I have trained my model on yolo format and now I would like to use hailo module for my project my question is how to start with it , I have a virtual machine with ubuntu 20.04 but whenever I try to install the software suit it with .run file it throws an error saying chmod +x hailo_ai_sw_suite_2025-01.run
sudo ./hailo_ai_sw_suite_2025-01.run
HAILO System requirements check - log
| Available RAM (21 GB) below recommended amount (32 GB). V | CPU architecture x86_64 is supported. V | Required avx CPU flag is supported. | Missing package: python3-tk. | Missing package: graphviz. | Missing package: libgraphviz-dev. | Missing package: python3.8-dev. | GPU Requirements are not checked- no GPU connected. V | Environment variable CC is unset. V | Environment variable CXX is unset. V | Environment variable LD is unset. V | Environment variable AS is unset. V | Environment variable AR is unset. V | Environment variable LN is unset. V | Environment variable DUMP is unset. V | Environment variable CPY is unset. | Node.js is not available. It can disallow start of DFC Studio part of SDK package. | Could not test for SecureBoot, make sure it is disabled on this machine. V | Package linux-headers-5.15.0-136-generic found. | Missing package: build-essential. V | Python package pip found. | Missing python package: virtualenv.
my question is can I use diffrent os for installation or do I ahve to use only ubuntu 20.04 and with 22,04 also suit is not installing at all
Hi @KlausK ,
Thanks for your detailed reply , so my next question would be instead of VM can I use The Raspberry Pi AI Kit bundles the Raspberry Pi M.2 HAT+ with a Hailo AI acceleration module Raspi forthis purpose what do you think ? Would it work ?
To convert a model (Model Build Computer), you need an x86-based machine running Ubuntu 20.04 or 22.04 with a lot of RAM. For certain optimizations, a high-performance NVIDIA GPU is recommended.
Inference can run on a wide range of hosts, including a Raspberry Pi with an AI HAT+.
one more thing , I am using now a different machine with ubuntu 22,04 and docker container works their , now what should be the next steps for running my yolo model over the Hailo-8 M2 AI Accerelation module , how should I proceeed with it , and device is in the UGREEN house casing as a usb connector
Please work trough the tutorials to get familiar with the workflow. Inside the Docker call:
hailo tutorial
This will start a Jupyter Notebook server with notebooks for each step of the process including running the model in the emulator in native and quantized mode.
Please make sure you have a Thunderbolt port. See this guide:
Then try the TAPPAS apps in the Docker. The detection example can run your yolo model when you compile it with NMS. Compare your model HEF file with the example models.
hailortcli parse-hef model.hef
You can find more options on how to create your own application here:
