Enable sane naming of registered objects with defaults (#429)

# What does this PR do? 

This is a follow-up to #425. That PR allows for specifying models in the
registry, but each entry needs to look like:

```yaml
- identifier: ...
  provider_id: ...
  provider_resource_identifier: ...
```

This is headache-inducing.

The current PR makes this situation better by adopting the shape of our
APIs. Namely, we need the user to only specify `model-id`. The rest
should be optional and figured out by the Stack. You can always override
it.

Here's what example `ollama` "full stack" registry looks like (we still
need to kill or simplify shield_type crap):
```yaml
models:
- model_id: Llama3.2-3B-Instruct
- model_id: Llama-Guard-3-1B
shields:
- shield_id: llama_guard
  shield_type: llama_guard
```

## Test Plan

See test plan for #425. Re-ran it.
This commit is contained in:
Ashwin Bharambe 2024-11-12 11:18:05 -08:00 committed by GitHub
parent d9d271a684
commit 09269e2a44
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
17 changed files with 295 additions and 207 deletions

View file

@ -10,15 +10,13 @@ from llama_models.llama3.api.datatypes import URL
from llama_models.schema_utils import json_schema_type, webmethod
from pydantic import Field
from pydantic import BaseModel, Field
from llama_stack.apis.common.type_system import ParamType
from llama_stack.apis.resource import Resource
from llama_stack.apis.resource import Resource, ResourceType
@json_schema_type
class Dataset(Resource):
type: Literal["dataset"] = "dataset"
class CommonDatasetFields(BaseModel):
schema: Dict[str, ParamType]
url: URL
metadata: Dict[str, Any] = Field(
@ -27,6 +25,26 @@ class Dataset(Resource):
)
@json_schema_type
class Dataset(CommonDatasetFields, Resource):
type: Literal[ResourceType.dataset.value] = ResourceType.dataset.value
@property
def dataset_id(self) -> str:
return self.identifier
@property
def provider_dataset_id(self) -> str:
return self.provider_resource_id
@json_schema_type
class DatasetInput(CommonDatasetFields, BaseModel):
dataset_id: str
provider_id: Optional[str] = None
provider_dataset_id: Optional[str] = None
class Datasets(Protocol):
@webmethod(route="/datasets/register", method="POST")
async def register_dataset(