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>
|
|
|
|
|
|
|
|
|
|
|
|
using namespace nd4j;
|
|
|
|
|
|
|
|
|
|
|
|
class DeclarableOpsTests14 : public testing::Test {
|
|
|
|
public:
|
|
|
|
|
|
|
|
DeclarableOpsTests14() {
|
|
|
|
printf("\n");
|
|
|
|
fflush(stdout);
|
|
|
|
}
|
|
|
|
};
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_Validation_Edge_1) {
|
|
|
|
auto x = NDArrayFactory::create<int>('c', {2}, {2, 2});
|
|
|
|
auto exp = NDArrayFactory::create('c', {2, 2}, Environment::getInstance()->defaultFloatDataType());
|
|
|
|
exp.assign(4.0f);
|
|
|
|
|
|
|
|
nd4j::ops::fill op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {4.0f});
|
2019-06-06 14:21:15 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(exp, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_Reshape_CF_1) {
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {2, 3}, {1.0, 4.0, 2.0, 5.0, 3.0, 6.0});
|
|
|
|
auto e = NDArrayFactory::create<double>('f', {3, 2}, {1.0, 3.0, 5.0, 2.0, 4.0, 6.0});
|
|
|
|
|
2019-11-13 15:15:18 +01:00
|
|
|
auto r = x.reshape('c', {3, 2});;
|
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
|
|
|
r.streamline('f');
|
2019-06-06 14:21:15 +02:00
|
|
|
|
|
|
|
nd4j::ops::reshape op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {3, 2});
|
2019-06-06 14:21:15 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_Inf_Comparison_1) {
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {5}, {1, 2, 3, std::numeric_limits<double>::infinity(), 5});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {5}, {1, 2, 3, std::numeric_limits<double>::infinity(), 5});
|
|
|
|
|
|
|
|
ASSERT_EQ(x, y);
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_Inf_Comparison_2) {
|
2020-02-13 18:59:35 +01:00
|
|
|
#ifdef FFAST_MATH
|
|
|
|
if (1 > 0)
|
|
|
|
return;
|
|
|
|
#endif
|
|
|
|
|
2019-06-06 14:21:15 +02:00
|
|
|
auto x = NDArrayFactory::create<double>('c', {5}, {1, 2, 3, std::numeric_limits<double>::infinity(), 5});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {5}, {1, 2, 3, -std::numeric_limits<double>::infinity(), 5});
|
|
|
|
|
|
|
|
ASSERT_NE(x, y);
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Multiply_test) {
|
|
|
|
|
|
|
|
for(int k=2;k<10;k++){
|
2019-11-13 15:15:18 +01:00
|
|
|
//nd4j_printf("k=%d\n", k);
|
2019-06-06 14:21:15 +02:00
|
|
|
NDArray x = NDArrayFactory::create<double>('c', {k, 1});
|
|
|
|
NDArray y = NDArrayFactory::create<double>('c', {k});
|
|
|
|
NDArray e = NDArrayFactory::create<double>('c', {k, k});
|
|
|
|
x.assign(1.0);
|
|
|
|
y.assign(1.0);
|
|
|
|
e.assign(1.0);
|
|
|
|
|
|
|
|
nd4j::ops::multiply op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &y});
|
2019-06-06 14:21:15 +02:00
|
|
|
auto f = result->at(0);
|
|
|
|
NDArray r = *f;
|
|
|
|
|
|
|
|
ASSERT_EQ(e, r);
|
|
|
|
ASSERT_EQ(e, *f);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_EvalReductionShape_1) {
|
|
|
|
auto x = NDArrayFactory::create<int>('c', {3}, {5, 3, 4});
|
|
|
|
auto y = NDArrayFactory::create<int>('c', {1}, {1});
|
|
|
|
auto e = NDArrayFactory::create<Nd4jLong>('c', {2}, {5, 4});
|
|
|
|
|
|
|
|
nd4j::ops::evaluate_reduction_shape op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &y}, {}, {}, {false, false});
|
2019-06-06 14:21:15 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
ASSERT_EQ(e, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_EvalReductionShape_2) {
|
|
|
|
auto x = NDArrayFactory::create<int>('c', {3}, {5, 3, 4});
|
|
|
|
auto y = NDArrayFactory::create<int>('c', {1}, {1});
|
|
|
|
auto e = NDArrayFactory::create<Nd4jLong>('c', {3}, {5, 1, 4});
|
|
|
|
|
|
|
|
nd4j::ops::evaluate_reduction_shape op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &y}, {}, {}, {true, false});
|
2019-06-06 14:21:15 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
ASSERT_EQ(e, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_Reduce_Min_Small_0) {
|
|
|
|
auto x = NDArrayFactory::create<float>('c', {3, 4}, {-999.f, 0.2236f, 0.7973f, 0.0962f, 0.7231f, 0.3381f, -0.7301f, 0.9115f, -0.5094f, 0.9749f, -2.1340f, 0.6023f});
|
|
|
|
auto z = NDArrayFactory::create<float>('c', {4});
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {4}, {-999.f, 0.2236f, -2.1340f, 0.0962f});
|
|
|
|
|
|
|
|
nd4j::ops::reduce_min op;
|
|
|
|
op.execute({&x}, {&z}, {}, {0}, {});
|
|
|
|
|
|
|
|
//z.printIndexedBuffer("Z");
|
|
|
|
|
|
|
|
ASSERT_EQ(e, z);
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_Reduce_Min_Small_1) {
|
|
|
|
auto x = NDArrayFactory::create<float>('c', {3, 4}, {-999.f, 0.2236f, 0.7973f, 0.0962f, 0.7231f, 0.3381f, -0.7301f, 0.9115f, -0.5094f, 0.9749f, -2.1340f, 0.6023f});
|
|
|
|
auto z = NDArrayFactory::create<float>('c', {3});
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {3}, {-999.f, -0.7301f, -2.1340f});
|
|
|
|
|
|
|
|
nd4j::ops::reduce_min op;
|
|
|
|
op.execute({&x}, {&z}, {}, {1}, {});
|
|
|
|
|
|
|
|
//z.printIndexedBuffer("Z");
|
|
|
|
|
|
|
|
ASSERT_EQ(e, z);
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_Diag_Zeros_1) {
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {2}, {1, 2});
|
|
|
|
auto z = NDArrayFactory::create<double>('c', {2, 2}, {-119, -119, -119, -119});
|
|
|
|
auto exp = NDArrayFactory::create<double>('c', {2, 2}, {1, 0, 0, 2});
|
|
|
|
|
|
|
|
nd4j::ops::diag op;
|
|
|
|
auto status = op.execute({&x}, {&z}, {}, {}, {});
|
|
|
|
ASSERT_EQ(Status::OK(), status);
|
|
|
|
|
|
|
|
ASSERT_EQ(exp, z);
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_scalar_broadcast_1) {
|
|
|
|
auto x = NDArrayFactory::create<float>(1.0f);
|
|
|
|
auto y = NDArrayFactory::create<float>('c', {5, 10});
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {5, 10});
|
|
|
|
e.assign(1.0);
|
|
|
|
|
|
|
|
|
|
|
|
nd4j::ops::add op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &y});
|
2019-06-06 14:21:15 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
ASSERT_EQ(e, *result->at(0));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_scalar_broadcast_2) {
|
|
|
|
