Commit Graph

256 Commits (c32acb2ec72bfb57492518318431311cc300c702)

Author SHA1 Message Date
Alexander Stoyakin 010744ef9c Lu wrapper and tests fixes (#144)
* Tests fixed

* Lu added

* Test fixed

* Default timeout

* Tests timeouts fixed.

* TF import fix

* Timeouts added

* Timeout fixed.

* Test corrected

* rgb and yiq conversion ops added

* Converter ops added

* Header

* Yuv converters

* API added

* Empty test for matmul

* Explanation

* skip gemm/gemv on empty inputs

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

* Test added

* Correct test

* one more empty pass-through for mmul

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

* Cleanup

* Test added

* Test fixed

* Added missing mapping

* Added missing mapping

Co-authored-by: raver119 <raver119@gmail.com>
2019-12-30 15:06:12 +03:00
Alex Black 1f9e1b6022
SameDiff profiler analysis improvements (#141)
* #8555 SameDiff profiler analysis improvements

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

* Fix TF sub-op aggregation

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

* Small filtering tweak

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

* Copyright headers

Signed-off-by: Alex Black <blacka101@gmail.com>
2019-12-23 15:24:20 +11:00
Alex Black ce02b6fae7
Small fixes (#140)
* Allow scalar op result array auto allocation

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

* Don't swallow underlying exception for calculateOutputShape execution failures

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

* Ignore for known keras failure

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-12-21 17:00:46 +11:00
Alexander Stoyakin 6d8a063c9b nd4j-tests cleanup (#137)
* Fixed tests

* Invalid test removed
2019-12-20 16:38:33 +03:00
Alex Black 3d8f6d50a1
SameDiff profiler / tracing and profile analysis/comparison (#133)
* Profiler

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

* Next steps, polishing, and loading SD/TF format JSON

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

* Next steps

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

* Profile comparison method

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

* Make profiling result writing async to reduce main thread overhead

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

* Profiling polishing

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

* Fix

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

* Profile analyzer fixes

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

* Polish

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

* Cleanup

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

* Small formatting improvement

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

* Formatting tweak

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

* License headers

Signed-off-by: Alex Black <blacka101@gmail.com>
2019-12-19 23:43:58 +11:00
Alexander Stoyakin f5068f3980 Added missing Java ops wrappers (#122)
* Timeouts added

* Added some ops

* Ops added

* Fixed tests

* Minor fix

* Some fixes

* Digamma added

* Small fixes

* Timeouts added

* Added some ops

* Ops added

* Fixed tests

* Minor fix

* Some fixes

* Digamma added

* Small fixes

* Fused batch norm fixes-

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

* Tests switched off.

* Added test for resize_bicubic.

* Eliminated wasted in test of bicubic resize.

* Switched off multithreading explicit.

* HsvToRgb and RgbToHsv added

* Eliminated waste comments and conform proper float constants.

Signed-off-by: shugeo <sgazeos@gmail.com>

* Fixed multithreading with resize_bicubic helper for cpu platform.

Signed-off-by: shugeo <sgazeos@gmail.com>

* ResizeBicubic was fixed.

* Some fixes

* Fix op name

* Validation fixed.

* Clarifications for tests

* Wrappers and small fixes for new ops.
2019-12-19 20:15:48 +11:00
Alex Black bfd9e3692a
Add op counting to TensorFlowImportValidator (#128)
* Add op counting to TensorFlowImportValidator

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

* Test tweak

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-12-17 10:23:37 +11:00
AlexDBlack 0df1b46c8c Merge 2019-12-10 15:08:50 +11:00
raver119 a5f5ac72b1
reduce bool changes (#118)
* reduce bool changes

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

* reduce bool tweaks

Signed-off-by: raver119 <raver119@gmail.com>
2019-12-09 20:08:59 +03:00
Alex Black 0175ace4c3
Small tweaks (#119)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-12-09 23:08:00 +11:00
Alexander Stoyakin 927d591421 ResizeBicubic added (#117)
* ResizeBicubic added
Some fixes.

* Test fixed

* Narrowed argument type changed to boolean

* Clean up
2019-12-09 18:25:39 +11:00
Alex Black b66154a9d4
Add ArraySavingListener for debugging (#114)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-12-09 14:16:11 +11:00
raver119 b32dd1bf92
[WIP] resize_bicubic types (#116)
* resize_bicubic: allow more dtypes

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

* resize_bicubic: allow less dtypes

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

* Refactored resize_bicubic op to full conform with TF1.5 and tests.

* Corrected test to proper data type output.

Signed-off-by: shugeo <sgazeos@gmail.com>

* Corrected double input test to float constant outputs.

