MLECO-4072: update MLEK dependencies to 23.08
* version updates in cmakelists file
* update submodules to 23.08 versions
* updates to documentation - vela 3.9.0
Change-Id: I5e24d5729ecbd2289b04fcc845fb44f455ab38d5
Signed-off-by: Nina Drozd <nina.drozd@arm.com>
diff --git a/CMakeLists.txt b/CMakeLists.txt
index a6e41f9..91f31f3 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -84,7 +84,7 @@
endif()
project(arm_ml_embedded_evaluation_kit
- VERSION 23.05.0
+ VERSION 23.08.0
DESCRIPTION "ARM ML Embedded Evaluation Kit"
LANGUAGES C CXX ASM)
diff --git a/dependencies/cmsis b/dependencies/cmsis
index e94a962..a75f017 160000
--- a/dependencies/cmsis
+++ b/dependencies/cmsis
@@ -1 +1 @@
-Subproject commit e94a96201a97be3e84d3d6ef081d2f0f7db9b5fd
+Subproject commit a75f01746df18bb5b929dfb8dc6c9407fac3a0f3
diff --git a/dependencies/cmsis-dsp b/dependencies/cmsis-dsp
index 3a04f81..8f8bb8d 160000
--- a/dependencies/cmsis-dsp
+++ b/dependencies/cmsis-dsp
@@ -1 +1 @@
-Subproject commit 3a04f817a4380c00ff51f9bcdc6c4e4bb7a18b80
+Subproject commit 8f8bb8d56b346274c9b2918928f2d5c3dae51758
diff --git a/dependencies/cmsis-nn b/dependencies/cmsis-nn
index 61d1bb6..ebd73de 160000
--- a/dependencies/cmsis-nn
+++ b/dependencies/cmsis-nn
@@ -1 +1 @@
-Subproject commit 61d1bb6c4428ad730e525d0164e12df12b789985
+Subproject commit ebd73deaa9a57a8838cfb2c9dde6cc682794d2ff
diff --git a/dependencies/core-platform b/dependencies/core-platform
index 4db21c9..8d3074b 160000
--- a/dependencies/core-platform
+++ b/dependencies/core-platform
@@ -1 +1 @@
-Subproject commit 4db21c986a8aec9a1ee31479fd3ca9806781865f
+Subproject commit 8d3074b0e82ed46e45d13c22a39e40388a9df8c7
diff --git a/dependencies/tensorflow b/dependencies/tensorflow
index 384dd27..80cb11b 160000
--- a/dependencies/tensorflow
+++ b/dependencies/tensorflow
@@ -1 +1 @@
-Subproject commit 384dd272f28ea7a99f5f7c211c9791f73cdb301b
+Subproject commit 80cb11b131e9738dc60b2db3e2f1f8e2425ded52
diff --git a/docs/quick_start.md b/docs/quick_start.md
index cad0e8c..d9c3712 100644
--- a/docs/quick_start.md
+++ b/docs/quick_start.md
@@ -37,14 +37,14 @@
[supported by Vela](https://review.mlplatform.org/plugins/gitiles/ml/ethos-u/ethos-u-vela/+/HEAD/SUPPORTED_OPS.md).
```commandline
- python3 ./build_default.py
+ python3.9 ./build_default.py
```
> **Note** The preceding command assumes you are using the GNU Arm Embedded toolchain. If you are using the Arm
> Compiler instead, you can override the default selection by executing:
```commandline
- python3 ./build_default.py --toolchain arm
+ python3.9 ./build_default.py --toolchain arm
```
5. Launch the project as explained in the following section: [Deployments](sections/deployment.md#deployment). In quick
@@ -71,7 +71,7 @@
python3 -m venv env
env/bin/python3 -m pip install --upgrade pip
env/bin/python3 -m pip install --upgrade setuptools
-env/bin/python3 -m pip install ethos-u-vela==3.8.0
+env/bin/python3 -m pip install ethos-u-vela==3.9.0
cd ..
