[Chore] refactor gbm (#37)

rektgbm/gbm.py

@@ -49,10 +49,8 @@ def fit(
         )
         if self._task_type in {TaskType.binary, TaskType.multiclass, TaskType.rank}:
             self.label_encoder = dataset.fit_transform_label()
-            self._is_label_encoder_used = True
-
-        if valid_set is not None and self.__is_label_encoder_used:
-            valid_set.transform_label(label_encoder=self.label_encoder)
+            if valid_set:
+                valid_set.transform_label(label_encoder=self.label_encoder)
 
         _objective = self.rekt_objective.get_objective_dict(method=self.method)
         _metric = self.rekt_metric.get_metric_dict(method=self.method)
@@ -69,16 +67,8 @@ def predict(self, dataset: RektDataset):
         if self._task_type in {TaskType.binary, TaskType.regression, TaskType.rank}:
             return preds
 
-        if self._task_type == TaskType.multiclass:
-            if self.method == MethodName.lightgbm:
-                preds = np.argmax(preds, axis=1).astype(int)
-            else:
-                preds = np.around(preds).astype(int)
-
-        if self.__is_label_encoder_used:
-            preds = self.label_encoder.inverse_transform(series=preds)
-        return preds
-
-    @property
-    def __is_label_encoder_used(self) -> bool:
-        return getattr(self, "_is_label_encoder_used", False)
+        if self.method == MethodName.lightgbm:
+            preds = np.argmax(preds, axis=1).astype(int)
+        else:
+            preds = np.around(preds).astype(int)
+        return self.label_encoder.inverse_transform(series=preds)

tests/test_gbm.py

@@ -5,6 +5,7 @@
 
 from rektgbm.base import MethodName
 from rektgbm.dataset import RektDataset
+from rektgbm.encoder import RektLabelEncoder
 from rektgbm.engine import RektEngine
 from rektgbm.gbm import RektGBM
 from rektgbm.task import TaskType
@@ -90,6 +91,8 @@ def test_rektgbm_predict_multiclass(mock_dataset, mock_engine):
     gbm.engine = mock_engine
     gbm._task_type = TaskType.multiclass
     gbm._is_fitted = True
+    gbm.label_encoder = RektLabelEncoder()
+    gbm.label_encoder.fit_label([0, 1, 2])
 
     mock_engine.predict.return_value = np.array(
         [[0.1, 0.7, 0.2], [0.3, 0.4, 0.3], [0.2, 0.2, 0.6]]