auto x = NDArrayFactory::create<float>(1.0f);
|
|
|
|
auto y = NDArrayFactory::create<float>('c', {5, 10});
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {5, 10});
|
|
|
|
y.assign(2.0f);
|
|
|
|
e.assign(-1.0f);
|
|
|
|
|
|
|
|
|
|
|
|
nd4j::ops::subtract op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &y});
|
2019-06-06 14:21:15 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
ASSERT_EQ(e, *result->at(0));
|
|
|
|
|
|
|
|
delete result;
|
2019-06-15 13:34:34 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_fill_1) {
|
|
|
|
auto x = NDArrayFactory::empty<int>();
|
|
|
|
auto y = NDArrayFactory::create<int>(1);
|
|
|
|
|
|
|
|
nd4j::ops::fill op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &y});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
ASSERT_EQ(y, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_lstmBlockCell_1) {
|
[WIP] build time improvements (#106)
* 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
* 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>
* 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>
* 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
* 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>
* Fix functions of OpaqueVariablesSet
* 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>
* couple of legacy groups reorganized into separate compialtion units
Signed-off-by: raver119 <raver119@gmail.com>
* wrong include
Signed-off-by: raver119 <raver119@gmail.com>
* wrong include
Signed-off-by: raver119 <raver119@gmail.com>
* ReductionLoops_float split
Signed-off-by: raver119 <raver119@gmail.com>
* maximum
Signed-off-by: raver119 <raver119@gmail.com>
* some more rearrangements
Signed-off-by: raver119 <raver119@gmail.com>
* spare ifdef
Signed-off-by: raver119 <raver119@gmail.com>
* mirror pad
Signed-off-by: raver119 <raver119@gmail.com>
* - reduce_float split
- mcmodel
Signed-off-by: raver119 <raver119@gmail.com>
* bad include fix
Signed-off-by: raver119 <raver119@gmail.com>
* norelax
Signed-off-by: raver119 <raver119@gmail.com>
* norelax
Signed-off-by: raver119 <raver119@gmail.com>
* norelax
Signed-off-by: raver119 <raver119@gmail.com>
* norelax
Signed-off-by: raver119 <raver119@gmail.com>
* norelax
Signed-off-by: raver119 <raver119@gmail.com>
* norelax gone
Signed-off-by: raver119 <raver119@gmail.com>
* get back sm
Signed-off-by: raver119 <raver119@gmail.com>
* fix couple of tests for msvc
Signed-off-by: raver119 <raver119@gmail.com>
* fix couple of tests for msvc
Signed-off-by: raver119 <raver119@gmail.com>
* compress-all
Signed-off-by: raver119 <raver119@gmail.com>
* reduced arch list
Signed-off-by: raver119 <raver119@gmail.com>
* compress-all
Signed-off-by: raver119 <raver119@gmail.com>
* reduced arch list
Signed-off-by: raver119 <raver119@gmail.com>
* all compute capabilities option for tests
Signed-off-by: raver119 <raver119@gmail.com>
2019-08-07 16:49:13 +02:00
|
|
|
auto a = NDArrayFactory::create<double>('c', {1, 5}, {0.7787856f, 0.80119777f, 0.72437465f, 0.23089433f, 0.72714126f});
|
|
|
|
auto b = NDArrayFactory::create<double>('c', {1, 3});
|
|
|
|
auto c = NDArrayFactory::create<double>('c', {1, 3});
|
|
|
|
auto d = NDArrayFactory::create<double>('c', {8, 12}, {-0.15320599,-0.120416045,0.33126968,0.13921785,-0.32313538,-0.43956736,0.4756174,0.4335605,-0.5450856,-0.3943429,-0.28687626,0.068032146,-0.2793799,0.17298919,-0.36553562,-0.097853184,-0.2544747,-0.39872527,-0.14556861,-0.31479517,0.2559092,0.47166896,-0.31330687,0.47313118,0.5134543,-0.4678212,-0.12853557,0.26142156,0.43472284,-0.42842552,-0.1895876,0.538689,0.508651,-0.020272732,0.112327516,0.2704304,-0.046546757,0.32570732,-0.15148133,-0.19145513,0.18631572,-0.024152994,0.41603214,-0.3421499,0.0106860995,-0.2966229,-0.36713937,0.25841123,0.0843398,0.49082482,0.10800403,0.1874243,-0.26379472,-0.22531849,0.24924624,0.23119557,0.49940765,-0.051413506,0.20315129,-0.41888732,0.44097036,0.40453392,0.013338983,0.23434466,0.23942488,0.47894,-0.19898453,0.09253675,-0.032358468,-0.15213022,-0.3441009,-0.15600958,-0.08235118,0.12165731,-0.4481289,-0.4842423,-0.45797008,-0.4606034,0.08163166,-0.2981107,0.50207126,0.44195646,0.13850057,0.072246075,-0.34388685,0.030900061,0.35821778,0.47900867,0.5094063,0.23683065,0.18020362,-0.1369732,0.015235603,0.2786904,0.07954317,0.12543976});
|
|
|
|
auto e = NDArrayFactory::create<double>('c', {3});
|
|
|
|
auto f = NDArrayFactory::create<double>('c', {3});
|
|
|
|
auto g = NDArrayFactory::create<double>('c', {3});
|
|
|
|
auto h = NDArrayFactory::create<double>('c', {12});
|
|
|
|
|
|
|
|
auto z0 = NDArrayFactory::create<double>('c', {1, 3});
|
|
|
|
auto z1 = NDArrayFactory::create<double>('c', {1, 3});
|
|
|
|
auto z2 = NDArrayFactory::create<double>('c', {1, 3});
|
|
|
|
auto z3 = NDArrayFactory::create<double>('c', {1, 3});
|
|
|
|
auto z4 = NDArrayFactory::create<double>('c', {1, 3});
|
|
|
|
auto z5 = NDArrayFactory::create<double>('c', {1, 3});
|
|
|
|
auto z6 = NDArrayFactory::create<double>('c', {1, 3});
|
2019-06-15 13:34:34 +02:00
|
|
|
|
|
|
|
nd4j::ops::lstmBlockCell op;
|
|
|
|
auto result = op.execute({&a, &b, &c, &d, &e, &f, &g, &h}, {&z0, &z1, &z2, &z3, &z4, &z5, &z6}, {1.0, -1.0}, {0}, {});
|
|
|
|
ASSERT_EQ(Status::OK(), result);
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_stack_1) {
|
|
|
|
auto x = NDArrayFactory::create<float>('c', {0});
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {1, 0});
|
|
|
|
|
|
|
|
nd4j::ops::stack op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {}, {0});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
ASSERT_EQ(e, *z);
|
|
|
|
nd4j::ops::reduce_min sumOp;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto res2 = sumOp.evaluate({&e}, {1.}, {1});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(res2->status(), Status::OK());
|
|
|
|
auto out = res2->at(0);
|
2019-08-02 19:01:03 +02:00
|
|
|
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(out->e<float>(0), DataTypeUtils::infOrMax<float>());
|
|
|
|
delete res2;
