io> * - correct the rest of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * - further correction of reduce_ stuff Signed-off-by: Yurii <yurii@skymind.io> * Added test for Cbow op. Also added cuda implementation for cbow helpers. * - improve code of stack operation for scalar case Signed-off-by: Yurii <yurii@skymind.io> * - provide cuda kernel for gatherND operation Signed-off-by: Yurii <yurii@skymind.io> * Implementation of cbow helpers with cuda kernels. * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * minor tests tweaks Signed-off-by: raver119 <raver119@gmail.com> * - further correction of cuda stuff Signed-off-by: Yurii <yurii@skymind.io> * Implementatation of cbow op helper with cuda kernels. Working edition. * Skip random testing for cudablas case. * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for ELU and ELU_BP ops. * Added tests for eq_scalar, gt_scalar, gte_scalar and lte_scalar ops. * Added tests for neq_scalar. * Added test for noop. * - further work on clipbynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * - get rid of concat op call, use instead direct concat helper call Signed-off-by: Yurii <yurii@skymind.io> * lstmBlockCell context fix Signed-off-by: raver119 <raver119@gmail.com> * Added tests for lrelu and lrelu_bp. * Added tests for selu and selu_bp. * Fixed lrelu derivative helpers. * - some corrections in lstm Signed-off-by: Yurii <yurii@skymind.io> * operator * result shape fix Signed-off-by: raver119 <raver119@gmail.com> * - correct typo in lstmCell Signed-off-by: Yurii <yurii@skymind.io> * few tests fixed Signed-off-by: raver119 <raver119@gmail.com> * CUDA inverse broadcast bool fix Signed-off-by: raver119 <raver119@gmail.com> * disable MMAP test for CUDA Signed-off-by: raver119 <raver119@gmail.com> * BooleanOp syncToDevice Signed-off-by: raver119 <raver119@gmail.com> * meh Signed-off-by: raver119 <raver119@gmail.com> * additional data types for im2col/col2im Signed-off-by: raver119 <raver119@gmail.com> * Added test for firas_sparse op. * one more RandomBuffer test excluded Signed-off-by: raver119 <raver119@gmail.com> * Added tests for flatten op. * Added test for Floor op. * bunch of tests fixed Signed-off-by: raver119 <raver119@gmail.com> * mmulDot tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Implemented floordiv_bp op and tests. * Fixed scalar case with cuda implementation for bds. * - work on cuda kernel for clip_by_norm backprop op is completed Signed-off-by: Yurii <yurii@skymind.io> * Eliminate cbow crach. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Eliminated abortion with batched nlp test. * more tests fixed Signed-off-by: raver119 <raver119@gmail.com> * Fixed shared flag initializing. * disabled bunch of cpu workspaces tests Signed-off-by: raver119 <raver119@gmail.com> * scalar operators fix: missing registerSpecialUse call Signed-off-by: raver119 <raver119@gmail.com> * Fixed logdet for cuda and tests. * - correct clipBynorm_bp Signed-off-by: Yurii <yurii@skymind.io> * Fixed crop_and_resize shape datatype. * - correct some mmul tests Signed-off-by: Yurii <yurii@skymind.io>
