Turn skeleton motion into action predictions.
Trains, evaluates, and exports skeleton-action classifiers and runs predictions on pose-keypoint sequences.
When to use it
Use for training, evaluation, export, or inference with a TAO pose-classification model.
Give it pose-keypoint data and a requested model operation; it produces training results, evaluation results, predictions, or an exported model.
What you provide
Requires Docker to run the packaged TAO container; no host Python or TAO SDK is needed.
Requires NVIDIA Container Toolkit alongside Docker.
Requires at least one GPU with 8GB or more VRAM per GPU.
Must run first for a standalone installation when the session was not initialized by the TAO skill bank plugin.
No additional actions listed.