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_stack_2) {
|
|
|
|
auto x = NDArrayFactory::empty<float>();
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {0});
|
|
|
|
|
|
|
|
nd4j::ops::stack op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {}, {0});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
ASSERT_EQ(e, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_stack_3) {
|
|
|
|
auto x = NDArrayFactory::empty<float>();
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {2, 0});
|
|
|
|
|
|
|
|
nd4j::ops::stack op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &x}, {}, {0});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
ASSERT_EQ(e, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_stack_4) {
|
|
|
|
auto x = NDArrayFactory::create<float>('c', {0});
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {2, 0});
|
|
|
|
|
|
|
|
nd4j::ops::stack op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &x}, {}, {0});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
ASSERT_EQ(e, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_reduce_min_1) {
|
|
|
|
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {1, 0});
|
|
|
|
nd4j::ops::reduce_min sumOp;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto res2 = sumOp.evaluate({&e}, {1.}, {1});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(res2->status(), Status::OK());
|
|
|
|
auto out = res2->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(out->e<float>(0), DataTypeUtils::infOrMax<float>());
|
|
|
|
delete res2;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_reduce_max_1) {
|
|
|
|
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {1, 0});
|
|
|
|
nd4j::ops::reduce_max sumOp;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto res2 = sumOp.evaluate({&e}, {1.}, {1});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(res2->status(), Status::OK());
|
|
|
|
auto out = res2->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(out->e<float>(0), -DataTypeUtils::infOrMax<float>());
|
|
|
|
delete res2;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_reduce_sum_1) {
|
2020-02-13 18:59:35 +01:00
|
|
|
#ifdef FFAST_MATH
|
|
|
|
if (1 > 0)
|
|
|
|
return;
|
|
|
|
#endif
|
2019-06-15 13:34:34 +02:00
|
|
|
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {1, 0});
|
|
|
|
nd4j::ops::reduce_sum sumOp;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto res2 = sumOp.evaluate({&e}, {1.}, {1});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(res2->status(), Status::OK());
|
|
|
|
auto out = res2->at(0);
|
|
|
|
ASSERT_EQ(out->e<float>(0), 0.f);
|
|
|
|
delete res2;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_reduce_mean_1) {
|
2020-02-13 18:59:35 +01:00
|
|
|
#ifdef FFAST_MATH
|
|
|
|
if (1 > 0)
|
|
|
|
return;
|
|
|
|
#endif
|
2019-06-15 13:34:34 +02:00
|
|
|
|
|
|
|
auto e = NDArrayFactory::create<float>('c', {1, 0});
|
|
|
|
nd4j::ops::reduce_mean sumOp;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto res2 = sumOp.evaluate({&e}, {1.}, {1});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(res2->status(), Status::OK());
|
|
|
|
auto out = res2->at(0);
|
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
|
|
|
// out->printShapeInfo("ReduceMean empty shape with keep dims");
|
|
|
|
// out->printIndexedBuffer("ReduceMean scalar");
|
|
|
|
ASSERT_TRUE(std::isnan(out->e<float>(0)));
|
2019-06-15 13:34:34 +02:00
|
|
|
delete res2;
|
|
|
|
}
|
|
|
|
|
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(DeclarableOpsTests14, Test_StridedSliceZeros_1) {
|
|
|
|
auto matrix = NDArrayFactory::create<double>('c', {1, 2, 0, 4});
|
|
|
|
auto b = NDArrayFactory::create<int>('c', {3}, {0, 0, 0});
|
|
|
|
auto e = NDArrayFactory::create<int>('c', {3}, {2,0,2});
|
|
|
|
auto s = NDArrayFactory::create<int>('c', {3}, {1,1,1});
|
|
|
|
|
|
|
|
auto exp = NDArrayFactory::create<double>('c', {1,0,0,4});
|
|
|
|
|
|
|
|
matrix.linspace(1);
|
|
|
|
|
|
|
|
nd4j::ops::strided_slice op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&matrix, &b, &e, &s}, {}, {0, 0, 0, 0, 0});
|
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
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_StridedSliceZeros_2) {
|
|
|
|
auto matrix = NDArrayFactory::create<double>('c', {1, 2, 0, 4});
|
|
|
|
auto b = NDArrayFactory::create<int>('c', {3}, {0, 0, 0});
|
|
|
|
auto e = NDArrayFactory::create<int>('c', {3}, {2,0,2});
|
|
|
|
auto s = NDArrayFactory::create<int>('c', {3}, {1,1,1});
|
|
|
|
|
|
|
|
auto exp = NDArrayFactory::create<double>('c', {0,0,4});
|
|
|
|
|
|
|
|
matrix.linspace(1);
|
|
|
|
|
|
|
|
nd4j::ops::strided_slice op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&matrix, &b, &e, &s}, {}, {0, 0, 0, 0, 1});
|
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
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
2019-06-15 13:34:34 +02:00
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_argmax_1) {
|
|
|
|
auto x = NDArrayFactory::create<float>('c', {1, 0});
|
|
|
|
auto y = NDArrayFactory::create<int>(0);
|
|
|
|
auto e = NDArrayFactory::create<Nd4jLong>('c', {0});
|
|
|
|
|
|
|
|
nd4j::ops::argmax op;
|
|
|
|
//nd4j::ops::reduce_max op;
|
|
|
|
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x, &y}, {}, {});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(e, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_argmax_2) {
|
|
|
|
auto x = NDArrayFactory::create<float>('c', {1, 0});
|
|
|
|
auto y = NDArrayFactory::create<int>(1);
|
|
|
|
|
|
|
|
nd4j::ops::argmax op;
|
|
|
|
try {
|
|
|
|
auto result = op.execute({&x, &y}, {&y}, {}, {}, {});
|
|
|
|
ASSERT_TRUE(false);
|
|
|
|
} catch (std::exception &e) {
|
|
|
|
//
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, test_empty_tanh_5) {
|
|
|
|
auto x = NDArrayFactory::create<float>('c', {32, 0});
|
|
|
|
|
|
|
|
nd4j::ops::tanh op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {}, {});
|
2019-06-15 13:34:34 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(x.isSameShape(z));
|
|
|
|
