SK-TDA4VM: custom train and deploy model on sk-tda4vm

Hi Nar,

To get started on model training and deployment, I would suggest looking at the following documentation: https://software-dl.ti.com/jacinto7/esd/processor-sdk-linux-edgeai/TDA4VM/08_06_00/exports/docs/common/inference_models.html

Particular points of interest from the documentation are:

1. If not done already, try to see if you can run some of our out-of-box examples with the pretrained/precompiled models, and reference our example code to get an understanding of how models can be deployed.
2. Try EdgeAI-ModelMaker as a starting point, as this is the tool that is most user-friendly that is currently available in terms of a step-by-step instruction to train your own model. The limitation is this tool supports limited usecases such as image classification and object detection.
3. For more advanced custom model training/compiling/deployment, try the edgeai-tidl-tools. This would give you more flexibility in terms of the number of usecases the tool supports.

Regards,

Takuma