tao-train-image-classification - Train and deploy TAO PyTorch image classifiers
Train, evaluate, distill, quantize, export, and run inference for PyTorch-based TAO image-classification models.
Tags
Updated: 2026-09-28Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Train image classifiers
- Evaluate model performance
- Distill classification models
- Quantize models
- Export trained models
- Run model inference
Inputs
- Training image folders
- Validation image folders
- Test image folders
- Class label file
- Backbone weights
- Pretrained model checkpoint
- Action specification
- Writable results directory
Outputs
- Trained model checkpoints
- Evaluation metrics
- Quantized model artifacts
- Exported model artifacts
- TensorRT engine
- Inference results
Requirements
- Docker
- NVIDIA Container Toolkit
- TAO skill bank setup when standalone
- PyTorch model support