Update README.md

README.md

@@ -35,6 +35,32 @@ The model was trained for 5 epochs with the following training parameters:
 
 The training loss consistently decreased, indicating successful learning.
 
+## Use case: 
+```python
+  from transformers import BartTokenizer, BartForConditionalGeneration
+  
+  # Load the trained model
+  model = BartForConditionalGeneration.from_pretrained("NepaliAI/NFT-6.9k")
+  
+  # Load the tokenizer for generating new output
+  tokenizer = BartTokenizer.from_pretrained("NepaliAI/NFT-6.9k")
+  
+  # Example text
+  input_text = "के म मेरो महिनावारीको समयमा स्ट्रेप थ्रोटको लागि डाक्टरले तोकेको औषधि लिन सक्छु?"
+  
+  # Tokenize the input
+  inputs = tokenizer(input_text, return_tensors="pt", max_length=128, truncation=True)
+  
+  # Generate text
+  generated_text = model.generate(**inputs, max_length=256, top_p=0.95, top_k=50, do_sample=True, temperature=0.7, num_return_sequences=1, no_repeat_ngram_size=2)
+  
+  # Decode the generated text
+  generated_response = tokenizer.batch_decode(generated_text, skip_special_tokens=True)[0]
+  
+  # Print the generated response
+  print("Generated Response:", generated_response)
+
+```
 ## Evaluation
 
 ### Metrics