2019-08-02 19:01:03 +02:00
TEST_F(DeclarableOpsTests15, test_lstmBlock_3) {
int seqLen = 3;
int bS = 2;
int nIn = 4;
NDArray f('f', {bS, nIn, seqLen}, nd4j::DataType::FLOAT32);
NDArray cLast('f', {bS, nIn}, nd4j::DataType::FLOAT32);
f = 2;
cLast = 3;
for (int t = 0; t < seqLen; ++t) {
//section 1
//auto ft = f({0,0, 0,0, t,t+1});
//auto temp = ft * cLast;
// section 2
auto ft = f({0,0, 0,0, t,t+1});
auto temp1 = ft.reshape('f', {bS, nIn});
auto temp2 = temp1 * cLast;
}
}
TEST_F(DeclarableOpsTests15, test_empty_increasing_1) {
auto x = NDArrayFactory::create<float>('c', {1, 0, 3});
auto z = NDArrayFactory::create<bool>(false);
Context ctx(1);
ctx.setInputArray(0, &x);
ctx.setOutputArray(0, &z);
nd4j::ops::is_strictly_increasing op;
auto status = op.execute(&ctx);
ASSERT_EQ(Status::OK(), status);
ASSERT_EQ(true, z.e<bool>(0));
}
TEST_F(DeclarableOpsTests15, test_empty_decreasing_1) {
auto x = NDArrayFactory::create<float>('c', {1, 0, 3});
auto z = NDArrayFactory::create<bool>(false);
Context ctx(1);
ctx.setInputArray(0, &x);
ctx.setOutputArray(0, &z);
nd4j::ops::is_non_decreasing op;
auto status = op.execute(&ctx);
ASSERT_EQ(Status::OK(), status);
ASSERT_EQ(true, z.e<bool>(0));
}
Oleh rgb to gray scale (#138) * libnd4j: RgbToGrayscale op #8536 - raw implementation in user branch, need checks for integration and adding other orders Signed-off-by: Oleg <oleg.semeniv@gmail.com> * libnd4j: RgbToGrayscale op #8536 next step of merging images Signed-off-by: Oleg <oleg.semeniv@gmail.com> * libnd4j: RgbToGrayscale op #8536, Revert merge of hsv_to_rgb and rgb_to_hsv as cause conflicts in naming need refactoring before merge, implementation of rbg_to_grs added * libnd4j: RgbToGrayscale op #8536 imlementation and conflict resolve * libnd4j: RgbToGrayscale op #8536 merged operations with images into image, renamed methods and files * libnd4j: RgbToGrayscale op #8536 added test for rgbToGrayScale, need clarification and fixed tests case run Signed-off-by: Oleg <oleg.semeniv@gmail.com> * libnd4j: RgbToGrayscale op #8536 bug fixing and need review * libnd4j: RgbToGrayscale op #8536 some additional corrections after review Signed-off-by: Oleg <oleg.semeniv@gmail.com> * - minor corrections in rgbToGrs test1 Signed-off-by: Yurii <iuriish@yahoo.com> * libnd4j: RgbToGrayscale op #8536, corrected tests and rbf_to_grs, fixed problems, refactoring, need review * libnd4j: RgbToGrayscale op #8536 fix for 'f' order in rgbToGrs * libnd4j: RgbToGrayscale op #8536 fixed several bugs with dimC, test case refactoring and improve Signed-off-by: Oleg <oleg.semeniv@gmail.com> * - add cuda kernel for rgbToGrs op Signed-off-by: Yurii <iuriish@yahoo.com> * - fix linkage errors Signed-off-by: Yurii <iuriish@yahoo.com> Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
