The class_labels should be ground truth data

Based on pytorch docs, the parameter "labels" of add_pr_curve() should be
ground truth data. But in Line 141, class_labels append our own
predictions. Therefore, I substitute "labels" for "predicted".

Author: gyes00205

16_tensorboard.py

@@ -138,7 +138,7 @@ def forward(self, x):
         class_probs_batch = [F.softmax(output, dim=0) for output in outputs]
 
         class_preds.append(class_probs_batch)
-        class_labels.append(predicted)
+        class_labels.append(labels)
 
     # 10000, 10, and 10000, 1
     # stack concatenates tensors along a new dimension