Age | Commit message (Collapse) | Author |
|
In the bug there were 4 outputs from the final layer. The de-serialized
layer bindings were incorrectly assigning the tensor info of one output
to all 4 outputs. The solution is to use outputSlotIndex.
One other minor fix: The debug text referred to an Input when dealing
with an output.
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: I6c68b781a450ae4a5cf1d0e8671bb96ff01862b2
|
|
* Change layer storage of ConstTensors to std::shared_ptr<ConstCpuTensorHandle>
* Change clone to share ConstTensor rather than copy
* Remove uses of non-const GetTensor() call
* Reduce scope of non-optimized network in ExeNet, so memory can be released after use
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: Ibb2c7309d12411d21405bd6024c76bcdf5404545
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ic3efd8106be43ff2c1d250238d162e51e3a86a57
|
|
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: Ie51ce20540e5e7a704ce2b4be4e8cf64f91ea990
|
|
* Added AsyncExecution to the BackendCapability enum class.
* Logged a warning if backends do not support AsyncExecution capability if AsyncNetwork is created.
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I49f8467297f4b6b8e414cb6a3638a7d3f1bb886a
|
|
* Enable GPU profiling on ExecuteNetwork
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I57bb4eeb45674e5218fce7e67b9bddf16ba0894d
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: If0921a9f4d5c9a61dac30a43650fbb5a875b8ce7
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I11c1a61a6ff7a7252f086b61a9fc14ec48dd9efc
|
|
* Implemented Pimpl Idiom for IAsyncNetwork
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Ic7311880563568b014a27f6347f8d41f2ad96df6
|
|
Fix break for DockerFile
Signed-off-by: Keith Mok <ek9852@gmail.com>
Change-Id: I0c0b10fa212583b8861301fb9e1e926237b7867d
|
|
* Adding ref backend "ifdef" around ref test cases in
NeonLayerTests_NDK_Bug.cpp
* Removing unnecessary includes from NeonLayerTests_NDK_Bug.cpp.
* Removing unnecessary include from NeonLayerTests.cpp
* Breaking up Backends_Capability_Test into one per backend to allow
for conditional compilation.
* Remove unnecessary printout in src/backends/neon/test/CMakeLists.txt
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: I9a36cd197e684ed55af244e5c998ee67bb8da88c
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I096f8f6f3c7352fdfa11eca5423c461082a1f163
|
|
* Added thread safe execution mechanism for armnn
* Removed duplicate function bool Compare(T a, T b, float tolerance)
* Added StridedSliceAsyncEndToEndTest
* Fixed memory leak
Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I2d367fc77ee7c01b8953138543e76af5e691211f
|
|
According to the header description - include/armnn/IRuntime.hpp, Each level should have below value.
-----------------------------------------------------------------------------------------------------
"TuningLevel" : int [0..3] (0=UseOnly(default) | 1=RapidTuning | 2=NormalTuning | 3=ExhaustiveTuning)
-----------------------------------------------------------------------------------------------------
And also src/backends/cl/ClBackendContext.cpp uses below enumeration values internally
for the tuning level,
----------------------
enum class TuningLevel
{
None,
Rapid,
Normal,
Exhaustive
};
----------------------
So this patch corrects TuningLevel enumeration values - which is exposed to user - to be consistent
with ones internally used.
Change-Id: I9a8aeea0115579bfd16cbe01f39183b96329cdfd
Signed-off-by: Inki Dae <inki.dae@samsung.com>
|
|
!armnn:5180
* Remove unnecessary memcopy for non permeuted const tensors
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: Idc3ce2ac001e7d6be61819279de486f093730383
|
|
* Creating a new NeonLayerTests_NDK_Bug test class.
* Moving a subset of tests effected by an NDK bug from NeonLayerTests
into NeonLayerTests_NDK_Bug.
* Excluding NeonLayerTests_NDK_Bug from the build if it's an Android
debug build and NDK is less than r21.
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: Icb16d8bbb784fb0357deef281d7aa713e11af8ac
|
|
TfLiteDelegate support for FullyConnected Operator'
* Added front-end support for non-const weights for FULLY_CONNECTED operator
* Added FULLY_CONNECTED end-to-end test
* Updated FULLY_CONNECTED operator support in TfLite Arm NN Delegate for non-const weights
* Updated the version numbers
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Iffa5b9aa9297aca4c02d923cce4636c88ac21faa
|
|
abs is int type, use fabsf instead for float type
Signed-off-by: Keith Mok <ek9852@gmail.com>
Change-Id: I157d471da25d66cbe39fa9809ff9ee536d4e95a1
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I482d3d6aebc409478c524811ebd41fc1a6c37e5a
|
|
This reverts commit 51ce7d487c761358de105f82ff90553570aedac0.
Reason for revert: https://jira.arm.com/browse/IVGCVSW-5798 LargeGraph_TENSOR_FLOAT32 CTS tests failures
Change-Id: Ib031a47f605340b2202ecf074ce96a8b54c51075
|
|
* Add new ClImportTensorHandlefactory for tensor import.
* Add unit tests.
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: I61884fed65e764ebd6985fe0833e43a7296d0641
|
|
This reverts commit 4c10dfc213a59ec7bbf56b645e313083638b81a4.
