👀 cv-model-pt_to_hef Overview

Convert Computer Vision model .pt to .hef for Rasberry Pi 5 Hailo AI HAT

🔎 Preparation

1. Prepare docker enviroment

Follow these steps:

sudo apt update
sudo apt upgrade -y

sudo apt install -y apt-transport-https ca-certificates curl software-properties-common

curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o /usr/share/keyrings/docker-archive-keyring.gpg

echo "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/docker-archive-keyring.gpg] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

sudo apt update
sudo apt install -y docker-ce docker-ce-cli containerd.io

sudo systemctl start docker
sudo systemctl enable docker

docker --version
sudo docker run hello-world




sudo systemctl stop docker.socket
sudo systemctl stop docker.service

sudo systemctl status docker
sudo systemctl status docker.socket

sudo mv /var/lib/docker /home/$USER/docker_data
sudo ln -s /home/$USER/docker_data /var/lib/docker

sudo systemctl start docker
sudo systemctl enable docker

Docker Root Dir: /home/$USER/docker_data

2. Go to theHAILO AI DEVELOPER ZONEadress and download .zip file with this configration: Software Package[AI Software Suite], Software Sub-Package[AI Software Suite], Architecture[x86], OS[Linux], Python Version[3.8]
3. In model training, place 60%–80% of the photos you use into the train/images folder. The photos should be raw, unlabelled images — there should be no labeling process applied to them. You don’t need labels or a classes.txt file; only the images folder is required.

3. Get your .pt model and convert it to .onnx

1. Run this .py code at the same directory with your .pt model:

!pip install ultralytics
from ultralytics import YOLO

model = YOLO("model.pt")
model.export(format="onnx")

📦 Setup

1. unzip hailo8_ai_sw_suite_2025-10_docker.zip -d /home/$USER/docker_hailo
2. cd /home/$USER/docker_hailo/
3. Edit your .sh document and delete these lines:

   -v /etc/timezone:/etc/timezone:ro
   -v /etc/localtime:/etc/localtime:ro`

4. ./hailo_ai_sw_suite_docker_run.sh --override

   If you want to continue with your already configured project:./hailo_ai_sw_suite_docker_run.sh --resume

5. /home/$USER/docker_hailo/shared_with_docker/train/images/(60%–80% of your photos)
   /home/$USER/docker_hailo/shared_with_docker/models/model.onnx
   /home/bob/docker_hailo/shared_with_docker/doc/
6. In normal terminal(run): cd /home/$USER/docker_hailo/ mv train/ shared_with_docker/
7. git clone https://github.com/LukeDitria/RasPi_YOLO.git
8. cd RasPi_YOLO/ Then you shold be in '/local/workspace/RasPi_YOLO/' directory
9. python hailo_calibration_data.py --data_dir /local/shared_with_docker/train/images/ --target_dir /local/shared_with_docker/doc

🎉 Run

``