Error when trying demo

[朱朱]

Hi, I have installed Openvino recently and when I try to run the demo_squeezenet_download_convert_run.bat to test my installation, it keeps telling me the error like below.

[ WARNING ] Failed to import Inference Engine Python API in: PYTHONPATH
[ WARNING ] [WinError 2] 系统找不到指定的文件。: 'D:\\Intel\\openvino_2021\\python\\python3.8\\openvino\\libs'
[ WARNING ] Failed to import Inference Engine Python API in: D:\Intel\openvino_2021.3.394\python\python3.8
[ WARNING ] [WinError 2] 系统找不到指定的文件。: 'D:\\Intel\\openvino_2021\\python\\python3.8\\openvino\\libs'
[ WARNING ] Failed to import Inference Engine Python API in: D:\Intel\openvino_2021.3.394\python\python3.8
[ WARNING ] [WinError 2] 系统找不到指定的文件。: 'D:\\Intel\\openvino_2021\\python\\python3.8\\openvino\\libs'
WARNING: Package(s) not found: openvino
[ WARNING ] Could not find the Inference Engine Python API. Installing OpenVINO (TM) toolkit using pip
Defaulting to user installation because normal site-packages is not writeable
Collecting openvino==2021.3

thanks for your answer~

系统：win10

wechat:zssjiushiwo

[Syamimi_Intel]

Hi Zhu Song,

Thank you for reaching out. I have validated demo_squeezenet_download_convert_run.bat using OpenVINO 2021.3 with python 3.8 and it works fine for me.

Thus, what I can suggest to fix the error is, there are four scripts required to modify. Please refer to this link to remove the codes under the red-highlighted regions and added the codes under the green-highlighted regions. Please be aware of sequence and space when pasting the codes.

The four scripts are in the following directories:

1. <INSTALL_DIR>\openvino_2021.3.394\python\python3.8\openvino\inference_engine\__init__.py

2. <INSTALL_DIR>\openvino_2021.3.394\python\python3.8\openvino\offline_transformations\__init__.py

3. <INSTALL_DIR>\openvino_2021.3.394\python\python3.8\ngraph\impl\__init__.py

4. <INSTALL_DIR>\openvino_2021.3.394\bin\setupvars.bat

Regards,

Syamimi

[朱朱]

thank you for your answer.

I corrected the four scripts as you said，the previous mistakes disappeared。

But there are new mistakes，as shown below:

Inference Engine version: 2.1.2021.3.0-2787-60059f2c755-releases/2021/3
Model Optimizer version: 2021.3.0-2787-60059f2c755-releases/2021/3
2021-05-18 17:03:02.757069: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'cudart64_101.dll'; dlerror: cudart64_101.dll not found
2021-05-18 17:03:02.757245: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
Please expect that Model Optimizer conversion might be slow. You are currently using Python protobuf library implementation.
However you can use the C++ protobuf implementation that is supplied with the OpenVINO toolkitor build protobuf library from sources.
Navigate to "install_prerequisites" folder and run: python -m easy_install protobuf-3.5.1-py($your_python_version)-win-amd64.egg
set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=cpp

For more information please refer to Model Optimizer FAQ, question #80. (https://docs.openvinotoolkit.org/latest/openvino_docs_MO_DG_prepare_model_Model_Optimizer_FAQ.html?question=80#question-80)
[ FRAMEWORK ERROR ] Exception message: cannot mmap an empty file

Possible reasons:
1. C:\Users\Tj\Documents\Intel\OpenVINO\openvino_models\models\public\squeezenet1.1/squeezenet1.1.caffemodel does not exist
2. C:\Users\Tj\Documents\Intel\OpenVINO\openvino_models\models\public\squeezenet1.1/squeezenet1.1.caffemodel does not have a valid structure

[ FRAMEWORK ERROR ] Model Optimizer is not able to parse C:\Users\Tj\Documents\Intel\OpenVINO\openvino_models\models\public\squeezenet1.1/squeezenet1.1.caffemodel

The hardware I use is NCS2，so i think i do not need the cudart64_101.dll file.

squeezenet1.1.caffemodel is downloaded by downloader.py, and I didn't modify the file.

I installed openvino according to the manual without any errors，so i do not know why the demo can not run .

Thank you again for your patient response。

微信:zssjiushiwo

[Vladimir_Dudnik · Accepted solution]

the message regarding cudart54_101.dll file is a standard tensorflow warning, you can just ignore this, as message suggest: "Ignore above cudart dlerror if you do not have a GPU set up on your machine.".

NCS2 hardware is not involved into conversion of model to OpenVINO IR.

and according to the message: "[ FRAMEWORK ERROR ] Exception message: cannot mmap an empty file" it seems model files either were not downloaded or downloaded files are corrupted.

Please try re download the model

[Syamimi_Intel]

Hi Zhu Song,

This thread will no longer be monitored since we have provided a solution. If you need any additional information from Intel, please submit a new question.

Regards,

Syamimi