Mobile Inference

Hello all,

I’m a newbie to PyTorch and I would like to run inference on a mobile device (ARM-based). I have my PyTorch model that needs to run outside my development environment where the model prediction will be executed.
Is there a PyTorch interpreter for embedded devices or what is workflow when the model needs to be run on embedded devices? How should I proceed to execute the PyTorch models on memory and computational constraint devices?

Thank you.