ASSERT_EQ(x, *z);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
2019-08-21 20:10:29 +02:00
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, repeat_1) {
|
|
|
|
|
|
|
|
NDArray x('c', {2, 3}, {1, 2, 3, 4, 5, 6});
|
|
|
|
NDArray e('c', {4, 3}, {1, 2, 3, 1, 2, 3, 4, 5, 6, 4, 5, 6});
|
|
|
|
|
|
|
|
nd4j::ops::repeat op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {}, {2, 0});
|
2019-08-21 20:10:29 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(e.isSameShape(z));
|
|
|
|
ASSERT_TRUE(e.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, repeat_2) {
|
|
|
|
|
|
|
|
NDArray x('c', {2, 3}, {1, 2, 3, 4, 5, 6});
|
|
|
|
NDArray e('c', {2, 6}, {1, 1, 2, 2, 3, 3,4, 4, 5, 5, 6, 6});
|
|
|
|
|
|
|
|
nd4j::ops::repeat op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {}, {2, 1});
|
2019-08-21 20:10:29 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(e.isSameShape(z));
|
|
|
|
ASSERT_TRUE(e.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, repeat_3) {
|
|
|
|
|
|
|
|
NDArray x('c', {2, 3}, {1, 2, 3, 4, 5, 6});
|
|
|
|
NDArray e('c', {2, 6}, {1, 2, 2, 3, 3, 3,4, 5, 5, 6, 6, 6});
|
|
|
|
|
|
|
|
nd4j::ops::repeat op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {}, {1,2,3, 1});
|
2019-08-21 20:10:29 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(e.isSameShape(z));
|
|
|
|
ASSERT_TRUE(e.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, repeat_4) {
|
|
|
|
|
|
|
|
NDArray x('c', {2, 3}, {1, 2, 3, 4, 5, 6});
|
|
|
|
NDArray e('c', {7, 3}, {1, 2, 3, 1, 2, 3, 1, 2, 3, 4, 5, 6, 4, 5, 6, 4, 5, 6, 4, 5, 6});
|
|
|
|
|
|
|
|
nd4j::ops::repeat op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {}, {3,4, 0});
|
2019-08-21 20:10:29 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(e.isSameShape(z));
|
|
|
|
ASSERT_TRUE(e.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, repeat_5) {
|
|
|
|
|
|
|
|
NDArray x('c', {2, 3, 4}, {1, 2, 3, 4, 5, 6, 7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24});
|
|
|
|
NDArray e('c', {2, 4, 4}, {1, 2, 3, 4, 5, 6, 7, 8, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 17, 18, 19, 20, 21, 22, 23, 24});
|
|
|
|
|
|
|
|
nd4j::ops::repeat op;
|
2020-01-30 08:07:24 +01:00
|
|
|
auto result = op.evaluate({&x}, {}, {1,2,1, 1});
|
2019-08-21 20:10:29 +02:00
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(e.isSameShape(z));
|
|
|
|
ASSERT_TRUE(e.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
2020-02-12 12:12:17 +01:00
|
|
|
/////////////////////////////////////////////////////////////////////////
|
2020-02-14 10:04:38 +01:00
|
|
|
TEST_F(DeclarableOpsTests14, Test_broadcast_SpecialCaseTest) {
|
2020-02-12 12:12:17 +01:00
|
|
|
|
|
|
|
auto y = NDArray('c', { 3 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto x = NDArray('c', { 5, 2, 1 }, nd4j::DataType::FLOAT32);
|
|
|
|
|
|
|
|
auto e = NDArray('c', { 5, 2, 3 }, { 2., 2., 2., 3., 3., 3., 4., 4., 4., 5., 5., 5., 6., 6., 6., 7., 7., 7., 8., 8., 8., 9., 9., 9., 10., 10., 10., 11., 11., 11. }, nd4j::DataType::FLOAT32);
|
2020-02-14 10:04:38 +01:00
|
|
|
|
2020-02-12 12:12:17 +01:00
|
|
|
y.assign(1.0);
|
|
|
|
x.linspace(1.0);
|
|
|
|
|
|
|
|
nd4j::ops::add op;
|
|
|
|
auto result = op.evaluate({ &x, &y });
|
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto res = *result->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(e, res);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
2020-02-14 10:04:38 +01:00
|
|
|
/////////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_broadcast_SpecialCaseTest2) {
|
|
|
|
|
|
|
|
auto y = NDArray('c', { 1, 3 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto x = NDArray('c', { 5, 2, 1 }, nd4j::DataType::FLOAT32);
|
|
|
|
|
|
|
|
auto e = NDArray('c', { 5, 2, 3 }, { 2., 2., 2., 3., 3., 3., 4., 4., 4., 5., 5., 5., 6., 6., 6., 7., 7., 7., 8., 8., 8., 9., 9., 9., 10., 10., 10., 11., 11., 11. }, nd4j::DataType::FLOAT32);
|
|
|
|
|
|
|
|
y.assign(1.0);
|
|
|
|
x.linspace(1.0);
|
|
|
|
|
|
|
|
nd4j::ops::add op;
|
|
|
|
auto result = op.evaluate({ &x, &y });
|
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
auto res = *result->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(e, res);
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
///////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_broadcast_SpecialCaseTest3) {
|
|
|
|
|
|
|
|
auto x = NDArray('c', { 3, 5, 1 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto y = NDArray('c', { 3, 1, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto z = NDArray('c', { 3, 5, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
// recieved by main algorithm
|
|
|
|
auto e = NDArray('c', { 3, 5, 4 }, { 10., 11., 12., 13., 20., 22., 24., 26., 30., 33., 36., 39., 40., 44., 48., 52., 50., 55., 60., 65., 84., 90., 96., 102., 98., 105., 112., 119., 112., 120., 128., 136., 126., 135., 144., 153., 140., 150., 160., 170., 198., 209., 220., 231., 216., 228., 240., 252., 234., 247., 260., 273., 252., 266., 280., 294., 270., 285., 300., 315. }, nd4j::DataType::FLOAT32);
|
|
|
|
|
|
|
|
x.linspace(1.f);
|
|
|
|
y.linspace(10.f);
|
|
|
|
z.assign(0.f);
|
|
|
|
|
|
|
|
x.applyTrueBroadcast(BroadcastOpsTuple::Multiply(), y, z);
|
|
|
|
ASSERT_EQ(e, z);
|
|
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_broadcast_SpecialCaseTest4) {
|
|
|
|
|
|
|
|
auto x = NDArray('c', { 2, 3, 5, 1 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto y = NDArray('c', { 2, 3, 1, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto z = NDArray('c', { 2, 3, 5, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
// recieved by main algorithm
|
|
|
|
auto e = NDArray('c', { 2, 3, 5, 4 }, { 10., 11., 12., 13.,20., 22., 24., 26.,30., 33., 36., 39.,40., 44., 48., 52.,50., 55., 60., 65.,84., 90., 96., 102.,98., 105., 112., 119.,112., 120., 128., 136.,126., 135., 144., 153.,140., 150., 160., 170.,198., 209., 220., 231.,216., 228., 240., 252.,234., 247., 260., 273.,252., 266., 280., 294.,270., 285., 300., 315.,352., 368., 384., 400.,374., 391., 408., 425.,396., 414., 432., 450.,418., 437., 456., 475.,440., 460., 480., 500.,546., 567., 588., 609.,572., 594., 616., 638.,598., 621., 644., 667.,624., 648., 672., 696.,650., 675., 700., 725.,780., 806., 832., 858.,810., 837., 864., 891.,840., 868., 896., 924.,870., 899., 928., 957.,900., 930., 960., 990. }, nd4j::DataType::FLOAT32);
