aboutsummaryrefslogtreecommitdiff
path: root/ConversionUtils.hpp
AgeCommit message (Collapse)Author
2020-01-15IVGCVSW-4241 Reenabling support for NCHW.v19.08.1branches/android-nn-driver_19_08Mike Kelly
* Reenabling support for NCHW for Convolution, Pooling and Resize. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I6ceb1ab0983e19d200ba353b0017b0f0cd1628e0
2020-01-15IVGCVSW-4315 Fix Fully Connected infer output shape bugFinnWilliamsArm
Change-Id: If4fd1abdedf7de2046435d418fb1ee95ceb73419 Signed-off-by: FinnWilliamsArm <Finn.Williams@Arm.com>
2019-12-20Revert "IVGCVSW-4305 - Re-enable NCHW on Pooling2D"Kevin May
This reverts commit b3680df4298a2f4d4ef8b637f35c3bc343b05339. Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: Iad15bb7360817eb91428681c05cfb065e317b4d0
2019-12-19IVGCVSW-4305 - Re-enable NCHW on Pooling2DKevin May
Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: I85ed84b818c15f402b220f223fb9e728e08dceed
2019-12-19IVGCVSW-4301 Correctly validate reshape for broadcastable inputsDerek Lamberti
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com> Change-Id: I4db6ea4ed0a192c85f124c4a9ced60b1666a3870
2019-12-19IVGCVSW-4303 Re-enable NCHW on Depthwise ConvolutionKevin May
* Also re-enable NCHW on BatchToSpace, SpaceToBatch and Transpose Conv Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: I386eb3d852978e7cc611efad4f9f0c982140865a
2019-12-17MLCE-144 Resolve CTS failuresMike Kelly
* Re-enabled Float16 * Re-enabled Concatenation Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I03bbe978947d5282d50daa2ed9795d042dc93cb0
2019-12-13MLCE-144 Disabling NCHW and Float16Mike Kelly
* Disabled support for NCHW * Disabled support for Float16 Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I0f5ee8a8ddda1edc7800f167d6a82e4a15c6069b
2019-12-12IVGCVSW-4262 Add Calls to IsReshapeSupported and IsPermuteSupportedKevin May
!armnn:2466 * Add calls before addition of these layers in ConvertConcatenation * Add outputInfo parameter wherever needed for IsReshapeSupported Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: Ic5d142ea046161960ff2fc137bd261ebb4e6ac0c
2019-12-11Catch std::exception instead of armnn::ExceptionMike Kelly
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com> MLCE-143 Fix for reference implementation for service stopping when running CTS. *ConvertDepthwise reports that weights are not supported with the first dimension of the tensor is not 1 *ConvertConcat was missing one case for reporting unsupported inputs. Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com> MLCE-144 Cts NNAPI test cases failed * Fixed numerous CTS/VTS failures related to Quantization !armnn:2457 Signed-off-by: Mike Kelly <mike.kelly@arm.com> MLCE-144 Fix 2d pooling convert function Signed-off-by: FinnWilliamsArm <Finn.Williams@Arm.com> Change-Id: I054635ebfd52cb5575490c3bfaae0104eb1685cc
2019-08-19IVGCVSW-3618 Fix VTS/Neural Netwok Test FailuresSadik Armagan
* Added IsReshapeSupported check on available backend(s) after adding Reshape layer. Signed-off-by: Sadik Armagan <sadik.armagan@arm.com> Change-Id: I38a36d5a3db0d67f0e4b4be2ef672dedb75bff22
2019-08-15IVGCVSW-3643 Fix Ref Normalization failuresMike Kelly
* Added check for unsupported optional parameters for L2Normalization and LocalResponseNormalization to skip the related tests rather than fail them. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: Ia17be706b14b2e2faf8eb5a0ec335c2e3863bf16
2019-08-14IVGCVSW-3633 Refactor HalPolicy to fully support V1.2 modelsMike Kelly
* Templated and moved V1.0 and V1.1 Convert methods to ensure they can work with later versions of models, operations and operands. * The V1.2 HalPolicy no longer converts V1.2 models, operations and operands to earlier versions. * The V1.2 HalPolicy no longer passes operations to the V1.1 or V1.0 HalPolicies for conversion. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I5de59d43a3abb1f8ac0253dc637ad68318960c76
2019-08-07IVGCVSW-3540 Fixed PReLU Being SkippedMike Kelly
* PReLU layers were being skipped due to logic error. Resubmitting due to accidental reversion. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I6aeed11519d19ba7ec8b4993694e414f58799028
2019-08-02IVGCVSW-3604 Fix TransposeConv2d padding calculationNarumol Prangnawarat