curl -L https://github.com/ARM-software/ML-zoo/raw/7c32b097f7d94aae2cd0b98a8ed5a3ba81e66b18/models/anomaly_detection/micronet_medium/tflite_int8/ad_medium_int8.tflite \
diff --git a/docs/sections/troubleshooting.md b/docs/sections/troubleshooting.md
index e6b921b..6841e62 100644
--- a/docs/sections/troubleshooting.md
+++ b/docs/sections/troubleshooting.md
@@ -6,7 +6,7 @@
- [NPU configuration mismatch error when running inference](./troubleshooting.md#npu-configuration-mismatch-error-when-running-inference)
- [Errors when cloning the repository](./troubleshooting.md#errors-when-cloning-the-repository)
- [Problem installing Vela](./troubleshooting.md#problem-installing-vela)
- - [No matching distribution found for ethos-u-vela==3.8.0](./troubleshooting.md#no-matching-distribution-found-for-ethos_u_vela)
+ - [No matching distribution found for ethos-u-vela==3.9.0](./troubleshooting.md#no-matching-distribution-found-for-ethos_u_vela)
- [How to update Python3 package to 3.9 version](./troubleshooting.md#how-to-update-python3-package-to-newer-version)
- [Error trying to build on Arm Virtual Hardware](./troubleshooting.md#error-trying-to-build-on-arm-virtual-hardware)
- [Internal Compiler Error](./troubleshooting.md#internal-compiler-error)
@@ -129,15 +129,15 @@
## No matching distribution found for ethos-u-vela
-Vela 3.8.0 increases Python requirement to at least version 3.9, if not installed on your system the following error will occur:
+Vela 3.9.0 increases Python requirement to at least version 3.9, if not installed on your system the following error will occur:
```log
-python3 -m pip install ethos-u-vela==3.8.0
-ERROR: Could not find a version that satisfies the requirement ethos-u-vela==3.8.0 (from versions: 0.1.0, 1.0.0, 1.1.0, 1.2.0, 2.0.0, 2.0.1, 2.1.1, 3.0.0, 3.1.0, 3.2.0)
-ERROR: No matching distribution found for ethos-u-vela==3.8.0
+python3 -m pip install ethos-u-vela==3.9.0
+ERROR: Could not find a version that satisfies the requirement ethos-u-vela==3.9.0 (from versions: 0.1.0, 1.0.0, 1.1.0, 1.2.0, 2.0.0, 2.0.1, 2.1.1, 3.0.0, 3.1.0, 3.2.0)
+ERROR: No matching distribution found for ethos-u-vela==3.9.0
```
-Ensure that the minimum Python 3.9 requirement is installed and it's the default version.
+Ensure that the minimum Python 3.9 requirement is installed, and it's the default version.
Check your current installed version of Python by running:
```commandline
diff --git a/download_dependencies.py b/download_dependencies.py
index b07ef0d..33c7eec 100755
--- a/download_dependencies.py
+++ b/download_dependencies.py
@@ -24,12 +24,12 @@
from zipfile import ZipFile
from pathlib import Path
-TF = "https://github.com/tensorflow/tflite-micro/archive/384dd272f28ea7a99f5f7c211c9791f73cdb301b.zip"
-CMSIS = "https://github.com/ARM-software/CMSIS_5/archive/e94a96201a97be3e84d3d6ef081d2f0f7db9b5fd.zip"
-CMSIS_DSP = "https://github.com/ARM-software/CMSIS-DSP/archive/refs/tags/v1.14.4.zip"
-CMSIS_NN = "https://github.com/ARM-software/CMSIS-NN/archive/refs/tags/v4.1.0.zip"
-ETHOS_U_CORE_DRIVER = "https://git.mlplatform.org/ml/ethos-u/ethos-u-core-driver.git/snapshot/ethos-u-core-driver-23.05.tar.gz"
-ETHOS_U_CORE_PLATFORM = "https://git.mlplatform.org/ml/ethos-u/ethos-u-core-platform.git/snapshot/ethos-u-core-platform-23.05.tar.gz"
+TF = "https://github.com/tensorflow/tflite-micro/archive/80cb11b131e9738dc60b2db3e2f1f8e2425ded52.zip"
+CMSIS = "https://github.com/ARM-software/CMSIS_5/archive/a75f01746df18bb5b929dfb8dc6c9407fac3a0f3.zip"
+CMSIS_DSP = "https://github.com/ARM-software/CMSIS-DSP/archive/refs/tags/v1.15.0.zip"
+CMSIS_NN = "https://github.com/ARM-software/CMSIS-NN/archive/refs/tags/23.08.zip"
+ETHOS_U_CORE_DRIVER = "https://git.mlplatform.org/ml/ethos-u/ethos-u-core-driver.git/snapshot/ethos-u-core-driver-23.08.tar.gz"
+ETHOS_U_CORE_PLATFORM = "https://git.mlplatform.org/ml/ethos-u/ethos-u-core-platform.git/snapshot/ethos-u-core-platform-23.08.tar.gz"
def download(url_file: str, post_process=None):
diff --git a/release_notes.txt b/release_notes.txt
index cc34c0a..7296739 100644
--- a/release_notes.txt
+++ b/release_notes.txt
@@ -1,3 +1,7 @@
+Changes in 23.08
+ * Support for 23.08 NPU components and dependencies (core-driver, core-platform, Vela 3.9.0, CMSIS, CMSIS-NN and TensorFlow Lite Micro).