|
|
|
|
x.linspace(1.f);
|
|
|
|
y.linspace(10.f);
|
|
|
|
z.assign(0.f);
|
|
|
|
|
|
|
|
x.applyTrueBroadcast(BroadcastOpsTuple::Multiply(), y, z);
|
|
|
|
ASSERT_EQ(e, z);
|
|
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_broadcast_SpecialCaseTest5) {
|
|
|
|
|
|
|
|
auto x = NDArray('c', { 3, 5, 1 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto y = NDArray('c', { 3, 1, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto z = NDArray('c', { 3, 5, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
// recieved by main algorithm
|
|
|
|
auto e = NDArray('c', { 3, 5, 4 }, { 0.1, 0.090909, 0.083333, 0.076923,0.2, 0.181818, 0.166667, 0.153846,0.3, 0.272727, 0.250000, 0.230769,0.4, 0.363636, 0.333333, 0.307692,0.5, 0.454545, 0.416667, 0.384615, 0.428571, 0.400000, 0.375000, 0.352941, 0.500000, 0.466667, 0.437500, 0.411765, 0.571429, 0.533333, 0.500000, 0.470588, 0.642857, 0.600000, 0.562500, 0.529412, 0.714286, 0.666667, 0.625000, 0.588235, 0.611111, 0.578947, 0.550000, 0.523810, 0.666667, 0.631579, 0.600000, 0.571429, 0.722222, 0.684211, 0.650000, 0.619048, 0.777778, 0.736842, 0.700000, 0.666667, 0.833333, 0.789474, 0.750000, 0.714286 }, nd4j::DataType::FLOAT32);
|
|
|
|
x.linspace(1.f);
|
|
|
|
y.linspace(10.f);
|
|
|
|
z.assign(0.f);
|
|
|
|
|
|
|
|
x.applyTrueBroadcast(BroadcastOpsTuple::Divide(), y, z);
|
|
|
|
ASSERT_EQ(e, z);
|
|
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_broadcast_SpecialCaseTest6) {
|
|
|
|
|
|
|
|
auto x = NDArray('c', { 2, 3, 5, 1 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto y = NDArray('c', { 2, 3, 1, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto z = NDArray('c', { 2, 3, 5, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
// recieved by main algorithm
|
|
|
|
auto e = NDArray('c', { 2, 3, 5, 4 }, { 0.1, 0.090909, 0.083333, 0.076923,0.2, 0.181818, 0.166667, 0.153846,0.3, 0.272727, 0.250000, 0.230769,0.4, 0.363636, 0.333333, 0.307692,0.5, 0.454545, 0.416667, 0.384615, 0.428571, 0.400000, 0.375000, 0.352941, 0.500000, 0.466667, 0.437500, 0.411765, 0.571429, 0.533333, 0.500000, 0.470588, 0.642857, 0.600000, 0.562500, 0.529412, 0.714286, 0.666667, 0.625000, 0.588235,0.611111, 0.578947, 0.550000, 0.523810,0.666667, 0.631579, 0.600000, 0.571429,0.722222, 0.684211, 0.650000, 0.619048,0.777778, 0.736842, 0.700000, 0.666667,0.833333, 0.789474, 0.750000, 0.714286, 0.727273, 0.695652, 0.666667, 0.64, 0.772727, 0.739130, 0.708333, 0.68, 0.818182, 0.782609, 0.750000, 0.72, 0.863636, 0.826087, 0.791667, 0.76, 0.909091, 0.869565, 0.833333, 0.80, 0.807692, 0.777778, 0.750000, 0.724138, 0.846154, 0.814815, 0.785714, 0.758621, 0.884615, 0.851852, 0.821429, 0.793103, 0.923077, 0.888889, 0.857143, 0.827586, 0.961538, 0.925926, 0.892857, 0.862069, 0.866667, 0.838710, 0.812500, 0.787879, 0.900000, 0.870968, 0.843750, 0.818182, 0.933333, 0.903226, 0.875000, 0.848485, 0.966667, 0.935484, 0.906250, 0.878788, 1.000000, 0.967742, 0.937500, 0.909091 }, nd4j::DataType::FLOAT32);
|
|
|
|
|
|
|
|
x.linspace(1.f);
|
|
|
|
y.linspace(10.f);
|
|
|
|
z.assign(0.f);
|
|
|
|
|
|
|
|
x.applyTrueBroadcast(BroadcastOpsTuple::Divide(), y, z);
|
|
|
|
ASSERT_EQ(e, z);
|
|
|
|
}
|
|
|
|
|
|
|
|
///////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_broadcast_SpecialCaseTest7) {
|
|
|
|
|
|
|
|
auto x = NDArray('c', { 3, 5, 1 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto y = NDArray('c', { 3, 1, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto z = NDArray('c', { 3, 5, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
// recieved by main algorithm
|
|
|
|
auto e = NDArray('c', { 3, 5, 4 }, { -9., -10., -11., -12.,-8., -9., -10., -11., -7., -8., -9., -10.,-6., -7., -8., -9.,-5., -6., -7., -8.,-8., -9., -10., -11.,-7., -8., -9., -10.,-6., -7., -8., -9.,-5., -6., -7., -8.,-4., -5., -6., -7.,-7., -8.000000, -9.000000, -10.00,-6.000000, -7.000000, -8.000000, -9.000,-5.000000, -6.000000, -7.000000, -8.000,-4.000000, -5.000000, -6.000000, -7.000,-3.000000, -4.000000, -5.000000, -6.000 }, nd4j::DataType::FLOAT32);
|
|
|
|
x.linspace(1.f);
|
|
|
|
y.linspace(10.f);
|
|
|
|
z.assign(0.f);
|
|
|
|
|
|
|
|
x.applyTrueBroadcast(BroadcastOpsTuple::Subtract(), y, z);
|
|
|
|
ASSERT_EQ(e, z);
|
|
|
|
}
|
|
|
|
///////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, Test_broadcast_SpecialCaseTest8) {
|
|
|
|
|
|
|
|
auto x = NDArray('c', { 2, 3, 5, 1 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto y = NDArray('c', { 2, 3, 1, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
auto z = NDArray('c', { 2, 3, 5, 4 }, nd4j::DataType::FLOAT32);
|
|
|
|
// recieved by main algorithm
|
|
|
|
auto e = NDArray('c', { 2, 3, 5, 4 }, { -9.0, -10., -11., -12.,-8., -9., -10., -11.0,-7., -8., -9., -10.,-6., -7., -8., -9.,-5., -6., -7., -8.,-8., -9., -10., -11.,-7., -8., -9., -10.,-6., -7., -8., -9.,-5., -6., -7., -8.,-4., -5., -6., -7.,-7., -8., -9., -10.,-6., -7., -8., -9.,-5., -6., -7., -8.,-4., -5., -6., -7.,-3., -4., -5., -6.,-6., -7., -8., -9.,-5., -6., -7., -8.,-4., -5., -6., -7.,-3., -4., -5., -6.,-2., -3., -4., -5.,-5., -6., -7., -8.,-4., -5., -6., -7.,-3., -4., -5., -6.,-2., -3., -4., -5.,-1., -2., -3., -4.,-4., -5., -6., -7.,-3., -4., -5., -6.,-2., -3., -4., -5.,-1., -2., -3., -4., 0., -1., -2., -3. }, nd4j::DataType::FLOAT32);
|
|
|
|
|
|
|
|
x.linspace(1.f);
|
|
|
|
y.linspace(10.f);
|
|
|
|
z.assign(0.f);
|
|
|
|
|
|
|
|
x.applyTrueBroadcast(BroadcastOpsTuple::Subtract(), y, z);
|
|
|
|
ASSERT_EQ(e, z);
|
|
|
|
}
|
2020-02-18 06:58:01 +01:00
