aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2018-11-02APPBROWSER-391: Fix GLES COMPUTE alignment issuesFrank Lei
APPBROWSER-402: Performance optimization for squeezenet/xray model Change-Id: If31b186b99a6d6087164019fe94d3ac9279e3204 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/119526 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
2018-11-02COMPMID-945: Fix GEMM CL FP32 mismatchMichele Di Giorgio
The mismatch is due to loss of precision of the FMA operations when dealing with large matrices containing small values. The solution proposed in this patch modifies the validation method so that for floating point value, the number is truncated to the 4th decimal value. Change-Id: I4a206a023cbcc23971e4988a554217677e3a96ad Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121375 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-582: Add validation to channel_extract kernels.Ioan-Cristian Szabo
Change-Id: I5022d02f06f9d849dad76e3d9b8e48632c236429 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121191 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-846: Create a ConvolutionLayer for NEONIsabella Gottardi
Change-Id: I98bbef40bfac5b05134be4ef9fb54d14c0c9e8e8 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118806 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-764 Removed reciprocal QS testsAnthony Barbier
Also fixed issues with some test names Change-Id: I7f35932eb24bedbb645b4b534d7d1bb1342f3696 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121259 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-765 - Fix get_convolution_method in order to return the correct method.Isabella Gottardi
Change-Id: Ia4be053b9f5399fe7e241cebb4292890e957ae54 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121141 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-936: Convolution failure in NEON Convolution Layer.Georgios Pinitas
Change-Id: I68a98eff57c8db719a501b68541666e8bc5f2081 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121180 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-853 Use tile 2 for CL depthwise convolution QASYM8Giorgio Arena
Change-Id: I91f6a0b057f5eb84c6ac7db5abbc05c7520ed5d2 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120760 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-784: Fixed SAME paddding in WinogradLayerPablo Tello
There were mismatches when using kernel size 5 and padding = SAME Change-Id: Id834e96ebcf665616f99c995b48e302dcff8dc48 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121144 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02Revert "COMPMID-582: Add validation to channel_extract kernels."Anthony Barbier
This reverts commit 9a0875951d43dda035f32d2e0728cf59d80cb4d3. Change-Id: I6af0bc64c656f91cf1e0357f8760defa08f2e78d Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121190 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-578: Faster common keypoint validation routine for FAST and Harris ↵Abe Mbise
corners Change-Id: I1f3038ccfa6be5e0a19bec11a89af5ed2247c42f Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/119936 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-942: Created 'embed_only' build target for ArmNNAnthony Barbier
Change-Id: I579943339fc77c31ce29253cdbc3d8654ac0c6f0 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121167 Reviewed-by: Pablo Tello <pablo.tello@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02Revert "COMPMID-915: Create ResNet50 example"Anthony Barbier
This reverts commit 2e8c7ee2ecebd9783c97bbd602a61989e1247d6b. Change-Id: Id90691f427a68d01480889f8d5fff190fd72c5a3 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121176 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-913: Fix preprocessing step for TF models.Georgios Pinitas
Change-Id: If0fbb6bbe5384038124d3dc189274b8266f796ca Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120771 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Reviewed-by: Pablo Tello <pablo.tello@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-939 Fix mismatches and finalize CLSoftmaxLayer optimizationGiorgio Arena
Change-Id: I4404f91a270e0ba7bbb7451c4c43a485fd4a3f6c Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121105 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-915: Create ResNet50 exampleAlex Gilday
ResidualLayer node (COMPMID-916) also created as required for the ResNet architecture. Change-Id: I3aef0b6d6fd5bfcd4916fed4d8d4466b8a92b70d Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120562 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
2018-11-02COMPMID-909: Enabling in-place computation for batchnormalization and ↵Michele Di Giorgio
activation at graph level Change-Id: I84d4a212629b21794451ab5fb5c5b187b5e28f98 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120127 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-925: Enabling OpenCL tuner in the graph examplesMichele Di Giorgio