2019-12-20 18:59:29 +01:00
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_1) {
// rank 1
NDArray rgbs('c', { 3 }, { 10, 50, 200 }, nd4j::DataType::INT32);
NDArray expected('c', { 1 }, { 55 }, nd4j::DataType::INT32);
nd4j::ops::rgb_to_grs op;
auto result = op.execute({&rgbs}, {}, {});
auto output = result->at(0);
ASSERT_EQ(Status::OK(), result->status());
ASSERT_TRUE(expected.isSameShape(output));
ASSERT_TRUE(expected.equalsTo(output));
delete result;
}
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_2) {
// rank 1
auto rgbs = NDArrayFactory::create<int>('f', { 3 }, { 1, 120, -25 });
auto expected = NDArrayFactory::create<int>('f', { 1 }, { 67 });
nd4j::ops::rgb_to_grs op;
auto result = op.execute({ &rgbs }, {}, {});
auto output = result->at(0);
ASSERT_EQ(Status::OK(), result->status());
ASSERT_TRUE(expected.isSameShape(output));
ASSERT_TRUE(expected.equalsTo(output));
delete result;
}
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_3) {
// rank 2
NDArray rgbs('c', { 4, 3 }, { -94, 99, 97, 90, 114, 101, 111, 96, 105, 100, 103, 102 }, nd4j::DataType::INT32);
NDArray expected('c', { 4, 1 }, { 41, 105, 101, 101 }, nd4j::DataType::INT32);
nd4j::ops::rgb_to_grs op;
auto result = op.execute({ &rgbs }, {}, {});
auto output = result->at(0);
ASSERT_EQ(Status::OK(), result->status());
ASSERT_TRUE(expected.isSameShape(output));
ASSERT_TRUE(expected.equalsTo(output));
delete result;
}
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_4) {
NDArray rgbs('c', { 3, 2 }, {14, 99, 207, 10, 114, 201 }, nd4j::DataType::INT32);
rgbs.permutei({1,0});
NDArray expected('c', { 2, 1 }, { 138, 58 }, nd4j::DataType::INT32);
nd4j::ops::rgb_to_grs op;
auto result = op.execute({ &rgbs }, {}, {});
auto output = result->at(0);
ASSERT_EQ(Status::OK(), result->status());
ASSERT_TRUE(expected.isSameShape(output));
ASSERT_TRUE(expected.equalsTo(output));
delete result;
}
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_5) {
// rank 2
NDArray rgbs('c', { 3, 4 }, { -94, 99, 97, 90, 114, 101, 111, 96, 105, 100, 103, 102 }, nd4j::DataType::INT32);
NDArray expected('c', { 1, 4 }, { 50, 100, 105, 94 }, nd4j::DataType::INT32);
nd4j::ops::rgb_to_grs op;
auto result = op.execute({ &rgbs }, {}, {0});
auto output = result->at(0);
ASSERT_EQ(Status::OK(), result->status());
ASSERT_TRUE(expected.isSameShape(output));
ASSERT_TRUE(expected.equalsTo(output));
delete result;
}
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_6) {
// rank 3
auto rgbs = NDArrayFactory::create<float>('c', { 5,4,3 }, {1.7750e+01f, -7.1062e+01f, -1.0019e+02f,-2.3406e+01f, 5.2094e+01f, 9.5438e+01f, -6.7461e+00f, 3.8562e+01f, 6.5078e+00f,3.3562e+01f, -5.8844e+01f, 2.2750e+01f, -1.0477e+01f, 7.7344e+00f, 9.5469e+00f,2.1391e+01f, -8.5312e+01f, 7.5830e-01f,2.3125e+01f, 1.8145e+00f, 1.4602e+01f,-4.5859e+00f, 3.9344e+01f, 1.1617e+01f,-8.6562e+01f, 1.0038e+02f, 6.7938e+01f,5.9961e+00f, 6.7812e+01f, 2.9734e+01f,2.9609e+01f, -6.1438e+01f, 1.7750e+01f,6.8562e+01f, -7.4414e+00f, 3.9656e+01f,1.1641e+01f, -2.7516e+01f, 6.7562e+01f,7.8438e+01f, 5.4883e+00f, 2.9438e+01f,-3.1344e+01f, 6.5125e+01f, 1.2695e+01f,4.0531e+01f, -6.1211e+00f, 6.2219e+01f,4.6812e+01f, 5.2250e+01f, -1.1414e+01f,1.5404e-02f, 2.9938e+01f, 5.6719e+00f,-2.0125e+01f, 2.1531e+01f, 6.2500e+01f,7.2188e+01f, 9.3750e+00f, -4.8125e+01f});
