Age | Commit message (Collapse) | Author |
|
Added support for FP16 in CLWinogradConvolutionLayer: 5x5 kernels and 3x3 kernels(COMPMID-937)
Change-Id: I0f394cbdc978dd04176416e9f612aca3986b09e6
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145537
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
|
|
Change-Id: I36b119f27c87e2ce3d82a46089a29e298509772c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/146625
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
|
|
COMPMID-1515: Add FP16 support to NEFloor and CLFloor
Change-Id: Ib63a62c7681056ee13be99ce081b4d3949da4217
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/146547
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
|
|
Change-Id: I0dbc4fd7f640d31daa1970eb3da0e941cb771f2b
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/146145
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
|
|
Change-Id: I12ba4c0c35f086ea3f395970b85af5bf8f94850b
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145052
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I4342d4240fe5b1aab234c015684a1216c3990a5f
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145631
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I33079e07acad83e0740c87b06546cc0f0f51301c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145164
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Jenkins <bsgcomp@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>
|
|
Still failing on S7 edge. Using absolute tolerance when relative tolerance is
not enough.
Change-Id: Iadcbc6e58eb80261c3f8e75e6fa43701c6ec854c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145973
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Setting the relative tolerance to the same value as for CLConvolutionLayer and
CLFullyConnectedLayer.
Change-Id: Ia4b483f07df5f2de8a1e3ceec51d80798881929e
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145909
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I7dee44deb360d057a689e92ad4c72c9a4309a789
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145715
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
The output of NECol2Im is already auto-initialized.
This patch is about calling ShapeCalculator instead of computing the shape inside the kernel, adding validate_and_configure_window, and standardize the way convolved dims are passed (now NEON uses Size2D, while CL passes a pair of uint values: using Size2D for both implementations)
Change-Id: I795696e1b6532f57847c3186c1b532c09f5a25da
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145345
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I46965aeb1fffba8cbf083cab7284c549b0e94d00
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145334
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
|
|
Change-Id: I99ebae61024a7bce9d17292a02c28626ae6c29d5
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/144872
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
|
|
On GEMM we had accuracy issue
On Magnitude we have disabled the fp16 acceleration since we do not have feature parity with CL
and this function is not used for ML
Change-Id: Iaebe3bbbd2a9f45db0c714aa5ebaf48eb0b65741
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/145467
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@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>
|
|
The mismatches are due by the FP16 reference implementation which does not match exactly the fp16
acceleration we have on Mali. The only workaround for this is to introduce a tolerance number.
So if the number of mismatches is greater than 7%, the test fails
Change-Id: Iab7bbee6d77eb5b5fda410c350025060c1cc7724
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/144964
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Tolerance issue
Change-Id: I50f05f63617bfb364648bb20f6ad41676a79174e
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/144493
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Using same quantization info and input values range as for ConvolutionLayer.
This needs further investigation to understand why there are mismatches when
using the entire range.
Change-Id: I8c20a341b29a1ac03c811d014911e7efc484c3a6
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/144340
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: I15c7df21773145b03f42b6f78bd7ad2e5b8a5219
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/144126
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Increasing the absolute tolerance as values seem to differ by at most 2.
Change-Id: I7f70f432760b64ee6c96a5fdeb34865c0f8f4796
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/144154
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I13ec79b6668e2b9559d3fa789ae0b51ab6975289
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/139126
Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I9e9b267ea58fd2339467af6f49ae76e9195cbc61
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143682
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I5188a2163e7341f1915d98c21464fea13a9a7faf
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143330
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
|
|
Change-Id: I4afb19751520a90fee27fb49b775cd10e92a94f5
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/140476
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
This causes problems when ACL is used as a shared library on Android.
Fixes some problems related to creation / destruction order between the Graph's CL backend and core / runtime
Change-Id: I716d63fd42f4586df1ffbb6fa97e4db06d3a781b
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143228
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
|
|
Changed RelativeTolerance to Asbsolute for F16/F32 as the values can
be very close to zero for large inputs.
