diff --git a/scripts/ML/BERT/Trainer.py b/scripts/ML/BERT/Trainer.py
index 06851eef22cfce4f575beaf358abde39d713e53d..70c2a8f4b75b068193e12ed343ae6e7e4cc8b580 100644
--- a/scripts/ML/BERT/Trainer.py
+++ b/scripts/ML/BERT/Trainer.py
@@ -3,7 +3,8 @@ import datetime
 from loaders import set_random
 import time
 import torch
-from transformers import AdamW, get_linear_schedule_with_warmup
+from torch.optim import AdamW
+from transformers import get_linear_schedule_with_warmup
 
 def chrono(f):
     def wrapped(*args, **kwargs):