I am trying to integrate object detection with prodigy. I found the link below from official documentation.
prodigy-recipes/image/tf_odapi at master · explosion/prodigy-recipes (github.com)
However, a lot of APIs used in the scripts are outdated. Is there an update version of this?
Hi and welcome! 
We don't have an updated version of this at the moment, since it was contributed by a community member and is already a bit older – but if you do end up adapting it, we'd definitely appreciate a PR!
For a newer approach that uses a model via an API (Segment Anything), you might find prodigy-segment useful. It's a slightly different approach with different trade-offs, but more modern:
Thank you!