Change-Id: Ibeab9f4e4d218e4ceaad00b1725acc34e80c7afb
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143576
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change CLReductionOperation border to be multiple of 64 instead of 16.
The opencl kernel works only with local_size(0) being a power of 2. This will
generate a padding of 63 if input_width % 64 = 1, but I don't think it's a
big issue and it keeps the border calculation pretty simple.
Also, increased tolerance for fp32 because there were mismatches
for the 4K image.
Change-Id: Id44990a262b2d6eff4c8ce56eb7c886274d9847e
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143415
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
This changes help to prevent errors like passing a matrix
with less elements than required into the warp functions.
Change-Id: I863f933a5e0568258717cffed3a20788d3d03083
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143044
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
QASYMM8
Fixed also a bug in the graph API related to the bias shape in DepthWiseConvolution for NHWC
Change-Id: I275141a42e51f6747b77db1c31d1bc69e8685af5
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143454
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
The flag "ChannelsFirstOutputNHWC" was not set
Change-Id: Id5f64a839d4e86638a07090e971a4f7ee82af349
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143457
Reviewed-by: Michele DiGiorgio <michele.digiorgio@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
+ validate() function
Change-Id: I6808de0254a7c4bca440322cc14b795b3b32465b
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/142427
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
This patch includes:
- Im2Col optimizations for NHWC using a new data layout
- Refactoring of CLIm2ColKernel adding validation method and auto-init
- Removed im2col_reduced from CLIm2ColKernel and created a new kernel CLFlattenLayerKernel
Change-Id: I1620640b6796baa268324b33ae92cdd8de53e27c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/141241
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
|
|
Change-Id: Idde333308db71087ec234b3fd1eb4e36a44db46c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/143049
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I55f0018ac7214775ebbca63f58a3bf5c93732fec
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/142632
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@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>
|
|
-Enables cell-to-input weights when !cifg and peephole
-Makes projection bias conditional
Change-Id: Iee866db9f5d8479c2dfd95d74a2d42492bf07a8d
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/140543
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Les Bell <les.bell@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I429087f8aa436cf0877c3abec8fd7201bec1b81c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/141661
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I0404df6d369855e2f458f2db8f26e81c80a1ee87
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/140148
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Jenkins <bsgcomp@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>
|
|
Change-Id: I3ed55bdb95d888aff0b0b76fb841bf1669659308
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/139963
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Added support for different quantization info in the operands
and the output tensors in CLArithmeticAddition.
Change-Id: I7704baccc3b609fcc514f947f1b5b5331745ed9e
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/139947
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
|
|
Change-Id: I8738aca2cc0104e4c4d7c9605762ab59fce10a33
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/137333
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Tested-by: Jenkins <bsgcomp@arm.com>
|
|
Change-Id: I2c6a744f174cfb6c78a9923b737f06537debaa0d
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/139758
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: I25b42c5b74567bb50444da393d723b7f5cedafad
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/138694
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
- Extend support for FP16 in CLReduction.
- For F16/F32 MeanStdDev we perform one reduction operation for mean
and one for stddev and we calculate the final result in the host CPU.
Change-Id: Iad2099f26c0ba7969737d22f00c6c275634d875c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/135870
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
Change-Id: Id5e0795238f77c049df9c109dafc5ef878c1897d
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/139234
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
Change-Id: Ibabce61cf5427de80078a6468023bed05f5e7c2c
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/139006
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
|
|
(part 2)
Change-Id: I4371c4b6403f55ea7d7baf39df91a45f45b21d5a
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/137975
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Anthony Barbier <anthony.barbier@arm.com>
|
|
OpenCL NHWC
Change-Id: Ia07e0dfcbcd07366c4bcb956e298369fb12a0369
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/138759
Tested-by: Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
|