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com> Signed-off-by: Aron Virginas-Tar <aron.virginas-tar@arm.com> Change-Id: I5c10ab18343ecf0ebeab24a436e5be2b6c2831c7
2019-08-01Fix explanatory comment around SanitizeBiasQuantizationScale()Aron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I92700da149a800f1d8215e964d1e67056628f067
2019-07-31IVGCVSW-3601 Fix skipped VTS Concatenate TestsMike Kelly
* Fixed Skipped VTS Concatenate Tests. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I29e7dcdedefc0e9c54f86fa5de23aa714c469585
2019-07-31IVGCVSW-3586 Fix Skipped Space_To_Batch Hal 1.2 VTS FailuresFinn Williams
Signed-off-by: Finn Williams <Finn.Williams@arm.com> Change-Id: I9d0d0a75d54c25075d8d87c6265e350486157f5c
2019-07-31IVGCVSW-3583 Correct data layout index for ConvertBatchToSpaceNdFinn Williams
Signed-off-by: Finn Williams <Finn.Williams@arm.com> Change-Id: Ibd683c147b6f2fb966a8f90105c9213880100cf8
2019-07-30IVGCVSW-3583 Fix Skipped Batch_To_Space Hal 1.2 TestsFinn Williams
Signed-off-by: Finn Williams <Finn.Williams@arm.com> Change-Id: I3f2928ba86a9d306a7eb400db3a420e42cf3fa7e
2019-07-30IVGCVSW-3591 Fixed unexpectedly skipped SUB testsMike Kelly
* A model that has Inputs with different quantized scales is not compliant with 1.1 Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: Ifb8277d78f05b5ef017effa879322a08c0efc851
2019-07-29IVGCVSW-3593 Remove HAL1.2-specific code from HAL1.0 ConvertConv2d() and ↵Aron Virginas-Tar
ConvertDepthwiseConv2d() * Removed code for reading data layout flag (only available in HAL1.2) * Removed code for reading dilation parameters (only available in HAL1.2) * Added more restrictive input size checks (== instead of >=, as none of the inputs are optional in HAL1.0/1.1) * Removed superfluous input size validation from 1.0/HalPolicy.cpp Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I8387120b9abcc8458f7fac6df577cb6453051be3
2019-07-29IVGCVSW-3540 Fixed PReLU Being SkippedMike Kelly
* PReLU layers were being skipped due to logic error. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: Ic0ec707da784b44d72958dee02901b0173b25e5c
2019-07-29IVGCVSW-3582 Fix Skipped Pooling Hal 1.2 TestsSadik Armagan
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com> Change-Id: I9f9d11b19de591ec8936822ca229b4090ed318fe
2019-07-26IVGCVSW-3578 Do not attempt to infer dynamic output shapesAron Virginas-Tar
* Report dynamic output tensors as unsupported for all operations, regardless of HAL level Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I2341dc96be965886666b75515e9a226d813a1591
2019-07-25IVGCVSW-3572 Fix tests failing due to usupported input tensorAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I55502cf2abcf28a1c1810269f1955ca78b23c39a
2019-07-25IVGCVSW-3530 Fix DynamicOutput Tests for Android Q NeuralNetworks 1.0 & 1.1Sadik Armagan
* Fixed for failing Conv2d, DepthwiseConv2d, and Activation tests on Hal 1.0 and 1.1 in Q Signed-off-by: Sadik Armagan <sadik.armagan@arm.com> Signed-off-by: Aron Virginas-Tar <aron.virginas-tar@arm.com> Change-Id: I435338b90b6c501320083f2fd9372e3a4ac3c32c
2019-07-25Fix verbose logging in ConvertPadAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I9b1b45f7723147456d2fb9c5b94822d581cf446e
2019-07-25IVGCVSW-3521 CpuAcc V1.2 pad FailuresMike Kelly
* Fixed Pad and PadV2 failures and skips. * Templated ConvertPad to enable float16 tests to run. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I50ded84fe44ea5d5949e877f383f32adff88680d
2019-07-23IVGCVSW-3530 Fix DynamicOutput Tests for Android Q NeuralNetworks 1.0 & 1.1Sadik Armagan
* Updated ConvertToActivation, ConvertConv2d, and ConvertDepthwiseConv2d functions to infer output shape from input if it is dynamic Signed-off-by: Sadik Armagan <sadik.armagan@arm.com> Change-Id: Ie24fbfd87c6186c69c3ecba5c68a6866507fb449
2019-07-23IVGCVSW-3553 Fix failing zero_sized testsAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: Idd10f34babc0d2552d599872b853ba5fb5c98351
2019-07-16IVGCVSW-3477 Refactor android-nn-driver to use armnn ILayerSupportedFerran Balaguer
!armnn:1508 Signed-off-by: Ferran Balaguer <ferran.balaguer@arm.com> Change-Id: Ica5fcb683f101bde9e651f0be0f5b9b4c409d1aa
2019-07-12IVGCVSW-3454 Fix VTS convolution test failures where output shape is incorrectMike Kelly
