The repository contains python modules used by the Fiji macros to run tensorflow training and inference from a general workstation on our deep learning workstation, volta.
model_runner.py
runs on deep learning workstation listening for requests on port 8080.
It handles running training and inference, and can also kill an existing process.