Commit Graph

72 Commits (3967e039a5b8dbaea1d5ae4221a068aee89856f8)

Author SHA1 Message Date
Alex Black 133223e865 Merge remote-tracking branch 'eclipse/master' 2020-04-14 22:24:17 +10:00
raver119 3e2dbc65dd
MatMul for gemm/gemv calls (#365)
* libnd4j added optional alpha and beta support to matmul

Signed-off-by: Oleg <oleg.semeniv@gmail.com>

* libnd4j typos fixes

Signed-off-by: Oleg <oleg.semeniv@gmail.com>

* libnd4j add optional alpha and beta to matmul_bp

Signed-off-by: Oleg <oleg.semeniv@gmail.com>

* libnd4j one more typo fix

Signed-off-by: Oleg <oleg.semeniv@gmail.com>

* libnd4j added optional alpha and beta to mkl implementation

Signed-off-by: Oleg <oleg.semeniv@gmail.com>

* MatMul alpha/beta on java side

Signed-off-by: raver119 <raver119@gmail.com>

* alpha/beta fix in libnd4j

Signed-off-by: raver119 <raver119@gmail.com>

* alpha/beta fix in matmul_bp

Signed-off-by: raver119 <raver119@gmail.com>

* restored view validation

Signed-off-by: raver119 <raver119@gmail.com>

* gemv/gemm now use MatMul op

Signed-off-by: raver119 <raver119@gmail.com>

* few tests fixed

Signed-off-by: raver119 <raver119@gmail.com>

* additional INDArray.mmul signature

Signed-off-by: raver119 <raver119@gmail.com>

* make C order default for INDArray.mmul, unless both A/B have F order

Signed-off-by: raver119 <raver119@gmail.com>

* Nd4j.gemm validation fix

Signed-off-by: raver119 <raver119@gmail.com>

* disable mkldnn matmul for xxf with beta != 0 case

Signed-off-by: raver119 <raver119@gmail.com>

* SimpleRnn workspace fix + timeouts

Signed-off-by: Alex Black <blacka101@gmail.com>

* two more tests + minor fix in matmul platform check

Signed-off-by: raver119 <raver119@gmail.com>

* Flaky test fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* propagate testresources profile

Signed-off-by: raver119 <raver119@gmail.com>

* Resources fix + flaky test fix

Signed-off-by: Alex Black <blacka101@gmail.com>

Co-authored-by: Oleg <oleg.semeniv@gmail.com>
Co-authored-by: Alex Black <blacka101@gmail.com>
2020-04-10 17:57:02 +03:00
Andrii T d86dd5b131
DL4J and SameDiff integration tests + LSTMLayer java op class (#353)
* init in this branch

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* Lenetet Mnist workflow

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* small fix for calculations

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* for Alex to check placeholder null pointer issue

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* CNN3D workflow

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* state for launching on dxg to regenterate dl4j examples

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* SD RNN test case workflow

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* small fixes

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* checkpoint at lstmBlock: Input array 1 (x) rank must be got input with rank 2 issue

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* Fix LSTMLayer inputs order

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* lstm mismatch with c++ op issue

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* LSTMLayer config draft

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* LSTMLayer config draft v2

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* have doubt I had to do this

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* NDRNN generated by codegen

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* LSTMLayerTestCases draft

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* minor fixes again

* added LSTMLayer testcases to nd4j-tests + setted Preconditions in LSTMLayer constructors

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* added lost SDCNNtestcases

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* overrided getNumOutputs from DynamicCustomOp in LSTMLayer and reorganized LSTMLayerOutputs according to cpp op

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* finished with LSTMLayerOutputs

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* Fix MKLDNN platform checks (i.e., when MKLDNN can be used vs. not)

Signed-off-by: Alex Black <blacka101@gmail.com>

* Fix LSTMLayerWeights input order

Signed-off-by: Alex Black <blacka101@gmail.com>

* More fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* minor fixes

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* fixed LSTMLayer testcases

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* finished SameDiffRNNTestCase

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* finished all testcases + minor fixes

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* Multiple generation-related fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* Fix multiple issues

Signed-off-by: Alex Black <blacka101@gmail.com>

* More fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* LSTM fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* Regenerate ND4J namespaces and fix multiple issues

Signed-off-by: Alex Black <blacka101@gmail.com>

* changed SameDiffRNNTestCase

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* Small fix

Signed-off-by: Alex Black <blacka101@gmail.com>

* added  Nd4j.getRandom().setSeed(12345) where needed

Signed-off-by: Andrii Tuzhykov <andrewtuzhykov@gmail.com>

* #8828 Fix ND4J profiler NaN/Inf checks when using OpContext

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8828 Fix ND4J profiler NaN/Inf checks when using OpContext

Signed-off-by: Alex Black <blacka101@gmail.com>

* Tweak to weight init for SameDiff CNN test case

Signed-off-by: Alex Black <blacka101@gmail.com>

* Tweaks for test cases

Signed-off-by: Alex Black <blacka101@gmail.com>

* Ignore failing tests until fixed

Signed-off-by: Alex Black <blacka101@gmail.com>

* Fix

Signed-off-by: Alex Black <blacka101@gmail.com>

Co-authored-by: Alex Black <blacka101@gmail.com>
2020-04-09 00:20:48 +10:00
Alexander Stoyakin 0a27e9f41d
Fix incompatibilities with generated code (#303)
* Cholesky fixed

* Constructors added

* MatMul wrapper

* Constructor added

* Missing wrappers added

* Generate Linalg namespace added

* Output data types

* Unit tests

* Added mmul

* Code generation

* Code generated

* Build fixed

* Fixing signatures

* Tests fixed

* Tests fixed

* Added enum

* Fix tests

* Some fixes

* Eye test fixed

* SameDiff: small fix for renameVariable - also replace variable name in lossVariable list if necessary

