Commit Graph

437 Commits (9124974e3bf4bcaa8b698ed9acc21a87ddd93d6b)

Author SHA1 Message Date
AlexDBlack b7226bdd7a Merge
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-05 00:54:11 +10:00
Alex Black 03c52ef9dd
Add SameDiff.bitwise namespace (#232)
* #8196 add SameDiff.bitwise namespace

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

* Add BitsHammingDistance, add remaining bitwise ops to bitwise namespace

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

* fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-04 22:34:31 +10:00
shugeo 548044a1e2 Shugeo doc (#235)
* Actualized doc to tnse ops.

* Added comments for dynamic_stitch op.

* Added comments to dynamic_stitch op implementation.

* Modified comment for unstack_list op.

* Added doc for space_to_depth and depth_to_space ops.

* Added doc for space_to_batch op.

* Enlarge test type for adjustSaturation.

* Added doc for runner.
2019-09-04 14:57:59 +03:00
raver119 a90c7dd995
[WIP] Last set of changes (#234)
* mmul op instead of cublasSgemm

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

* transB

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

* jcpp handles

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

* bitwise and/or/xor

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

* bitwise and/or/xor mapping

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

* cuda/cublas version check

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

* add expected version

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

* cuda/cublas version check in java

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

* one more error check

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

* build fix

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

* build fix

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

* build fix

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

* one more fix

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

* skip CUDA version check for now

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

* better wording

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

* few more tweaks

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

* few more tweaks

Signed-off-by: raver119 <raver119@gmail.com>
2019-09-04 14:41:08 +03:00
Alex Black d41018751b
Small fix (#233)
Signed-off-by: Alex Black <blacka101@gmail.com>
2019-09-04 19:11:17 +10:00
Ryan Nett e9454b8882
SDCNN cleanup pass (#230)
* SDCNN cleanup

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

* NonNull annotations

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

* better javadoc, NonNull fix for sconv

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

* update builders to fix names

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

* fixes

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

* even more fixes

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

* fix for null bias

Signed-off-by: Ryan Nett <rnett@skymind.io>
2019-09-04 00:44:01 -07:00
Alex Black 6cc887bee9
Rename flatbuffers DataType to DType (#228)
* Rename flatbuffers DataType enum to DType

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

* Rename flatbuffers DataType enum to DType

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

* Updates for flatbuffers datatype enum renaming

Signed-off-by: Alex Black <blacka101@gmail.com>
2019-09-04 16:36:11 +10:00
raver119 64eaafb4cd remove unwanted noexcept
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-04 08:38:22 +03:00
Robert Altena 25b01f7850 javadoc and remove deprecated methods. (#231)
Signed-off-by: Robert Altena <Rob@Ra-ai.com>
2019-09-04 13:29:02 +10:00
raver119 7abc574eeb
Snapshot update (#8194)
* fix double consumption of rng on cpu

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

* Shyrma docs (#222)

* - documenting and profiling matrix_set_diag cuda kernel

Signed-off-by: Yurii <yurii@skymind.io>

* - correct formula of pnorm pooling in cuda 2d/3d kernels
- remove helper matrix_diag which duplicates work of helper matrix_set_diag

Signed-off-by: Yurii <yurii@skymind.io>

* cublasHandle sharing + lock

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

* cublasHandle sharing + lock

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

* Documentation from serialization/deserialization in NLP (#221)

* refactoring

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

* Javadocs

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

* Javadoc fixed

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

* Cleanup

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

* dedicated lock for getCudaCublasHandle

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

* Small fixes (#223)

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

* ELU DL4J fixes (#224)

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

* javadoc (#225)

Signed-off-by: Robert Altena <Rob@Ra-ai.com>

* Small test compilation fix (#226)

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

* #8182 remove spark version suffix (#227)

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

* [WIP] Thread safety (#229)