Signed-off-by: shugeo <sgazeos@gmail.com>

* Finished with correction of tests for bicubic interpolated resizes expected.

Signed-off-by: shugeo <sgazeos@gmail.com>

* Fixed adjust_contrast ops to allow non-RGB inputs.

Signed-off-by: shugeo <sgazeos@gmail.com>

* Refactored adjust_contrast_v2 to conform with TF one.

Signed-off-by: shugeo <sgazeos@gmail.com>

* AdjustContrast tests activated

* two typos fixed

Signed-off-by: raver119 <raver119@gmail.com>
2019-12-06 18:58:37 +03:00
raver119 972fae60dc
Update master (#8511)
* cleaned up bert iterator tests (#110)

Signed-off-by: eraly <susan.eraly@gmail.com>

* Various pre-release fixes (#111)

* Various fixes

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

* Fix default dtypes for MaxPoolWithArgmax

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

* Small pre-release tweak (#112)

* Log UI address on launch as in previous Play-based UI

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

* Logging level tweak for UI

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

* http not https

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

* datavec python ensure host (#113)

* ensure host

* one more host ensure

* info->debug

* [WIP] reverse improvements (#115)

* initial commit

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

* reverse draft

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

* reverse kernel

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

* reverse kernel

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

* 2 micro fixes

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

* Shugeo resize fix5 (#102)

* Refactored resize images ops to use TF-like bool args as input.

* Refactored helpers for cpu implementation of resize_bilinear and resize_nearest_neighbor ops.

* Refactored cuda implementation for image.resize_bilinear and image.resize_nearest_neighbor ops helpers.

* Refactored nearest_neighbor resize op.

* Added a pair of tests for special case of resize_bilinear algorithm.

* Fixed issue with resize_bilinear op.

* Refactored cpu implementation for helpers with resize_nearest_neighbor op.

* Final fixed for resize ops to conform TF v.1.5

* Refactored cuda helpers for resize_neares_neighbor op.

* Fixed resize_bilinear to accept proper data.

* Fixed issue with non-float input for resize_bilinear op.

* Refactored cuda helper for resize_bilinear to proper process non-float inputs.

* Added tests for resize_bilinear to int inputs.

* Fixed ResizeBilinear wrapper

* Tests fixed

* Fixed float and bool constant to avoid overflow for some kind of compilers.

* Corrected float constants with float data type.

* Added f suffix for float constants.

* Corrected float constant to avoid overflow with initializing lists.

* Corrected float initializing list with float input.

* Corrected bool constant with initalizing list.

* Corrected float and bool values with initializing lists.

* Fixed wrong constant.

* Fixed issue with 1x1 input picture for resize.

* ResizeBilinear default values on import fix

Signed-off-by: raver119 <raver119@gmail.com>
2019-12-06 11:10:44 +03:00
Robert Altena e7730eded4 delete unused and refactor. (#8262)
Signed-off-by: Robert Altena <Rob@Ra-ai.com>
2019-12-05 22:25:41 -05:00
shugeo e09a785232 Shugeo resize fix5 (#102)
* Refactored resize images ops to use TF-like bool args as input.

* Refactored helpers for cpu implementation of resize_bilinear and resize_nearest_neighbor ops.

* Refactored cuda implementation for image.resize_bilinear and image.resize_nearest_neighbor ops helpers.

* Refactored nearest_neighbor resize op.

* Added a pair of tests for special case of resize_bilinear algorithm.

* Fixed issue with resize_bilinear op.

* Refactored cpu implementation for helpers with resize_nearest_neighbor op.

* Final fixed for resize ops to conform TF v.1.5

* Refactored cuda helpers for resize_neares_neighbor op.

* Fixed resize_bilinear to accept proper data.

* Fixed issue with non-float input for resize_bilinear op.

* Refactored cuda helper for resize_bilinear to proper process non-float inputs.

* Added tests for resize_bilinear to int inputs.

* Fixed ResizeBilinear wrapper

* Tests fixed

* Fixed float and bool constant to avoid overflow for some kind of compilers.

* Corrected float constants with float data type.

* Added f suffix for float constants.

* Corrected float constant to avoid overflow with initializing lists.

* Corrected float initializing list with float input.

* Corrected bool constant with initalizing list.

* Corrected float and bool values with initializing lists.

* Fixed wrong constant.

* Fixed issue with 1x1 input picture for resize.