Fix for IVGCVSW-5768
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I7197067a7bb407ff9d35059289a31912592d3af4
|
|
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ia6442ead871c829ce17b27fa225e3739c675d91f
|
|
Some models would add a PAD layer before a pooling when they can't
express their padding configuration as SAME or VALID.
Arm NN can merge the two because pooling layers are described with
explicit padding.
Signed-off-by: Diego Lopez Recas <diego.lopez.recas@gmail.com>
Change-Id: Id048186db6a005e0257bfbc1406c3b0dab2cdd58
|
|
* Doxygen can't handle code blocks right after a headline.
It needs text or an empty line first
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: Ida56737b08cdd24046f17a1845ea31a12a60cf96
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I18493c36ad87e87f836c42a0e72946332b8bcfa2
|
|
from the 21.08 release
Change-Id: I89ef90fe696706323715caa5f1a86b6dde978181
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
|
|
Optimisation when the input is a const tensor
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I8b1357bdefc45880d064d7e448af364ac8644c0d
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I815f8d596caab053dc16d3c655bb1b92fbfbd531
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Id0c1ce0ae8946bbc061f5c1f7a7798d3033e6ce4
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I59bf2eade949043d1f498772fad66eda7d64a87f
|
|
* Update ReduceLayer to use descriptor parameters
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I2ff240f63534de0fd647428cf411763829412443
|
|
* Contains a dummy import function as that will be implemented separately
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: If551b69e832c045c76775a7e5fa25647c2313908
|
|
Updated many minor points in the document to correct errors or out of date
references.
Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: Ie42c186916d9dbff3114a810a97ed25da5a06ca5
|
|
Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I127e6923f257bb4094485aa1716ead1067cf8d34
|
|
Instead of using the predefined backend ids,
we should just pass the backend string directly to the optimator,
otherwise we cannot use vendor specific backends or GpuAcc
Signed-off-by: Keith Mok <ek9852@gmail.com>
Change-Id: Ic52c81d48364a19f0fcc59bc3c975cf3bd740e21
|
|
* fix delegate perchannel quantization
* change delegate to check reshape options before inputs
* Add int8 "qsymms8" option to ExecuteNetwork
* Add option to run ExecuteNetwork on tflite w/o delegate
!referencetests:301301
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: If3e12599b17aff1199d7ab0a55e1c901e480083d
|
|
* excluding our top level readme
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: I45b156c248178a141b8497dcc4d09148b67bf406
|
|
The input tensor data pointer was passed incorrectly
Signed-off-by: Keith Mok <ek9852@gmail.com>
Change-Id: I03f872c57ec588fde0f7d444c80b38823ea4f9b4
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I8e617a6b67cb7188e389146875e41fe15542143b
|
|
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I79694713bf67bccc61c376d439333c096448e258
|
|
The subgraphs produced by SubgraphViewSelector were not deterministic
as the order of the input slots, outputs slots and layers within each
subgraph were determined by the pointer values of those objects, which
are not guaranteed to be the same for each execution.
This patch adds a post-processing sorting step based on the GUIDs of the
layers and the slot indices so that the results will be the same for
each execution.
This makes debugging the optimised graph much easier as subsequent
stages can also be deterministic.
Change-Id: Ifbcb199733066f99e9f95808a8da22023e5944f1
Signed-off-by: Rob Hughes <robert.hughes@arm.com>
|
|
inference time.
Usage: armnn_tfl_benchmark -m <model .tflite>
-m --model_file <.tflite file path>: .tflite model to be executed
-b --backend <device>: preferred backend device to run
layers on by default. Possible
choices: CpuAcc, CpuRef
-l --loops <int>: provide the number of time the
inference will be executed
(by default nb_loops=1)
Signed-off-by: Vincent ABRIOU <vincent.abriou@st.com>
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
Change-Id: Ia26fafd4f382f0ad03856436dcae6e71b5abbd26
|
|
* Added FILL operator to TfLite ArmNN Delegate
* Added unit tests
Signed-off-by: David Monahan <david.monahan@arm.com>
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I335ef469ff773fa4305eb87f6e93ae9c03fc6997
|
|
* add link to new wav2letter model from ModelZoo in Readme
* update model input size
* update to match new model's output
Signed-off-by: Nina Drozd <nina.drozd@arm.com>
Change-Id: I8e85d025610a458e9ae7be93fd7179c71bac5b18
|
|
* Add updates for Boost, Caffe and TF parsers
* Update path
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Ice9e5393ab47a678b5fe7c1ba4070460ef5b473c
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
|
|
* Updating README
* Updating android ndk build guide
Change-Id: I09f5e4fe09d590b6b482800730b78cbc84cee9e9
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
|
|
* Update supported operators for the delegate, parsers,
serializer and deserializer
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: I33ac99a29d894eec055cd05411014075d78b3168
|
|
* For Quantizer, Tensorflow Parser and Caffe Parser
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: If30daa54947c417c0d700f11e630546b1745b72e
|
|
Debug layers are given names based on the layer they take their input from.
If a layer has multiple outputs then one debug layer will be attached to
each output. Therefore all these debug layers would have identical names
making them hard to distinguish when debugging. This patch includes the
output slot index which the debug layer takes its input from when creating
the name.
Change-Id: I09eaa8a7edad9bfdf678b4778cf740340013126c
Signed-off-by: Rob Hughes <robert.hughes@arm.com>
|