<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: Two models in a neural compute stick2 in Intel® Distribution of OpenVINO™ Toolkit</title>
    <link>https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Two-models-in-a-neural-compute-stick2/m-p/645076#M1563</link>
    <description>&lt;P&gt;Hi @SamuelShin &lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Yes, OpenVINO has samples that run multiple models one 1 NCS sequentially. The &lt;A href="https://docs.openvinotoolkit.org/2018_R5/_samples_interactive_face_detection_demo_README.html" target="_self"&gt;interactive_face_detection_demo&lt;/A&gt; is one sample that can handle multiple models. We also have more sample programs that handle multiple models. Take a look at them &lt;A href="https://docs.openvinotoolkit.org/2018_R5/_samples_interactive_face_detection_demo_README.html" target="_self"&gt;here&lt;/A&gt;. &lt;A href="https://github.com/PINTO0309/OpenVINO-EmotionRecognition" target="_self"&gt;Here&lt;/A&gt; is a python example created by one of our NCS community members that loads multiple models one one NCS. &lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;Please let me know if you have any further questions!&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Sincerely,&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;Sahira</description>
    <pubDate>Wed, 29 May 2019 01:17:01 GMT</pubDate>
    <dc:creator>idata</dc:creator>
    <dc:date>2019-05-29T01:17:01Z</dc:date>
    <item>
      <title>Two models in a neural compute stick2</title>
      <link>https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Two-models-in-a-neural-compute-stick2/m-p/645075#M1562</link>
      <description>&lt;P&gt;I'm working on implementing inference codes of two models which are optimized by OpenVINO with one neural compute stick2.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;I wrote some codes referring python sample codes which are provided with OpenVINO. The directory is like this /intel/openvino/inference_engine/samples/python_samples&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;1        detector_bin = os.path.splitext(detector_path)[0] + ".bin"&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;2        plugin_detector = IEPlugin(device="MYRIAD")&lt;P&gt;&amp;nbsp;&lt;/P&gt;3        net_detector = IENetwork(model=detector_path, weights=detector_bin)&lt;P&gt;&amp;nbsp;&lt;/P&gt;4        self.detector = plugin_detector.load(network=net_detector)&lt;P&gt;&amp;nbsp;&lt;/P&gt;5&lt;P&gt;&amp;nbsp;&lt;/P&gt;6        skeleton_bin = os.path.splitext(skeleton_path)[0] + ".bin"&lt;P&gt;&amp;nbsp;&lt;/P&gt;7        plugin_skeleton = IEPlugin(device="MYRIAD")&lt;P&gt;&amp;nbsp;&lt;/P&gt;8        net_skeleton = IENetwork(model=skeleton_path, weights=skeleton_bin)&lt;P&gt;&amp;nbsp;&lt;/P&gt;9        self.skeleton = plugin_skeleton.load(network=net_skeleton)&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;[Error messages]&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;self.skeleton = plugin_skeleton.load(network=net_skeleton)&lt;P&gt;&amp;nbsp;&lt;/P&gt;  File "ie_api.pyx", line 395, in openvino.inference_engine.ie_api.IEPlugin.load&lt;P&gt;&amp;nbsp;&lt;/P&gt;  File "ie_api.pyx", line 406, in openvino.inference_engine.ie_api.IEPlugin.load&lt;P&gt;&amp;nbsp;&lt;/P&gt;RuntimeError: Can not init USB device: NC_ERROR&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;In the first part 1~4 where the first model is being mounted, there is no problem.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;But in the second part 6~9 where the second model is being mounted, it causes an error.&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Is there any way to mount two models at the same time in a neural compute stick2 ???&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;P.S&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;I googled it and I found this discussion. &lt;A href="https://ncsforum.movidius.com/discussion/795/can-i-run-more-than-one-models-with-one-stick"&gt;https://ncsforum.movidius.com/discussion/795/can-i-run-more-than-one-models-with-one-stick&lt;/A&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;In that discussion, Tome_at_Intel gave a link. &lt;A href="https://github.com/movidius/ncappzoo/blob/ncsdk2/apps/birds/birds.py"&gt;https://github.com/movidius/ncappzoo/blob/ncsdk2/apps/birds/birds.py&lt;/A&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;But I am familiar with OpenVINO. Isn't there a sample code for using two models in a neural compute stick2 with OpenVINO??&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Thank you for reading my questions!&lt;/P&gt;</description>
      <pubDate>Tue, 28 May 2019 13:56:46 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Two-models-in-a-neural-compute-stick2/m-p/645075#M1562</guid>
      <dc:creator>idata</dc:creator>
      <dc:date>2019-05-28T13:56:46Z</dc:date>
    </item>
    <item>
      <title>Re: Two models in a neural compute stick2</title>
      <link>https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Two-models-in-a-neural-compute-stick2/m-p/645076#M1563</link>
      <description>&lt;P&gt;Hi @SamuelShin &lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Yes, OpenVINO has samples that run multiple models one 1 NCS sequentially. The &lt;A href="https://docs.openvinotoolkit.org/2018_R5/_samples_interactive_face_detection_demo_README.html" target="_self"&gt;interactive_face_detection_demo&lt;/A&gt; is one sample that can handle multiple models. We also have more sample programs that handle multiple models. Take a look at them &lt;A href="https://docs.openvinotoolkit.org/2018_R5/_samples_interactive_face_detection_demo_README.html" target="_self"&gt;here&lt;/A&gt;. &lt;A href="https://github.com/PINTO0309/OpenVINO-EmotionRecognition" target="_self"&gt;Here&lt;/A&gt; is a python example created by one of our NCS community members that loads multiple models one one NCS. &lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;Please let me know if you have any further questions!&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Sincerely,&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;Sahira</description>
      <pubDate>Wed, 29 May 2019 01:17:01 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Two-models-in-a-neural-compute-stick2/m-p/645076#M1563</guid>
      <dc:creator>idata</dc:creator>
      <dc:date>2019-05-29T01:17:01Z</dc:date>
    </item>
    <item>
      <title>Re: Two models in a neural compute stick2</title>
      <link>https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Two-models-in-a-neural-compute-stick2/m-p/645077#M1564</link>
      <description>&lt;P&gt;Thank you so much @Sahira_at_Intel&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;I solved the problem referring the python example you mentioned.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;And I got to know that my codes were flawed because I made IEPlugin(device="MYRIAD") instance again at line 7 in my codes.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;If I change the line 7 like "plugin_skeleton = plugin_detector",&lt;P&gt;&amp;nbsp;&lt;/P&gt;the above codes run well.&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Anyway, thank you again @Sahira_at_Intel &lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Best regards,&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;Samuel Shin</description>
      <pubDate>Thu, 30 May 2019 14:45:04 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/Two-models-in-a-neural-compute-stick2/m-p/645077#M1564</guid>
      <dc:creator>idata</dc:creator>
      <dc:date>2019-05-30T14:45:04Z</dc:date>
    </item>
  </channel>
</rss>

