This example builds a simple neural network. It is built and exported in make_model.py.
inference.py implements the required two functions load_model and call_model that define how to load a model and how the model should be called. The call_model function correctly follows the task schemas.
test.py shows how Backprop uses the model in a production environment.
The upload .zip will contain inference.py, model.h5, config.json and requirements.txt. This can be uploaded on Backprop's Dashboard.
The uploaded model can be invoked by making POST requests with the appropriate body to:
api.backprop.co/customapi.backprop.co/image-classification