Update README.md

README.md

@@ -16,7 +16,7 @@ metrics:
 
 Examples with `noAnswer` and `badQuestion` are not included in the training process.
 
-```shell
+```bash
 $ cd ~/projects/transformers/examples/legacy/question-answering
 $ mkdir bert_base_uncased_finetuned_newsqa
 $ python run_newsqa.py \
@@ -39,7 +39,7 @@ $ python run_newsqa.py \
 
 Results:
 
-```shell
+```bash
 {'exact': 60.19350380096752, 'f1': 73.29371985128037, 'total': 4341, 'HasAns_exact': 60.19350380096752, 'HasAns_f1': 73.29371985128037, 'HasAns_total': 4341, 'best_exact': 60.19350380096752, 'best_exact_thresh': 0.0, 'best_f1': 73.29371985128037, 'best_f1_thresh': 0.0}
 ```