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test1) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {3, 4});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {4, 3});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {3, 3}, {35., 79., 123., 40., 92., 144., 45., 105., 165.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test2) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {3, 4});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {4, 3});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {3, 3}, {35., 79., 123.,40., 92., 144.,45.,105., 165.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test3) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {3, 4});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {4, 3});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {3, 3}, {35., 79., 123.,40., 92., 144.,45.,105., 165.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test4) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double> ('f', {3, 4});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {4, 3});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {3, 3}, {35., 79., 123.,40., 92., 144.,45.,105., 165.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test5) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {4, 3});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {4, 3});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {3, 3}, {83., 94., 105., 94., 107., 120., 105., 120., 135.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test6) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {4, 3});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {3, 4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {3, 3}, {35., 40., 45., 79., 92., 105., 123., 144., 165.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test7) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {5, 3,4});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {5, 3,4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{5, 3,3}, {3. , 84.6, 281.4, 593.4, 1020.6, 7. , 107.8, 323.8, 655. , 1101.4,11. , 131. , 366.2, 716.6, 1182.2,
|
|
|
|
7. , 107.8, 323.8, 655. , 1101.4,17.4, 137.4, 372.6, 723. , 1188.6,27.8, 167. , 421.4, 791. , 1275.8,
|
|
|
|
11. , 131. , 366.2, 716.6, 1182.2,27.8, 167. , 421.4, 791. , 1275.8,44.6, 203. , 476.6, 865.4, 1369.4,});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {0, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test8) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {2,5, 3,4});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {2,5, 3,4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{2,5, 3,3}, {3. , 1563. , 84.6, 2220.6, 281.4, 2993.4, 593.4, 3881.4,1020.6, 4884.6, 7. , 1663. , 107.8, 2339.8, 323.8, 3131.8, 655. , 4039. ,1101.4, 5061.4,
|
|
|
|
11. , 1763. , 131. , 2459. , 366.2, 3270.2, 716.6, 4196.6,1182.2, 5238.2, 7. , 1663. , 107.8, 2339.8, 323.8, 3131.8, 655. , 4039. ,1101.4, 5061.4,
|
|
|
|
17.4, 1769.4, 137.4, 2465.4, 372.6, 3276.6, 723. , 4203. ,1188.6, 5244.6, 27.8, 1875.8, 167. , 2591. , 421.4, 3421.4, 791. , 4367. ,1275.8, 5427.8,
|
|
|
|
11. , 1763. , 131. , 2459. , 366.2, 3270.2, 716.6, 4196.6,1182.2, 5238.2, 27.8, 1875.8, 167. , 2591. , 421.4, 3421.4, 791. , 4367. ,1275.8, 5427.8,
|
|
|
|
44.6, 1988.6, 203. , 2723. , 476.6, 3572.6, 865.4, 4537.4,1369.4, 5617.4});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {0, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test9) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {2,5, 4,3});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {2,5, 3,4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{2,5, 3,3}, {7. , 1639. , 103. , 2311. , 314.2, 3098.2, 640.6, 4000.6,1082.2, 5018.2, 8. , 1664. , 108.8, 2340.8, 324.8, 3132.8, 656. , 4040. ,1102.4, 5062.4,
|
|
|
|
9. , 1689. , 114.6, 2370.6, 335.4, 3167.4, 671.4, 4079.4,1122.6, 5106.6, 15.8, 1743.8, 131. , 2435. , 361.4, 3241.4, 707. , 4163. ,1167.8, 5199.8,
|
|
|
|
18.4, 1770.4, 138.4, 2466.4, 373.6, 3277.6, 724. , 4204. ,1189.6, 5245.6, 21. , 1797. , 145.8, 2497.8, 385.8, 3313.8, 741. , 4245. ,1211.4, 5291.4,
|
|
|
|
24.6, 1848.6, 159. , 2559. , 408.6, 3384.6, 773.4, 4325.4,1253.4, 5381.4, 28.8, 1876.8, 168. , 2592. , 422.4, 3422.4, 792. , 4368. ,1276.8, 5428.8,
|
|
|
|
33. , 1905. , 177. , 2625. , 436.2, 3460.2, 810.6, 4410.6,1300.2, 5476.2});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test10) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create_<float>('c', {3, 5});
|
|
|
|
x->linspace(1);
|
|
|
|
|
|
|
|
auto y = NDArrayFactory::create_<float>('c', {5, 3});
|
|
|
|
y->linspace(1);
|
|
|
|
|
|
|
|
float _expB[]{135.0f, 310.0f, 485.0f, 150.0f, 350.0f, 550.0f, 165.0f, 390.0f, 615.0f};
|
|
|
|
Nd4jLong _expS[] {2, 3, 3, 1, 3, 0, 1, 102}; // expected shape
|
|
|
|
ArrayOptions::setDataType(_expS, nd4j::DataType::FLOAT32);
|
|
|
|
NDArray exp(_expB, _expS);
|
|
|
|
|
|
|
|
auto variableSpace = new VariableSpace();
|
|
|
|
variableSpace->putVariable(-1, x);
|
|
|
|
variableSpace->putVariable(-2, y);
|
|
|
|
variableSpace->putVariable(1, new Variable());
|
|
|
|
|
|
|
|
auto block = new Context(1, variableSpace, false);
|
|
|
|
block->fillInputs({-1, -2});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
|
|
|
|
Nd4jStatus status = op.execute(block);
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, status);
|
|
|
|
ASSERT_TRUE(variableSpace->hasVariable(1));
|
|
|
|
|
|
|
|
auto result = variableSpace->getVariable(1)->getNDArray();
|
|
|
|
|
|
|
|
ASSERT_TRUE(result->equalsTo(&exp));
|
|
|
|
|
|
|
|
delete block;
|
|
|
|
delete variableSpace;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test11) {
|
|
|
|
auto A = NDArrayFactory::create<float>('c', {3, 3});
|
|
|
|
auto B = NDArrayFactory::create<float>('c', {3, 1});
|
|
|
|
auto exp = NDArrayFactory::create<float>('c', {3, 1}, {14.00f, 32.00f, 50.00f});
|
|
|
|
|
|
|
|
A.linspace(1);
|
|
|
|
B.linspace(1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
|
|
|
|
auto result = op.evaluate({&A, &B}, {}, {});
|
|
|
|
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test12) {
|
|
|
|
auto x= NDArrayFactory::create<double>('c', {3, 4}, {1, 2, 3, 4, 5, 6, 7, 8 , 9, 10, 11, 12});
|
|
|
|
