Age | Commit message (Collapse) | Author |
|
kernels
Change-Id: I98183f95814442b6f3dbb67a1bdae99df05b9b01
|
|
Change-Id: I5d2ed5dcc342abff8124762f7bdee587cdf20032
|
|
- Fixing a bug for which we did not scale the boxes before transforming them
- Adding the correct_transform_coords option to BoundingBoxTransformInfo
Change-Id: I40281254bcf87e7c8583c119e99562414fe59822
|
|
Introduced F32 accumulation for F16 winograd gemm and output transform
WinogradConvolution will be available for F16 only if fast math flag is enabled
Change-Id: I215593c205236a0f9669218437bb40b184ec6a4f
|
|
Change-Id: I99e1c3939cfea4b9cb0ddfa313706f31b213ca89
|
|
Change-Id: Ib0798cc17496b7817f5b5769b25d98913a33a69d
|
|
Change-Id: I5bf5d751ec7c02d96c26a769f49d03ea23a248b7
|
|
Some systems don't have enough memory to run the VGG networks, for example
on systems with only 2GB memory the VGG example fails throwing a bad_alloc exception.
This patch introduces the concept of global memory policy in ACL, the policy
is a mechanism which could be used by the library's functions to try to reduce
memory consumption on systems with limited memory.
In this specific case the VGG examples set the policy to MINIMIZE. The GEMM
function checks if the policy is MINIMIZE and in this case does not use the
pretransposed weights path as this requires considerable more memory.
Change-Id: I53abc3c9c64d045d8306793ffc9d24b28e228b7b
|
|
The issue was related to CLIm2Col when the number of input channels was less than
the number of elements processed by each thread.
The bug has been fixed in the validate_and_configure_window() function setting the correct number of elements accessed
in the output tensor.
Also fixed an issue GEMM3D when we have a single output channel
Change-Id: I094292d0c7662599c4a4c3916ec5f5821df5faef
|
|
Change-Id: I5aae537372bf797fbb2a2bae81038f8963b041a9
|
|
OpenCL
COMPMID-1424 - Add dot product support for CLDepthwise QASYMM8 3x3 NHWC non-unit stride
With this patch we are able to improve the performance of MobileNet v1-qasymm8 by 37 %
Tried to use the dot product instruction in CLDepthwise QASYMM8 3x3 NHWC non-unit stride
but I have not seen any benefit (maybe because we have few arithemtic operation and we
do not have more load instructions). However Depthwise convolution has been improved by
30%
Change-Id: Id768a99c2e53a04276707e427af5d0ec93419ada
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/155082
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I91865506166951b3bf7f06a0b2d4cde925cfefb6
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/153447
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
NEGEMMConvolutionLayer.
Change-Id: Iea5f2c5bcac8051c4c7655a6eabb2c43772eb31f
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/154104
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
|
|
Change-Id: Id331199f569f52a37280a9ada5bf84694580b93c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/152843
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
|
|
Change-Id: Ie215daacd10477309dbf8af1bb2b05b7a0a8f203
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/150773
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Isabella Gottardi <isabella.gottardi@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
Change-Id: Ideead99410e5e0bda1035030af1bbcd0a65ea15e
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/144792
Tested-by: bsgcomp <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: If2b44da31fae528c76be742b4b3a21fb0eb06b49
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145284
Reviewed-by: Giuseppe Rossini <giuseppe.rossini@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: Id978c34889b86fa8b9184d3349cc9b12837141a2
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145403
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I3b8a6c00e61ba6da459ca5fc7275393f9d073aed
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/142533
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Skipped im2col in CLGEMMConvolutionLayer for 1x1 convolutions with NHWC data layout
Change-Id: I894e6b952ed8605e8f3ffc0ffc25c24730d4664c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/141909
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I7670f79209a1e4439d57e05c1f5c576f600971cb
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/142299
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I62e3ead903366baeeb1488f233a9b8b0c388c9de
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/140403
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Removed QS32 references
Change-Id: Ic7df02c08ae7aa1b7dcae15bdda113321af851b8
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/138703
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Removed fixed point related code.
Change-Id: I487acf138dace3b0450e0d72ca7071eaec254566
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/137678
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I8c4823a0d909e19e9ef548f00b9ae98c66de61dd
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/123569
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Disables Wnoexcept in tests
Used gcc 8.2 to reproduce.
Change-Id: I06f4636d96bd3ac98dace72341bb808edcd23750
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/134490
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
CLFullyConnectedLayer
Change-Id: I1c3b2197906cd4b905309bbd5f2012bbae6a7dba
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/133730
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
COMPMID-1014 - Refactoring Winograd's dataset
Change-Id: I6abdcbf9a90d663f4db666cd410afece9f1d034d
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/125899
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Ie73d8771f85d1f5b059f3a56f1bbd73c98e94a38
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/124723
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: Ic1685de4e19e0ac79669ef2da64e1dc96c7ea0bf
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/115248
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I89bdab7dc63a437eb7d60c0ae652c53c3875e503
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/122524
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
Change-Id: Ifcc406d2d0a99c911d6b6c875657b0e0028255d5
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/119148
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
padding
Currently an assert gets fired in debug mode, and we just ignore the asymmetric padding in release mode.
Change-Id: Ia6278b5722f7e93f356a975ab3243e6bb07e44a8
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120840
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I9a607fe620f795cdea1a99fdd3f5f8c2fc76f980
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/119234
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: Ida1e9a836bc518bfe5563e16bf7f92bde5fc13f7
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118472
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
Change-Id: Ib8100e7c659c49694c746fa3f36ce20f44f6929f
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/117804
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I4f2cca52caf210fdb7d6bb7e9436ac51cb5088b4
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/112398
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
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>
|
|
Change-Id: I91e39713ffa580e9d2213988ad3517a8a41bf4e8
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/114013
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Ic3881e97b4fcbae0ac287a1e010cfc6f0fd8d7d1
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/112139
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: Id69df4ce98d1d89bdf9c9aa5c4d909659909b30f
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110456
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I6272a36636c5d9baff6d35dee0a50dc847f65bfa
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110266
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: I8f6adc67f6e6157710fca540f217d70e6f297217
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/111721
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: If1dcca724e5e5f5ab363ffc16b0ef8c943e0b657
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110105
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Adds a is_scaled parameter in the NormalizationLayerInfo that flags if
the alpha parameter should be scaled by the normalization size of not.
Unscaled parameter is used by [Krichevksy 2012] which is used in
AndroidNN and TensorFlow LRN layer.
Change-Id: Iad2aa5e688cf4dcd6cc77a6e28c0663764f34ccb
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/96102
Reviewed-by: Diego Lopez Recas <diego.lopezrecas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@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>
|
|
Change-Id: I7a4126f96aa7ef7ed768ebe5b4e2b1f84228f8e6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/95060
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I32ba6afbac6694ffa053dd16f03a1b3d14627a19
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/94857
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Adds support for excluding the padding pixels from the average scaling
factor calculation.
Change-Id: Ia13fbfeae235aff564db74191613921848231a01
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/93715
Reviewed-by: Robert Hughes <robert.hughes@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
|
|
Change-Id: I3c39ecbd36f06d5376c35ed4eb38dd73533ef97e
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/93686
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|