auto expected = NDArrayFactory::create<float>('c', { 5,4,1 }, {-47.82958221f, 34.46305847f, 21.36137581f, -21.91625023f,2.49686432f, -43.59792709f, 9.64180183f, 23.04854202f,40.7946167f, 44.98754883f, -25.19047546f, 20.64586449f,-4.97033119f, 30.0226841f, 30.30688286f, 15.61459541f,43.36166f, 18.22480774f, 13.74833488f, 21.59387016f});
nd4j::ops::rgb_to_grs op;
auto result = op.execute({ &rgbs }, {}, {});
auto output = result->at(0);
ASSERT_EQ(Status::OK(), result->status());
ASSERT_TRUE(expected.isSameShape(output));
ASSERT_TRUE(expected.equalsTo(output));
delete result;
}
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_7) {
// rank 3
auto rgbs = NDArrayFactory::create<float>('c', { 5,3,4 }, { 1.7750e+01f, -7.1062e+01f, -1.0019e+02f,-2.3406e+01f, 5.2094e+01f, 9.5438e+01f, -6.7461e+00f, 3.8562e+01f, 6.5078e+00f,3.3562e+01f, -5.8844e+01f, 2.2750e+01f, -1.0477e+01f, 7.7344e+00f, 9.5469e+00f,2.1391e+01f, -8.5312e+01f, 7.5830e-01f,2.3125e+01f, 1.8145e+00f, 1.4602e+01f,-4.5859e+00f, 3.9344e+01f, 1.1617e+01f,-8.6562e+01f, 1.0038e+02f, 6.7938e+01f,5.9961e+00f, 6.7812e+01f, 2.9734e+01f,2.9609e+01f, -6.1438e+01f, 1.7750e+01f,6.8562e+01f, -7.4414e+00f, 3.9656e+01f,1.1641e+01f, -2.7516e+01f, 6.7562e+01f,7.8438e+01f, 5.4883e+00f, 2.9438e+01f,-3.1344e+01f, 6.5125e+01f, 1.2695e+01f,4.0531e+01f, -6.1211e+00f, 6.2219e+01f,4.6812e+01f, 5.2250e+01f, -1.1414e+01f,1.5404e-02f, 2.9938e+01f, 5.6719e+00f,-2.0125e+01f, 2.1531e+01f, 6.2500e+01f,7.2188e+01f, 9.3750e+00f, -4.8125e+01f});
auto expected = NDArrayFactory::create<float>('c', { 5,1,4 }, { 36.626545f, 38.607746f, -40.614971f, 18.233341f, -51.545094f,2.234142f, 20.913160f, 8.783220f, 15.955761f, 55.273506f, 36.838833f, -29.751089f, 8.148357f, 13.676106f, 1.097548f, 68.766457f, 38.690712f, 27.176361f, -14.156269f, 7.157052f });
Oleh rgb to gray scale (#138) * libnd4j: RgbToGrayscale op #8536 - raw implementation in user branch, need checks for integration and adding other orders Signed-off-by: Oleg <oleg.semeniv@gmail.com> * libnd4j: RgbToGrayscale op #8536 next step of merging images Signed-off-by: Oleg <oleg.semeniv@gmail.com> * libnd4j: RgbToGrayscale op #8536, Revert merge of hsv_to_rgb and rgb_to_hsv as cause conflicts in naming need refactoring before merge, implementation of rbg_to_grs added * libnd4j: RgbToGrayscale op #8536 imlementation and conflict resolve * libnd4j: RgbToGrayscale op #8536 merged operations with images into image, renamed methods and files * libnd4j: RgbToGrayscale op #8536 added test for rgbToGrayScale, need clarification and fixed tests case run Signed-off-by: