exadel-inc / CompreFace

Leading free and open-source face recognition system
https://exadel.com/accelerator-showcase/compreface/
Apache License 2.0
5.71k stars 775 forks source link

GPU - two of "uwsgi" process stay at 46MB and not changed. Its need to be like that? #1189

Open antonbek89 opened 1 year ago

antonbek89 commented 1 year ago

Hi So I downloaded from the release the last version 1.2.0 and took the "docker-compose-gpu.yml" and changed the name to docker-compose.yml and ran it with docker compose up. Until here all working well

When I check the nvidia-smi with watch -n 0.1 The process"uwsgi" which belongs to Comparface stays at 46MB ALWAYS With the face detection or not, I guess something wrong here. I need help with this issue please

+---------------------------------------------------------------------------------------+ | NVIDIA-SMI 530.30.02 Driver Version: 530.30.02 CUDA Version: 12.1 | |-----------------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+======================+======================| | 0 NVIDIA GeForce GTX 1050 Ti Off| 00000000:01:00.0 Off | N/A | | 0% 55C P0 N/A / 72W| 1064MiB / 4096MiB | 11% Default | | | | N/A | +-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=======================================================================================| | 0 N/A N/A 667 G /usr/lib/xorg/Xorg 9MiB | | 0 N/A N/A 757 G /usr/bin/gnome-shell 6MiB | | 0 N/A N/A 2630 C ffmpeg 79MiB | | 0 N/A N/A 2650 C ffmpeg 75MiB | | 0 N/A N/A 2652 C ffmpeg 93MiB | | 0 N/A N/A 2679 C ffmpeg 150MiB | | 0 N/A N/A 3273 C ffmpeg 77MiB | | 0 N/A N/A 3980 C ffmpeg 83MiB | | 0 N/A N/A 3982 C ffmpeg 83MiB | | 0 N/A N/A 4432 C ffmpeg 77MiB | | 0 N/A N/A 5840 C uwsgi 46MiB | | 0 N/A N/A 5841 C uwsgi 46MiB | | 0 N/A N/A 11232 C ffmpeg 77MiB | | 0 N/A N/A 16590 C ffmpeg 77MiB | | 0 N/A N/A 25295 C ffmpeg 77MiB | +---------------------------------------------------------------------------------------+

This of containers was installed 89d33b05c114 exadel/compreface-fe:1.2.0
9ed09a738f42 exadel/compreface-admin:1.2.0 19ffe976b985 exadel/compreface-api:1.2.0 26bcc26b474b exadel/compreface-core:1.2.0 170c6a40166e exadel/compreface-postgres-db:1.2.0

antonbek89 commented 1 year ago

My GPU was with 4GB I replace it with 8GB now it’s working well

someone can explain why CompreFace take 5.4GB GPU memory please ?

antonbek89 commented 1 year ago

It is possible to limit the GPU memory which CompreFace uses?