+ * Update Python package requirements for compatibility reasons.
+
Changes in 23.05
* Support for 23.05 NPU components and dependencies (core-driver, core-platform, Vela 3.8.0, CMSIS, CMSIS-NN and TensorFlow Lite Micro).
* Minor updates to CMSIS-pack dependencies.
diff --git a/set_up_default_resources.py b/set_up_default_resources.py
index b442432..f983508 100755
--- a/set_up_default_resources.py
+++ b/set_up_default_resources.py
@@ -382,7 +382,7 @@
metadata_file_path = download_dir / "resources_downloaded_metadata.json"
metadata_dict = dict()
- vela_version = "3.8.0"
+ vela_version = "3.9.0"
py3_version_minimum = (3, 9)
# Is Python minimum requirement matched?
diff --git a/source/application/api/common/include/TensorFlowLiteMicro.hpp b/source/application/api/common/include/TensorFlowLiteMicro.hpp
index 17dfd64..6957b1b 100644
--- a/source/application/api/common/include/TensorFlowLiteMicro.hpp
+++ b/source/application/api/common/include/TensorFlowLiteMicro.hpp
@@ -30,19 +30,16 @@
#pragma clang diagnostic ignored "-Wunused-parameter"
#include "tensorflow/lite/micro/micro_mutable_op_resolver.h"
#include "tensorflow/lite/micro/micro_interpreter.h"
- #include "tensorflow/lite/micro/all_ops_resolver.h"
#pragma clang diagnostic pop
#elif defined(__GNUC__)
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wunused-parameter"
#include "tensorflow/lite/micro/micro_mutable_op_resolver.h"
#include "tensorflow/lite/micro/micro_interpreter.h"
- #include "tensorflow/lite/micro/all_ops_resolver.h"
#pragma GCC diagnostic pop
#else
#include "tensorflow/lite/micro/micro_mutable_op_resolver.h"
#include "tensorflow/lite/micro/micro_interpreter.h"
- #include "tensorflow/lite/micro/all_ops_resolver.h"
#endif
#include "tensorflow/lite/c/common.h"
diff --git a/source/application/api/common/source/Model.cc b/source/application/api/common/source/Model.cc
index da8f46b..b344a53 100644
--- a/source/application/api/common/source/Model.cc
+++ b/source/application/api/common/source/Model.cc
@@ -193,7 +193,7 @@
for (size_t i = 0; i < nOperators; ++i) {
const tflite::Operator* op = subgraph->operators()->Get(i);
const tflite::OperatorCode* opcode = opcodes->Get(op->opcode_index());
- const TfLiteRegistration_V1* reg = nullptr;
+ const TFLMRegistration* reg = nullptr;
tflite::GetRegistrationFromOpCode(opcode, this->GetOpResolver(), ®);
std::string opName;
diff --git a/source/application/api/use_case/inference_runner/include/MicroMutableAllOpsResolver.hpp b/source/application/api/use_case/inference_runner/include/MicroMutableAllOpsResolver.hpp
new file mode 100644
index 0000000..df55a83
--- /dev/null
+++ b/source/application/api/use_case/inference_runner/include/MicroMutableAllOpsResolver.hpp
@@ -0,0 +1,134 @@
+/*
+ * SPDX-FileCopyrightText: Copyright 2023 Arm Limited and/or its affiliates <open-source-office@arm.com>
+ * SPDX-License-Identifier: Apache-2.0
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+#ifndef INF_RUNNER_MICRO_MUTABLE_ALLOPS_RESOLVER_HPP
+#define INF_RUNNER_MICRO_MUTABLE_ALLOPS_RESOLVER_HPP
+
+#include <tensorflow/lite/micro/micro_mutable_op_resolver.h>
