liveness_vit

Liveness detection using ViT model

ViT model:

• cpu: < 10 FPS
• gpu: 60 FPS

Resnet50 model:

• cpu: < 30 fps
• gpu: < 70 FPS

Set up NGINX

Here I have 2 servers in localhost:8091 and localhost:8092. Then I will use NGINX as an API gatway or load balancer

First, I have config to run the server using wsgi server called waitress pip install waitress

Then setup NGINX docker container run -p 8090:80 --name nginx_tutorial nginx docker container exec -it nginx_tutorial bash apt-get update apt-get install nano nano nginx.conf

The set up is as this link: https://viblo.asia/p/cau-hinh-nginx-server-nhu-the-nao-ByEZko0xZQ0

Then we modify the nginx.conf file

  events {
  
  }
  
  http {
  
      upstream backend_servers {
          server your_host:8091;
          server your_host:8092;
      } 
  
      server {
          listen 80;
          server_name nginx-tutorial.test;
  
          location / {
              proxy_pass http://backend_servers;
              proxy_set_header Host $host;
              proxy_set_header X-Real-IP $remote_addr;
              proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
              proxy_set_header X-Forwarded-Proto $scheme;
          }       
      }
  }

We have to add 1 more line in the file etc/hosts like this

127.0.0.1 nginx-tutorial.test

Reference: https://viblo.asia/p/load-balancing-voi-nginx-bWrZnVxnZxw

Then run:

  python src/api.py
  python src/api_2.py
  python src/test_api.py 

We can see that the port is changing over time -> success