Age | Commit message (Collapse) | Author |
|
This patch introduces a new GEMM capable to improve the mac utilisation
of 10% compared to the GEMM without reshape. However this implementation
is not faster in all cases as we need to take into account the time for
reshaping the matrices. For this reason an heuristic solution to select
the optimal GEMM to use has been added to the function. More information
about the heuristic implementation can be found at COMPMID-852.
With this new patch, GoogleNet, MobileNet, VGG16 and SqueezeNet can
improved the performance of 1.5x.
More information about the performance uplift can be found here:
https://confluence.arm.com/display/MLENG/GEMM+FP32+performance%3A+ACL+18.02
Change-Id: I024563c06b9aed02a211a974e452bae5c233b04c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/117140
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
when not running in a terminal
Change-Id: I4ec90803c5dc41b0cee05c36113ae3f189564d58
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/117831
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: Ib21de61fe39d2768638af11c067dfc7bcf63aae2
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/117112
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
|
|
Change-Id: I389e0d4104b7dde60b7cdd612a83f3328517e44c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/115804
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Ic76b3b6adaff8c84ba4d2ca5283d9291c69344f0
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/114466
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: Ib178a97c080ff650094d02ee49e2a0aa22376dd0
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/115717
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I180281e796e1670b9ad391d82d66ecde0119ef78
Note: this is for internal use only which is why I think the hackiness of RunExample.cpp is acceptable.
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/115154
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
-Propagates hints to subgraph.
-Fixes dispatching of apropriate optimized DepthwiseConvolution kernel
for OpenCL backend. NEON backend is altered to default to the generic
case until COMPMID-769 is addressed.
Change-Id: I544f05cd99a9ac253f1b19aa4e4bb222b8fdd087
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/114781
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
to memory_barrier
Also fix the synchronisation issues between different kernels.
Change-Id: Ib59d83ae8d5cc8b0bdf13e6f4958edccdab91ca4
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/114594
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: Ic59b2d852d59abb3d149e29760a1e16978d41bdc
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/114593
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Joel Liang <joel.liang@arm.com>
Reviewed-by: Ioan-Cristian Szabo <ioan-cristian.szabo@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I56333ed23d30c5ec3094f64b78a023589064fe06
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/113375
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Jim He <jim.he@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Symbols from translation units of arm_compute_graph were stripped during
static linkage.
Forces to include all symbols of arm_compute_graph.
Change-Id: Ib66f513792c8796fca10f8deaca887db474f2bed
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/113187
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
- Fixed data type issue in cl_sgemm
- Added support for NEON and OpenCL targets in graph examples. Before we
could run only OpenCL target
- Add auto_init() in NEDepthwiseVectorToTensorKernel
Change-Id: I4410ce6f4992b2375b980634fe55f1083cf3c471
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/112850
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
|
|
Revert back to static linkage for android as dynamic linkage causes
error within libc++
Change-Id: I448dd5ff4e527b81d702e331bd070101431fb7de
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/111995
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
|
|
- alpha and beta were integer values whilst should be
float.