+
+constexpr int kNumberOperators = 97;
+
+namespace arm {
+namespace app {
+
+ inline tflite::MicroMutableOpResolver<kNumberOperators> get_resolver() {
+ tflite::MicroMutableOpResolver<kNumberOperators> micro_op_resolver;
+
+ micro_op_resolver.AddAbs();
+ micro_op_resolver.AddAdd();
+ micro_op_resolver.AddAddN();
+ micro_op_resolver.AddArgMax();
+ micro_op_resolver.AddArgMin();
+ micro_op_resolver.AddAssignVariable();
+ micro_op_resolver.AddAveragePool2D();
+ micro_op_resolver.AddBatchToSpaceNd();
+ micro_op_resolver.AddBroadcastArgs();
+ micro_op_resolver.AddBroadcastTo();
+ micro_op_resolver.AddCallOnce();
+ micro_op_resolver.AddCast();
+ micro_op_resolver.AddCeil();
+ micro_op_resolver.AddCircularBuffer();
+ micro_op_resolver.AddConcatenation();
+ micro_op_resolver.AddConv2D();
+ micro_op_resolver.AddCos();
+ micro_op_resolver.AddCumSum();
+ micro_op_resolver.AddDepthToSpace();
+ micro_op_resolver.AddDepthwiseConv2D();
+ micro_op_resolver.AddDequantize();
+ micro_op_resolver.AddDetectionPostprocess();
+ micro_op_resolver.AddDiv();
+ micro_op_resolver.AddElu();
+ micro_op_resolver.AddEqual();
+ micro_op_resolver.AddEthosU();
+ micro_op_resolver.AddExp();
+ micro_op_resolver.AddExpandDims();
+ micro_op_resolver.AddFill();
+ micro_op_resolver.AddFloor();
+ micro_op_resolver.AddFloorDiv();
+ micro_op_resolver.AddFloorMod();
+ micro_op_resolver.AddFullyConnected();
+ micro_op_resolver.AddGather();
+ micro_op_resolver.AddGatherNd();
+ micro_op_resolver.AddGreater();
+ micro_op_resolver.AddGreaterEqual();
+ micro_op_resolver.AddHardSwish();
+ micro_op_resolver.AddIf();
+ micro_op_resolver.AddL2Normalization();
+ micro_op_resolver.AddL2Pool2D();
+ micro_op_resolver.AddLeakyRelu();
+ micro_op_resolver.AddLess();
+ micro_op_resolver.AddLessEqual();
+ micro_op_resolver.AddLog();
+ micro_op_resolver.AddLogicalAnd();
+ micro_op_resolver.AddLogicalNot();
+ micro_op_resolver.AddLogicalOr();
+ micro_op_resolver.AddLogistic();
+ micro_op_resolver.AddLogSoftmax();
+ micro_op_resolver.AddMaxPool2D();
+ micro_op_resolver.AddMaximum();
+ micro_op_resolver.AddMean();
+ micro_op_resolver.AddMinimum();
+ micro_op_resolver.AddMirrorPad();
+ micro_op_resolver.AddMul();
+ micro_op_resolver.AddNeg();
+ micro_op_resolver.AddNotEqual();
+ micro_op_resolver.AddPack();
+ micro_op_resolver.AddPad();
+ micro_op_resolver.AddPadV2();
+ micro_op_resolver.AddPrelu();
+ micro_op_resolver.AddQuantize();
+ micro_op_resolver.AddReadVariable();
+ micro_op_resolver.AddReduceMax();
+ micro_op_resolver.AddRelu();
+ micro_op_resolver.AddRelu6();
+ micro_op_resolver.AddReshape();
+ micro_op_resolver.AddResizeBilinear();
+ micro_op_resolver.AddResizeNearestNeighbor();
+ micro_op_resolver.AddRound();
+ micro_op_resolver.AddRsqrt();
+ micro_op_resolver.AddSelectV2();
+ micro_op_resolver.AddShape();
+ micro_op_resolver.AddSin();
+ micro_op_resolver.AddSlice();
+ micro_op_resolver.AddSoftmax();
+ micro_op_resolver.AddSpaceToBatchNd();
+ micro_op_resolver.AddSpaceToDepth();
+ micro_op_resolver.AddSplit();
+ micro_op_resolver.AddSplitV();
