Age | Commit message (Collapse) | Author |
|
* Implemented Serialization and Deserialization of CLContext.
* Fixed flatbuffers android-nn-driver dependency.
!android-nn-driver:4772
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: If806f050535ffaa70922ba0f1ffe7bb10f902329
|
|
* Added CL implementation of Rank Workload
* Removed references to memcpy_s as it's a windows only function
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: Ia63666b9640d76a775f2ab98b3cd7e9f77b5a507
|
|
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I85f0c30757043f8c27c78d607f0f9dbbdd35b9fb
|
|
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I5a85597e75d2b879ae234c6929686fabe99d7bc8
|
|
* Injected CLCompileContext object to each CL workload.
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I4837dbd3d5b56cf743b3b89c944e3cdf8b11a42a
|
|
Sigmoid Activation"
* This reverts commit be25d94aefe53f221304b1f5f344913b708f808b.
* Add Unit Test: any receiver layer + any activation layer in float and QAsymmU8
* Tidy up fuse activation tests
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ie059d03b85cd17eaaafe5188bb173672a1fb9ae0
|
|
* Convolution
* Depthwise Convolution
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I61b356fbffb176e9a05e08d9b6867d082b6712c8
|
|
* Introduced CLCompileContext to ClWorkloadFactory
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Ied38f4336210502e5f518b9955ae6a5ba3d242b3
|
|
* Correct the id when EthosN is enable
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I5203e615f809e56c7597ffeeec56b5ad38d4ff17
|
|
* In favour of ElementwiseUnary functions which
are the currently used code path.
Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: I34964d2dcabd4b7ddf0b455df17c48e3c6812ee4
|
|
* Add supported activations check to Neon FullyConected validate
Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I67a36eb83d0568d000e928e27eba3c84e32cdc72
|
|
* Add CL Logical workloads for NOT,
AND and OR.
* Enable Layer and IsSupported tests on CL.
Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: I8b7227b2487fdbbb55a4baf6e61f290313947de1
|
|
* Add NEON Logical workloads for NOT,
AND and OR.
* Enable Layer and IsSupported tests on NEON.
Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: Ibca59530457a664ca3d77751825642f8daf52fab
|
|
* Add Boolean support for Reshape
* Use LogicalUnary factory and data type for LogicalNot
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I8e072fde200b7716556ae67f79616458cf98ff20
|
|
* GetOverriddenDataType was returning incorrect quantization data
* Optimized CpuAcc and GpuAcc SubGraphs fail validation on debug versions
of ArmNN
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: Ie97935cc2af67bd9aeebc94b63dafa458bd1aa8c
|
|
* Added FP16 DataType support to DetectionPostProcess
* For DetectionPostProcess layer output is always Float32 regardless of input type
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I21f63dd08f0863e9a98e105b3009bab3da1ab0c3
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I144ebfca524f4cdee9cc82eef3995c6b32bfc40b
|
|
* Correct datatype of QAsymmS8
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Id4987b91e06d87735254d3cdd5c9adbe11cc8870
|
|
* Resolved the review items in the main review.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I5da34b74ac204569ea2d210fb5a069beb7d0835b
|
|
* Added Fused Activation Optimization to both CL and Neon backends.
* Added Fused Activation support to all the CL and Neon workloads
that support it.
* Changed ProfilingTest network to be a Convolution layer
followed by an Abs layer rather than an Activation layer.
* Added IBackendInternal::OptimizeSubgraphView function that can accept a
ModelOptions.
* Network will now call OptimizeSubgraphView passing in the ModelOptions.
Signed-off-by: Keith Davis <keith.davis@arm.com>
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ib536ac3cbafc7d9b35c139ad9a65b7735262cd9d
|
|
* Now enter if block if bias OR weights have
multiple quantization scales.
Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: I5eba0ceac9b347d0e3467e86d72d587b749b9521
|
|
* Use NEConvolutionLayer
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ieb81fafaf34a63be8daf297ebe1bb0e4079daf4e
|
|
* Add frontend and reference implementation for logical
ops NOT, AND, OR.
* Unary NOT uses existing ElementwiseUnary layer and
ElementwiseUnary descriptor.
* Binary AND/OR uses new layer LogicalBinary and new
LogicalBinary descriptor.
* Add serialization/deserializion support and add missing
ElementwiseUnary deserializer code.
* Add additional Boolean decoder in BaseIterator.hpp.
Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: Id343b01174053a166de1b98b6175e04a5065f720
|
|
Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: I0a07dea96a86849701ba387dbea148909a6d729b
|
|
* Fallback to memory copy if memory import is not supported
* Remove direct compatibility between Neon and Cl Tensors
* Unit tests fallback from Neon to Cl and Cl to Neon
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Iec00a77423fb23b37a6b1aefee1b2ec4d649efca
|
|
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I28aace7092cff5743353df1b1de8e7a4691554d3
|
|
If visited flag set true, it should not be visited any more.
For example, if we put 10 boxes (ordered by score) into NonMaxSuppression:
* Step1: Suppose Box 2/3/6/8 are suppressed by Box 1. Box 4/5/7/9/10 survived.