Signed-off-by: Alex Black <blacka101@gmail.com>

* Some fixes

* Tests fixed

* Revert wrong fix

* Some fixes

* Some fixes

* Extending base test class

* Added pad

* Fixed for generated signatures

* Fixes due to nd4j codegen

* Backwards compatibility fixes

* Fixed errors in tests, reverted wrong changes

* Test fixed

* Added missing operations used for nd4s operators

* Compilation fixed

* Added meshgrid

* Fixed constructors

* fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* Fix bad commit (incorrectly reverted change from master)

Signed-off-by: Alex Black <blacka101@gmail.com>

* Fixed test

Co-authored-by: Alex Black <blacka101@gmail.com>
2020-04-01 12:00:38 +11:00
Alex Black 55a3d9bb2c Fix loading both model and serializer at once from stream + re-add checks
Signed-off-by: Alex Black <blacka101@gmail.com>
2020-03-31 11:56:56 +11:00
Alex Black e3a8629214 2 more try-with-resources (buffered + close)
Signed-off-by: Alex Black <blacka101@gmail.com>
2020-03-31 11:38:12 +11:00
Alex Black 3f49d646fc API fix + try with resources (close file stream + buffered)
Signed-off-by: Alex Black <blacka101@gmail.com>
2020-03-31 11:34:43 +11:00
Alex Black 2e52f0217f Merge branch 'master' of https://github.com/horschi/deeplearning4j into horschi-master 2020-03-31 10:20:31 +11:00
Alex Black 63c9223bc2
Assorted fixes (#318)
* #8777 MultiLayerNetwork.evaluate(MultiDataSetIterator) overload

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8768 SameDiff.equals

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8750 shade freemarker library and switch to it in DL4J UI

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8704 DL4J UI redirect

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8776 RecordReaderDataSetIterator builder collectMetaData fix

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8718 Fix DL4J doEvaluation metadata

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8715 ArchiveUtils - Add option to not log every extracted file

Signed-off-by: Alex Black <blacka101@gmail.com>

* No exception for evaluations that don't support metadata

Signed-off-by: Alex Black <blacka101@gmail.com>

* Fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8765 CompGraph+MDS fix for SharedTrainingMaster

Signed-off-by: Alex Black <blacka101@gmail.com>

* small fix

Signed-off-by: Alex Black <blacka101@gmail.com>

* Timeout

Signed-off-by: Alex Black <blacka101@gmail.com>

* Ignore

Signed-off-by: Alex Black <blacka101@gmail.com>

* Revert freemarker shading

Signed-off-by: Alex Black <blacka101@gmail.com>

* Ignore

Signed-off-by: Alex Black <blacka101@gmail.com>
2020-03-27 00:33:13 +11:00
Fariz Rahman b1bc7df160
tf.keras model import (#258)
* tf op initial

* ..

* protobuf parsing working

* model build working

* test passing

* headers

* conffix

* service loader + tests

* revert cuda version

* msg

* override

* refacc

* pom

* rem bad import

* dtype fix + const cast caaching

* rem unnecessary fields

* rem println

* rem dep

* refacc

* rem redundant arg

* Ignore TFOpLayer in DTypeTests

Signed-off-by: Alex Black <blacka101@gmail.com>

Co-authored-by: Alex Black <blacka101@gmail.com>
2020-03-24 20:37:27 +11:00
Alex Black a80fb99a5f
DL4J integrations tests updates + add SameDiff support (#298)
* Revive and start updating DL4J integration tests

Signed-off-by: Alex Black <blacka101@gmail.com>

* Add SameDiff support - first pass

Signed-off-by: Alex Black <blacka101@gmail.com>

* SameDiff test case generation

Signed-off-by: Alex Black <blacka101@gmail.com>

* SameDiff integration tests polishing

Signed-off-by: Alex Black <blacka101@gmail.com>

* More SameDiff integration test fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* Final polish

Signed-off-by: Alex Black <blacka101@gmail.com>

* Small test tweak

Signed-off-by: Alex Black <blacka101@gmail.com>
2020-03-07 22:44:41 +11:00
Alex Black 483c3d7b8c
Assorted SameDiff/DL4J fixes (#279)
* #8565 Normalizer toString/hashcode

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8731 ImagePreProcessingScaler lables/segmentation fix

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8691 Fix SameDiffLayer/Vertx finetuning and parameter setting support

Signed-off-by: Alex Black <blacka101@gmail.com>

* #8663 DL4J embedding layer weight init - don't depend on vocab size

Signed-off-by: Alex Black <blacka101@gmail.com>

* EmbeddingLayer test tweak

Signed-off-by: Alex Black <blacka101@gmail.com>
2020-03-02 16:15:49 +11:00
cspriegel 1df8899fb1 restoreMultiLayerNetwork() needlessly writes temp-file #8735
Signed-off-by: cspriegel <christian.spriegel@gmail.com>
2020-02-26 16:24:32 +01:00
Shams Ul Azeem f165160edb
MultiLayerConfiguration should be ComputationGraphConfiguration for validating a cg model conf (#240) 2020-02-14 12:06:00 +11:00
Alex Black 569a46f87d
Fixes (#213)
* Increase timeouts for 2 tests occasionally failing on CI

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Explicitly set character encoding via argline for maven surefire tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* CUDA gradient check timeout fix + simple rnn masking fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2020-02-05 17:07:36 +11:00
raver119 5d28e6143d
OpContext handling (#214)
* nano tweaks

Signed-off-by: raver119 <raver119@gmail.com>

* OpContext tweaks

Signed-off-by: raver119 <raver119@gmail.com>

* OpContext deallocators

Signed-off-by: raver119 <raver119@gmail.com>

* get rid of few mkldnn safety checks

Signed-off-by: raver119 <raver119@gmail.com>

* databuffer setSpecial fix

Signed-off-by: raver119 <raver119@gmail.com>
2020-02-05 07:27:24 +03:00
Alex Black 57d5eb473b
Fixes for global pooling + masking with different mask datatypes (#212)
* Fixes for global pooling + masking with different mask datatypes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Global pooling backprop dtype fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2020-02-04 15:38:06 +11:00
raver119 7ef0ef907e
Packages fix (#193)
* packages fix

