AI Training Server Setup

Follow these steps to set up your AI Training Server:

  1. Prepare a Ubuntu os system with nvidia gpu (recommend version: ubuntu-24.04)

  2. Install nvidia driver

  • Open Software & Updates

  • Navigate to the Additional Drivers tab

  • Select the driver labeled “proprietary, tested” (e.g., nvidia-driver-560)

  • Click Apply Changes and reboot

  • After rebooting, verify the installation by running nvidia-smi

../../_images/ai_train_server_setup_2_1.png

  1. Install docker

sudo usermod -aG docker $USER

  • Use docker ps to verify the installation

  1. Create a folder to put the scripts and tar.gz files inside (ex. AI_train_server), the folder structure will be similar as follow

AI_train_server/
|-- docker_images/
|   |-- IMAGES.txt --> docker images list
|   |-- load_docker_images.sh  --> installation scripts
|   |-- acuity_converter_v1.1.tar.gz  --> docker image file
|   |-- training-server-train_latest.tar.gz  --> docker image file
|   |-- training-server-importer_latest.tar.gz  --> docker image file
|   |-- nvidia_cuda_12.1.1-cudnn8-runtime-ubuntu22.04.tar.gz  --> docker image file
|-- base
|-- base-20260109-165208.tar.gz
|-- workspaces_example
|-- workspaces-example-20251223-135111.tar.gz
|-- INSTALLATION.md

  1. Install the scripts

tar -xzf base-<timestamp>.tar.gz
tar -xzf workspaces-example-<timestamp>.tar.gz
cd docker_images && ./load_docker_images.sh
cd ../base
sudo ./install.sh
cd ../workspaces_example && ./install_workspaces_example.sh (optional, generate default example)

  1. After installation, there will an shortcut on the desktop, or you can login by “http://localhost:8080/login”.

../../_images/ai_train_server_setup_6_1.jpg

Fig. 19 System login interface

  1. On the model training interface, log in with the default username “admin@realtek.com” and password “admin123” After logging in, you can change the password if you want.

AI Training Server Run

  1. Log into the Server

    Ensure you have access to the server and log in with the appropriate credentials.

  2. Start the Training

    Once you have successfully logged into the server, you can upload your own dataset or download the example datasets from hugging face, user can also adjust the training configuration.

../../_images/ai_train_server_run_2.png

Fig. 20 Importing dataset

  1. Download the Model

    When the training is completed, a download button will appear. Click this button to download the trained model, which you can use on AmebaPro2.

../../_images/ai_train_server_run_3.png

Fig. 21 ‘Run’ and ‘Download Model’ button