Intel® Edge Software Hub
Get answers from community peers to your questions about building Edge Software Hub solutions for edge compute node.
Announcements
Welcome to the Intel Community! If you like the answer to your question, please mark it as 'Accepted Solution' to help others.

The Edge Software Vision Package for Red Hat Enterprise Linux is now available here.

Learn more about developing on Intel® Hardware and Software here.
394 Discussions

Cannot load library 'libcpu_extention_avx2.so' in Edge insight for vision

Shigeyoshi_O_Intel
1,085 Views

Hi,

I followed the instruction in get started to install edge insight for vision. The installation completed successfully, but I got an error when running single detection application on CPU.

libcpu_extention_avx2.so doesn't exist. It seems that the guide assume OpenVINO 2019.3, but 2020_R3 was installed.

I tried libMKLDNNPlugin.so instead, but I saw another error - 'undefined symbol: CreateExtension'.

May I ask for suggestion on how to solve the situation?

 

Regards,

0 Kudos
1 Solution
Shigeyoshi_O_Intel
1,031 Views

Hi Nadia,

Thank you so much! It works now after following your suggestion.

Regards,

Ohara

View solution in original post

0 Kudos
4 Replies
Nadia_J_Intel
Moderator
1,052 Views

Hi Shigeyoshi, 

Since you have OpenVINO 2020 installed, I assumed you downloaded the 2020.3 Version for the Edge Insights Vision, is this correct ?

Is it possible for you to share what else is inside your openvino/deployment_tools/inference_engine/lib/intel64/ directory ? 

Thanks.

0 Kudos
Shigeyoshi_O_Intel
1,043 Views

Hi Nadia,

Thank you so much. Here are the list of files in the folder. May I ask if you could take a look?

Regards,

Ohara

-----------------

total 71412
drwxr-xr-x 4 root root 4096 7月 14 17:35 ./
drwxr-xr-x 3 root root 4096 7月 14 17:35 ../
-rw-r--r-- 1 root root 9095407 5月 20 04:43 cache.json
drwxr-xr-x 2 root root 4096 7月 14 17:35 cldnn_global_custom_kernels/
-rwxr-xr-x 1 root root 194728 5月 20 05:04 compile_tool*
-rwxr-xr-x 1 root root 80040 5月 20 05:04 hddl_perfcheck*
-rwxr-xr-x 1 root root 9078576 5月 20 05:04 libclDNNPlugin.so*
-rwxr-xr-x 1 root root 1038584 5月 20 05:04 libGNAPlugin.so*
-rwxr-xr-x 1 root root 4827248 7月 14 17:58 libHDDLPlugin.so*
-rwxr-xr-x 1 root root 444424 5月 20 05:04 libHeteroPlugin.so*
-rwxr-xr-x 1 root root 194560 5月 20 05:04 libinference_engine_c_api.so*
-rwxr-xr-x 1 root root 2447376 5月 20 05:04 libinference_engine_legacy.so*
-rwxr-xr-x 1 root root 514040 5月 20 05:04 libinference_engine_lp_transformations.so*
-rwxr-xr-x 1 root root 636920 5月 20 05:04 libinference_engine_nn_builder.so*
-rwxr-xr-x 1 root root 878584 5月 20 05:04 libinference_engine_preproc.so*
-rwxr-xr-x 1 root root 2341000 7月 14 17:58 libinference_engine.so*
-rwxr-xr-x 1 root root 845816 5月 20 05:04 libinference_engine_transformations.so*
-rwxr-xr-x 1 root root 31287984 5月 20 05:04 libMKLDNNPlugin.so*
-rwxr-xr-x 1 root root 309328 5月 20 05:04 libMultiDevicePlugin.so*
-rwxr-xr-x 1 root root 2914600 5月 20 05:04 libmyriadPlugin.so*
-rwxr-xr-x 1 root root 227496 5月 20 05:04 myriad_compile*
-rwxr-xr-x 1 root root 80048 5月 20 05:04 myriad_perfcheck*
-rw-r--r-- 1 root root 1797520 5月 20 04:42 pcie-ma248x.mvcmd
-rw-r--r-- 1 root root 654 5月 20 05:01 plugins.xml
-rw-r--r-- 1 root root 1793452 5月 20 04:42 usb-ma2450.mvcmd
-rw-r--r-- 1 root root 2037808 5月 20 04:42 usb-ma2x8x.mvcmd
drwxr-xr-x 2 root root 4096 7月 14 17:35 vpu_custom_kernels/

0 Kudos
Nadia_J_Intel
Moderator
1,037 Views

Hi Shigeyoshi, 

In OpenVINO 2020_R3, the CPU extension is no longer needed, thus it is not in the folder. You can run the demo by removing the extension command as below to run the demo: 

$ python3 object_detection_demo_yolov3_async.py -i $HOME/Downloads/YOLOv3/Sample_videos/one-by-one-person-detection.mp4 -m $HOME/Downloads/YOLOv3/tensorflow-yolo-v3/FP32/frozen_darknet_yolov3_model.xml

You can also do the same for the Multi-Detection Application

python3 object_detection_demo_yolov3_async.py -i $HOME/Downloads/YOLOv3/Sample_videos/person-bicycle-car-detection.mp4 -m $HOME/Downloads/YOLOv3/tensorflow-yolo-v3/FP32/frozen_darknet_yolov3_model.xml

 

Thanks.

Shigeyoshi_O_Intel
1,032 Views

Hi Nadia,

Thank you so much! It works now after following your suggestion.

Regards,

Ohara

0 Kudos
Reply