Signed-off-by: raver119 <raver119@gmail.com>

* few imports fixed

Signed-off-by: raver119 <raver119@gmail.com>

* few imports fixed

Signed-off-by: raver119 <raver119@gmail.com>
2020-01-27 23:04:21 +03:00
Alexander Stoyakin 4db28a9300 Cleanup of multiple projects (#175)
* Cleanup modules

* Moving subprojects to nd4j-api

* Project cleanup

* Dropped AWS sub-project

* dl4j-util moved to core

* dl4j-perf moved to core

* Tests coverage

* Revert "Moving subprojects to nd4j-api"

This reverts commit bc6eb573c6b60c407ade47172c5d204725077e6b.

* Moved nd4j-buffer and nd4j-context to nd4j-api

* Rolled back change

* Revert "Project cleanup"

This reverts commit 64ac7f369b2d968f7be437718034f093fc886ffc.

* Datavec cleaned up

* Revert "Moved nd4j-buffer and nd4j-context to nd4j-api"

This reverts commit 75f4e8da80d2551e44e1251dd6c5923289fff8e1.

# Conflicts:
#	nd4j/nd4j-backends/nd4j-tests/src/test/java/org/nd4j/autodiff/opvalidation/ReductionBpOpValidation.java

* Resolve conflict

* Compilation fixed.

* nd4j-context and nd4j-buffer moved to nd4j-api

* Fixed TF mapping for mmul

* Fix for dl4j-cuda tests

Signed-off-by: Alex Black <blacka101@gmail.com>

* Move last few tests from deeplearning4j-nn to -core

Signed-off-by: Alex Black <blacka101@gmail.com>

* Remove incorrect TF import mapping for TensorMmul op

Signed-off-by: Alex Black <blacka101@gmail.com>

* Cleaned TF mapping

* Fix path for test results on windows

* Remove old dependency

Signed-off-by: Alex Black <blacka101@gmail.com>

* One more attempt to fix path for test results on windows

* fixup! One more attempt to fix path for test results on windows

* fixup! One more attempt to fix path for test results on windows

Co-authored-by: Alex Black <blacka101@gmail.com>
Co-authored-by: Serhii Shepel <9946053+sshepel@users.noreply.github.com>
Co-authored-by: raver119 <raver119@gmail.com>
2020-01-24 22:35:00 +03:00
Alex Black a25bb6a11c
Unit/integration test split + test speedup (#166)
* Add maven profile + base tests methods for integration tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Switch from system property to environment variable; seems more reliable in intellij

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add nd4j-common-tests module, and common base test; cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Ensure all ND4J tests extend BaseND4JTest

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Test spam reduction, import fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add test logging to nd4j-aeron

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix unintended change

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Reduce sprint test log spam

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More test spam cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Significantly speed up TSNE tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* W2V iterator test unit/integration split

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More NLP test speedups

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Avoid debug/verbose mode leaking between tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* test tweak

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Arbiter extends base DL4J test

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Arbiter test speedup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* nlp-uima test speedup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More test speedups

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix ND4J base test

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Few small ND4J test speed improvements

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J tests speedup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More tweaks

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Even more test speedups

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More tweaks

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Various test fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* More test fixes

Signed-off-by: Alex Black <blacka101@gmail.com>

* Add ability to specify number of threads for C++ ops in BaseDL4JTest and BaseND4JTest

Signed-off-by: Alex Black <blacka101@gmail.com>

* nd4j-aeron test profile fix for CUDA

Signed-off-by: Alex Black <blacka101@gmail.com>
2020-01-22 22:27:01 +11:00
raver119 29e8e09db6
String changes (#3)
* initial commit

* additional data types & tensor type

Signed-off-by: raver119 <raver119@gmail.com>

* next step

Signed-off-by: raver119 <raver119@gmail.com>

* missing include

* sparse_to_dense

Signed-off-by: raver119 <raver119@gmail.com>

* few more tests files

Signed-off-by: raver119 <raver119@gmail.com>

* draft

Signed-off-by: raver119 <raver119@gmail.com>

* numeric sparse_to_dense

Signed-off-by: raver119 <raver119@gmail.com>

* comment

Signed-off-by: raver119 <raver119@gmail.com>

* string sparse_to_dense version

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA DataBuffer expand

Signed-off-by: raver119 <raver119@gmail.com>

* few tweaks for CUDA build

Signed-off-by: raver119 <raver119@gmail.com>

* shape fn for string_split

Signed-off-by: raver119 <raver119@gmail.com>

* one more comment

Signed-off-by: raver119 <raver119@gmail.com>

* string_split indices

Signed-off-by: raver119 <raver119@gmail.com>

* next step

Signed-off-by: raver119 <raver119@gmail.com>

* test passes

Signed-off-by: raver119 <raver119@gmail.com>

* few rearrangements for databuffer implementations

Signed-off-by: raver119 <raver119@gmail.com>

* DataBuffer: move inline methods to common implementations

Signed-off-by: raver119 <raver119@gmail.com>

* add native DataBuffer to Nd4j presets

Signed-off-by: raver119 <raver119@gmail.com>

* DataBuffer creation

Signed-off-by: raver119 <raver119@gmail.com>

* use DataBuffer for allocation

Signed-off-by: raver119 <raver119@gmail.com>

* cpu databuffer as deallocatable

Signed-off-by: raver119 <raver119@gmail.com>

* DataBuffer setters for bufers

Signed-off-by: raver119 <raver119@gmail.com>

* couple of wrappers

Signed-off-by: raver119 <raver119@gmail.com>

* DataBuffers being passed around

Signed-off-by: raver119 <raver119@gmail.com>

* Bunch of ByteBuffer-related signatures gone

Signed-off-by: raver119 <raver119@gmail.com>

* - few more Nd4j signatures removed
- minor fix for bfloat16

Signed-off-by: raver119 <raver119@gmail.com>

* nullptr pointer is still a pointer, but 0 as address :)

