What would you like to happen?

Currently we use the tensorflow model handler in tfx_bsl which has some constraints listed in #23584. Implement a new model handler with support for more types like numpy, tensors, etc.

• Load an entire saved model
• Load model with saved weights

Issue Priority

Priority: 2 (default / most feature requests should be filed as P2)

Issue Components

• Component: Python SDK
• Component: Java SDK
• Component: Go SDK
• Component: Typescript SDK
• Component: IO connector
• 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