I have a DL model which sizes around 3.5 gigs and I'm creating a HTTP trigger to hit the model, all works fine when I test it in my local machine. When I start deploying the model as Function App into Azure, deployment breaks midway.
Also how can I link a gpu compute to the function app for the model to run.