* Step2: Correct way: We use Box 4 to suppress the survive boxes.
Prior to this commit: Box 4 may be suppressed by Box 2,
even Box 2 is already suppressed by Box 1...
Signed-off-by: Antkillerfarm <antkillerfarm@gmail.com>
Change-Id: I38d7a84287649827a16565748592fb562b4df5d5
|
|
* Added Documentation for fast_math to CLBackendModelContext
* Added Documentation for fast_math to NeonBackendModelContext
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I43a0568ae6914e074a80130a051e5d9bb849f2ba
|
|
* Replaced with fmt::format in Descriptors.cpp.
* Removed remaining boost/format headers in ArmNN codebase.
* Removed additional boost header in Network.cpp
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ib98b83bf4ec99ef98ce7a3635ec0dd478c3e43e1
|
|
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
Change-Id: Ie5ecfa67e4763d0c058905592fe2e2fd7315f85c
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I054f0b71d4e9581c637fa09e40f6b661e58e39f3
|
|
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Id2227c58809b84c7a7af61f7c0d88ad7d45ce558
|
|
* Set use_padding to false in neon workload
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ia5367de331efe1d28dea4dfbefc0713720da81f9
|
|
* Floating point casts now use armnn::numeric_cast.
* Also removed remaining header imports.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I2d37847d67f164fc0a0ae17f34d49ff3d2210c30
|
|
* replaced with fmt::format
* one case required std:stringstream instead
Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: Ife7c4cf5f143e43373f42edf6124158af132abc5
|
|
Change-Id: I1a6df60683cc51fcd9739b6dc98f1e722becf045
Signed-off-by: Robert Hughes <robert.hughes@arm.com>
|
|
* LargeGraph_TENSOR_INT32_Rank4/26
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I9d07444db56e26c13a77bf022938644ed7953d6b
|
|
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: Ie0cf69b2cd76d6ecedab43d3d9ae267d23bbc052
|
|
* INetwork::Optimize(...) states that the function should throw an exception
if it fails but the implementation in Network.cpp returned null in some
scenarios instead. This has led to some confusion amongst users.
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I358d1293232c9464772aa0e39ab3355e3570c823
|
|
Change-Id: Id8bd0a0418be31d975b944b54bbacb25051ffb2e
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
|
|
*Cosmetic changes on ClQLstmWorkload
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I61f55343263e623aaae042d8dfe8c294540e98f1
|
|
* Change tensor to non const to fix build error caused by ACL fix for QLSTM
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I7ab0f644dfb3cb3cf21bda73028e9368f3354f4a
|
|
This patch adds int32 and int64 ArgMax op support.
Current ARMNN already has ArgMax op but not used, and
it doesn't support int64 output type.
So this patch adds a new type, Signed64, and also adds
ArgMinMax computation function for int64 type support.
In default, output tensor type of ArgMax op is int64 in case of
tensorflow lite model so this patch makes a proper function - ArgMax op
for int64 or int32 - to be called according to parsed output_type value.
With this patch, ARMNN supports both types - int64 and int32 - for
ArgMinMax op.
Changelog v1:
- Check if output data type of ArgMinMax op is valid or not.
- Use template function to support int32 and int64 types of ArgMinMax function.
- Keep using Signed32 as default data type of m_Output_Type.
Change-Id: I7a8e7e38dd9e5acc81464571d8b4d51378fc7f14
Signed-off-by: Inki Dae <inki.dae@samsung.com>
|
|
* Unit test implemented to make sure it returns WINOGRAD
* Updated the enable-fast-math option in ExecuteNetwork to be consistent
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Id64f114ae47966def69a9eef0770a4251ee56a41
|
|
* Replaced with armnn/utility/NumericCast.hpp
* Some exclusions in reference backend
* Excluded as requires float implementation in NumericCast.hpp
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I9e4e9cd502c865452128fa04415fd6f250baa855
|
|
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I4eb3e27837aea926593d49f9ccea07bab8388d5b
|
|
factory'
* Pass ModelOptions to WorkloadFactory
* Updated signature of CL and NEON Convolution2d workloads added FastMathEnabled param.
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I536178be8e4dd4083489e69febadaf0feeba46d2
|
|
* Introduced ModelOptions to IBackendInternal
* Introduced ModelOptions to Network
* Added FastMathEnabled parameter to Conv2d Validate function in CL and NEON
* Added Optimizer tests
Signed-off-by: Ryan OShea <Ryan.OShea2@arm.com>
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Ib54c1e82cb3d89a52756ed499cf91b6a7fdb2063
|
|
* Axis for LogSoftMax and SoftMax can be either positive or negative
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I36b0507ad7600c0a98c3b8be3c0350045ee05b84
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
|
|
* Moved DynamicBackend tests to only build when ArmnnRef is enabled
due to a dependency on them dynamically loading the ArmnnRef backend
object
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: Iee0480e7d0cf505bbb5c26629829d3d20fb60051
|