aboutsummaryrefslogtreecommitdiff
path: root/1.0
AgeCommit message (Collapse)Author
2020-03-26IVGCVSW-4447 Add Hal 1_3 SupportKevin May
* Add new 1.3 files HalPolicy, ArmnnDriver, ArmnnDriverImpl * Add new .rc file for 1.3 service * Add ArmnnPreparedModel_1_3 and implement new functions * Update Android.mk with 1.3 driver and service * Refactor ifdef to include ARMNN_ANDROID_NN_V1_3 * Create Utils getMainModel for new 1.3 Model Main Subgraph * Use android Utils to convertToV1_X in ArmnnPrepapredModel_1_3 * Refactor HAL 1.2 convert functions into ConversionUtils_1_2.hpp * Replace ArmnnBurstExecutorWithCache with call to ExecutionBurstServer Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: I514069e9e1b16bcd1c4abfb5d563d25ac22d02e3
2020-03-03IVGCVSW-4473 Android R pre Hal 1_3 build changesKevin May
* Update ErrorStatus to V1_0::ErrorStatus * Update Request to V1_0::Request * Update OperandType to V1_2::OperandType * Add namespace android::nn::hal in ArmnnDriverImpl for R only * Add missing g_RelaxedFloat32toFloat16PerformancePowerUsageName * Add namespace V1_0 or V1_1 where necessary * Update Android.mk with R macro and android.hardware.neuralnetworks@1.3 * Remove androidnn.go * include IAllocator in DriverTestHelpers * Remove unused LOCAL_CFLAGS Signed-off-by: Kevin May <kevin.may@arm.com> Change-Id: I1787f1ed6784b3bbec017536d87d49197405e853 Signed-off-by: Kevin May <kevin.may@arm.com>
2020-01-09IVGCVSW-4315 Fix Fully Connected infer output shape bugFinnWilliamsArm
Change-Id: If4fd1abdedf7de2046435d418fb1ee95ceb73419 Signed-off-by: FinnWilliamsArm <Finn.Williams@Arm.com>
2019-10-02IVGCVSW-3737 Add support for converting DEPTH_TO_SPACEAron Virginas-Tar
* Added ConvertDepthToSpace() to hal_1_0::HalPolicy and hal_1_2::HalPolicy * Implemented ConvertDepthToSpace() template inside ConversionUtils.hpp * Changed unsupported operation from DEPTH_TO_SPACE to HASHTABLE_LOOKUP in GenericLayerTests/GetSupportedOperations Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I12bf73ea721e7b6d49cc4a76000b43a3f274c6f5
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
2019-07-31IVGCVSW-3601 Fix skipped VTS Concatenate TestsMike Kelly
* Fixed Skipped VTS Concatenate Tests. Signed-off-by: Mike Kelly <mike.kelly@arm.com> Change-Id: I29e7dcdedefc0e9c54f86fa5de23aa714c469585
2019-07-29IVGCVSW-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-3578 Report dynamic output unsupported for further operations in ↵Aron Virginas-Tar
hal_1_0::HalPolicy * LocalResponseNormalization * Mul * SpaceToDepth * ResizeBilinear Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: Ie52dd7b274d721f0ca715ad92277e12e4c2b1913
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-3529 Fix UnknownCombinationsTest for Android Q NeuralNetworks 1.0 & 1.1Aron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: Ifb6875c8445027f597ad13f29a0c0ae5b94e30ac
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
2019-07-24IVGCVSW-3567 Add more verbose logging to conversion methodsAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: If1ce2aff656ff995f41d33777cdc6fb966392cca
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-3498 Fix Vts 1.2 fully_connected dynamic output testsFinnWilliamsArm
Signed-off-by: FinnWilliamsArm <Finn.Williams@arm.com> Change-Id: Id28e4bd1d55f02c5011f8a364d4b715d1a9190db
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-3460 Fix VTS l2_normalization dynamic_output_shape test failuresFinnWilliamsArm