auto y= NDArrayFactory::create<double>('c', {4, 3}, {1, 2, 3, 4, 5, 6, 7, 8 , 9, 10, 11, 12});
|
|
|
|
auto exp= NDArrayFactory::create<double>('f', {4, 4}, {38.0, 44.0, 50.0, 56.0, 83.0, 98.0, 113.0, 128.0, 128.0, 152.0, 176.0, 200.0, 173.0, 206.0, 239.0, 272.0});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&x, &y}, {}, {1, 1});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test13) {
|
|
|
|
auto x= NDArrayFactory::create<double>('c', {1, 3}, {1, 2, 3});
|
|
|
|
auto y= NDArrayFactory::create<double>('c', {1, 4}, {1, 2, 3, 4});
|
|
|
|
auto exp= NDArrayFactory::create<double>('f', {3, 4}, {1.0, 2.0, 3.0, 2.0, 4.0, 6.0, 3.0, 6.0, 9.0, 4.0, 8.0, 12.0});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&x, &y}, {}, {1, 0});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
//z->printIndexedBuffer("z");
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test14) {
|
|
|
|
auto x= NDArrayFactory::create<double>('c', {3, 1}, {1, 2, 3});
|
|
|
|
auto y= NDArrayFactory::create<double>('c', {4, 1}, {1, 2, 3, 4});
|
|
|
|
auto exp= NDArrayFactory::create<double>('f', {3, 4}, {1.0, 2.0, 3.0, 2.0, 4.0, 6.0, 3.0, 6.0, 9.0, 4.0, 8.0, 12.0});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&x, &y}, {}, {0, 1});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
//z->printIndexedBuffer("z");
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test15) {
|
|
|
|
auto x= NDArrayFactory::create<double>('c', {3, 1}, {1, 2, 3});
|
|
|
|
auto y= NDArrayFactory::create<double>('c', {1, 4}, {1, 2, 3, 4});
|
|
|
|
auto exp= NDArrayFactory::create<double>('f', {3, 4}, {1.0, 2.0, 3.0, 2.0, 4.0, 6.0, 3.0, 6.0, 9.0, 4.0, 8.0, 12.0});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&x, &y}, {}, {});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
//z->printIndexedBuffer("z");
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test16) {
|
|
|
|
auto x= NDArrayFactory::create<double>('c', {4, 1}, {1, 2, 3, 4});
|
|
|
|
auto y= NDArrayFactory::create<double>('c', {1, 4}, {1, 2, 3, 4});
|
|
|
|
auto exp= NDArrayFactory::create<double>('f', {4, 4}, {1,2, 3, 4,2,4, 6, 8,3,6, 9,12,4,8,12,16});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&x, &y});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
//z->printIndexedBuffer("z");
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test17) {
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {1, 2}, {2.0f, 2.0f});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {2, 1}, {2.0f, 2.0f});
|
|
|
|
auto exp = NDArrayFactory::create<double>('c', {1, 1}, {8.0f});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&x, &y}, {}, {});
|
|
|
|
ASSERT_EQ(Status::OK(), result->status());
|
|
|
|
|
|
|
|
ASSERT_EQ(exp, *result->at(0));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test18) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {1, 4, 3});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {1, 3, 4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {1, 3, 3}, {35., 40., 45., 79., 92., 105., 123., 144., 165.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test19) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {4, 1});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {1, 4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {1, 1}, {15});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 1});
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
|
|
|
|
auto z = results->at(0);
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test20) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {1, 4, 1});
|
|
|
|
auto y = NDArrayFactory::create<double>('f', {1, 1, 4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {1, 1, 1}, {15});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 1});
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
auto z = results->at(0);
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test21) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {2, 3});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {3, 5});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {5, 2}, {23. , 26. , 29. , 32. , 35., 50. , 57.5, 65. , 72.5, 80.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {0, 0, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test22) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {3, 2});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {3, 5});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {5, 2}, {37. , 41.5, 46. , 50.5, 55., 46. , 52. , 58. , 64. , 70.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 0, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test23) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {3, 2});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {3, 5});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f', {5, 2}, {37. , 41.5, 46. , 50.5, 55., 46. , 52. , 58. , 64. , 70.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.5, 0.5);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 0, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test24) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {2,2, 3,5});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {2,2, 4,3});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{2,2, 4,5}, {4.6, 281.8, 89.2, 582.4, 10. , 314.2,108.1, 628.3, 15.4, 346.6,127. , 674.2, 20.8, 379. ,145.9, 720.1, 5.2, 289.6, 93.4, 593.8,
|
|
|
|
11.5, 322.9,113.2, 640.6, 17.8, 356.2,133. , 687.4, 24.1, 389.5,152.8, 734.2, 5.8, 297.4, 97.6, 605.2, 13. , 331.6,118.3, 652.9,
|
|
|
|
20.2, 365.8,139. , 700.6, 27.4, 400. ,159.7, 748.3, 6.4, 305.2,101.8, 616.6, 14.5, 340.3,123.4, 665.2, 22.6, 375.4,145. , 713.8,
|
|
|
|
30.7, 410.5,166.6, 762.4, 7. , 313. ,106. , 628. , 16. , 349. ,128.5, 677.5, 25. , 385. ,151. , 727. , 34. , 421. ,173.5, 776.5});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 1, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test25) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {4, 3});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{3}, {7., 8., 9.});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 0});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test26) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {3});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {4, 3});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{4}, {1.4, 3.2, 5., 6.8});
