Fine-tune a TAO model on your labeled dataset.
What it does
Trains a TAO model on a labeled dataset, with conditional evaluation and optional export.
When to use it
Use for standard TAO training without iterative data augmentation, AutoML, or DEFT loops; explicitly disable AutoML for a plain single run.
How to use it
Give it a compatible TAO model and training dataset URI; it runs training after platform and image confirmation, with evaluation when configured and export on request.
What you provide
- a compatible TAO model
- the training dataset URI
What you need · 10
Requires Docker for the container-backed training workflow.
Requires NVIDIA Container Toolkit; individual workflows declare additional requirements.
Must run first for a standalone installation when the session was not initialized by the TAO skill bank plugin.
Routes platform, credential, and dataset intake through this skill as part of the workflow.
Access · 0
No additional actions listed.