diff --git a/modelcenter/PLSC-ViT/APP/app.py b/modelcenter/PLSC-ViT/APP/app.py
index 89e38d59d07cd94de59629dbc631f35df8cd8bbb..7dc1791b13b83edfb30f29f7aaad69cdcbe6fde4 100644
--- a/modelcenter/PLSC-ViT/APP/app.py
+++ b/modelcenter/PLSC-ViT/APP/app.py
@@ -15,8 +15,12 @@ def model_inference(image):
             model_path=model_path,
             params_path=params_path,
             label_path=label_path)
-    scores, labels = predictor.predict(image)
-    json_out = {"scores": scores.tolist(), "labels": labels.tolist()}
+    class_ids, scores, labels = predictor.predict(image)
+    json_out = {
+        "class_ids": class_ids.tolist(),
+        "scores": scores.tolist(),
+        "labels": labels.tolist()
+    }
     return image, json_out
 
 
diff --git a/modelcenter/PLSC-ViT/APP/predictor.py b/modelcenter/PLSC-ViT/APP/predictor.py
index 46fd81a799f4790c5ddbf99fb4ec29c1d4d5f7b7..6d3782f86d06eb19dc139352dc8a987d03df8476 100644
--- a/modelcenter/PLSC-ViT/APP/predictor.py
+++ b/modelcenter/PLSC-ViT/APP/predictor.py
@@ -4,6 +4,7 @@ import numpy as np
 import paddle
 from download import get_model_path, get_data_path
 
+
 class Predictor(object):
     def __init__(self,
                  model_type="paddle",
@@ -42,8 +43,8 @@ class Predictor(object):
         self.predictor.run()
         outputs = []
         for output_idx in range(len(self.output_names)):
-            output_tensor = self.predictor.get_output_handle(
-                self.output_names[output_idx])
+            output_tensor = self.predictor.get_output_handle(self.output_names[
+                output_idx])
             outputs.append(output_tensor.copy_to_cpu())
         if self.postprocess is not None:
             output_data = self.postprocess(outputs)
@@ -69,7 +70,7 @@ class Predictor(object):
             for line in f:
                 if len(line) < 2:
                     continue
-                label = line.strip().split(',')[0].split(' ')[2]
+                label = line.strip().split(',')[1].strip()
                 labels.append(label)
         return labels
 
@@ -83,6 +84,5 @@ class Predictor(object):
         pred = np.array(logits).squeeze()
         pred = self.softmax(pred)
         class_idx = pred.argsort()[::-1]
-        return pred[class_idx[:5]], np.array(self.labels)[class_idx[:5]]
-
-
+        return class_idx[:5], pred[class_idx[:5]], np.array(self.labels)[
+            class_idx[:5]]