* ResizeBilinear default values on import fix

Signed-off-by: raver119 <raver119@gmail.com>
2019-12-05 22:05:33 +03:00
Alex Black 2052ce7026
Various pre-release fixes (#111)
* Various fixes

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

* Fix default dtypes for MaxPoolWithArgmax

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-12-05 14:20:03 +11:00
Fariz Rahman 0d14032d26 TF Updates (#87)
* tf updates

* pom

* copyright

* graphrunner tests

* gpu test

* getSessionOptionsConfigProto

* dtype fix

* Small fix

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

* cast graphs

* savemodel test fix

* testresource instead of local

* Logging level

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

* gson dependency issue fix; fix GraphRunnerTest for no session options config case

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

* Final tweaks

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

* few minor fixes

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

* one more fix

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

* Tweak configuration for GraphRunnerTest

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

* nd4j align config

* tf warmup
2019-12-04 17:11:03 +11:00
raver119 25b3cd9b80
[WIP] CUDA tests (#95)
* one more CI test

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

* export additional symbols

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

* few more tweaks

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

* one more tweak for linux

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

* fix dtype in few tests

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

* missing sync and memset in couple of tests

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

* copy step for libnd4j cuda

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

* no-op on empty for adjust hue/contrast/saturation

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

* CUDA_VERBOSE Off

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

* BroadcastBool fix + few tests

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

* trigger jenkins

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

* trigger jenkins

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

* - ignore couple of warnings
- remove redundant compiler options

Signed-off-by: raver119 <raver119@gmail.com>
2019-12-02 21:37:21 +03:00
Alexander Stoyakin 5e152c0d9a TF import tests - adding missing operations (#65)
* Add and fix mappings.

* Intermediate

* Added and fixed some mappings

* Added op

* Missing constructors added.

* Added new mappings

* SDImage wrappers and minor tweaks.

* Added missing constructor

* Some corrections

* Cleanup

* Small fixes

* Ops wrappers

* Minor fixes.

* Max Pooling

* MaxPoolWithArgmax

* Some fixes

* Ignores for failures

* Some ops fixed.

* Some fixes

* Missing package added

* Some fixes

* Ignored tests fixed.

* Some fixes

* Merge master

* bitcast fix

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

* Bitcast fixed
2019-12-02 21:23:06 +11:00
Alex Black 8123d9fa9b
SameDiff: Add Java-level assertion check/exception (#96)
Signed-off-by: Alex Black <blacka101@gmail.com>
2019-12-02 18:07:54 +11:00
Alex Black 2be47082c9
#8470 TrainingConfig json fix for Evaluation instances (#93)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-30 20:08:30 +11:00
Alex Black 35ab4a72ba
TF import test resources loading precision fixes (#92)
* Fix precision issues when loading from CSV

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

* Small tweak

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-30 18:58:37 +11:00
Alex Black 4fb9fa7748
Add ND4J namespaces (#83)
* Add NDValidation

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

* Add bitwise namespace

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

* Math namespace op constructor fixes

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

* Constructor fixes

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

* Add Math namespace

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

* Update NDBitwise

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

* Add random namespaces

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

* Update

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

* NN namespace

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

* Small cleanup

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-30 18:39:32 +11:00
Yurii Shyrma d19eeaec52 Shyrma casual conv1d (#90)
* - add causal mode of padding to convolutions

Signed-off-by: Yurii <iuriish@yahoo.com>

* - add additional tests for causal conv1d

Signed-off-by: Yurii <iuriish@yahoo.com>

* - add causal mode for cuda conv kernels

Signed-off-by: Yurii <iuriish@yahoo.com>

* Java side of Conv1D changes

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

* Add Conv1DDerivative op

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

* Causal Conv1D gradient checks

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

* Tweaks

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

* - add causal padding mode to conv2d_bp

Signed-off-by: Yurii <iuriish@yahoo.com>

* More thorough causal conv1d tests

Signed-off-by: Alex Black <blacka101@gmail.com>
2019-11-29 14:14:30 +03:00
Samuel Audet 5e07998e59 Add support for CUDA 10.2 (#89) 2019-11-29 16:31:03 +11:00
shugeo 009007120b Shugeo_release_fixes3 (#81)
* Implementation for non_max_suppression_v3 was added. Initial version

* Added check for overcome threshold.

* Added definition for V3 method.

* java remapping for NonMaxSuppressionV3

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

* Fixed proporly processing of an empty output and test.

* Refactored op to less threshold data to float.

* Implemented cuda-based helper for non_max_suppression_v3 op.

* Fixed fake_quant_with_min_max_vars op.

* Fixed tests with float numbers.

* - assert now stops execution
- sortByKey/sortByValue now have input validation

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

* missing var

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

* Fixed proper processing for zero max_size inputs.

