Loading a .pb model in Kapacitor UDF for inference

Hi. See I want to do Tf Inference in Kapacitor UDF https://docs.influxdata.com/kapacitor/v1.5/guides/anomaly_detection/ . I have written the code, but how can I load my .pb trained model.. I am using Container, should I just place the trained model in the docker itself?

The issue points to the Kapacitor anomaly-detection guide and a .pb model, but names no repository file, test, or entry point. Start by reading that guide and examining how the UDF runs in the container. Done would be a documented, working model-loading workflow, including how the trained model is supplied to the container.

Written by the indexing model from the issue text.