Use of TFLite as backend

Hello, Please let me know if it is possible to use tflite as backend? In my case, performance is not the main concern. I like to have smaller tf library footprint to optimize my entire library size.

It’s not supported now. Could you tell us more on your scenario? For inference or edge device?

thanks for your reply, I would say it is more similar to an edge device since the device architecture can not support huge libraries like tensorflow. With other models that I developed, I can call Keras tflite convertor and run the model fully using tflite.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.