Update main.py

main.py

@@ -45,6 +45,10 @@ async def predict(input: InputFeatures):
 
 
     full_pipeline=ColumnTransformer([('num_pipe',num_pipeline,num_attr)])
+        full_pipeline.fit(df)  # Fit the full pipeline on the DataFrame
+
+    final_input = np.array(full_pipeline.transform(df), dtype=np.str)
+        prediction = model_input.predict(final_input)
 
     XGB = Pipeline([
         ("col_trans", full_pipeline),  # You need to define full_pipeline
@@ -52,9 +56,19 @@ async def predict(input: InputFeatures):
         ("model", BaggingClassifier(base_estimator=XGBClassifier(random_state=42)))
     ])
 
-    df = pd.DataFrame([input])
-    final_input = np.array(full_pipeline.fit_transform(df), dtype=np.str)  # Check predict_input, maybe it should be XGB
-    prediction = model_input.predict(np.array([final_input]).reshape(1, -1))
+    # df = pd.DataFrame([input])
+    # final_input = np.array(full_pipeline.fit_transform(df), dtype=np.str)  # Check predict_input, maybe it should be XGB
+    # #prediction = model_input.predict(np.array([final_input]).reshape(1, -1))
+    # prediction = model_input.predict(final_input)
+
+    # full_pipeline = ColumnTransformer([('num_pipe', num_pipeline, num_attr)])
+    #     full_pipeline.fit(df)  # Fit the full pipeline on the DataFrame
+
+    #     #final_input = np.array(full_pipeline.transform(df), dtype=np.str)
+
+    
+        
+
 
     return prediction