Agent skill · NVIDIA
tao-train-single-step
Standard single-step train/eval/export workflow for any TAO model. Use when training a TAO model on a dataset
What it needs
About 2k tokens when loaded.
What this skill does
Normal Train Standalone install? If this session was not initialized by the TAO skill bank plugin, run the tao-setup skill first (host preflight, credentials, cross-skill discovery). Standard supervised fine-tuning: train a model on a labeled dataset, optionally evaluate, then optionally export. The most common TAO workflow for adapting a pretrained model to a new dataset. Steps 1. train — executed through AutoML when the selected model has automlenabled: true and automlpolicy is on; set automlpolicy=off for a plain single training run 2. eval — executed if evaldataseturi is resolved 3. export — optional, on user request after training Prerequisites The selected model skill's resolved containerimage is the default training runtime. Do not replace it with a host venv, uv environment, generic training image, or hand-written trainer unless the user explicitly requests that execution mode. SDK/controller Python environments are control-plane-only; the model action remains container-backed. Required model: A compatible TAO model (e.g., clip, nvdinov2, groundingdino) traindataseturi: URI of the training dataset (e.g., s3://bucket/train/) platform: Ask from the generated supported-platform list: ${TAOSKILLBANKPATH:-~/tao-skills-external}/scripts/listtaoplatforms.py --format text container image confirmation: resolve the default image from the selected model/action config, show it to the user, and require confirmation or image=<override> before creating runner files or submitting training. Optional evaldataseturi: Some model skills mark this as required — check the resolved model skill before treating it as optional. basecheckpoint: If not provided, defaults to the NGC pretrained checkpoint listed in the model skill, or trains from scratch if no NGC checkpoint exists. automlpolicy: on by default; set off to bypass model-level AutoML for this run while leaving model metadata unchanged. Use only on / off in new launch settings. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills NVIDIA/tao-train-single-step