Skip to main content

Deploy_Your_Model

You can get this Notebook on GitHub.

Step 1: Prepare you environment on your raspberry pi​

Note：This part of code run on your raspberry pi5 or Recomputer R Series with your AI kit

Update time​




sudo date -s "$(wget -qSO- --max-redirect=0 google.com 2>&1 | grep Date: | cut -d' ' -f5-8)Z"



Upgrade raspberry pi​




sudo apt update && sudo apt upgrade



Install jupyter​




sudo apt install jupyter



Open jupyter​




jupyter notebook



Note

Please install hailo8l to your raspberrypi follow this, and for reComputerR series follow this.

# Install hailo-all on raspberrypi

! sudo apt install hailo-all -y
# Check if the hailo-all is installed

! hailortcli fw-control identify
# Install hailo-rpi5-examples

! git clone https://github.com/hailo-ai/hailo-rpi5-examples.git

Step 2: Configure labels​

The output of yolov8n hef model will be three classes 0, 1, 2. And we use json file to parse the output.

import json

data = {
    "detection_threshold": 0.1,
    "max_boxes": 200,
    "labels": [
        " ",
        "apple",
        "banana",
        "orange"
    ]
}

output_file = "config.json"

with open(output_file, "w") as file:
    json.dump(data, file, indent=4)

print(f"JSON file saved as {output_file}")
# Install necessary libs

! source ./hailo-rpi5-examples/setup_env.sh && pip install setproctitle
# Install postprocess .so lib

! source ./hailo-rpi5-examples/setup_env.sh && cd ./hailo-rpi5-examples && bash ./install.sh

Step 3: Run your model​

! source ./hailo-rpi5-examples/setup_env.sh && python ./hailo-rpi5-examples/basic_pipelines/detection_pipeline.py --labels-json ./config.json --hef ../../models/Chapter5/yolov8n.hef -i /dev/video0