Drop training argument from fit_class_random_forest

per Open-EO/openeo-processes#354

openeo_driver/ProcessGraphDeserializer.py

@@ -734,12 +734,6 @@ def fit_class_random_forest(args: dict, env: EvalEnv) -> DriverMlModel:
 
     # TODO: get defaults from process spec?
     # TODO: do parameter checks automatically based on process spec?
-    training = extract_arg(args, 'training')
-    if not isinstance(training, float) or training < 0.0 or training > 1.0:
-        raise ProcessParameterInvalidException(
-            parameter="training", process="fit_class_random_forest",
-            reason="should be a float between 0 and 1."
-        )
     num_trees = args.get("num_trees", 100)
     if not isinstance(num_trees, int) or num_trees < 0:
         raise ProcessParameterInvalidException(
@@ -760,7 +754,7 @@ def fit_class_random_forest(args: dict, env: EvalEnv) -> DriverMlModel:
         )
 
     return predictors.fit_class_random_forest(
-        target=target, training=training,
+        target=target,
         num_trees=num_trees, mtry=mtry, seed=seed,
     )
 

openeo_driver/_version.py

@@ -1 +1 @@
-__version__ = '0.17.5a1'
+__version__ = '0.17.6a1'

openeo_driver/dummy/dummy_backend.py

@@ -252,13 +252,12 @@ def prepare_for_json(self):
         return self.data
 
     def fit_class_random_forest(
-            self, target: dict,
-            training: float, num_trees: int, mtry: Optional[int] = None, seed: Optional[int] = None
+            self, target: dict, num_trees: int, mtry: Optional[int] = None, seed: Optional[int] = None
     ) -> DriverMlModel:
         # Fake ML training: just store inputs
         return DummyMlModel(
             process_id="fit_class_random_forest",
-            data=self.data, target=target, training=training, num_trees=num_trees, mtry=mtry, seed=seed,
+            data=self.data, target=target, num_trees=num_trees, mtry=mtry, seed=seed,
         )
 
 

openeo_driver/save_result.py

@@ -532,8 +532,7 @@ def write_assets(self, directory: str) -> Dict[str, StacAsset]:
         return {str(Path(filename).name): asset}
 
     def fit_class_random_forest(
-            self, target: dict,
-            training: float, num_trees: int, mtry: Optional[int] = None, seed: Optional[int] = None
+            self, target: dict, num_trees: int, mtry: Optional[int] = None, seed: Optional[int] = None
     ) -> DriverMlModel:
         # TODO: this method belongs eventually under DriverVectorCube
         raise NotImplementedError

tests/test_views_execute.py

@@ -2378,7 +2378,6 @@ def test_fit_class_random_forest(api100):
             "process_id": "fit_class_random_forest",
             "data": [[100.0, 100.1, 100.2, 100.3], [101.0, 101.1, 101.2, 101.3]],
             "target": DictSubSet({"type": "FeatureCollection"}),
-            "training": 0.5,
             "num_trees": 200,
             "mtry": None,
             "seed": None,