Day 5 (04:00): revert session changes (could not fix build)

Author: Test User

src/backend/api/chat.py

@@ -38,7 +38,6 @@
 # Helpers
 # ---------------------------------------------------------------------------
 
-
 def _load_dataset(dataset_id: str, session: Session) -> tuple[Dataset, pd.DataFrame]:
     dataset = session.get(Dataset, dataset_id)
     if not dataset:
@@ -54,7 +53,6 @@ def _load_dataset(dataset_id: str, session: Session) -> tuple[Dataset, pd.DataFr
 # 1. Suggestions
 # ---------------------------------------------------------------------------
 
-
 @router.get("/{dataset_id}/suggestions")
 def get_suggestions(dataset_id: str, session: Session = Depends(get_session)):
     """Return AI-generated feature transformation suggestions for a dataset."""
@@ -82,7 +80,6 @@ def get_suggestions(dataset_id: str, session: Session = Depends(get_session)):
 # 2. Apply transformations
 # ---------------------------------------------------------------------------
 
-
 class ApplyRequest(BaseModel):
     transformations: list[dict]  # list of {column, transform_type, params?}
 
@@ -120,7 +117,9 @@ def apply_feature_transforms(
     session.refresh(feature_set)
 
     preview = transformed_df.head(5).to_dict(orient="records")
-    new_columns = sorted(set(transformed_df.columns) - set(df.columns))
+    new_columns = sorted(
+        set(transformed_df.columns) - set(df.columns)
+    )
 
     return {
         "feature_set_id": feature_set.id,
@@ -135,7 +134,6 @@ def apply_feature_transforms(
 # 3. Preview a feature set
 # ---------------------------------------------------------------------------
 
-
 @router.get("/{feature_set_id}/preview")
 def preview_feature_set(feature_set_id: str, session: Session = Depends(get_session)):
     """Return a preview of the transformed dataset for a given FeatureSet."""
@@ -162,7 +160,6 @@ def preview_feature_set(feature_set_id: str, session: Session = Depends(get_sess
 # 4. Set target variable
 # ---------------------------------------------------------------------------
 
-
 class TargetRequest(BaseModel):
     target_column: str
     feature_set_id: str | None = None  # update an existing FeatureSet if provided
@@ -211,7 +208,6 @@ def set_target(
 # 6. List pipeline steps for a FeatureSet
 # ---------------------------------------------------------------------------
 
-
 @router.get("/{feature_set_id}/steps")
 def list_pipeline_steps(feature_set_id: str, session: Session = Depends(get_session)):
     """Return the ordered list of transformation steps in the pipeline."""
@@ -230,7 +226,6 @@ def list_pipeline_steps(feature_set_id: str, session: Session = Depends(get_sess
 # 7. Append a single step to the pipeline
 # ---------------------------------------------------------------------------
 
-
 class AddStepRequest(BaseModel):
     column: str
     transform_type: str
@@ -264,7 +259,6 @@ def add_pipeline_step(
     df = pd.read_csv(file_path)
 
     from core.feature_engine import apply_transformations
-
     transformed_df, column_mapping = apply_transformations(df, steps)
 
     feature_set.transformations = json.dumps(steps)
@@ -287,7 +281,6 @@ def add_pipeline_step(
 # 8. Remove (undo) a step by index
 # ---------------------------------------------------------------------------
 
-
 @router.delete("/{feature_set_id}/steps/{step_index}", status_code=200)
 def remove_pipeline_step(
     feature_set_id: str,
@@ -318,7 +311,6 @@ def remove_pipeline_step(
     df = pd.read_csv(file_path)
 
     from core.feature_engine import apply_transformations
-
     transformed_df, column_mapping = apply_transformations(df, steps)
 
     feature_set.transformations = json.dumps(steps)
@@ -341,7 +333,6 @@ def remove_pipeline_step(
 # 9. Feature importance
 # ---------------------------------------------------------------------------
 
-
 @router.get("/{dataset_id}/importance")
 def get_feature_importance(
     dataset_id: str,
@@ -355,9 +346,7 @@ def get_feature_importance(
     type_result = detect_problem_type(df, target_column)
     problem_type = type_result.get("problem_type", "regression")
 
-    importance = compute_feature_importance(
-        df, target_column, problem_type, column_stats
-    )
+    importance = compute_feature_importance(df, target_column, problem_type, column_stats)
 
     return {
         "dataset_id": dataset_id,