Dataset Preparation - Visualization

Once data collection is complete, you can preview and inspect your recorded dataset using the following steps.

Launch the Dataset Visualization Tool:

Enter the Physical AI Tools Docker container

ROBOT PC

cd ~/physical_ai_tools && ./docker/container.sh enter

Navigate to the physical_ai_tools/lerobot directory inside the container:

ROBOT PC 🐋 PHYSICAL AI TOOLS

cd /root/ros2_ws/src/physical_ai_tools/lerobot/src

Run the following command to launch the dataset visualization tool:

python lerobot/scripts/visualize_dataset_html.py \
  --host 0.0.0.0 \
  --port 9091 \
  --repo-id YOUR_REPO_ID

• Replace YOUR_REPO_ID with the dataset ID found under the User ID section in the right sidebar.
• For example: ROBOTIS/ffw_sg2_rev1_Test

You should see an output similar to the following:

WARNING:root:'torchcodec' is not available in your platform, falling back to 'pyav' as a default decoder
INFO 2025-07-28 06:55:31 set_html.py:364 Output directory already exists. Loading from it: '/tmp/lerobot_visualize_dataset_772b0wso'
 * Serving Flask app 'visualize_dataset_html'
 * Debug mode: off
INFO 2025-07-28 06:55:31 _internal.py:97 WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
 * Running on all addresses (0.0.0.0)
 * Running on http://127.0.0.1:9091
 * Running on http://192.168.50.127:9091
INFO 2025-07-28 06:55:31 _internal.py:97 Press CTRL+C to quit

Access one of the following URLs to view the dataset:

• http://ffw-{serial number}.local:9091
• http://{IP address of ROBOT PC}:9091

You should see a web interface similar to the one below: