[yaml] Normalize RunInference transform

[Polber]

What would you like to happen?

This bug is part of the Beam YAML milestone. More information about Beam YAML can be found here: https://github.com/apache/beam/tree/master/sdks/python/apache_beam/yaml#beam-yaml-api

A Beam YAML contribution guide can be found here: https://s.apache.org/beam-yaml-contribute

---

The RunInference transform should be normalized to be compatible with Beam YAML. There is a doc for normalizing the Beam IO's for Beam YAML, but the general process can be carried over to this work. That doc can be found here.

Issue Priority

Priority: 3 (nice-to-have improvement)

Issue Components

• Component: Python SDK
• Component: Java SDK
• Component: Go SDK
• Component: Typescript SDK
• Component: IO connector
• Component: Beam YAML
• Component: Beam examples
• Component: Beam playground
• Component: Beam katas
• Component: Website
• Component: Spark Runner
• Component: Flink Runner
• Component: Samza Runner
• Component: Twister2 Runner
• Component: Hazelcast Jet Runner
• Component: Google Cloud Dataflow Runner

[smeet07]

.take-issue

[smeet07]

@Polber can you guide me here

[Polber]

I would suggest taking a look at one of the other transforms that need normalizing:
https://github.com/apache/beam/issues?q=is%3Aopen+is%3Aissue+label%3Ayaml+normalize

TFRecord would be a good candidate if your interest is in ML transforms

[derrickaw]

#33406 covers the initial part with VertexAI Handler.

Probably need to add more handlers and/or documentation.

[derrickaw]

#38696 added huggingface model handler support