Change-Id: I4fe501281f527e20e8fdd0253d59ea2c4629056b Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120354 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
2018-11-02COMPMID-845: Create a ConvolutionLayer for CLIsabella Gottardi
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>
2018-11-02COMPMID-934: Return an error in Validate when we don't support asymmetric ↵Anthony Barbier
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>
2018-11-02COMPMID-828 - Add support for pool widths 4, 5 & 6 and for non square data ↵Isabella Gottardi
sizes - Part 2 (NEON) Change-Id: I64bc8e3f71236edb71494f431ee34077eb8814ca Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118203 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-938: OCLgrind: Mismatches in depthwise convolution on BifrostGeorgios Pinitas
Invalid conversions in oclgrind when clamp is used. Removed call to clamp in CL kernel and replace with convert_sat. Change-Id: I3cd9b87dc10c65d307fbf6eb0aec1b671fba6e97 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/121062 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-784: Productise Winograd.Pablo Tello
a) Added support for kernel size 5. b) Templatised data type for transforms and batched gemms kernels. Change-Id: Idb83dda7a5eec19e015888ab31902bd791913297 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120540 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-582: Add validation to channel_extract kernels.Ioan-Cristian Szabo
Change-Id: I6413a05f6870a0d04f12d7348269b15297ae8493 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/114696 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-882 - Optimizing GEMMLowp on OpenCL reshaping matricesGian Marco
This new optimization allows to achieve 36.3 % of MAC utilisation on Mate 9 @ 1GHz. The performance have been reported here https://confluence.arm.com/display/MLENG/GEMMLowp+performance%3A+ACL+18.02 Change-Id: I71b6a217068763dfdc11bbf3574ee0eb94f93679 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118531 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-930 Disable warning about implicit-fallthrough in new versions of GCCAnthony Barbier
Change-Id: Ibbacb65cb84ccfb64960ac00c482040353be468c Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120875 Reviewed-by: Pablo Tello <pablo.tello@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-905 Optimize CLSoftmaxLayer for QASYMM8Giorgio Arena
Change-Id: I3512d67b8a72b17db1381842ca42780e39cc511c Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120605 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
2018-11-02COMPMID-765 Move direct convolution output stage to the right fileGiorgio Arena
Change-Id: Ifb4d27ba05aa618babb79b1f8e95fbfa689c5f3a Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120792 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-856: CL Depthwise Convolution QASYMM8 supportGeorgios Pinitas
Change-Id: Ic6097e7cf160e8b829fb521b7b99d9a57d9799d3 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118774 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-931 Bugfix for NumPyBinLoader with SubTensor output. (#345)hakanardo
* Bugfix for NumPyBinLoader with SubTensor output. When a SubTensor is used as output from a NumPyBinLoader, data should not be written to the begining of the buffer. * Reduce the buffer size with the same amount * SubTensor's has to be accessed through execution windows Change-Id: Ib15b19f7873d73d8cff0497adfe60805c56c640d Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120733 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Pablo Tello <pablo.tello@arm.com>
2018-11-02COMPMID-906: Use fused activation in NEON Batch normalizationGeorgios Pinitas
Change-Id: I5a6413548b2c9b8972c91ddba57395509dffd87e Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120656 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-765: Fix CPPPermute error when permuting the strides.Georgios Pinitas
Change-Id: I4ea57579d997dd6a2e248634e3b7cb58bb3e2838 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120693 Reviewed-by: Pablo Tello <pablo.tello@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-765 : NEON Wrapper initial traits and overloadsGeorgios Pinitas
Change-Id: Iea4c4732d19e8cf9b245ac2a9f75b2aa70a5839e Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118149 Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-765: Remove wrong shape in LargeShapesBroadcastMichele Di Giorgio
An additional shape was added by mistake which caused tests to fail. Change-Id: Ic0300f2dff336b4a13b2d647bf5507427a4a939a Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120588 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-765: Switch 1x1 DeconvolutionLayer to use the ConvolutionLayerGeorgios Pinitas