* Refactored kernel callers.

* Fixed return statement for logdet op helper.

* Refactored unsorted segment SqrtN op.

* get back 8 tail bytes on CUDA

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

* Refactored segment prod ops and helpers for cuda and tests.

* Additional test.

* CudaWorkspace tests updated for 8 tail bytes

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

* special atomic test

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

* atomicMul/atomicDiv fix for 16bit values

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

* Eliminated waste prints.
2019-11-28 21:08:51 +03:00
Alex Black abd2017a0a
Add ignore for known issue with non_max_suppression_v2/float16 test (#85)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-27 16:35:05 +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
raver119 aa44fd6850 one more BitCast test
Signed-off-by: raver119 <raver119@gmail.com>
2019-11-25 08:52:11 +03:00
Alex Black e910ce75ec
Various Fixes (#75)
* #8431 Cast loss function weights array automatically

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

* Add 'regex verbose mode' printing (ExecDebugListener) for TFGraphTestAllSameDiff'

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

* Class import mapping fix

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

* Reshape fixes

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

* Don't swallow first exception in NativeOpExecutioner.exec(CustomOp)

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-23 20:06:12 +11:00
shugeo 4187190609 Shugeo release fix2 (#70)
* Corrected input checking and tests for bitcast op.

* Fixed an issue with non_max_suppression form generation and processing with score threshold given.

* Fixed bilinear resize kernel and tests.

* push for Serhii

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

* Added test for nearest_neighbor resize with int input.

* Added data type check for input/output match.

* Eliminate error in macros.

* Improved output message for type checking.

* Fixed input/output types for op.

* Eliminated waste logging.

* Refactored resize_bilinear helper for multithreading for cpu platform.

* Cosmetic changes only.

* Fixed error for string substitution.

* Skip test for cbow_batch with cuda.

* fix for resizeNearestNeighbor output dtype

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

* Refactored non_max_suppression helper.

* Refactored shape generation and input handling.

* Added additional test.
2019-11-22 22:42:44 +03:00
Alex Black 4a2fedf3e7
DL4J: Add Sparse multi-class cross entropy loss function (#72)
* #8432 Add sparse mcxent loss

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

* Fixes for LossSparseMCXENT

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

* add simple debugging listener for SameDiff exec debugging

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

* Extra gradient check + header polishing

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-22 18:54:31 +11:00
Samuel Audet ff73e6da3f ND4J: Fix OpenBLAS loading for nd4j-native (#64)
* ND4J: Fix OpenBLAS loading for nd4j-native and remove bundling of OpenMP

Signed-off-by: Samuel Audet <samuel.audet@gmail.com>

* Bundle back libgomp.so.1 for Linux

Signed-off-by: Samuel Audet <samuel.audet@gmail.com>

* Readd preload directories for ARM

Signed-off-by: Samuel Audet <samuel.audet@gmail.com>

* Add back preloads for GCC on Windows

Signed-off-by: Samuel Audet <samuel.audet@gmail.com>

* Add explicit preloadpaths for ARM and POWER to bundle correct library

Signed-off-by: Samuel Audet <samuel.audet@gmail.com>
2019-11-21 15:54:41 +03:00
raver119 064a56ccf1
Few fixes (#66)
* skip legacy transforms execution in case of empty input arrays

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

* - BroadcastBool ops now accept extraParams to make MatchCondition possible
- TrueBroadcastHelper now uses samediff::threads

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

* java side

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

* trigger jenkins

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

* update LessThanOrEqual opNum mapping

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

* update LessThanOrEqual opNum mapping

Signed-off-by: raver119 <raver119@gmail.com>
2019-11-21 15:43:03 +03:00
raver119 83cb0d9329
[WIP] Create and small fix (#67)
* - create op
- skip exec for empty inputs for non_max_suppression
- EmptyHandling idea

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

* Create op and mapping for it

Signed-off-by: raver119 <raver119@gmail.com>
2019-11-21 13:31:20 +03:00
raver119 3f38900c33
J9+ -> J8 ByteBuffer fix (#59)
Signed-off-by: raver119 <raver119@gmail.com>
2019-11-20 07:43:17 +03:00
Yurii Shyrma 66b84b38cf
Shyrma mmul (#58)
* - get rid of some copy procedures in mmulHelper ops

Signed-off-by: Yurii <iuriish@yahoo.com>

* - further work on embedding cuda api for batched gemm (cublasGemmBatchedEx) in our mmulHelper class

