Logs

app.py

@@ -80,12 +80,10 @@ print(raw_dataset.column_names)
 # --> if tokens are inside a word, replace 'B-' with 'I-' 
 def align_labels_with_tokens(labels):
     aligned_label_ids = []
-    aligned_label_ids.append(-100)
     for i, label in enumerate(labels):
         if label.startswith("B-"):
             label = label.replace("B-", "I-")
         aligned_label_ids.append(label2id[label])
-    aligned_label_ids.append(-100)
     return aligned_label_ids
 
 # create tokenize function
@@ -151,10 +149,8 @@ def evaluate_model(model, dataloader, device):
     print("evaluate_model sizes")
     print(len(all_preds))
     print(len(all_labels))
-    for i in range(min(100, len(all_labels))):
-        print("Pred vs Label")
-        print(id2label[all_preds[i]])
-        print(id2label[all_labels[i]])
+    print(id2label[all_preds[0]])
+    print(id2label[all_labels[0]])
     all_preds = np.asarray(all_preds, dtype=np.float32)
     all_labels = np.asarray(all_labels, dtype=np.float32)
     accuracy = accuracy_score(all_labels, all_preds)