* sync after cublas*gemm

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

* mutex for CublasHelper

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

* don't store cublasHandle in LaunchContext, it's per-device anyway

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

* some printout

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

* check for field instead

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

* pew-pew

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

* don't release ContextBuffers until device changed

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

* small tweak

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

* some logging in sgemm

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

* stream sync

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

* some more logging

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

* some more error checks

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

* one fancy test

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

* one fancy test

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

* minor AffinityManager fix

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

* cudaEvent error logging improvement

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

* ConstantHelper thread safety

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

* - minor corrections in ConstantTadHelper

Signed-off-by: Yurii <yurii@skymind.io>

* ConstantShapeHelper thread safety

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

* ConstantTadHelper.cu updated

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

* logging off

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

* logging off

Signed-off-by: raver119 <raver119@gmail.com>
2019-09-03 22:02:02 +03:00
raver119 dddc8a1143
[WIP] Thread safety (#229)
* sync after cublas*gemm

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

* mutex for CublasHelper

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

* don't store cublasHandle in LaunchContext, it's per-device anyway

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

* some printout

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

* check for field instead

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

* pew-pew

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

* don't release ContextBuffers until device changed

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

* small tweak

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

* some logging in sgemm

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

* stream sync

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

* some more logging

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

* some more error checks

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

* one fancy test

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

* one fancy test

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

* minor AffinityManager fix

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

* cudaEvent error logging improvement

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

* ConstantHelper thread safety

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

* - minor corrections in ConstantTadHelper

Signed-off-by: Yurii <yurii@skymind.io>

* ConstantShapeHelper thread safety

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

* ConstantTadHelper.cu updated

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

* logging off

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

* logging off

Signed-off-by: raver119 <raver119@gmail.com>
2019-09-03 22:00:38 +03:00
raver119 9d03bb9425 allow atomicAdd for CUDA 10 only
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-03 13:30:16 +03:00
Alex Black 5be43e7253
#8182 remove spark version suffix (#227)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-03 18:54:19 +10:00
raver119 f6f9437a36 get back cc 7.0 support for cuda 9.2
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-03 09:26:35 +03:00
Alex Black f076a8b285
Small test compilation fix (#226)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-03 14:17:53 +10:00
Robert Altena c64b340975 javadoc (#225)
Signed-off-by: Robert Altena <Rob@Ra-ai.com>
2019-09-03 14:06:42 +10:00
Alex Black 364a6e1a2a
ELU DL4J fixes (#224)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-03 13:35:02 +10:00
Alex Black ba269a26ab
Small fixes (#223)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-03 10:48:59 +10:00
raver119 6ce620709a Merge remote-tracking branch 'origin/master' 2019-09-02 20:01:27 +03:00
raver119 d3253aff3f dedicated lock for getCudaCublasHandle
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-02 20:01:13 +03:00
Alexander Stoyakin 90b62c4579 Documentation from serialization/deserialization in NLP (#221)
* refactoring

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

* Javadocs

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

* Javadoc fixed

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

* Cleanup

Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-09-02 17:17:55 +03:00
raver119 2129d5bcac cublasHandle sharing + lock
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-02 16:52:28 +03:00
raver119 18828f9725 cublasHandle sharing + lock
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-02 16:52:10 +03:00
Yurii Shyrma cb4c9377b1 Shyrma docs (#222)
* - documenting and profiling matrix_set_diag cuda kernel

Signed-off-by: Yurii <yurii@skymind.io>

* - correct formula of pnorm pooling in cuda 2d/3d kernels
- remove helper matrix_diag which duplicates work of helper matrix_set_diag

Signed-off-by: Yurii <yurii@skymind.io>
2019-09-02 16:25:58 +03:00
raver119 106524663b fix double consumption of rng on cpu
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-02 15:24:51 +03:00
raver119 0e05cba2f9
change pointer reference for cudnn (#220)
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-02 12:40:32 +03:00
AlexDBlack 7ded4416cb Merge remote-tracking branch 'fork/master' 2019-09-02 18:52:12 +10:00
raver119 5b8ea3e830 one more tiny cuda fux
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-02 11:49:13 +03:00
Alex Black 82c9dc5743
ELU fix (#219)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-02 18:37:05 +10:00
raver119 e42c34ca55
[WIP] minor (#218)
* - initial docs commit
- merge* cuda fix

