Merge remote-tracking branch 'upstream/master' into arm64

.github/workflows/linux.yml

@@ -171,6 +171,8 @@ jobs:
       - name: Build Debian packages
         run: |
           cmake -DENABLE_PYTHON_PACKAGING=ON \
+                -UPython3* \
+                -DPython3_EXECUTABLE=/usr/bin/python3 \
                 -DENABLE_TESTS=OFF \
                 -DCPACK_GENERATOR=DEB \
                 ${BUILD_DIR}
@@ -243,25 +245,20 @@ jobs:
           echo "deb [trusted=yes] file:${DEBIAN_PACKAGES_DIR} ./" | tee /etc/apt/sources.list.d/openvino-local.list
           apt-get update -y
           apt-get install openvino -y
-          # install python3.9
-          apt install --assume-yes libpython3.9 python3.9 python3.9-distutils wget
-          wget https://bootstrap.pypa.io/get-pip.py
-          python3.9 get-pip.py
-          python3.9 -m pip install numpy
         working-directory: ${{ env.DEBIAN_PACKAGES_DIR }}
 
       - name: Test debian packages
         run: |
           /usr/share/openvino/samples/cpp/build_samples.sh
           /usr/share/openvino/samples/c/build_samples.sh
           ~/openvino_cpp_samples_build/aarch64/Release/hello_query_device
-          python3.9 /usr/share/openvino/samples/python/hello_query_device/hello_query_device.py
-          python3.9 -c 'from openvino import Core; Core().get_property("CPU", "AVAILABLE_DEVICES")'
-          python3.9 -c 'from openvino import Core; Core().get_property("AUTO", "SUPPORTED_METRICS")'
-          python3.9 -c 'from openvino import Core; Core().get_property("MULTI", "SUPPORTED_METRICS")'
-          python3.9 -c 'from openvino import Core; Core().get_property("HETERO", "SUPPORTED_METRICS")'
-          python3.9 -c 'from openvino import Core; Core().get_property("BATCH", "SUPPORTED_METRICS")'
-          python3.9 -c 'from openvino.frontend import FrontEndManager; assert len(FrontEndManager().get_available_front_ends()) == 6'
+          python3 /usr/share/openvino/samples/python/hello_query_device/hello_query_device.py
+          python3 -c 'from openvino import Core; Core().get_property("CPU", "AVAILABLE_DEVICES")'
+          python3 -c 'from openvino import Core; Core().get_property("AUTO", "SUPPORTED_METRICS")'
+          python3 -c 'from openvino import Core; Core().get_property("MULTI", "SUPPORTED_METRICS")'
+          python3 -c 'from openvino import Core; Core().get_property("HETERO", "SUPPORTED_METRICS")'
+          python3 -c 'from openvino import Core; Core().get_property("BATCH", "SUPPORTED_METRICS")'
+          python3 -c 'from openvino.frontend import FrontEndManager; assert len(FrontEndManager().get_available_front_ends()) == 6'
           benchmark_app --help
           ovc --help
 
@@ -885,7 +882,7 @@ jobs:
           # Find and install OV dev wheel
           pushd ${INSTALL_DIR}/tools
             ov_dev_wheel_name=$(find . -name 'openvino_dev*.whl')
-            python3.9 -m pip install $ov_dev_wheel_name[mxnet,caffe,kaldi,onnx,tensorflow2]
+            python3.9 -m pip install $ov_dev_wheel_name[mxnet,caffe,kaldi,onnx,tensorflow2,pytorch]
           popd
 
       #
@@ -923,13 +920,15 @@ jobs:
       # E   OSError: libarmpl_lp64_mp.so: cannot open shared object file: No such file or directory
       - name: Model Optimizer unit tests
         if: ${{ always() }}
-        run: python3.9 -m pytest -s ${INSTALL_TEST_DIR}/mo/unit_tests \
-          --junitxml=${INSTALL_TEST_DIR}/TEST-ModelOptimizer.xml
+        run: |
+          python3.9 -m pytest -s ${INSTALL_TEST_DIR}/mo/unit_tests \
+            --ignore=${INSTALL_TEST_DIR}/mo/unit_tests/mo/front/mxnet \
+            --junitxml=${INSTALL_TEST_DIR}/TEST-ModelOptimizer.xml
 
