Update README.md

README.md

@@ -1,6 +1,6 @@
 ---
 library_name: transformers
-tags: []
+tags: [spell_correction]
 ---
 
 # Model Card for Model ID
@@ -13,24 +13,39 @@ Persian spelling correction model based on LLama3.1 instruct. This model was tra
 ### Model Description
 
 Base_prompt = """You are tasked with correcting spelling mistakes in the queries that users submitted to a Persian marketplace.
+
 Output the corrected query in the following JSON format:
+
 - If the input requires correction, use:
+  
   {"correction": "<corrected version of the query>"}
+  
 - If the input is correct, use:
+
+  
   "correction": ""}
+
+  
 Here are some examples:
+
 "query": "ندل چسبی زنانه" Your answer: {"correction": "صندل چسبی زنانه"}
+
 "query": "بادکنک جشن تواد"  Your answer: {"correction": "بادکنک جشن تولد"}
+
 "query": "صندلی بادی"  Your answer: {"correction": ""}\n"""
 
+
+
 ## Uses
 
 It should be used for spelling correction in a setting with Persian language around.
+
 ### Direct Use
 
+```python 
 //output structring
 def extract_json(text):
-    try:
+    try: 
         correction = None
         pos = 0
         decoder = json.JSONDecoder()
@@ -76,10 +91,12 @@ with torch.no_grad():
                                             )
     output = tokenizer.decode(output_tokens[0][prompt_length:], skip_special_tokens=True)
     return extract_json(output)
-
+```
 
 ## Model Card Contact
 
 Majid F. Sadi
+
 [email protected]
+
 https://www.linkedin.com/in/mfsadi/