-Swithes the 1x1 DeconvolutionLayer to use the ConvolutionLayer instead of the DirectConvolutionLayer. Change-Id: I3ffe152c42c3b1c7ea572f264cd3215df01aedc2 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120292 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-765: Sanitize permutation vector for Permute.Georgios Pinitas
If permutation vector is bigger than the tensorshape to permute then infer dimensions of size one for the extra dimensions. Change-Id: I5addb292f770d925f47f756902e16073039e8f71 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120473 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Stefana Simion <stefana.simion@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-905 Asymm functions support for all vec sizesGiorgio Arena
Change-Id: Ie0c5885a60771f728f80a8c4bdb7f1e4085fa3ee Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120267 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-850: Increase tolerance while issue is being investigatedAnthony Barbier
Change-Id: Ibcb230898db03f5cbb1a84987fef8862aa127d79 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120279 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
2018-11-02COMPMID-589: Refactor HOGDescriptor validation testJohn Richardson
Change-Id: Iee93f17a8ce303cff878b1c5d8904947fb910838 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118892 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-765 - Used GEMM-based convolution in VGG16Gian Marco
In order to use GEMM-based convolution in VGG16, it has been created a function which allocates 1.8 GB. If the function fails, will be used DIRECT convolution instead Change-Id: Ibec8928ee6fe6684d6dc24b7df380beeb671bf27 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/119490 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-903: Implements NEPermute for NHWC conversionsGeorgios Pinitas
Change-Id: I4083e8d16bb23933634f229a1408dfd0e8f2922a Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120069 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-918: AArch32 SGEMM generic kernel fix.David Mansell
This kernel was missing the test for zero iterations of main loop, added it. Change-Id: I661066ee4644f83a847ee7cbd9b8c9642d175744 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120278 Reviewed-by: Pablo Tello <pablo.tello@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-765: Fix CLDeconvolutionLayerUpsampleKernel access window.Georgios Pinitas
Change-Id: I4893060ee2fe46db16aac6ee762c45dd30f35cc0 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/120216 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-792: Update doctrings of CL kernels supporting broadcast operationsMichele Di Giorgio
Change-Id: I71146a83c67c4b193ef1e79d78bd80f9449781e2 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118748 Reviewed-by: Anthony Barbier <anthony.barbier@arm.com> Tested-by: Jenkins <bsgcomp@arm.com>
2018-11-02COMPMID-875: Deconvolution 4x4 not workingGeorgios Pinitas
-Enforces the use of the ConvolutionLayer function in the DeconvolutionLayer. -Adds tests for 4x4 Deconvolution. -Alters the ConvolutionLayer validation to support even kernels. Change-Id: Id27e285f078e690b8dd58490dd8ea6d875b3cec6 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118632 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-884: Valgrind: NEDirectConvolutionLayerKernel invalid readGeorgios Pinitas
Change-Id: I258f03b61446e8333645efe80f2857e8c725b9de Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118943 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-828 - Add support for pool widths 4, 5 & 6 and for non square data ↵Isabella Gottardi
sizes - Part 2 (CL) Change-Id: I004906b9b1f11158fe17b4aa2640a7f4685fb929 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118462 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-765 - Extended GEMM benchmarkGian Marco
Added new benchmarks GEMM in order to evaluate the performance when the input matrix B has to be reshaped only once Change-Id: I1c4790213704ce57ea7b28f6f362c56edccd1eb9 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118910 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com> Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
2018-11-02COMPMID-897 Merge batch normalization with bounded reluGiorgio Arena
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>
2018-11-02COMPMID-892: OCLGrind failures on both validation and benchmarkGeorgios Pinitas
-Adds quantization info to the ActivationLayer benchmark fixture -Replaces clamp with convert_sat in depthwise conv kernel -Fixes ROIPooling execution slice Change-Id: Ie9bbe08abcfb8278456964e476b0948247c7ecba Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/118957 Tested-by: Jenkins <bsgcomp@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
generated by cgit v1.2.1 (git 2.23.0) at 2024-06-29 13:03:36 +0000