Signed-off-by: raver119 <raver119@gmail.com>

* one special test

Signed-off-by: raver119 <raver119@gmail.com>

* empty string array init

Signed-off-by: raver119 <raver119@gmail.com>

* one more test in cpp

Signed-off-by: raver119 <raver119@gmail.com>

* memcpy instead of databuffer swap

Signed-off-by: raver119 <raver119@gmail.com>

* special InteropDataBuffer for front-end languages

Signed-off-by: raver119 <raver119@gmail.com>

* few tweaks for java

Signed-off-by: raver119 <raver119@gmail.com>

* pointer/indexer actualization

Signed-off-by: raver119 <raver119@gmail.com>

* CustomOp returns list for inputArumgents and outputArguments instead of array

Signed-off-by: raver119 <raver119@gmail.com>

* redundant call

Signed-off-by: raver119 <raver119@gmail.com>

* print_variable op

Signed-off-by: raver119 <raver119@gmail.com>

* - view handling (but wrong one)
- print_variable java wrapper

Signed-off-by: raver119 <raver119@gmail.com>

* one more test

Signed-off-by: raver119 <raver119@gmail.com>

* - empty arrays handling

Signed-off-by: raver119 <raver119@gmail.com>

* - deserialization works now

Signed-off-by: raver119 <raver119@gmail.com>

* minor fix

Signed-off-by: raver119 <raver119@gmail.com>

* meh

Signed-off-by: raver119 <raver119@gmail.com>

* one more fix

Signed-off-by: raver119 <raver119@gmail.com>

* initial cuda commit

Signed-off-by: raver119 <raver119@gmail.com>

* print_variable message validation

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA views

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA special buffer size

Signed-off-by: raver119 <raver119@gmail.com>

* minor update to match master changes

Signed-off-by: raver119 <raver119@gmail.com>

* - consider arrays always actual on device for CUDA
- additional PrintVariable constructor
- CudaUtf8Buffer now allocates host buffer by default

Signed-off-by: raver119 <raver119@gmail.com>

* meh

Signed-off-by: raver119 <raver119@gmail.com>

* - print_variable now allows print from device

Signed-off-by: raver119 <raver119@gmail.com>

* InteropDataBuffer data type fix

Signed-off-by: raver119 <raver119@gmail.com>

* ...

Signed-off-by: raver119 <raver119@gmail.com>

* disable some debug messages

Signed-off-by: raver119 <raver119@gmail.com>

* master pulled in

Signed-off-by: raver119 <raver119@gmail.com>

* couple of new methods for DataBuffer interop

Signed-off-by: raver119 <raver119@gmail.com>

* java side

Signed-off-by: raver119 <raver119@gmail.com>

* offsetted constructor

Signed-off-by: raver119 <raver119@gmail.com>

* new CUDA deallocator

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA backend torn apart

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA backend torn apart 2

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA backend torn apart 3

Signed-off-by: raver119 <raver119@gmail.com>

* - few new tests
- few new methods for DataBuffer management

Signed-off-by: raver119 <raver119@gmail.com>

* few more tests + few more tweaks

Signed-off-by: raver119 <raver119@gmail.com>

* two failing tests

Signed-off-by: raver119 <raver119@gmail.com>

* one more test

Signed-off-by: raver119 <raver119@gmail.com>

* two failing tests pass

Signed-off-by: raver119 <raver119@gmail.com>

* now we pass DataBuffer to legacy ops too

Signed-off-by: raver119 <raver119@gmail.com>

* Native DataBuffer for legacy ops, Java side

Signed-off-by: raver119 <raver119@gmail.com>

* CPU java side update

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA java side update

Signed-off-by: raver119 <raver119@gmail.com>

* no more prepare/register action on java side

Signed-off-by: raver119 <raver119@gmail.com>

* NDArray::prepare/register use now accepts vectors

Signed-off-by: raver119 <raver119@gmail.com>

* InteropDataBuffer now has few more convenience methods

Signed-off-by: raver119 <raver119@gmail.com>

* java bindings update

Signed-off-by: raver119 <raver119@gmail.com>

* tick device in NativeOps

Signed-off-by: raver119 <raver119@gmail.com>

* Corrected usage of OpaqueBuffer for tests.

* Corrected usage of OpaqueBuffer for java tests.

* NativeOpsTests fixes.

* print_variable now returns scalar

Signed-off-by: raver119 <raver119@gmail.com>

* one more test

Signed-off-by: raver119 <raver119@gmail.com>

* compat_string_split fix for CUDA

Signed-off-by: raver119 <raver119@gmail.com>

* - CUDA execScalar fix
- CUDA lazyAllocateHostPointer now checks java indexer/pointer instead of native pointer

Signed-off-by: raver119 <raver119@gmail.com>

* legacy ops DataBuffer migration prototype

Signed-off-by: raver119 <raver119@gmail.com>

* ignore device shapeinfo coming from java

Signed-off-by: raver119 <raver119@gmail.com>

* minor fix

Signed-off-by: raver119 <raver119@gmail.com>

* minor transformAny fix

Signed-off-by: raver119 <raver119@gmail.com>

* minor tweak for lazy host allocation

Signed-off-by: raver119 <raver119@gmail.com>

* - DataBuffer::memcpy method
- bitcast now uses memcpy

Signed-off-by: raver119 <raver119@gmail.com>

* - IndexReduce CUDA dimension buffer fix

Signed-off-by: raver119 <raver119@gmail.com>

* views for CPU and CUDA

Signed-off-by: raver119 <raver119@gmail.com>

* less spam

Signed-off-by: raver119 <raver119@gmail.com>

* optional memory init

Signed-off-by: raver119 <raver119@gmail.com>

* async memset

Signed-off-by: raver119 <raver119@gmail.com>

* - SummaryStats CUDA fix
- DataBuffer.sameUnderlyingData() impl
- execBroadcast fix

