I am researching edge ML option for text based ML models but ost of what I see is around vision, image categorization, etc.
Thank you for reaching out.
It may not be possible to run non-vision ML on the NCS2. On the other hand, you can make an image representation of the data you are trying to use, train a model with such data (images), and then use your model to classify that way.
Please let me know if you have any additional questions.
Intel Customer Support Technician
A Contingent Worker at Intel