From 3a8444185486fe747853f243e4f0e9d9e4a3bebd Mon Sep 17 00:00:00 2001 From: michael vincerra <37549381+mvincerx@users.noreply.github.com> Date: Tue, 19 May 2020 11:54:20 -0700 Subject: [PATCH] Deprecate OpenVINO tutorial; CL no longer supports computer-vision-openvino (#1177) * Deprecate OpenVINO tutorial; CL no longer supports computer-vision-openvino. - Closes #1166 - openvino was also removed from containers-basic bundle - See also: https://community.clearlinux.org/t/openvino-in-clear-linux-os-moving-to-docker/4566 Signed-off-by: Michael Vincerra * Remove ref to computer-vision-openvino in developer-workstation. - Remove ref to openVINO in tutorials index. Signed-off-by: Michael Vincerra --- .../maintenance/developer-workstation.rst | 3 - source/tutorials/index.rst | 1 - source/tutorials/openvino.rst | 290 ------------------ 3 files changed, 294 deletions(-) delete mode 100644 source/tutorials/openvino.rst diff --git a/source/guides/maintenance/developer-workstation.rst b/source/guides/maintenance/developer-workstation.rst index c3e0bb97..5333fedc 100644 --- a/source/guides/maintenance/developer-workstation.rst +++ b/source/guides/maintenance/developer-workstation.rst @@ -78,9 +78,6 @@ tools you need to start. Consider these profiles as a starting point. * - Work with deep learning and edge-optimized models. - `computer-vision-models `_ - * - Basic OpenVINO™ toolkit. - - `computer-vision-openvino `_ - * - API helper for cloud access. - `cloud-api `_ diff --git a/source/tutorials/index.rst b/source/tutorials/index.rst index ec35151c..6773db3d 100644 --- a/source/tutorials/index.rst +++ b/source/tutorials/index.rst @@ -37,7 +37,6 @@ sorted by difficulty level! Learn about :ref:`how we evaluate tutorials -o - - .. note:: - - * Where :file:`` is the one you chose from previous step - - * Where :file:`` is your project directory - -#. For this example, enter: - - .. code-block:: bash - - python3 downloader.py --name squeezenet1.1 -o $HOME/. - -#. After running this command, the model appears as downloading at your - :file:`$HOME/classification/squeezenet/1.1/caffe` as follows: - - .. code-block:: console - - ###############|| Downloading topologies ||############### - - ========= Downloading /$HOME/classification/squeezenet/1.1/caffe/squeezenet1.1.caffemodel - ... 100%, 4834 KB, 2839 KB/s, 1 seconds passed - - ... - -Convert model to IR format -========================== - -#. As necessary, follow the instruction on :ref:`convert-dl-models` - to convert deep learning models. - -#. Navigate to the model: - - .. code-block:: bash - - cd $HOME/classification/squeezenet/1.1/caffe - -#. Enter the command: - - .. code-block:: bash - - python3 /usr/share/openvino/model-optimizer/mo.py --input_model squeezenet1.1.caffemodel - - - The output will show these files being generated: - - .. code-block:: console - - squeezenet1.1.xml - - squeezenet1.1.bin - -#. Finally, enter :command:`ls` to view the newly added model and files. - -Run image classification -======================== - -This sample application demonstrates how to run the Image Classification in asynchronous mode on supported devices. In this example, we use the image of a specific type of automobile to test the inference engine. Squeezenet 1.1 is designed to perform image classification and has been trained on the `ImageNet`_ database. - -#. We provide an image of an automobile, shown in Figure 1. For ease of use, - save this image into the :file:`classification` model directory. - - .. figure:: ../_figures/openvino/automobile.png - :height: 375 px - :width: 500 px - :scale: 100 % - :alt: Photo by Goh Rhy Yan on Unsplash - - Figure 1: Photo by Goh Rhy Yan on Unsplash - -#. To execute the sample application enter the command: - - .. code-block:: bash - - classification_sample_async -i -m -d - - .. note:: - - * Where :file:`` is the image that you selected - - * Where :file:`` is the path to the IR model file - - * Where :file:`` is your choice of CPU, GPU, etc. - -#. In this case, we replace the :file:`` with the previously - saved image for CPU inferencing. - - .. code-block:: bash - - classification_sample_async -i ./automobile.png -m squeezenet1.1.xml - - .. note:: - - If you do not specify the :file:`device`, the CPU is used by default. - -#. The results show the highest probability is 67% for a sports car. - - .. code-block:: bash - - classid probability - ------- ----------- - 817 0.6717085 - 511 0.1611409 - - +-----------------------+-----------------------------------+ - |:command:`classid` 817 | :command:`sports car, sport car` | - +-----------------------+-----------------------------------+ - |:command:`classid` 511 |:command:`convertible` | - +-----------------------+-----------------------------------+ - - .. note: - - Label definitions are provided by `ImageNet`_. - -#. Next, add :command:`-d GPU` to the end of the above command for GPU - inferencing. - - .. code-block:: bash - - classification_sample_async -i ./automobile.png -m squeezenet1.1.xml -d GPU - -Run benchmark_app -***************** - -This sample application demonstrates how to use benchmark application to -estimate deep learning inference **performance** on supported devices. -We use the same image of an automobile, Figure 1, from the previous section. - -#. To execute this sample application, enter: - - .. code-block:: bash - - benchmark_app -i -m -d - - .. note:: - - * Where :file:`` is the image that you selected - - * Where :file:`` is the path to the IR model file - - * Where :file:`` is local your choice of CPU, GPU, etc. - -#. Change directory: - - .. code-block:: bash - - cd $HOME/classification/squeezenet/1.1/caffe - -#. Enter the following command for CPU inferencing. - - .. code-block:: bash - - benchmark_app -i ./automobile.png -m squeezenet1.1.xml - -#. For the CPU, the results show a :guilabel:`Throughput` of 243.202 FPS. - - .. code-block:: console - :linenos: - :emphasize-lines: 4 - - Count: 1464 iterations - Duration: 60196.8 ms - Latency: 164.104 ms - Throughput: 243.202 FPS - -#. Next, add :command:`-d GPU` to the end of the same command for GPU - inferencing. - - .. code-block:: bash - - benchmark_app -i ./automobile.png -m squeezenet1.1.xml -d GPU - -#. For the GPU, the results show a :guilabel:`Throughput` of 372.677 FPS. - - .. code-block:: console - :linenos: - :emphasize-lines: 4 - - Count: 2240 iterations - Duration: 60105.7 ms - Latency: 107.554 ms - Throughput: 372.677 FPS - -.. _ImageNet: http://image-net.org/