       - name: PyTorch Layer Tests
         timeout-minutes: 40
         run: |
-          export LD_PRELOAD=/usr/lib/aarch64-linux-gnu/libgomp.so.1
+          export LD_PRELOAD=/usr/local/lib/python3.9/dist-packages/torch/lib/../../torch.libs/libgomp-d22c30c5.so.1.0.0
           python3.9 -m pip install -r ${LAYER_TESTS_INSTALL_DIR}/requirements.txt
           python3.9 -m pytest ${LAYER_TESTS_INSTALL_DIR}/pytorch_tests -m precommit --junitxml=${INSTALL_TEST_DIR}/TEST-pytorch.xml
         env:
@@ -1010,7 +1009,7 @@ jobs:
         if: ${{ always() }}
         timeout-minutes: 40
         run: |
-          export LD_PRELOAD=/usr/lib/aarch64-linux-gnu/libgomp.so.1
+          export LD_PRELOAD=/usr/local/lib/python3.9/dist-packages/torch/lib/../../torch.libs/libgomp-d22c30c5.so.1.0.0
           python3.9 -m pip install -r ${LAYER_TESTS_INSTALL_DIR}/requirements.txt
           # TODO: remove setupvars.sh from here; currently, it's used for 'test_utils' installed in '<package>/python/openvino'
           source ${INSTALL_DIR}/setupvars.sh
@@ -1025,7 +1024,7 @@ jobs:
         if: ${{ always() }}
         timeout-minutes: 40
         run: |
-          export LD_PRELOAD=/usr/lib/aarch64-linux-gnu/libgomp.so.1
+          export LD_PRELOAD=/usr/local/lib/python3.9/dist-packages/torch/lib/../../torch.libs/libgomp-d22c30c5.so.1.0.0
           python3.9 -m pip install -r ${LAYER_TESTS_INSTALL_DIR}/requirements.txt
           # TODO: remove setupvars.sh from here; currently, it's used for 'test_utils' installed in '<package>/python/openvino'
           source ${INSTALL_DIR}/setupvars.sh
@@ -1232,8 +1231,8 @@ jobs:
     defaults:
       run:
         shell: bash
-    runs-on: ubuntu-20.04-16-cores-arm
     # TODO: Switch back to self-hosted runners
+    runs-on: aks-linux-16-cores-arm
     container:
       image: openvinogithubactions.azurecr.io/dockerhub/ubuntu:20.04
       volumes:

.github/workflows/mac.yml

@@ -482,15 +482,15 @@ jobs:
           # Find and install OV dev wheel
           pushd ${{ env.INSTALL_DIR }}/tools
             ov_dev_wheel_name=$(find . -name 'openvino_dev*.whl')
-            python3 -m pip install $ov_dev_wheel_name[mxnet,caffe,kaldi,onnx,tensorflow2]
+            python3 -m pip install $ov_dev_wheel_name[mxnet,caffe,kaldi,onnx,tensorflow2,pytorch]
           popd
 
       - name: nGraph and IE Python Bindings Tests
         run: |
           python3 -m pytest -s ${{ env.INSTALL_TEST_DIR }}/pyngraph \
             --junitxml=${{ env.INSTALL_TEST_DIR }}/TEST-Pyngraph.xml \
-            --ignore=${{ env.INSTALL_TEST_DIR }}/pyngraph/tests/test_onnx/test_zoo_models.py \
-            --ignore=${{ env.INSTALL_TEST_DIR }}/pyngraph/tests/test_onnx/test_backend.py
+            --ignore=${{ env.INSTALL_TEST_DIR }}/pyngraph/tests_compatibility/test_onnx/test_zoo_models.py \
+            --ignore=${{ env.INSTALL_TEST_DIR }}/pyngraph/tests_compatibility/test_onnx/test_backend.py
 
       - name: Python API 2.0 Tests
         run: |

.github/workflows/windows.yml

@@ -303,7 +303,7 @@ jobs:
           
           # Find and install the dev OV wheel
           $ovDevWheelPath=Get-ChildItem -Path "${{ env.INSTALL_DIR }}\tools" -Filter openvino_dev*.whl | % { $_.FullName }
-          python3 -m pip install "$ovDevWheelPath[mxnet,caffe,kaldi,onnx,tensorflow2]"
+          python3 -m pip install "$ovDevWheelPath[mxnet,caffe,kaldi,onnx,tensorflow2,pytorch]"
 
       - name: Python API 1.0 Tests
         shell: cmd

cmake/features.cmake

@@ -80,14 +80,16 @@ else()
     set(THREADING_DEFAULT "TBB")
 endif()
 
+set(THREADING_OPTIONS "TBB" "TBB_AUTO" "SEQ")
+if(NOT APPLE)
+    list(APPEND THREADING_OPTIONS "OMP")
+endif()
+
 set(THREADING "${THREADING_DEFAULT}" CACHE STRING "Threading")
-set_property(CACHE THREADING PROPERTY STRINGS "TBB" "TBB_AUTO" "OMP" "SEQ")
+set_property(CACHE THREADING PROPERTY STRINGS ${THREADING_OPTIONS})
 list (APPEND OV_OPTIONS THREADING)
-if (NOT THREADING STREQUAL "TBB" AND
-    NOT THREADING STREQUAL "TBB_AUTO" AND
-    NOT THREADING STREQUAL "OMP" AND
-    NOT THREADING STREQUAL "SEQ")
-    message(FATAL_ERROR "THREADING should be set to TBB (default), TBB_AUTO, OMP or SEQ")
+if(NOT THREADING IN_LIST THREADING_OPTIONS)
+    message(FATAL_ERROR "THREADING should be set to either ${THREADING_OPTIONS}")
 endif()
 
 if((THREADING STREQUAL "TBB" OR THREADING STREQUAL "TBB_AUTO") AND

docs/articles_en/get started/installing-openvino-overview.md

@@ -30,7 +30,7 @@
    all necessary components.
 
    The OpenVINO Development Tools is still available for older versions of OpenVINO,
-   as well as the current one, from the GitHub repository.
+   as well as the current one, from the GitHub repository and PyPI. :doc:`Learn more <openvino_docs_install_guides_install_dev_tools>`.
 
 
 .. tip::

docs/articles_en/learn_openvino.md

@@ -14,6 +14,7 @@
 
    Interactive Tutorials (Python) <tutorials>
    Sample Applications (Python & C++) <openvino_docs_OV_UG_Samples_Overview>
+   Generative AI Optimization and Deployment <gen_ai_guide>
 
 
 This section will help you get a hands-on experience with OpenVINO even if you are just starting