- Replaced CLImage with CLTensor
- Replaced Format with DataType
Change-Id: I19f81b52d2eab8976be689b601d8e8e2bedfc6aa
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/111725
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I1181abbd5785065f3d57e91844376a4b110938a9
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110701
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
- Added --no-allow-shlib-undefined to library builds to make sure we
don't leave any unresolved symbol
- Added --allow-shlib-undefined to the graph examples build lines so that
it doesn't matter whether or not OpenCL is supported by the Graph
librarylibrary
- Don't link tests statically anymore on Android (This was probably
required in the past because we were duplicating symbols in the
core and runtime libraries, but it doesn't seem to be an issue
anymore)
- Updated Doxygen accordingly
Change-Id: I9905f388e0838bc8e3369de52838bf980b992ead
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/95172
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: Iead7497cc03e1e7bde440d2965a7bf54cbfa88bf
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/95579
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Joel Liang <joel.liang@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
|
|
Last commit:
commit b25c5f68042b0c81bf611d59a1bb8535e1c42497
Author: Xinghang Zhou <xinghang.zhou@arm.com>
Date: Wed Oct 25 18:48:10 2017 +0800
Synced validation's tolerances of GCSoftmax from cl side
Change-Id: Ibe72054205c1c8721845d679a31af7ed0a7c5cf6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/93283
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I9440581e387ac5763b72444130c617424b14e1aa
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/91424
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: Ie26904a3b232ed614a3a063f7deb24995249e820
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/94657
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
|
|
Change-Id: I8baaee68b82e200a4829c3fc6c60dd211c06e14f
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/94534
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Ib6f2f9e73043d2c59b2698c243fb1a9f51c526e9
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/94363
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
|
|
Change-Id: Ied405a9c0e9746598d03ac6a944ad87e9b6494eb
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/93680
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I17478202348d34716d00c78e3cc21254c76b32e1
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/93304
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I40c3e2dfcde10c65ed9c86f9283a53f9e679d4fa
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/92437
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I5f35dbc31ddf821d777c8d8867d25f05743c73a5
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/92411
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I73d1433ee7a682aeabb7540aa2ea1f6564f90aae
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/91775
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Id233f0c1c329ee0d5ee93166d4aa0718f7d629b7
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/92337
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I778007c9221ce3156400284c4039b90245eb2b7f
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90043
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I7eb2463ab15425ef5681c2c871118964bcfd2179
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/91356
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: If912d8232e12cd496923d55d386898450dac09e2
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89897
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
GraphContext hold all the information about the hints that need to be
passed in the nodes. As these might expand, it serves as a centralized
class for such information.
Change-Id: I0b5527630fb97cc5fa500db0bac8307ff2ea36e6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/90300
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
-Adds grouping support in convolution layer
-Adds Normalization layer node in graph
-Adds alexnet example
-Fixes FullyConnectedLayer output autoconfigure (works only for 1d batch
space)
Change-Id: I5bd75f9a8b08cfd68f7c34745150266c2bc4221f
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89518
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
levels of library
Until now we had:
core = core_obj
arm_compute = core_obj + arm_compute_obj
graph = core_obj + arm_compute_obj + graph_obj
But if an application link against more than one of these libraries then bad things happen.
Added version strings in the runtime library too (As it used to only be in Core objects)
Updated doxygen for how to compile the examples
Change-Id: I7aad6ecf75cfa8dca59f2ea093e13fb0314a3eb4
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89743
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
In case the library has been built just for NEON, we should not run the example
Change-Id: I99f8d3efd20276dfeb73ca8362872111be1b1fd7
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89682
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I3bd91a9f234b8b42a96454c0d2e45f6098c543a5
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89481
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
example
Change-Id: I31839993bd75789566045c0db53dbf6dc3ae14e9
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/89202
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Ie841419bf65d0e06bdfe0bdd2d8d4e0bb3631e54
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/87931
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
* Add tensorflow_data_extractor script.
* Incorporate 3rdparty npy reader libnpy.
* Port AlexNet system test to validation_new.
* Port LeNet5 system test to validation_new.
* Update 3rdparty/ and data/ submodules.
Change-Id: I156d060fe9185cd8db810b34bf524cbf5cb34f61
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/84914
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: Ie4797e88fd2ad38f30cb69651ed37481b835fae8
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/80781
Reviewed-by: Moritz Pflanzer <moritz.pflanzer@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Add proper stride and paddding info to avoid incorrect tensor shapes.
Change-Id: Ib906e43cf4ef09364acfb03201e765642dbc25af
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/79555
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
against libgcc, libstdc++ and computer_vision.
This is needed to run 'computer_vision' on the ARM 8.2-a + FP16 Fast model.
Change-Id: I66d6427e4d4117b7995c35ec9df1dea5b3f22205
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/78429
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I32f7b84daa560e460b77216add529c8fa8b327ae
|