Update src/RAGSample.py

src/RAGSample.py

@@ -366,12 +366,15 @@ Answer:
     # Initialize a local Hugging Face model
     hf_pipeline = pipeline(
         "text-generation",
-        model="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",  # Good for Q&A tasks
+        model="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",  # Excellent for Q&A tasks
         tokenizer="deepseek-ai/DeepSeek-R1-0528-Qwen3-8B",
-        max_length=512,
+        max_new_tokens=150,  # Generate only 150 new tokens instead of max_length
         temperature=0.1,
         device=0 if torch.cuda.is_available() else -1,
-        return_full_text=False
+        return_full_text=False,
+        truncation=True,  # Truncate input if too long
+        do_sample=True,   # Enable sampling for better responses
+        pad_token_id=50256  # Add padding token to avoid warnings
     )
     
     # Wrap it in LangChain