Signed-off-by: raver119 <raver119@gmail.com>

* - reduce3All fix
switch to CUDA 10 temporarily

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA version

Signed-off-by: raver119 <raver119@gmail.com>

* proper memory deallocator registration

Signed-off-by: raver119 <raver119@gmail.com>

* HOST_ONLY workspace allocation

Signed-off-by: raver119 <raver119@gmail.com>

* temp commit

Signed-off-by: raver119 <raver119@gmail.com>

* few conflicts resolved

Signed-off-by: raver119 <raver119@gmail.com>

* few minor fixes

Signed-off-by: raver119 <raver119@gmail.com>

* one more minor fix

Signed-off-by: raver119 <raver119@gmail.com>

* NDArray permute should operate on JVM primitives

Signed-off-by: raver119 <raver119@gmail.com>

* - create InteropDataBuffer for shapes as well
- update pointers after view creation in Java

Signed-off-by: raver119 <raver119@gmail.com>

* - addressPointer temporary moved to C++

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA: don't account offset twice

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA: DataBuffer pointer constructor updated

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA NDArray.unsafeDuplication() simplified

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA minor workspace-related fixes

Signed-off-by: raver119 <raver119@gmail.com>

* CPU DataBuffer.reallocate()

Signed-off-by: raver119 <raver119@gmail.com>

* print_affinity op

Signed-off-by: raver119 <raver119@gmail.com>

* print_affinity java side

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA more tweaks for data locality

Signed-off-by: raver119 <raver119@gmail.com>

* - compat_string_split tweak
- CudaUtf8Buffer update

Signed-off-by: raver119 <raver119@gmail.com>

* INDArray.close() mechanic restored

Signed-off-by: raver119 <raver119@gmail.com>

* one more test fixed

Signed-off-by: raver119 <raver119@gmail.com>

* - CUDA DataBuffer.reallocate() updated
- cudaMemcpy (synchronous) restored

Signed-off-by: raver119 <raver119@gmail.com>

* one last fix

Signed-off-by: raver119 <raver119@gmail.com>

* bad import removed

Signed-off-by: raver119 <raver119@gmail.com>

* another small fix

Signed-off-by: raver119 <raver119@gmail.com>

* one special test

Signed-off-by: raver119 <raver119@gmail.com>

* fix bad databuffer size

Signed-off-by: raver119 <raver119@gmail.com>

* release primaryBuffer on replace

Signed-off-by: raver119 <raver119@gmail.com>

* higher timeout

Signed-off-by: raver119 <raver119@gmail.com>

* disable timeouts

Signed-off-by: raver119 <raver119@gmail.com>

* dbCreateView now validates offset and length of a view

Signed-off-by: raver119 <raver119@gmail.com>

* additional validation for dbExpand

Signed-off-by: raver119 <raver119@gmail.com>

* restore timeout back again

Signed-off-by: raver119 <raver119@gmail.com>

* smaller distribution for rng test to prevent timeouts

Signed-off-by: raver119 <raver119@gmail.com>

* CUDA DataBuffer::memcpy now copies to device all the time

Signed-off-by: raver119 <raver119@gmail.com>

* OpaqueDataBuffer now contains all required methods for interop

Signed-off-by: raver119 <raver119@gmail.com>

* some javadoc

Signed-off-by: raver119 <raver119@gmail.com>

* GC on failed allocations

Signed-off-by: raver119 <raver119@gmail.com>

* minoe memcpu tweak

Signed-off-by: raver119 <raver119@gmail.com>

* one more bitcast test

Signed-off-by: raver119 <raver119@gmail.com>

* - NDArray::deviceId() propagation
- special multi-threaded test for data locality checks

Signed-off-by: raver119 <raver119@gmail.com>

* DataBuffer additional syncStream

Signed-off-by: raver119 <raver119@gmail.com>

* DataBuffer additional syncStream

Signed-off-by: raver119 <raver119@gmail.com>

* one ignored test

Signed-off-by: raver119 <raver119@gmail.com>

* skip host alloc for empty arrays

Signed-off-by: raver119 <raver119@gmail.com>

* ByteBuffer support is back

Signed-off-by: raver119 <raver119@gmail.com>

* DataBuffer::memcpy minor fix

Signed-off-by: raver119 <raver119@gmail.com>

* few minor prelu/bp tweaks

Signed-off-by: raver119 <raver119@gmail.com>

* nullify-related fixes

Signed-off-by: raver119 <raver119@gmail.com>

* PReLU fixes (#157)

Signed-off-by: Alex Black <blacka101@gmail.com>

* Build fixed

* Fix tests

* one more ByteBuffer signature restored

Signed-off-by: raver119 <raver119@gmail.com>

* nd4j-jdbc-hsql profiles fix

Signed-off-by: raver119 <raver119@gmail.com>

* nd4j-jdbc-hsql profiles fix

Signed-off-by: raver119 <raver119@gmail.com>

* PReLU weight init fix

Signed-off-by: Alex Black <blacka101@gmail.com>

* Small PReLU fix

Signed-off-by: Alex Black <blacka101@gmail.com>

* - INDArray.migrate() reactivated
- DataBuffer::setDeviceId(...) added
- InteropDataBuffer Z syncToDevice added for views

Signed-off-by: raver119 <raver119@gmail.com>

* missed file

Signed-off-by: raver119 <raver119@gmail.com>

* Small tweak

Signed-off-by: Alex Black <blacka101@gmail.com>

* cuda 10.2

Signed-off-by: raver119 <raver119@gmail.com>

* minor fix

Signed-off-by: raver119 <raver119@gmail.com>

Co-authored-by: shugeo <sgazeos@gmail.com>
Co-authored-by: Alex Black <blacka101@gmail.com>
Co-authored-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2020-01-04 13:27:50 +03:00
Alex Black 29104083cc
Various fixes (#143)
* #8568 ArrayUtil optimization