Oleg <oleg.semeniv@gmail.com> * libnd4j: RgbToGrayscale op #8536 bug fixing and need review * libnd4j: RgbToGrayscale op #8536 some additional corrections after review Signed-off-by: Oleg <oleg.semeniv@gmail.com> * - minor corrections in rgbToGrs test1 Signed-off-by: Yurii <iuriish@yahoo.com> * libnd4j: RgbToGrayscale op #8536, corrected tests and rbf_to_grs, fixed problems, refactoring, need review * libnd4j: RgbToGrayscale op #8536 fix for 'f' order in rgbToGrs * libnd4j: RgbToGrayscale op #8536 fixed several bugs with dimC, test case refactoring and improve Signed-off-by: Oleg <oleg.semeniv@gmail.com> * - add cuda kernel for rgbToGrs op Signed-off-by: Yurii <iuriish@yahoo.com> * - fix linkage errors Signed-off-by: Yurii <iuriish@yahoo.com> Co-authored-by: Yurii Shyrma <iuriish@yahoo.com>
2019-12-20 18:59:29 +01:00
nd4j::ops::rgb_to_grs op;
auto result = op.execute({ &rgbs }, {}, {1});
auto output = result->at(0);
ASSERT_EQ(Status::OK(), result->status());
ASSERT_TRUE(expected.isSameShape(output));
ASSERT_TRUE(expected.equalsTo(output));
delete result;
}
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_8) {
// rank 3
auto rgbs = NDArrayFactory::create<float>('c', { 3,5,4 }, {1.7750e+01f, -7.1062e+01f, -1.0019e+02f,-2.3406e+01f, 5.2094e+01f, 9.5438e+01f, -6.7461e+00f, 3.8562e+01f, 6.5078e+00f,3.3562e+01f, -5.8844e+01f, 2.2750e+01f, -1.0477e+01f, 7.7344e+00f, 9.5469e+00f,2.1391e+01f, -8.5312e+01f, 7.5830e-01f,2.3125e+01f, 1.8145e+00f, 1.4602e+01f,-4.5859e+00f, 3.9344e+01f, 1.1617e+01f,-8.6562e+01f, 1.0038e+02f, 6.7938e+01f,5.9961e+00f, 6.7812e+01f, 2.9734e+01f,2.9609e+01f, -6.1438e+01f, 1.7750e+01f,6.8562e+01f, -7.4414e+00f, 3.9656e+01f,1.1641e+01f, -2.7516e+01f, 6.7562e+01f,7.8438e+01f, 5.4883e+00f, 2.9438e+01f,-3.1344e+01f, 6.5125e+01f, 1.2695e+01f,4.0531e+01f, -6.1211e+00f, 6.2219e+01f,4.6812e+01f, 5.2250e+01f, -1.1414e+01f,1.5404e-02f, 2.9938e+01f, 5.6719e+00f,-2.0125e+01f, 2.1531e+01f, 6.2500e+01f,7.2188e+01f, 9.3750e+00f, -4.8125e+01f});
try {
nd4j::ops::rgb_to_grs op;
auto result = op.execute({ &rgbs }, {}, {});
ASSERT_EQ(Status::THROW(), result->status());
delete result;
} catch (std::exception& e) {
nd4j_printf("Error should be here `%s'. It's OK.\n", e.what());
}
}
////////////////////////////////////////////////////////////////////////////////
TEST_F(DeclarableOpsTests15, test_rgb_to_grs_9) {
// rank 3
auto rgbs = NDArrayFactory::create<float>('f', { 2, 2, 3 }, { 1.7750e+01f,-7.1062e+01f, -1.0019e+02f, -2.3406e+01f,5.2094e+01f,9.5438e+01f, -6.7461e+00f,3.8562e+01f, 6.5078e+00f, 3.3562e+01f,-5.8844e+01f,2.2750e+01f});
auto expected = NDArrayFactory::create<float>('f', { 2,2,1 }, { 36.626545f, 38.607746f, -40.614971f, 18.233341f });
nd4j::ops::rgb_to_grs op;
auto result = op.execute({ &rgbs }, {}, {});
auto output = result->at(0);
ASSERT_EQ(Status::OK(), result->status());
ASSERT_TRUE(expected.isSameShape(output));
ASSERT_TRUE(expected.equalsTo(output));
delete result;
}