Cannot find NCS2 in IECore (OpenVINO 2022.1.0)

[vlimodya1002]

Dear Sir or Madam,

I have downloaded and installed OpenVINO 2022.1.0 through "pip install openvino-dev[tensorflow2]==2022.1.0". I have installed this in a conda environment, and my OS is Windows 10.

However, when I run this code: 

from openvino.inference_engine import IECore

# Create an instance of IECore
ie = IECore()

# Get the available devices
available_devices = ie.available_devices

# Print the list of available devices
print("Available Devices:")
for device in available_devices:
     print(device)

I only see the CPU as output, even though I have my Neural Compute Stick 2 plugged-in.

Do I have to configure something before-hand? If so, I could not find it in this link

Configurations for Intel® Neural Compute Stick 2 — OpenVINO™ documentationCopy to clipboardCopy to clipboardCopy to clipboardCopy to clipboardCopy to clipboard — Version(2022.3)

for Windows. And what is <INSTALL-DIR>? I mean I did not input anything during the installation with pip.

Thank you in advance for your reply.

[wearyhacker]

The instructions below are for Linux. But they should give you a clue of what to do on windoze. For the NCS2 stick to be visible you need the openvino runtime installed. I would suggest that you install version 2022.3.1. Follow the instructions here. You can leave out step 3 for the time being, but in step 4 select the NCS2 tab and follow the steps there.

When you have finished step 4 you should have a working environment for both python and c++ apis in the shell you ran the setup in. Every time you start a new shell you must activate the environment by running step 2 again.

source /opt/intel/openvino_2022/setupvars.sh

You can put the above line at the end of your .bashrc if you want the environment to be activated whenever you start a new shell instance.

This is what you should see.

roger@dragon:~$ python
Python 3.10.6 (main, May 29 2023, 11:10:38) [GCC 11.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from openvino.runtime import Core
>>> core = Core()
>>> print(core.available_devices)
['CPU', 'GNA', 'MYRIAD']
>>> 
roger@dragon:~$ 

If you want to use the development toolkit. You have to make sure that your ide is configured to run the setupvars.sh  script in the context of whatever shell it is using to run python.

If want to use python outside of an ide the same thing applies.

[Peh_Intel]

Hi vlimodya1002,

For your information, OpenVINO™ Runtime package install from PyPI only includes a set of libraries for Intel® CPU and Intel® GPU hardware.

To configure VPU (MYRIAD), please download OpenVINO™ Runtime package using Installer.

Regards,

Peh

[vlimodya1002]

Hello Peh,

thank you for the reply. I think you gave the link of Runtime where the instructions for PyPI and Github installations are described. Or do I misunderstand something here? I think I didn't see a link to download the installer.

[Peh_Intel]

Hi vlimodya1002,

From the Installer, select the ‘Runtime’ option (default is ‘Development Tools’). Once Runtime is chosen, you are able to see “Download Archives” button to access to the storage.openvinotoolkit.org.

You can access the link below directly as well.

https://storage.openvinotoolkit.org/repositories/openvino/packages/2022.3.1/

OpenVINO™ 2022.3.1 LTS is the latest LTS release that support VPU (NCS2/HDDL).

Regards,

Peh

[vlimodya1002]

Hello Peh,

thanks for the solution. It has worked. And sorry for the late reply!

Best,

Vernando

[Peh_Intel]

Hi Vernando,

Hello, thank you for confirming your problem is resolved. I am glad to hear you are up and running.

This thread will no longer be monitored since this issue has been resolved. If you need any additional information from Intel, please submit a new question.

Regards,

Peh