Update app.py with transformer embeddings and prediction pipeline

app.py

@@ -71,10 +71,10 @@ def predict_with_gpflow(model, X):
     print(model.signatures)
     # Convert input to TensorFlow tensor
     X_tensor = tf.convert_to_tensor(X, dtype=tf.float64)
-
+    print (X_tensor.shape)
     # Get predictions
     predict_fn = model.predict_f_compiled
-    mean, variance = predict_fn(X_tensor)
+    mean, variance = predict_fn(X_new=X_tensor)
 
     # Return mean and variance as numpy arrays
     return mean.numpy().flatten(), variance.numpy().flatten()
@@ -99,6 +99,9 @@ def process_target(target, selected_models, sequence, prediction_type):
         return target, round(y_pred, 2)
     else:
         # GPflow prediction
+        print (esm_model_name)
+        print (layer)
+        print (model)
         y_pred, y_uncertainty = predict_with_gpflow(model, embedding)
         return target, round(y_pred[0], 2), round(y_uncertainty[0], 2)