is there a google colab notebook to convert a yolov8 model to a .hef file

user921 · March 24, 2026, 4:53pm

hi, I have a raspberry pi 5 with a hailo 8L. Is there a google colab notebook script to convert a yolov8 object detection model to a .hef file to run on the raspberry pi. Thanks mike

Michael · March 24, 2026, 8:31pm

Hi @user921,

Please see here:

github.com/hailo-ai/hailo-apps

doc/developer_guide/retraining_example.md

main

# YOLOv8 Retraining Example

This guide demonstrates how to retrain a YOLOv8 model for barcode detection using the Kaggle barcode-detector dataset. After training, we'll convert the model to HEF format and deploy it on the Raspberry Pi 5 AI Kit (or any x86 platform with a Hailo accelerator).

#### Required compute resources:
- **Training phase**: Any cloud service with GPU resources (including Python notebook-based services like Google Colab)
- **Hailo compilation phase**: Conversion to HEF format is compute-intensive but manageable on standard PCs (e.g., overnight task)

For complete reference implementations, see the Jupyter notebooks (please note some of the stages, including installations, might take time): 
- [`retraining.ipynb`](retraining.ipynb)
- [`compilation.ipynb`](compilation.ipynb)

## Training Setup

### Set up environment

First, create a Python virtual environment:

```bash
python -m venv env

This file has been truncated. show original

github.com/hailo-ai/hailo-apps

doc/developer_guide/compilation.ipynb

main

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",

This file has been truncated. show original

Thanks,

user921 · March 27, 2026, 7:10am

hi Michael, Thanks for your reply. I created a google colab notebook script following the example notebook that you referenced. I got a error when I tried to run !hailomz compile \. The error indicated that “The zoo’s internal metadata check is hardcoded and rejects hailo8l for yolov8n". I have a raspberry pi 5 with a hailo 8L hat. Could you please advise how I can correct this issue. Many Thanks mike

Michael · March 27, 2026, 7:42am

Hi @user921,

The line that finds the uploaded DFC wheel is commented out (for example). Change:

dfc_whl = 'dfc'  # [f for f in uploaded.keys() if "hailo_dataflow_compiler" in f]

to:

dfc_whl = [f for f in uploaded.keys() if "hailo_dataflow_compiler" in f]

In the last cell change the target to Hailo8L:

!hailomz compile \
    --ckpt /content/barcode-detection.onnx \
    --calib-path /content/barcode-detect/valid/ \
    --yaml /content/hailo_model_zoo/hailo_model_zoo/cfg/networks/yolov8s.yaml \
    --classes 2 \
    --hw-arch hailo8l \
    --performance

Thanks,