Add /estimationprocedure/list (#121)

* Add /estimationprocedure/list

* [no ci] Add changes for 'Others'

specifically, estimationprocedure/list and evaluationmeasure/list

docs/migration.md

@@ -90,3 +90,13 @@ includes datasets which are private.
 
 The `limit` and `offset` parameters can now be used independently, you no longer need
 to provide both if you wish to set only one.
+
+## Others
+
+### `GET /estimationprocedure/list`
+The `ttid` field has been renamed to `task_type_id`.
+All values are now typed.
+Outer levels of nesting have been removed.
+
+#### `GET /evaluationmeasures/list`
+Outer levels of nesting have been removed.

src/core/formatting.py

@@ -6,6 +6,15 @@
 from core.errors import DatasetError
 
 
+def _str_to_bool(string: str) -> bool:
+    if string.casefold() in ["true", "1", "yes"]:
+        return True
+    if string.casefold() in ["false", "0", "no"]:
+        return False
+    msg = f"Could not parse {string=} as bool."
+    raise ValueError(msg)
+
+
 def _format_error(*, code: DatasetError, message: str) -> dict[str, str]:
     """Formatter for JSON bodies of OpenML error codes."""
     return {"code": str(code), "message": message}

src/database/evaluations.py

@@ -1,5 +1,7 @@
-from typing import Any
+from typing import Any, Iterable
 
+from core.formatting import _str_to_bool
+from schemas.datasets.openml import EstimationProcedure
 from sqlalchemy import Connection, CursorResult, text
 
 
@@ -14,3 +16,23 @@ def get_math_functions(function_type: str, connection: Connection) -> CursorResu
         ),
         parameters={"function_type": function_type},
     )
+
+
+def get_estimation_procedures(connection: Connection) -> Iterable[EstimationProcedure]:
+    rows = connection.execute(
+        text(
+            """
+            SELECT `id` as 'id_', `ttid` as 'task_type_id', `name`, `type` as 'type_',
+                   `repeats`, `folds`, `stratified_sampling`, `percentage`
+            FROM estimation_procedure
+            """,
+        ),
+    )
+    typed_rows = [
+        {
+            k: v if k != "stratified_sampling" or v is None else _str_to_bool(v)
+            for k, v in row.items()
+        }
+        for row in rows.mappings()
+    ]
+    return [EstimationProcedure(**typed_row) for typed_row in typed_rows]

src/main.py

@@ -4,6 +4,7 @@
 from fastapi import FastAPI
 from routers.mldcat_ap.dataset import router as mldcat_ap_router
 from routers.openml.datasets import router as datasets_router
+from routers.openml.estimation_procedure import router as estimationprocedure_router
 from routers.openml.evaluations import router as evaluationmeasures_router
 from routers.openml.qualities import router as qualities_router
 from routers.openml.tasktype import router as ttype_router
@@ -43,6 +44,7 @@ def create_api() -> FastAPI:
     app.include_router(mldcat_ap_router)
     app.include_router(ttype_router)
     app.include_router(evaluationmeasures_router)
+    app.include_router(estimationprocedure_router)
     return app
 
 

src/routers/openml/estimation_procedure.py

@@ -0,0 +1,17 @@
+from typing import Annotated, Iterable
+
+from database.evaluations import get_estimation_procedures as db_get_estimation_procedures
+from fastapi import APIRouter, Depends
+from schemas.datasets.openml import EstimationProcedure
+from sqlalchemy import Connection
+
+from routers.dependencies import expdb_connection
+
+router = APIRouter(prefix="/estimationprocedure", tags=["estimationprocedure"])
+
+
+@router.get("/list", response_model_exclude_none=True)
+def get_estimation_procedures(
+    expdb: Annotated[Connection, Depends(expdb_connection)],
+) -> Iterable[EstimationProcedure]:
+    return db_get_estimation_procedures(expdb)

src/schemas/datasets/openml.py

@@ -46,6 +46,17 @@ class Feature(BaseModel):
     nominal_values: list[str] | None
 
 
+class EstimationProcedure(BaseModel):
+    id_: int = Field(serialization_alias="id")
+    task_type_id: int
+    name: str
+    type_: str = Field(serialization_alias="type")
+    percentage: int | None
+    repeats: int | None
+    folds: int | None
+    stratified_sampling: bool | None
+
+
 class DatasetMetadata(BaseModel):
     id_: int = Field(json_schema_extra={"example": 1}, alias="id")
     visibility: Visibility = Field(json_schema_extra={"example": Visibility.PUBLIC})