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #6171 Keras ReLU and ELU support

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Keras softmax layer import

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8549 Webjars dependency management

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix for TF import names ':0' suffix issue / NPE

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* BiasAdd: fix default data format for TF import

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Update zoo test ignores

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8509 SameDiff Listener API - provide frame + iteration

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8520 ND4J Environment

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Deconv3d

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Deconv3d fixes + gradient check

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Conv3d fixes + deconv3d DType test

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix issue with deconv3d gradinet check weight init

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8579 Fix BaseCudaDataBuffer constructor fix for UINT16

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DataType.isNumerical() returns false for BOOL type

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8504 Reduce Spark log spam for tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Clean up DL4J gradient check test spam

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More Gradient check spam reduction

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* SameDiff test spam reduction

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fixes for FlatBuffers mapping

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* SameDiff log spam cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Tests should extend BaseNd4jTest

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Remove debug line in c++ op

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* ND4J test spam cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J test spam reduction

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More Dl4J and datavec test spam cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix for bad conv3d test

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Additional test

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Embedding layers: don't inherit global default activation function

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Trigger CI

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Consolidate all BaseDL4JTest classes to single class used everywhere; make timeout configurable per class

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Test fixes and timeout increases

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Timeouts and PReLU fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Restore libnd4j build threads arg for CUDA build

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Increase timeouts on a few tests to avoid spurious failures on some CI machines

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More timeout fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More test timeout fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Tweak timeout for one more test

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Final tweaks

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* One more ignore

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2020-01-04 13:45:07 +11:00
Alex Black 9cc8803b8d
DL4J + Keras import: Causal Conv1D support (#107)
* Keras causal conv1d support first steps

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Causal conv mode

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Gradient check and fixes for causal conv1d

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix Conv1D import and testing

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small keras test fix

Signed-off-by: Alex Black <blacka101@gmail.com>

* Don't allow setting causal convolution mode to conv2d/3d layers

Signed-off-by: Alex Black <blacka101@gmail.com>

* More robustly infer nIn for recurrent layers for ambiguous NCW and NWC cases

Signed-off-by: Alex Black <blacka101@gmail.com>

* Polish and cleanup

Signed-off-by: Alex Black <blacka101@gmail.com>
2019-12-04 22:52:06 +11:00
Samuel Audet 5e07998e59 Add support for CUDA 10.2 (#89) 2019-11-29 16:31:03 +11:00
Alex Black 8843c7377a
Update shaded Jackson version to 2.10.1 (#82)
* Update shaded Jackson version to 2.10.1

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Remove no longer needed scala compiler plugin from UI

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix op name for BitwiseAnd op

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* TimeDistributedLayer mask array fix + test

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-26 19:24:38 +11:00
Alex Black 5b2ee72673
DL4J Time Distributed + fixes + Vertx module profiles fix (#78)
* Add test profiles to vertx module

* Arbiter test tweaks

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add TimeDistributed wrapper layer

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Tests for TimeDistributed layer

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small test dependency exclusion for Spark module

* Fixes, more thorough tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-25 16:00:21 +11:00
Alex Black 8d87b078c2
Fix master (#74)
* Remove non-existent cases

Signed-off-by: Paul Dubs <paul.dubs@gmail.com>

* Add dependency for non-oracle jdk

Signed-off-by: Paul Dubs <paul.dubs@gmail.com>
2019-11-23 12:09:37 +11:00
Alex Black a856922fe9
#8409 Fix compgraph backprop issue with dual embedding layers from single input (#52)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-16 23:09:41 +11:00
Alex Black 09a827fb6d
Fixes and pre-release QA (#51)
* #8395 Keras import - support scaled identity weight init

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More Keras scaled weight init fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8352 Deprecate duplicate SamplingDataSetIterator class

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Remove /O2 optimization for faster CUDA build

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Tweak regression test precision for CUDA

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix edge cases for buffer creation

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Update MKLDNN validation tests to new helper enable/disable settings

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Delete debugging class

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* MKLDNN test - add proper skip for CUDA backend

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Align WeightInitUtil with weight init classes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix for SameDiff test layers weight init when using IWeightInit classes

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-16 17:04:29 +11:00
Alex Black 47d19908f4
Various fixes (#43)
* #8172 Enable DL4J MKLDNN batch norm backward pass

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8382 INDArray.toString() rank 1 brackets / ambiguity fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8308 Fix handful of broken links (inc. some in errors)

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Unused dependencies, round 1

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Unused dependencies, round 2

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Unused dependencies, round 3

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Uniform distribution TF import fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-14 19:38:20 +11:00
raver119 48df1acdfb
[WIP] ThreadPool (#8)
This PR removes OpenMP use in 95% of cases
2019-11-13 17:04:59 +03:00
Alex Black df8b4e607a
SameDiff: make use of DeviceLocal configurable (#32)
* #8340 make DeviceLocal configurable

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Javadoc

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J SameDiff layers: use SingleThreadArrayHolder to avoid assigns + DeviceLocalNDArray overhead

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Javadoc

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-06 18:52:41 +11:00
Alex Black 9efd811508
Use DL4J workspaces for SameDiff layers in MLN/CG (#23)
* #8329 DL4J workspace integration for SameDiff layers

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix bug for Nd4j.createUninitializedDetached for scalars (length 0 shape array)

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* SameDiff output layer, graph vertex, various fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Javadoc

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-02 17:42:01 +11:00
Alex Black d82877b18b
Various SameDiff fixes (#21)
* MKLDNN LSTM forward implementation (disabled pending #8331)

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8318 add SameDiff.calculateGradientsAndOutputs

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Disable mkldnn backprop for now - pending fix, issue #8335

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8337 Fix CudaExecutioner unnecessary result array allocation/replacement

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small FlatBuffers serde fix, UInt8

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8135 ImagePreProcessingScaler - add segmentation support

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8319 Ensure listeners are called when they are supposed to be called

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8214 UNet (non-pretrained) last conv layer kernal size fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-02 11:25:53 +11:00
Alexander Stoyakin 45a40c8a89
DL4J/ND4J: Do pass on integer casts (#15)
* Int cast fixes.