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

* one more fix

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

* one more fix

Signed-off-by: raver119 <raver119@gmail.com>
2019-09-02 11:25:48 +03:00
Alex Black 65c9f2a888
ELU fix (#217)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-02 17:42:12 +10:00
Alex Black acf559425a
Small test fix (#216)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-02 16:44:57 +10:00
Ryan Nett b3a134b608 New Nd4j backprop ops for activations (#211)
* new (for java at least) backprop ops

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

* update activation functions

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

* add differential functions for SameDiff

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

* deprecate old ops

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

* update correct old ops

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

* update ops backprop to use new ops

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

* misc updates for deprecated functions (mostly Nd4j.rand w/ vararg shape)

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

* remove old imports

Signed-off-by: Ryan Nett <rnett@skymind.io>
2019-09-02 16:15:23 +10:00
Robert Altena 6d04d30c94 INDArray.java javadoc (#215)
* javadoc

Signed-off-by: Robert Altena <Rob@Ra-ai.com>

* javadoc

Signed-off-by: Robert Altena <Rob@Ra-ai.com>
2019-09-02 16:06:20 +10:00
raver119 c34826da4d fixed args
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-01 22:06:01 +03:00
raver119 00cf28f477 get rid of builtin_popcount to please ppc
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-01 19:57:39 +03:00
raver119 3679e55c49 fix bits_hamming_distance for ppc
Signed-off-by: raver119 <raver119@gmail.com>
2019-09-01 19:33:23 +03:00
Robert Altena ef1de6a4aa rcorbish #8617 (#8188)
Signed-off-by: Robert Altena <Rob@Ra-ai.com>
2019-09-01 07:17:36 +03:00
Yurii Shyrma a35926c6e9 - add parameter alpha to elu and lrelu_bp (#213)
* - add parameter alpha to elu and lrelu_bp

Signed-off-by: Yurii <yurii@skymind.io>

* - forgot to correct header activations.h

Signed-off-by: Yurii <yurii@skymind.io>
2019-08-31 20:57:39 +03:00
raver119 b71c993ded
[WIP] maxpool_bp cuda fix (#212)
* one test for alex

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

* fix

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

* get rid of safety offset in cpp

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

* bfloat16

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

* minor test rearrangement to fastpath launch

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

* - atomicAdd/Mul/Div fix for float16/bfloat16 misalignment
- one special test for maxpoolbp java
- safety offset of 8 bytes is back to libnd4j legacy

Signed-off-by: raver119 <raver119@gmail.com>
2019-08-31 20:57:05 +03:00
Alex Black f00a7bb3f2
Small fix for clashing dependency (test scope only) (#214)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-09-01 00:16:38 +10:00
AlexDBlack b393d3fdb1 Merge
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-31 12:33:22 +10:00
Alex Black 12d14efcf2
ND4S test fix (#210)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-31 12:27:09 +10:00
Yurii Shyrma 00fd50cee2
Shyrma softmax (#209)
* - provide new cuda kernel for softmax

Signed-off-by: Yurii <yurii@skymind.io>

* - further work on cuda kernel for softmax

Signed-off-by: Yurii <yurii@skymind.io>

* - correction cuda kernel for softmax

Signed-off-by: Yurii <yurii@skymind.io>
2019-08-30 20:31:05 +03:00
raver119 bdc3eacafd one small playground test
Signed-off-by: raver119 <raver119@gmail.com>
2019-08-30 20:13:01 +03:00
Alex Black 6efffb727f
Import fix (#208)
Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-30 23:40:52 +10:00
Alex Black a7dca9fc87
Add java op class for relu derivative, and use in Activation ReLU (#207)
* Add java op class for relu derivative, and use in ACtivation ReLU

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

* Fix

Signed-off-by: AlexDBlack <blacka101@gmail.com>
2019-08-30 23:36:00 +10:00
raver119 70a9ae5068
[WIP] few tweaks (#206)
* scatter empty check

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

* scatter empty test

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

* one more test

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

* two tweaks

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

* dup tweak

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

* - put empty checking of indices array immediately prior  helper run

Signed-off-by: Yurii <yurii@skymind.io>

* minor tests fix

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

* minor tests fix

Signed-off-by: raver119 <raver119@gmail.com>
2019-08-30 16:32:01 +03:00
Alexander Stoyakin f414239ed5
Types fixed (#205)
Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-08-30 16:18:41 +03:00
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