Test inference of yolov5 on a aws ec2 instance

Test the model trained previously on a aws ec2 instance. More details here

Result

curl -w '\n' -F image=@helmet.jpg http://ip:8080/predict -w "\n %{time_connect}:%{time_starttransfer}:%{time_total} \n"

It works, not so pretty to visualize. Let’s see the response

Ec2 instance

I will post here the main configuration for the application, such as how to open the port

In the security group, open the 8080 port

Set up

Access the machine through SSH

• clone this repository
• create an environment with conda env
• in the root folder, execute pip install -r requirements.txt
• in the root folder, execute gunicorn --bind=0.0.0.0:8080 server:app
• you should see

[2021-09-09 12:52:20 +0000] [45903] [INFO] Starting gunicorn 20.1.0
[2021-09-09 12:52:20 +0000] [45903] [INFO] Listening at: http://0.0.0.0:8080 (45903)
[2021-09-09 12:52:20 +0000] [45903] [INFO] Using worker: sync
[2021-09-09 12:52:20 +0000] [45904] [INFO] Booting worker with pid: 45904

Our app is now listening to port 8080. The IP? Let’s find out on the ec2 dasboard

• select your instance on ec2 dashboard
• select ipv4 address

Call the endpoint

curl -F image=@helmet.jpg http://your_ipv4_address:8080/predict