Signed-off-by: FinnWilliamsArm <Finn.Williams@arm.com> Change-Id: I2ea4a92ee7ce6282c9fad35deefef99349fe291d
2019-07-11IVGCVSW-3461 Add support for dynamic output shape in ConvertDequantizeAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: Iad2673ffbeb334ab9a39ce09600f476ce96f4165
2019-07-11IVGCVSW-3459 Add support for dynamic output shape in ConvertSoftmaxAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: Ie1777ec774fe29369b61b0f881cc4b7cef0485a5
2019-07-10IVGCVSW-3482 Report operations with dynamic output size as unsupportedAron Virginas-Tar
Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: Ifafe2a6fbfd6019b3395d51ed9967db794d2b034
2019-07-09IVGCVSW-3396 Support joined lstm parametersFerran Balaguer
!armnn:1470 Signed-off-by: Ferran Balaguer <ferran.balaguer@arm.com> Change-Id: I67a393c1556f0b3022436e41f82f2bf1ab3a1d40
2019-07-08IVGCVSW-3197 Support SPACE_TO_DEPTH on AndroidKeith Davis
* Added support within 1.0/HalPolicy * Added support within 1.1/HalPolicy * Added support within 1.2/HalPolicy * Updated NnapiSupport.txt Signed-off-by: Keith Davis <keith.davis@arm.com> Change-Id: I2dc2743ee2f858d2b3496aae0f3e3e28a96ee96e
2019-07-03Fix bug in hal_1_0::HalPolicy::ConvertResizeBilinearAron Virginas-Tar
* Fix order in which target width and height parameters are read Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I2430abc50f4952a94024d55f8e024cd282d4e0ad
2019-07-02IVGCVSW-3382 Deprecate ResizeBilinear and use Resize with Bilinear methodAron Virginas-Tar
!armnn:1450 Signed-off-by: Aron Virginas-Tar <Aron.Virginas-Tar@arm.com> Change-Id: I901f5093933aa8c1cfa05d8af383ef06a1c1f203
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
2019-06-11IVGCVSW-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
2019-05-30IVGCVSW-3146 Update Android NN Driver to support DequantizeDavid Monahan
Signed-off-by: David Monahan <david.monahan@arm.com> Change-Id: I734336fe0487198ec03779d3a85706883c137163
2019-05-29IVGCVSW-3119 Rename Merger to ConcatJim Flynn
!armnn:1209 Change-Id: Ic493e5cdfe479e459342d7c7c9d77c63f859fa30 Signed-off-by: Jim Flynn <jim.flynn@arm.com>
2019-05-20IVGCVSW-3125 Deprecate CreateMergerDescriptorForConcatenation functionJim Flynn
!armnn:1182 Change-Id: I8687e33c51593810eef45a4de7d5b856bcab4e3b Signed-off-by: Jim Flynn <jim.flynn@arm.com>
2019-05-14Use the new deprecation APIMatteo Martincigh
* Refactored the code to no longer use the deprecated methods where applicable !armnn:1125 Change-Id: I954c8497c098ea3578986eede612c55637fcd3d9 Signed-off-by: Matteo Martincigh <matteo.martincigh@arm.com>
2019-05-14IVGCVSW-3076 Add ConcatLayer methods to public APIJim Flynn
!armnn:1119 Change-Id: Ic187f49427e38345f279537ed10a9275980c1360 Signed-off-by: Jim Flynn <jim.flynn@arm.com>
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>
2019-04-25MLCE-117 Handle more cases for implicit flattening of Fully Connected inputMatthew Bentham
Adds new unit test cases, and changes the implementation of FlattenFullyConnectedInput to more closely match the documentation of Android NNAPI. Change-Id: I7ca96b1168b9c7bc78db66f53b0cc776153fd780 Signed-off-by: Matthew Bentham <matthew.bentham@arm.com>
2019-04-25MLCE-117 Add a unit test for implicit flatten of FC layer inputMatthew Bentham
Change-Id: Ia4dd63927a54aa0cc24d5a378f30189c957f12e8 Signed-off-by: Matthew Bentham <matthew.bentham@arm.com>
