Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.
6392 Discussions

Cann't infer with Iris Pro Graphics P580

GangXiong
Novice
829 Views

Hello,  I have installed openvino on my computer and the two verification scripts are executed without any errors with default settings.  But when I specified the device as "GPU" with the following command:

demo_squeezenet_download_convert_run.bat -d GPU

I got the following error message no matter what the model precision is (FP32/FP16):

target = GPU
target_precision = FP32
Python 3.6.5
[setupvars.bat] OpenVINO environment initialized
INTEL_OPENVINO_DIR is set to C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194
Python 3.6.5
ECHO is off.
Requirement already satisfied: pyyaml in c:\users\admin\appdata\roaming\python\python36\site-packages (from -r C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\demo\..\open_model_zoo\tools\downloader\requirements.in (line 1)) (5.3.1)
Requirement already satisfied: requests in c:\users\admin\appdata\roaming\python\python36\site-packages (from -r C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\demo\..\open_model_zoo\tools\downloader\requirements.in (line 2)) (2.18.4)
Requirement already satisfied: urllib3<1.23,>=1.21.1 in c:\users\admin\appdata\roaming\python\python36\site-packages (from requests->-r C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\demo\..\open_model_zoo\tools\downloader\requirements.in (line 2)) (1.22)
Requirement already satisfied: chardet<3.1.0,>=3.0.2 in c:\users\admin\appdata\roaming\python\python36\site-packages (from requests->-r C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\demo\..\open_model_zoo\tools\downloader\requirements.in (line 2)) (3.0.4)
Requirement already satisfied: idna<2.7,>=2.5 in c:\users\admin\appdata\roaming\python\python36\site-packages (from requests->-r C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\demo\..\open_model_zoo\tools\downloader\requirements.in (line 2)) (2.6)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\admin\appdata\roaming\python\python36\site-packages (from requests->-r C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\demo\..\open_model_zoo\tools\downloader\requirements.in (line 2)) (2020.6.20)
WARNING: You are using pip version 20.2.4; however, version 20.3.1 is available.
You should consider upgrading via the 'c:\program files (x86)\python3.6.5\python.exe -m pip install --upgrade pip' command.
Download public squeezenet1.1 model
python "C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\open_model_zoo\tools\downloader\downloader.py" --name "squeezenet1.1" --output_dir "C:\Users\Admin\Documents\Intel\OpenVINO\openvino_models\models" --cache_dir "C:\Users\Admin\Documents\Intel\OpenVINO\openvino_models\cache"
################|| Downloading models ||################

========== Retrieving C:\Users\Admin\Documents\Intel\OpenVINO\openvino_models\models\public\squeezenet1.1\squeezenet1.1.prototxt from the cache

========== Retrieving C:\Users\Admin\Documents\Intel\OpenVINO\openvino_models\models\public\squeezenet1.1\squeezenet1.1.caffemodel from the cache

################|| Post-processing ||################

========== Replacing text in C:\Users\Admin\Documents\Intel\OpenVINO\openvino_models\models\public\squeezenet1.1\squeezenet1.1.prototxt
squeezenet1.1 model downloading completed

Waiting for 0 seconds, press a key to continue ...

Target folder C:\Users\Admin\Documents\Intel\OpenVINO\openvino_models\ir\public\squeezenet1.1\FP32 already exists. Skipping IR generation with Model Optimizer.
If you want to convert a model again, remove the entire C:\Users\Admin\Documents\Intel\OpenVINO\openvino_models\ir\public\squeezenet1.1\FP32 folder.

Waiting for 0 seconds, press a key to continue ...

###############|| Generate VS solution for Inference Engine samples using cmake ||###############


Waiting for 0 seconds, press a key to continue ...
Creating Visual Studio 16 2019 x64 files in C:\Users\Admin\Documents\Intel\OpenVINO\inference_engine_samples_build...
-- Selecting Windows SDK version 10.0.18362.0 to target Windows 10.0.14393.
-- The C compiler identification is MSVC 19.27.29111.0
-- The CXX compiler identification is MSVC 19.27.29111.0
-- Check for working C compiler: C:/Program Files (x86)/Microsoft Visual Studio/2019/Community/VC/Tools/MSVC/14.27.29110/bin/Hostx64/x64/cl.exe
-- Check for working C compiler: C:/Program Files (x86)/Microsoft Visual Studio/2019/Community/VC/Tools/MSVC/14.27.29110/bin/Hostx64/x64/cl.exe -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Detecting C compile features
-- Detecting C compile features - done
-- Check for working CXX compiler: C:/Program Files (x86)/Microsoft Visual Studio/2019/Community/VC/Tools/MSVC/14.27.29110/bin/Hostx64/x64/cl.exe
-- Check for working CXX compiler: C:/Program Files (x86)/Microsoft Visual Studio/2019/Community/VC/Tools/MSVC/14.27.29110/bin/Hostx64/x64/cl.exe -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Found InferenceEngine: C:/Program Files (x86)/IntelSWTools/openvino_2020.3.194/deployment_tools/inference_engine/lib/intel64/Release/inference_engine.lib (Required is at least version "2.1")
-- Configuring done
-- Generating done
-- Build files have been written to: C:/Users/Admin/Documents/Intel/OpenVINO/inference_engine_samples_build

Waiting for 0 seconds, press a key to continue ...