Signed-off-by: Yurii <iuriish@yahoo.com>

* - further work on cuda batched gamm api

Signed-off-by: Yurii <iuriish@yahoo.com>

* - write own cuda kernel performing batched gemm

Signed-off-by: Yurii <iuriish@yahoo.com>

* missing include in MmulHelper

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

* - forgot to keep in code previous correct kernels for mmulNxN, since it may happen that new onw will fail for some reason in future

Signed-off-by: Yurii <iuriish@yahoo.com>

* disable old tensordot

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

* - rewrite cuda kernels for usualGemm and usualGemv

Signed-off-by: Yurii <iuriish@yahoo.com>

* - profiling mmul helpers

Signed-off-by: Yurii <iuriish@yahoo.com>

* - prints to check shapes were added

Signed-off-by: Yurii <iuriish@yahoo.com>

* - correct type of output array Cin mmulNxN

Signed-off-by: Yurii <iuriish@yahoo.com>

* - take into account possible nans in C array

Signed-off-by: Yurii <iuriish@yahoo.com>

* slightly change numThreads message

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

* - make corrections in accordance to given notes in pr review

Signed-off-by: Yurii <iuriish@yahoo.com>
2019-11-19 15:39:36 +02:00
Alex Black da1944e8e1
SameDiff TF import (#49)
* Added implementation files for image_resize and resize_bicubic ops.

* Image resize and image.resize_bicubic ops implementation. Initial revision.

* Minor fix

* Some TF imports disabled.

* Finished with infrastructure development for image.resize_bilinear op and image_resizo op implementation.

* Refactored resize methods.

* Added processing for Mitchelcubic algorithm.

* adjust_contrast

* Small fix for TF import expected value loading when variable name starts with the test name

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

* Tests

* Tests added.

* Removed tf names absent in mapping.

* Some fixes.

* Small fixes

* Minor change

* Some failing tests.

* Disable failed test

* Ignore some tests

* Fix import class mapping

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

* Fix float property mapping (flatbuffers)

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

* Override equality function for model 'dropout'

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

* Fail tests

* Failed tests ignored temporarily.

* Minor fixes

* Small fix

* Conflict resolved

* Default implementations of tensorflowName and onnxName
2019-11-19 22:44:29 +11:00
raver119 ce2ef20f96 additional reverse signatures
Signed-off-by: raver119 <raver119@gmail.com>
2019-11-19 13:54:07 +03:00
raver119 db7ca956c5
[WIP] Mish (#55)
* Mish activation function and its derivative

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

* signature fix

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

* mish as activation for dl4j

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

* javadoc

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

* minor optimization

Signed-off-by: raver119 <raver119@gmail.com>
2019-11-18 13:21:26 +03:00
raver119 c5cbdcd8f4
[WIP] clang for jcpp (#53)
* clang as compiler for jcpp

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

* we don't need macos profile

Signed-off-by: raver119 <raver119@gmail.com>
2019-11-17 09:45:30 +03:00
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
raver119 1780dcc883
[WIP] Small fixes here and there (#50)
* one range test

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

* few Context convenience singatures

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

* one more range test

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

* "range" "fix"

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

* adjuct_contrast_v2 now allows scale factor to be provided via input_variable

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

* adjust_contrast now allows scale factor as variable too

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

* bitcast shape tests

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

* BitCast import dtype added

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

* few more BitCast signatures

Signed-off-by: raver119 <raver119@gmail.com>
2019-11-15 17:04:29 +03:00
raver119 c5b912bddf
few changes for openblas and jcpp preloads (on macos) (#46)
Signed-off-by: raver119 <raver119@gmail.com>
2019-11-14 19:50:24 +03:00
raver119 1eb3de90d7
[WIP] Platform helpers switches (#44)
* - platform helpers can be disabled on per-op basis now via Context::allowHelpers
- java has access to it as well

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

* global platform-helpers trigger

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

* few signatures renamed

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

* - few new env variables to follow
- maxThreads/masterThreads differentiation

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

* Javadoc update

Signed-off-by: raver119 <raver119@gmail.com>
2019-11-14 14:35:02 +03:00
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 1d96bb9e6e
SameDiff op runtime benchmarking listener (#42)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-12 22:51:09 +11:00
Alex Black 18c01f5bdc
Add SameDiff memory reuse memory manager (array cache) (#39)
* Attention op comments

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

* ArrayCacheMemoryMgr - first pass

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

* Tweak array cache for use with SameDiff identity arrays

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

* ArrayCacheMemoryMgr javadoc and properly get max memory

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

* LRU cache policy + add tests

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

* Fixes

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

* Resize arrays internally if required for ArrayCacheMemoryMgr

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

* Test improvement

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

* Small polish

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-11-12 21:15:44 +11:00