* Revert "Int cast fixes."

This reverts commit aa36e8ca

* Int casts

* Int cast

* Int casts

* Get rid of int casts. Dropping deprecated aggregate ops.

* java scatterUpdate changes

Signed-off-by: raver119 <raver119@gmail.com>

* c++ scatterUpdate changes

Signed-off-by: raver119 <raver119@gmail.com>

* Remove aggregated ops.

* Restored test

* Tests restored.

* Minor fixes
2019-10-31 11:23:09 +02:00
Alex Black d333d29099
SameDiff cleanup and fixes (#12)
* #8160 Remove resolvePrepertiesFromSameDiffBeforeExecution

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* SameDiff API cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More SameDiff cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8248 Switch SameDiff variable init from lazy to creation time for more predictable behaviour

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8252 TanhDerivative javadoc

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8225 Deconvolution2D input validation

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8265 Switch SameDiff.outputs() to user settable, instead of unreliable 'best guess'

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8224 SameDiff.zero and .one create constants, not variables

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More cleanup and fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small test fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J SameDiff fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Re-add hack for Deconvolution2DLayer until #8315 is resolved

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8270 Move CUDA device/version logging to Java; can be disabled via existing org.nd4j.log.initialization system property

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* All ND4J init logging checks system property

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small tweak

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Remove redundant device logging

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* One more fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* UX improvements

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Deconv fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add deconv tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Remove debug code

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-10-26 12:38:08 +11:00
Alexander Stoyakin d98784197a
[WIP] Some optimizations in dl4j (#11)
* Minor optimization.

* Reduce number of objects.

* Extend arrays when limit reached

* Test

* Some fixes.

* Small fix

* Wrong condition fixed.

* Fixes of reallocation.

* Small fix.

* Tests

* Clean up

* Test added.

* Tests and some fixes.

* Test

* Test fixed.

* Conflict fixed.

* UX improved
2019-10-24 14:15:50 +03:00
Alex Black 3f0b4a2d4c
SameDiff execution, TF and memory management overhaul (#10)
* SameDiff execution memory management improvements, round 1

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Round 2

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Round 3

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Clear node outputs closed array references; Slight change to OpValidation internals to not rely on cached op outputs

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Next steps

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Next step

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More polish

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add WeakIdentityHashmap

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Session fixes for control ops and next steps

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* First steps for training session + in-line updating

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Next steps

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix losses and history during training

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* BiasAdd and other fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Don't use SDVariable.getArr() in TFGraphTestAllHelper (import tests)

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* First steps for new dependency tracking approach

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Start integrating dependency tracking for memory management

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Non-control op dependency tracking works/passes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Switch/merge

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Next steps

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup and next steps

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix issue dependency tracking for initial variables/constants

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add check for aliases when determining if safe to close array

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* First pass on new TF graph import class

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Import fixes, op fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup and fixes for new TF import mapper

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Partial implementation of new dependency tracker

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Next steps

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* AbstractDependencyTracker for shared code

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Overhaul SameDiff graph execution (dependency tracking)

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More fixes, cleanup, next steps

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Ad no-op memory manager, cleanup, fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix switch dependency tracking

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* INDArray.toString: no exception on closed arrays, just note closed

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix enter and exit dependency tracking

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* TensorArray memory management fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add unique ID for INDArray instances

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix memory management for NextIteration outputs in multi-iteration loops

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Remove (now unnecessary) special case handling for nested enters

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Handle control dependencies during execution; javadoc for memory managers

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup, polish, code comments, javadoc

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup and more javadoc

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add memory validation for all TF import tests - ensure all arrays (except outputs) are released

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Clean up arrays waiting on unexecuted ops at the end of execution

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fixes for enter op memory managent in the context of multiple non-nested loops/frames

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix order of operation issues for dependency tracker

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Always clear op fields after execution to avoid leaks or unintended array reuse

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Re-implement dtype conversion

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix for control dependencies execution (dependency tracking)

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix TF import overrides and filtering

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix for constant enter array dependency tracking

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J Fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More DL4J fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup and polish

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More polish and javadoc

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More logging level tweaks, small DL4J fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fix to DL4J SameDiffLayer

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix empty array deserialization, add extra deserialization checks

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* FlatBuffers control dep serialization fixes; test serialization as part of all TF import tests

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Variable control dependencies serialization fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix issue with removing inputs for ops

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* FlatBuffers NDArray deserialization fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* FlatBuffers NDArray deserialization fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Final cleanup/polish

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-10-23 21:19:50 +11:00
Robert Altena f25e3e71e5 remove lengthLong (#236)
Signed-off-by: Robert Altena <Rob@Ra-ai.com>
2019-09-05 11:19:38 +10:00
Alex Black d41018751b
Small fix (#233)
Signed-off-by: Alex Black <blacka101@gmail.com>
2019-09-04 19:11:17 +10:00
Alex Black 3f3b676ce5
DL4J Fixes (#204)
* Fix issue with recently introduced exception handling system in MultiLayerNetwork/ComputationGraph

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix for SpaceToBatch layer

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* #8133 DL4J SpaceToBatch gradient fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-30 23:00:53 +10:00
Alex Black 130c9aa536
Fixes (#202)
* Remove unnecessary dup in MKLDNNSubsamplingHelper

