Update app.py

app.py

@@ -99,7 +99,7 @@ def predictmic(sequence):
     import torch
     from transformers import BertTokenizer, BertModel
     import numpy as np
-    import pickle
+    import joblib  # ✅ Use joblib instead of pickle
     from math import expm1
 
     # === Load ProtBert model ===
@@ -142,7 +142,7 @@ def predictmic(sequence):
         "K.Pneumonia": {
             "model": "pne_mlp_model.pkl",
             "scaler": "pne_scaler.pkl",
-            "pca": "pne_pca"
+            "pca": "pne_pca"  # Ensure this PCA file was also saved using joblib
         }
     }
 
@@ -151,21 +151,18 @@ def predictmic(sequence):
     for bacterium, cfg in bacteria_config.items():
         try:
             # === Load scaler and transform ===
-            with open(cfg["scaler"], "rb") as f:
-                scaler = pickle.load(f)
+            scaler = joblib.load(cfg["scaler"])
             scaled = scaler.transform(embedding)
 
             # === Apply PCA if exists ===
             if cfg["pca"] is not None:
-                with open(cfg["pca"], "rb") as f:
-                    pca = pickle.load(f)
+                pca = joblib.load(cfg["pca"])
                 transformed = pca.transform(scaled)
             else:
                 transformed = scaled
 
             # === Load model and predict ===
-            with open(cfg["model"], "rb") as f:
-                mic_model = pickle.load(f)
+            mic_model = joblib.load(cfg["model"])
             mic_log = mic_model.predict(transformed)[0]
             mic = round(expm1(mic_log), 3)  # Inverse of log1p used in training