[BEAM-14305] Add call in RunInference API

• Type: Sub-task
• Status: Resolved
• Priority: P2
• Resolution: Fixed
• Fix Version/s: 2.39.0
• Component/s: sdk-py-core
• Assignee: Andy Ye
• Reporter: Ryan Thompson
• Created: 13/Apr/22 17:15
• Updated: 21/Apr/22 21:25
• Resolved: 21/Apr/22 21:25

Description

Right now beam.ml.inference.api.RunInference does not call anything and is an empty API.

 

It should call base.RunInference.

 

We should also investigate changing the api to work more like a transform.

 

Right now it might look like this:

api.RunInference(pcollection, modelLoader)

 

but we might want it to follow other transforms and have it run like this:

 

pcollection | api.RunInference(modelLoader)

Issue Links

• links to: GitHub Pull Request #17377