diff --git a/image_ref/config.py b/image_ref/config.py
index 3676bcdc2f4702c4d0578e750798b8f69a7a4662..3f96b9eb85e5be7b488e249318e3a753377e08a5 100644
--- a/image_ref/config.py
+++ b/image_ref/config.py
@@ -4,7 +4,7 @@ import argparse
 def load_args_contrastive():
     parser = argparse.ArgumentParser()
 
-    parser.add_argument('--epoches', type=int, default=3)
+    parser.add_argument('--epoches', type=int, default=1)
     parser.add_argument('--save_inter', type=int, default=50)
     parser.add_argument('--eval_inter', type=int, default=1)
     parser.add_argument('--noise_threshold', type=int, default=0)
diff --git a/image_ref/main.py b/image_ref/main.py
index b9b861112b4e15bc4102cc3cf8e5dfdcc439eaf5..3d983bc765b526e5298f9d7b5aaaa26157f664b2 100644
--- a/image_ref/main.py
+++ b/image_ref/main.py
@@ -160,18 +160,12 @@ def make_prediction_duo(model, data, f_name, f_name2):
             img_ref = img_ref.cuda()
             label = label.cuda()
         output = model(imaer,imana,img_ref)
-        print(output, label)
         confidence_pred_list[specie].append(output[:,0].data.cpu().numpy())
         #Mono class output (only most postive paire)
         output = torch.argmax(output[:,0])
         label = torch.argmin(label)
-
-        print(output, label)
-        y_pred.extend(output)
-
-        label = torch.argmin(label)
-        print(output, label)
-        y_true.extend(label)  # Save Truth
+        y_pred.append(output.tolist())
+        y_true.append(label.tolist())  # Save Truth
     # constant for classes
 
     # Build confusion matrix