+ micro_op_resolver.AddSqrt();
+ micro_op_resolver.AddSquare();
+ micro_op_resolver.AddSquaredDifference();
+ micro_op_resolver.AddSqueeze();
+ micro_op_resolver.AddStridedSlice();
+ micro_op_resolver.AddSub();
+ micro_op_resolver.AddSum();
+ micro_op_resolver.AddSvdf();
+ micro_op_resolver.AddTanh();
+ micro_op_resolver.AddTranspose();
+ micro_op_resolver.AddTransposeConv();
+ micro_op_resolver.AddUnidirectionalSequenceLSTM();
+ micro_op_resolver.AddUnpack();
+ micro_op_resolver.AddVarHandle();
+ micro_op_resolver.AddWhile();
+ micro_op_resolver.AddZerosLike();
+
+ return micro_op_resolver;
+ }
+
+} /* namespace app */
+} /* namespace arm */
+
+#endif /* INF_RUNNER_MICRO_MUTABLE_ALLOPS_RESOLVER_HPP */
diff --git a/source/application/api/use_case/inference_runner/include/TestModel.hpp b/source/application/api/use_case/inference_runner/include/TestModel.hpp
index 80318e6..bc6a0fb 100644
--- a/source/application/api/use_case/inference_runner/include/TestModel.hpp
+++ b/source/application/api/use_case/inference_runner/include/TestModel.hpp
@@ -1,5 +1,5 @@
/*
- * SPDX-FileCopyrightText: Copyright 2021 Arm Limited and/or its affiliates <open-source-office@arm.com>
+ * SPDX-FileCopyrightText: Copyright 2021, 2023 Arm Limited and/or its affiliates <open-source-office@arm.com>
* SPDX-License-Identifier: Apache-2.0
*
* Licensed under the Apache License, Version 2.0 (the "License");
@@ -18,6 +18,7 @@
#define INF_RUNNER_TESTMODEL_HPP
#include "Model.hpp"
+#include "MicroMutableAllOpsResolver.hpp"
namespace arm {
namespace app {
@@ -26,7 +27,7 @@
protected:
/** @brief Gets the reference to op resolver interface class. */
- const tflite::AllOpsResolver& GetOpResolver() override;
+ const tflite::MicroMutableOpResolver<kNumberOperators>& GetOpResolver() override;
/** @brief Adds operations to the op resolver instance, not needed as using AllOpsResolver. */
bool EnlistOperations() override {return false;}
@@ -34,7 +35,7 @@
private:
/* No need to define individual ops at the cost of extra memory. */
- tflite::AllOpsResolver m_opResolver;
+ tflite::MicroMutableOpResolver<kNumberOperators> m_opResolver;
};
} /* namespace app */
diff --git a/source/application/api/use_case/inference_runner/src/TestModel.cc b/source/application/api/use_case/inference_runner/src/TestModel.cc
index 660c3e6..c69a98d 100644
--- a/source/application/api/use_case/inference_runner/src/TestModel.cc
+++ b/source/application/api/use_case/inference_runner/src/TestModel.cc
@@ -1,5 +1,5 @@
/*
- * SPDX-FileCopyrightText: Copyright 2021 Arm Limited and/or its affiliates <open-source-office@arm.com>
+ * SPDX-FileCopyrightText: Copyright 2021, 2023 Arm Limited and/or its affiliates <open-source-office@arm.com>
* SPDX-License-Identifier: Apache-2.0
*
* Licensed under the Apache License, Version 2.0 (the "License");
@@ -17,7 +17,7 @@
#include "TestModel.hpp"
#include "log_macros.h"
-const tflite::AllOpsResolver& arm::app::TestModel::GetOpResolver()
+const tflite::MicroMutableOpResolver<kNumberOperators>& arm::app::TestModel::GetOpResolver()
{
return this->m_opResolver;
}