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {0, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test27) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {1, 1});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {1, 1});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{1, 1}, {0.2});
|
|
|
|
|
|
|
|
x.linspace(2.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test28) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {1, 1});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {1, 1});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{1, 1}, {0.2});
|
|
|
|
|
|
|
|
x.linspace(2.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1,1,1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test29) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {1});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {1, 1});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{1}, {0.2});
|
|
|
|
|
|
|
|
x.linspace(2.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test30) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {1,1});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {1});
|
|
|
|
auto exp = NDArrayFactory::create<double>('f',{1}, {0.2});
|
|
|
|
|
|
|
|
x.linspace(2.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test31) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {4});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {4});
|
|
|
|
auto exp = NDArrayFactory::create<double>(3.);
|
|
|
|
|
|
|
|
x.linspace(1.);
|
|
|
|
y.linspace(0.1, 0.1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test32) {
|
|
|
|
|
|
|
|
auto x = NDArrayFactory::create<double>('f', {1}, {2.});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {1}, {3.});
|
|
|
|
auto exp = NDArrayFactory::create<double>(6.);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto results = op.evaluate({&x, &y}, {}, {1, 1});
|
|
|
|
auto z = results->at(0);
|
|
|
|
|
|
|
|
ASSERT_EQ(Status::OK(), results->status());
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete results;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test33) {
|
|
|
|
auto x = NDArrayFactory::create<double>('c', {4, 3});
|
|
|
|
auto y = NDArrayFactory::create<double>('c', {4, 1});
|
|
|
|
auto exp = NDArrayFactory::create<double>('c',{ 3, 1}, {70, 80, 90});
|
|
|
|
|
|
|
|
x.linspace(1);
|
|
|
|
y.linspace(1);
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&x, &y}, {}, {1, 0});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test34) {
|
|
|
|
auto a = NDArrayFactory::create<double>('c', {3, 4}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12});
|
|
|
|
auto b = NDArrayFactory::create<double>('c', {4}, {1, 2, 3, 4});
|
|
|
|
auto exp = NDArrayFactory::create<double>('c', {3}, {30, 70, 110});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&a, &b});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test35) {
|
|
|
|
auto a = NDArrayFactory::create<double>('c', {4}, {1, 2, 3, 4});
|
|
|
|
auto b = NDArrayFactory::create<double>('c', {4, 3}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12});
|
|
|
|
auto exp = NDArrayFactory::create<double>('c', {3}, {70, 80, 90});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&a, &b});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test36) {
|
|
|
|
auto a = NDArrayFactory::create<double>('c', {1, 4}, {1, 2, 3, 4});
|
|
|
|
auto b = NDArrayFactory::create<double>('c', {4, 3}, {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12});
|
|
|
|
auto exp = NDArrayFactory::create<double>('c', {1, 3}, {70, 80, 90});
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto result = op.evaluate({&a, &b});
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, result->status());
|
|
|
|
|
|
|
|
auto z = result->at(0);
|
|
|
|
|
|
|
|
ASSERT_TRUE(exp.isSameShape(z));
|
|
|
|
ASSERT_TRUE(exp.equalsTo(z));
|
|
|
|
|
|
|
|
delete result;
|
|
|
|
}
|
|
|
|
|
|
|
|
//////////////////////////////////////////////////////////////////////
|
|
|
|
TEST_F(DeclarableOpsTests14, matmul_test37) {
|
|
|
|
|
|
|
|
NDArray a('c', {32, 12, 128, 64}, nd4j::DataType::FLOAT32);
|
|
|
|
NDArray b('c', {32, 12, 128, 64}, nd4j::DataType::FLOAT32);
|
|
|
|
NDArray c('c', {32,12,128,128}, nd4j::DataType::FLOAT32);
|
|
|
|
NDArray cExp('c', {32,12,128,128}, nd4j::DataType::FLOAT32);
|
|
|
|
|
|
|
|
a = 1;
|
|
|
|
b = 1;
|
|
|
|
cExp = 64; //Each entry in output c is sum of 64 (1.0 x 1.0) multiplications
|
|
|
|
|
|
|
|
nd4j::ops::matmul op;
|
|
|
|
auto status = op.execute({&a, &b}, {&c}, {}, {0,1});
|
|
|
|
|
|
|
|
ASSERT_EQ(ND4J_STATUS_OK, status);
|
|
|
|
|
|
|
|
ASSERT_TRUE(cExp.isSameShape(c));
|
|
|
|
ASSERT_TRUE(cExp.equalsTo(c));
|
|
|
|
}
|
|
|
|
|
|
|
|
// @Test
|
|
|
|
// public void testMmulRank4_simple(){
|
|
|
|
|
|
|
|
// INDArray arr1 = Nd4j.ones(DataType.FLOAT, 32, 12, 128, 64);
|
|
|
|
// INDArray arr2 = Nd4j.ones(DataType.FLOAT, 32, 12, 128, 64);
|
|
|
|
|
|
|
|
// DynamicCustomOp op = DynamicCustomOp.builder("matmul")
|
|
|
|
// .addInputs(arr1, arr2)
|
|
|
|
// .addIntegerArguments(0, 1) //Transpose arr2 only
|
|
|
|
// .build();
|
|
|
|
|
|
|
|
// List<LongShapeDescriptor> shapes = op.calculateOutputShape();
|
|
|
|
// assertEquals(1, shapes.size());
|
|
|
|
// long[] shape = new long[]{32,12,128,128};
|
|
|
|
// assertArrayEquals(shape, shapes.get(0).getShape());
|
|
|
|
|
|
|
|
// INDArray out = Nd4j.create(DataType.FLOAT, shape);
|
|
|
|
|
|
|
|
// op.setOutputArgument(0, out);
|
|
|
|
// Nd4j.exec(op);
|
|
|
|
// // System.out.println(out);
|
|
|
|
|
|
|
|
// INDArray exp = Nd4j.valueArrayOf(shape, 64.0, DataType.FLOAT); //Each entry in output is sum of 64 (1.0 x 1.0) multiplications
|
|
|
|
// assertEquals(exp, out);
|
|
|
|
// }
|