Add pydantic model schema

[timothyjlaurent]

Is your feature request related to a problem? Please describe.

I want to have robust input and output shema validation. Similar to what is achieved with FastAPI https://fastapi.tiangolo.com/.

Describe the solution you'd like

Add support to the JsonInput to:

1. Accept lists of json items natively - allows for hitting such endpoints with a list, instead of needing to go through the pandas input. Allow the microbatcher to concat the lists of requests.
2. accept a schema parameter that can be a type definition or a pydantic model https://pydantic-docs.helpmanual.io/usage/models/ . This will be used to generate the response in the openapi schema https://pydantic-docs.helpmanual.io/usage/schema/
3. Allow for some way to turn off validation, as I'm told this is being done for custom FastAPI KFServing apps to be able to handle high load

Describe alternatives you've considered

Additional context

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[parano]

@larme would be great for you to look into this issue

[aarnphm]

BentoML/bentoml/_internal/io_descriptors/json.py

Line 133 in 9899d58

pydantic_model: t.Optional["pydantic.BaseModel"] = None,

will provide support for pydantic model schema in the upcoming 1.0 releases. Let me know if we need to reopen this issue. Closing for now