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* gson dep fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-30 16:36:14 +10:00
Alex Black dcc2baa676
Version upgrades (#199)
* DataVec fixes for Jackson version upgrade

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J jackson updates + databind version 2.9.9.3

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Shade snakeyaml along with jackson

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Version fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Switch DataVec legacy JSON format handling to mixins

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Next set of fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup for legacy JSON mapping

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Upgrade commons compress to 1.18; small test fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* New Jackson backward compatibility for DL4J - Round 1

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* New Jackson backward compatibility for DL4J - Round 2

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More fixes, all but legacy custom passing

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Provide an upgrade path for custom layers for models in pre-1.0.0-beta JSON format

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Legacy deserialization cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small amount of polish - legacy JSON

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Upgrade guava version

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* IEvaluation legacy format deserialization fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Upgrade play version to 2.7.3

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Update nd4j-parameter-server-status to new Play API

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Update DL4J UI for new play version

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More play framework updates

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Remove Spark 1/2 adapter code from DataVec

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* datavec-spark dependency cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J spark updates, pt 1

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J spark updates, pt 2

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J spark updates, pt 3

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J spark updates, pt 4

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Test fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Another fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Breeze upgrade, dependency cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add Scala 2.12 version to pom.xml

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* change-scala-versions.sh - add scala 2.12, remove 2.10

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Move Spark version properties to parent pom (now that only one spark version is supported)

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DataVec Play fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* datavec play dependency fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Clean up old spark/jackson stuff

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup jackson unused dependencies

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Dropping redundant dependency

Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>

* Removed scalaxy dependency

Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>

* DataVec fixes for Jackson version upgrade

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J jackson updates + databind version 2.9.9.3

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Shade snakeyaml along with jackson

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Version fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Switch DataVec legacy JSON format handling to mixins

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Next set of fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup for legacy JSON mapping

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Upgrade commons compress to 1.18; small test fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* New Jackson backward compatibility for DL4J - Round 1

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* New Jackson backward compatibility for DL4J - Round 2

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More fixes, all but legacy custom passing

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Provide an upgrade path for custom layers for models in pre-1.0.0-beta JSON format

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Legacy deserialization cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small amount of polish - legacy JSON

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Upgrade guava version

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* IEvaluation legacy format deserialization fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Upgrade play version to 2.7.3

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Update nd4j-parameter-server-status to new Play API

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Update DL4J UI for new play version

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* More play framework updates

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Remove Spark 1/2 adapter code from DataVec

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* datavec-spark dependency cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J spark updates, pt 1

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J spark updates, pt 2

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J spark updates, pt 3

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J spark updates, pt 4

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Test fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Another fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Breeze upgrade, dependency cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add Scala 2.12 version to pom.xml

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* change-scala-versions.sh - add scala 2.12, remove 2.10

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Move Spark version properties to parent pom (now that only one spark version is supported)

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DataVec Play fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* datavec play dependency fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Clean up old spark/jackson stuff

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Cleanup jackson unused dependencies

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Add shaded guava

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Dropping redundant dependency

Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>

* Removed scalaxy dependency

Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>

* Ensure not possible to import pre-shaded classes, and remove direct guava dependencies in favor of shaded

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* ND4J Shaded guava import fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DataVec and DL4J guava shading

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Arbiter, RL4J fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Build fixed

Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>

* Fix dependency

Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>

* Fix bad merge

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Jackson shading fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Set play secret, datavec-spark-inference-server

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Fix for datavec-spark-inference-server

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Arbiter fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Arbiter fixes

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* Small test fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-30 14:35:27 +10:00
Ryan Nett f40bdcf885 Dl4j LSTM and Dropout CuDNN fallback and options (#152)
* add fallback for Conv layer activation

Signed-off-by: Ryan Nett <rnett@skymind.io>

* add fallback and config option for LSTM layers

Signed-off-by: Ryan Nett <rnett@skymind.io>

* add fallback option and setting for dropout

Signed-off-by: Ryan Nett <rnett@skymind.io>

* fix comments and error messages

Signed-off-by: Ryan Nett <rnett@skymind.io>

* move helper fail count to layer instance

Signed-off-by: Ryan Nett <rnett@skymind.io>

* ignore helperCountFail for equals and json

Signed-off-by: Ryan Nett <rnett@skymind.io>

* typo fix (MLK -> MKL)

Signed-off-by: Ryan Nett <rnett@skymind.io>

* add MKLDNN to error messages

Signed-off-by: Ryan Nett <rnett@skymind.io>

* add helperAllowFallback to builders, deprecate cudnnAllowFallback

Signed-off-by: Ryan Nett <rnett@skymind.io>
2019-08-29 13:05:01 +10:00
Alex Black fd22a8ecc7
Small build fix, after last PR (#177)
Signed-off-by: Alex Black <blacka101@gmail.com>
2019-08-27 19:46:26 +10:00
Alex Black b85238a6df
MLN/CG: Don't swallow exceptions if a second exception occurs during workspace closing (#161)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-24 17:33:11 +10:00
Alex Black 8e3d569f18
Small fixes to subsampling layer (#158)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-23 22:50:07 +10:00
Alex Black 70ee8ba91d
Optimization / fix for DL4J SameDiff layers (#156)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-23 20:54:24 +10:00
Alex Black e855e47f73
More fixes (#148)
* Small batch norm fix (cuda/no-mkldnn)

Signed-off-by: Alex Black <blacka101@gmail.com>

* Dropout fix for RnnOutputLayer

Signed-off-by: Alex Black <blacka101@gmail.com>

* Allow block size < 2 in batch_to_space_nd and space_to_batch_nd for import, in spite of what TF docs say

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-22 19:55:27 +10:00
Alex Black 9c2bfc9863
Various fixes (DL4J, ND4J) (#147)
* Import fixes, IsMax dtype calc, small test fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* SubsamplingLayer fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>

* DL4J - SpaceToBatch layer updates

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-22 16:16:03 +10:00