###############|| Build Inference Engine samples using MS Visual Studio (MSBuild.exe) ||###############


Waiting for 0 seconds, press a key to continue ...
"C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Current\Bin\MSBuild.exe" Samples.sln /p:Configuration=Release /t:classification_sample_async /clp:ErrorsOnly /m
Microsoft (R) Build Engine version 16.7.0+b89cb5fde for .NET Framework
Copyright (C) Microsoft Corporation. All rights reserved.


Waiting for 0 seconds, press a key to continue ...

###############|| Run Inference Engine classification sample ||###############


Waiting for 0 seconds, press a key to continue ...
        1 file(s) copied.
classification_sample_async.exe -i "C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\demo\car.png" -m "C:\Users\Admin\Documents\Intel\OpenVINO\openvino_models\ir\public\squeezenet1.1\FP32\squeezenet1.1.xml" -d GPU
[ INFO ] InferenceEngine:
        API version ............ 2.1
        Build .................. 2020.3.0-3467-15f2c61a-releases/2020/3
        Description ....... API
[ INFO ] Parsing input parameters
[ INFO ] Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     C:\Program Files (x86)\IntelSWTools\openvino_2020.3.194\deployment_tools\demo\car.png
[ INFO ] Creating Inference Engine
        GPU
        clDNNPlugin version ......... 2.1
        Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3

[ INFO ] Loading network files
[ INFO ] Preparing input blobs
[ WARNING ] Image is resized from (787, 259) to (227, 227)
[ INFO ] Batch size is 1
[ INFO ] Loading model to the device
[ ERROR ] Error has occured for: pooling:pool1
Best_kernel.empty()(true) should be false
Cannot find a proper kernel with this arguments

Error

 

My environment is like this:

CPU: Intel Xeon E3-1515M V5

GPU: Iris(TM) Pro Graphics P580 (Driver version: 21.20.16.4627)

OS: Windows enterprise 2016 LTSB  (version 1607)

Openvino version: 2020.3.194

The official document says that openvino does support my hardware. Is there anyone can provide some useful information? Thanks in advance! 

0 Kudos
1 Solution
GangXiong
Novice
775 Views

Hello @Adli

 

Thanks for your kind reply. I think I've solved the problem now. It seemed that the version of my graphics driver is not high enough, since the GPU can be used as the inference device after I upgraded the driver version from 21.20.16.4627 to 23.20.16.4982. The following link may give the reason:

https://github.com/openvinotoolkit/openvino/issues/1256

Thank you again for your help!

Best regards,

Gang

 

View solution in original post

0 Kudos
6 Replies
Adli
Moderator
810 Views

Hi Gang,


Thank you for reaching out to us. We are currently investigating this issue and will get back to you soon.


Regards,

Adli


0 Kudos
Adli
Moderator
800 Views

Hi Gang,

 

If possible, could you install the OpenVINO toolkit 2021.1 version and try executing the demo on GPU. You can download the latest version of the OpenVINO toolkit at the following link:

https://software.intel.com/content/www/us/en/develop/tools/openvino-toolkit/download.html

 

Regards,

Adli


0 Kudos
GangXiong
Novice
793 Views

Hi @Adli ,

 

Thanks for your kind reply. In fact, the first thing I tried  is to upgrade the openvino toolkit version to 2021.1 when I met the problem. With this version, it also went fine when inferring with CPU,  while it would raise such kind of errors: "implicit declaration of function ’intel_sub_group_block_read_us8‘ " with GPU the inference device. Is it implies that my GPU doesn't support openvino toolkit?

 

Best regards,

 

Gang

0 Kudos
Adli
Moderator
786 Views

Hi Gang,


If possible, let's try some more troubleshootings first. Remove the 'openvino_models' file in '<user_directory>\Documents\Intel\OpenVINO' directory and run the 'demo_squeezenet_download_convert_run' demo infer in GPU on OpenVINO 2021.


If it doesn’t work, I will escalate this case. Please also share the full log error here.


Regards,

Adli


0 Kudos
GangXiong
Novice
776 Views

Hello @Adli

 

Thanks for your kind reply. I think I've solved the problem now. It seemed that the version of my graphics driver is not high enough, since the GPU can be used as the inference device after I upgraded the driver version from 21.20.16.4627 to 23.20.16.4982. The following link may give the reason:

https://github.com/openvinotoolkit/openvino/issues/1256

Thank you again for your help!

Best regards,

Gang

 

0 Kudos
Adli
Moderator
767 Views

Hi Gang,


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,

Adli



0 Kudos
Reply