Sir,
Is it possible to train the Transfer learning toolkit models with my own dataset on Jetson Nano. If not, how to train those models with my own dataset. Please help me. Thanks in advance.
Yes, it is possible to train the Transfer learning toolkit models with your own dataset.See Transfer Learning Toolkit — Transfer Learning Toolkit 3.0 documentation for more details.
The goal of TLT is to train and fine-tune a model using the user’s own dataset.
But please note that the training is running on host PC instead of Nano.
The inference can be host PC or nano. See https://docs.nvidia.com/metropolis/TLT/tlt-user-guide/text/requirements_and_installation.html#requirements-and-installation
The TLT is designed to run on x86 systems with an NVIDIA GPU (e.g., GPU-powered workstation, DGX system) or can be run in any cloud with an NVIDIA GPU. For inference, models can be deployed on any edge device such as an embedded Jetson platform or in a data center with GPUs like T4 or A100.