Update README.md

README.md

@@ -39,13 +39,15 @@ language:
 
 # Model Card for T5-LM-Large_Canard-HotpotQA-rephrase 
 This model is trained on three objectives: 
-(1) Generating answers for Canard dataset based on Wikipedia search results 
-(2) Generating answers for HotpotQA, 
-(3) Rephrasing questions by the conversation context.
+  1. Generating answers for Canard dataset based on Wikipedia search results 
+  2. Generating answers for HotpotQA, 
+  3. Rephrasing questions by the conversation context.
 
 ## Training
-The model was trained using the following script, exported from the corresponding Jupyter notebook. All details, including the request format, can be inferred without errors from the code.
+The model was trained using the following script, which can be copy-pasted and run as-is (with the installed `requirements.txt`). 
+All details, including the request format, can be inferred without errors from the code.
 The best checkpoint was picked by a maximum ROUGE on Canard conversational QA's ROUGE.
+
 ```python
 import datasets