martinenkoEduard

> Hi @martinenkoEduard, Is it implemented through python, JavaScript or others? Could you please help me in reproducing the same with complete steps at our end. Thank you! ``` #https://www.youtube.com/watch?v=jn1HSXVmIrA...

BTW I have suspect I have a similar issue in .deb standalone installation. What logs do I need to post a bug report? (what are their locations)

After thath I get etect?&face_plugins=calculator", host: "localhost:8000" compreface-ui | 172.19.0.1 - - [24/Aug/2022:18:20:15 +0000] "POST /api/v1/detection/detect?&face_plugins=calculator HTTP/1.1" 504 160 "-" "python-requests/2.25.1" compreface-api | 2022-08-24 18:20:36.200 ERROR 7 --- [nio-8080-exec-1] c.e.f.c.h.ResponseExceptionHandler...

More logs: 73.814 seconds (JVM running for 76.911) compreface-core | [18:15:38] ../src/nnvm/legacy_json_util.cc:208: Loading symbol saved by previous version v1.3.0. Attempting to upgrade... compreface-core | [18:15:38] ../src/nnvm/legacy_json_util.cc:216: Symbol successfully upgraded! compreface-core...

Images are not that big. And most of them contain only one face.. And on most of them CPU usage are close to 100% on all cores....

But there are more chances if there are several faces. I am using Python SDK, with the latest version of compre-face Winth 20gb of RAM SubCenter-ArcFace-r100

Log: [log.txt](https://github.com/exadel-inc/CompreFace/files/9418667/log.txt)

There is a chance that compre face recovers after that. But sometimes it just returns error, and need to be restarted.

The most annoying thing that in requires a restart to fix it. The model keeps at 100% CPU usage.

This is right after one of those group faces. But it still kept working, but consumes almost 100% of cpu