Today, NVIDIA Transfer Learning Toolkit has been renamed to the NVIDIA TAO Toolkit. The core technologies, performance, and roadmap remain unchanged.
NVIDIA Train, Adapt, and Optimize (TAO) is NVIDIA’s AI-model-adaptation platform that simplifies and accelerates the creation of AI.
The TAO platform brings together various NVIDIA technologies required to accelerate the model adaptation process. This includes Transfer Learning Toolkit or TLT, which will now be renamed to the TAO Toolkit. The NVIDIA TAO Toolkit is a CLI-based, Jupyter notebook driven solution that simplifies the process of transfer learning by abstracting away the AI/DL framework complexity.
The list below is our changes:
APIs, NGC containers, and other developer resources have all been updated as a result of the change.
Performance, achievements and new features that we have shared with you will continue to operate and stay the same.
CLI command will change from ‘tlt’ to ‘tao’. Other than that, all other options will continue to work as in TLT. To migrate to TAO Toolkit version, refer to the TLT to TAO Migration guide.
Existing models that you trained with TLT will continue to work with the TAO Toolkit.
To learn more about the recent change visit the TAO Toolkit product page.