2019-04-04IVGCVSW-2886 Support multiple backends in Android driverNattapat Chaimanowong
Change-Id: I4abe1f750801911570b6dc65c187b828c5929b5f Signed-off-by: Nattapat Chaimanowong <nattapat.chaimanowong@arm.com>
2019-02-22MLCE-91: Fix for LSTM not supporting optional inputv19.02branches/android-nn-driver_19_02David Monahan
Change-Id: Ie4b8d92723ecb9834c97c006f3c2f929508176d2 Signed-off-by: David Monahan <david.monahan@arm.com>
2019-02-22MLCE-87 Fix ConvertFullyConnected not supportedKevin May
Change-Id: If9f06b5b117972a43b5d1516c27ad350fafc4263 Signed-off-by: Kevin May <kevin.may@arm.com>
2019-01-23IVGCVSW-2394 Update Android Hal Policy for ResizeBilinear IsLayerSupportedSadik Armagan
* IsResizeBilinearSupported function requires outputInfo as well. Change-Id: Ib74241ef2db147bb64fa77287ad0d616a2e4159d
2019-01-11IVGCVSW-2454 Merge together the pluggable backends work (was in aMatteo Martincigh
separate branch) and master * Minor fixes to make the NN Driver build with the merged changes * Passed missing ReshapeDescriptor to the layer support function !armnn:491 Change-Id: I61fb6ed8df6895bd5a12c6f2b1c322929eebe176
2019-01-02MLCE-77 Depthwise Convolution with depth multiplier > 1 doesn't workMatteo Martincigh
* Changed the weight swizzling to [ M, I, H, W ] as now required by ArmNN !armnn:460 Change-Id: I7c25e7ab3e1efc47d7db3f2b57e17382ea8b36cf
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
2018-11-08IVGCVSW-1972 - Update HAL Policy Convolution2d to use NHWC data layoutexperimental/nhwc-previewnarpra01
Change-Id: I0e967b46b213a3b122e9d6aa564d909d2c546b00
2018-11-01IVGCVSW-2084: Edit HAL Policy NHWC DepthwiseConvolutionJames Conroy
Change-Id: Iee54c42e9abd927447d958377f3aee266a60ca5d
2018-10-29IVGCVSW-1986 Modify HAL Policy for Resize BilinearMohamed Nour Abouelseoud
Specified NHWC data layout in ResizeBilinear in HAL policy Change-Id: I0160ec992b595a39ece6d8b060393c5067e8d3a5
2018-10-25IVGCVSW-2065 - Modify HAL Policy for Normalization to use NHWC data layoutnarpra01
Change-Id: Ic60c4dacb55bcf2514f011a8e844e7b8f7b13560
2018-10-22IVGCVSW-1875 Proper support for NHWC L2NormalizationMatteo Martincigh
* Removed input/output swizzling from ConvertL2Normalization Change-Id: I623d878617614aec4d1f41556839808e6184a350
2018-10-12IVGCVSW-1964: Replace optional biases with home-grown Optionalarovir01
!armnn:151777 Change-Id: Ib7eba10f05810c03d7384256828bfd0c23bb6bba
2018-10-12IVGCVSW-1863 Support NHWC for L2NormalizationMatteo Martincigh
* Added L2NormalizationDescriptor to ConvertL2Normalization !armnn:150051 Change-Id: I622d7a8af2bfd27e7d4543a24c59e53988cb5d82
2018-09-18IVGCVSW-1806 More Android NN Driver refactoringMatteo Martincigh
* Changed #if defined to #ifdef * Simplified the Android ML namespace resolution * Fixed the relative path in some include directives Change-Id: I46e46faff98559c8042c1a4b8b82007f462df57d
2018-09-18IVGCVSW-1713 Create a minimum unit test to compare the resultsNikhil Raj
before and after passing the FP16 flag in the Android-nn-driver Change-Id: If8d4ca12421c3bee2526eec98f11d393af822373
2018-09-18IVGCVSW-1806: Refactor Android-NN-Driver ModelToINetworkConverterarovir01
* Moved conversion logic into new V1_0 and V1_1 HalPolicy classes * Extracted common helper functions into ConversionUtils class Change-Id: I1ab50edc266dd528c0cb22a5cd1aa65e103674d9
generated by cgit v1.2.1 (git 2.23.0) at 2024-04-29 16:23:03 +0000