add new models

app/main.py

@@ -182,10 +182,10 @@ async def get_answer(item: Item, token: str = Depends(verify_token)):
         raise HTTPException(status_code=500, detail=str(e))
 
 # naive bayes
-model = joblib.load('app/1713630229.4965415_trained_model.joblib')
+model = joblib.load('app/1713638595.315147_trained_model')
 pca = joblib.load('app/pca.pkl')
-scaler = joblib.load('app/scaler.pkl')
-label_encoder = joblib.load('app/label_encoder.pkl')
+scaler = joblib.load('app/1713638595.3178492_scaler.joblib')
+label_encoder = joblib.load('app/1713638744.044928_label_encoder.joblib')
 
 # def preprocess_audio(path, save_dir):
 #     y, sr = librosa.load(path)
@@ -332,23 +332,23 @@ async def handle_audio(file: UploadFile = File(...)):
         # features = extract_features(temp_filename)
         print("Extracted Features:", features)
 
-        # features = scaler.transform(features)
+        features = scaler.transform(features)
         # features = pca.transform(features)
         features = np.array(features).reshape(1, -1)
 
         # proceed with an inference
         results = model.predict(features)
-        print(results)
         # decoded_predictions = [label_encoder.classes_[i] for i in results]
 
         # # Decode the predictions using the label encoder
-        # decoded_predictions = label_encoder.inverse_transform(results)
+        decoded_predictions = label_encoder.inverse_transform(results)
+        print(decoded_predictions)
         # .tolist()
         # Clean up the temporary file
         os.remove(temp_filename)
 
         # Return a successful response with decoded predictions
-        return {"message": "File processed successfully", "prediction": results}
+        return {"message": "File processed successfully", "prediction": decoded_predictions}
     except Exception as e:
         print(e)
         # Handle possible exceptions