* Changed calculation when calculating implicit padding to take dilation into account. * Added CalcPadding function that takes dilation as a parameter. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I38981d9b2cfe8ef36cd0a14d090b541111e7bf34
2019-07-09IVGCVSW-3456 Add support for dynamic output shape in ConvertPreluAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I8fc7a716455be3f51b51177f6896a73790a41fc3
2019-07-05IVGCVSW-3301 Support PAD_V2 in HAL1.2 driverAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: If3da7b9015fb816bb27b85ab491102005859b3b0
2019-07-05IVGCVSW-3370 Add broadcasting support to PReLU to properly run the Android ↵Matteo Martincigh
VTS/NN tests * Updated ConvertPrelu to support input broadcasting * Updated the BroadcastTensor utility function so that it preserves the order of the inputs * Updated the documentation Signed-off-by: Matteo Martincigh <matteo.martincigh@arm.com> Change-Id: Ibbac6f7161132740e61c85f597f8be70cd5d7325
2019-07-05IVGCVSW-3295 Support ResizeNearestNeighbour in HAL1.2 driverAron Virginas-Tar
* Added ConvertNearestNeighbor to 1.2/HalPolicy Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I9858849b70be36e7e5ced3dae667c85c03e50d4f
2019-06-18IVGCVSW-3136 Run VTS tests with 1.2 DriverSadik Armagan
* Updated android-nn-driver to run VTS tests with 1.2 Driver Change-Id: I99a73b1e32dcf5bc655d88c2dbe0e610e245ea15 Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
2019-06-14IVGCVSW-3280 Refactor conversion methods to depend only on HalPolicyAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I162cdca922655d1bd71b18fc5d2937351f8879be
2019-06-12IVGCVSW-3265 Add support for optional Convolution2d dilation parametersAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I3968d682b67439d4b4aca670d074ea4f8993a39a
2019-06-12IVGCVSW-3272 Add support for optional DepthwiseConvolution2d dilation parametersAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: Idce519adcd4a2983e8c338b2e4483e040ad879d2
2019-06-11IVGCVSW-3181 Add HAL 1.2 support to android-nn-driverMike Kelly
* Updated Android.mk to build HAL 1.2 driver * Added 1.2 HalPolicy and ArmnnDriver * Added 1.2 ArmnnPreparedModel * Updated converters and utilities to accept new HAL 1.2 operands and operand types. Signed-off-by: Sadik Armagan <sadik.armagan@arm.com> Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I62856deab24e106f72cccce09468db4971756fa6
2019-05-07IVGCVSW-2911 Work towards compatibility with QMatthew Bentham
Explicitly use HAL V1_0 Operand, OperandType, and IPreparedModelCallback Change-Id: If6aa6e9dc1bd0b7673ec247ac3560e7f072ef699 Signed-off-by: Matthew Bentham <matthew.bentham@arm.com>
2019-04-10IVGCVSW-2947 Remove boost dependency from include/TypesUtils.hppAron Virginas-Tar
!armnn:962 Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: Ia70c4b8397e5d6422d1947fca8c8db29a423822e
2019-04-04IVGCVSW-2886 Support multiple backends in Android driverNattapat Chaimanowong
Change-Id: I4abe1f750801911570b6dc65c187b828c5929b5f Signed-off-by: Nattapat Chaimanowong <nattapat.chaimanowong@arm.com>
2019-03-15MLCE-91 LSTM doesn't support optional inputKevin May
* Add fix for optional NO_VALUE operands in ConversionUtils.hpp * Remove fail message for optional NO_VALUE in ConversionUtils.hpp * Add to existing tests and test helper to cover optional NO_VALUE Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: Icf36af1fc00d3fb33cdd77ff6d6618cc4700d3fd
2018-11-18IVGCVSW-2127- Update HAL Policy for mergerbranches/android-nn-driver_18_11narpra01
* Remove permutation when concat axis is inner most * Add additional parameter to IsMergerSupported as changed in armnn !armnn:151 Change-Id: Ie214c9573f242d8f04d58fc61621ad3831991d9a
2018-11-12IVGCVSW-1981: Edit HAL Policy for NHWC Pooling2DMatteo Martincigh
* Removes permutes of tensors for Pooling2D, as NHWC is now supported by Arm NN. Change-Id: I48417c91f387b6f73bc071e473828f2ee5949332
2018-10-29IVGCVSW-1065 Allow MODEL_OUTPUT as input to another layerMatthew Bentham
Change-Id: I24fe0916847ec713abf952b9ea911693a692d440
2018-10-22IVGCVSW-1930: Simplifying the driver conversion function for Pooling2d,Éanna Ó Catháin
removing the conversion to splitter layer if pool is 1x1. Change-Id: Iea3af08652dd6fd1e6820ef43d9101d38c230b5e
generated by cgit v1.2.1 (git 2.23.0) at 2024-05-16 01:33:56 +0000