wip api

llama_stack/providers/impls/meta_reference/evals/__init__.py

@@ -0,0 +1,19 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from .config import MetaReferenceEvalsImplConfig  # noqa
+from llama_stack.apis.inference import *  # noqa: F403
+from llama_stack.distribution.datatypes import Api, ProviderSpec
+
+
+async def get_provider_impl(
+    config: MetaReferenceEvalsImplConfig, deps: Dict[Api, ProviderSpec]
+):
+    from .evals import MetaReferenceEvalsImpl
+
+    impl = MetaReferenceEvalsImpl(config, deps[Api.inference])
+    await impl.initialize()
+    return impl

llama_stack/providers/impls/meta_reference/evals/config.py

@@ -0,0 +1,10 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from pydantic import BaseModel
+
+
+class MetaReferenceEvalsImplConfig(BaseModel): ...

llama_stack/providers/impls/meta_reference/evals/evals.py

@@ -0,0 +1,34 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from llama_stack.apis.inference import *  # noqa: F403
+from llama_stack.apis.evals import *  # noqa: F403
+
+from .config import MetaReferenceEvalsImplConfig
+
+
+class MetaReferenceEvalsImpl(Evals):
+    def __init__(self, config: MetaReferenceEvalsImplConfig, inference_api: Inference):
+        self.inference_api = inference_api
+
+    async def initialize(self) -> None:
+        pass
+
+    async def shutdown(self) -> None:
+        pass
+
+    async def run_evals(
+        self,
+        model: str,
+        dataset: str,
+        task: str,
+    ) -> EvaluateResponse:
+        print("hi")
+        return EvaluateResponse(
+            metrics={
+                "accuracy": 0.5,
+            }
+        )