Update app.py with transformer embeddings and prediction pipeline

app.py

@@ -75,10 +75,13 @@ def predict_with_gpflow(model, X):
     # Get predictions
     #predict_fn = model.predict_f_compiled
     predict_fn = model.signatures["serving_default"]
-    mean, variance = predict_fn(Xnew=X_tensor)
+    #mean, variance = predict_fn(Xnew=X_tensor)
+    mean = result["output_0"].numpy()  # Adjust output key names if needed
+    variance = result["output_1"].numpy()
 
     # Return mean and variance as numpy arrays
-    return mean.numpy().flatten(), variance.numpy().flatten()
+    #return mean.numpy().flatten(), variance.numpy().flatten()
+    return mean.flatten(), variance.flatten()