Can I train pretrained tensorflow lite model in matlab for object detection?

I am using latest deeplearning toolbox interface for Tensorflow Lite in matlab .For my project I want to train and create a custom object detection model in Tensorflow Lite format to use on raspiberrypi so I want to use the preprocessing capabilities of the matlab for that so can i do training a pretrained Tensorflow lite model in matlab with the help of this toolbox ??

Answer 1

[Muskan]

Hi Vikas,

As per my understanding of the question, yes, you can fine-tune a pre-trained TensorFlow Lite model in MATLAB for object detection using transfer learning. Transfer learning is a common technique used to retrain a pre-existing model on new data, allowing you to quickly and easily adapt a pre-trained model for new tasks.

You can refer to the following documentations for a better understanding.

Run TensorFlow Lite Models with MATLAB and Simulink Video - MATLAB (mathworks.com)

Pretrained Deep Neural Networks - MATLAB & Simulink (mathworks.com)

I hope the above information helps resolve your query.

Thanks.