Update model_loader.py

model_loader.py

@@ -193,8 +193,9 @@ def load_skin_models():
 #  DIABETES — Keras ANN
 # ══════════════════════════════════════════════════════════════════════════════
 
-DIABETES_MODEL_PATH  = "/app/diabetes_model.h5"
-DIABETES_SCALER_PATH = "/app/diabetes_scaler.json"
+DIABETES_REPO         = "SaswatML123/DiabetesModel"
+DIABETES_MODEL_FILE   = "diabetes_model.h5"
+DIABETES_SCALER_FILE  = "diabetes_scaler.json"
 DIABETES_FEATURE_COLS = [
     "Pregnancies", "Glucose", "BloodPressure", "SkinThickness",
     "Insulin", "BMI", "DiabetesPedigreeFunction", "Age"
@@ -203,6 +204,30 @@ _diabetes_model  = None
 _diabetes_scaler = None
 
 
+def _build_diabetes_model():
+    import tensorflow as tf
+    from tensorflow.keras import layers
+    model = tf.keras.Sequential([
+        layers.Input(shape=(8,)),
+        layers.Dense(256),
+        layers.BatchNormalization(),
+        layers.Activation("relu"),
+        layers.Dropout(0.3),
+        layers.Dense(128),
+        layers.BatchNormalization(),
+        layers.Activation("relu"),
+        layers.Dropout(0.3),
+        layers.Dense(64),
+        layers.BatchNormalization(),
+        layers.Activation("relu"),
+        layers.Dropout(0.2),
+        layers.Dense(32),
+        layers.Activation("relu"),
+        layers.Dense(1, activation="sigmoid")
+    ])
+    return model
+
+
 def load_diabetes_model():
     global _diabetes_model, _diabetes_scaler
     if _diabetes_model is not None:
@@ -211,13 +236,12 @@ def load_diabetes_model():
     import tensorflow as tf
 
     print("[Diabetes] Loading model...")
-    _diabetes_model = tf.keras.models.load_model(
-        DIABETES_MODEL_PATH,
-        compile=False,
-        custom_objects={"quantization_config": None}
-    )
+    model_path  = _download(DIABETES_REPO, DIABETES_MODEL_FILE)
+    scaler_path = _download(DIABETES_REPO, DIABETES_SCALER_FILE)
+    _diabetes_model = _build_diabetes_model()
+    _diabetes_model.load_weights(model_path)
 
-    with open(DIABETES_SCALER_PATH) as f:
+    with open(scaler_path) as f:
         params = json.load(f)
     _diabetes_scaler = {
         "mean":  np.array(params["mean"]),