Update chatbot_rag.py

chatbot_rag.py

@@ -25,21 +25,23 @@ def build_qa():
 
     # 3. Load LLM (Flan-T5 small for lightweight QA)
     print("🔹 Loading LLM...")
-    model_id = "google/flan-t5-small"
+    model_id = "microsoft/Phi-3-mini-4k-instruct"
     tokenizer = AutoTokenizer.from_pretrained(model_id)
     model = AutoModelForSeq2SeqLM.from_pretrained(model_id)
 
     pipe = pipeline(
-        "text2text-generation",
-        model=model,
-        tokenizer=tokenizer,
-        max_new_tokens=256,
+    "text-generation", 
+    model=model, 
+    tokenizer=tokenizer, 
+    max_new_tokens=300,
+    do_sample=True, # Set to True to enable sampling and use temperature
+    temperature=0.2 # This is the temperature parameter
     )
     llm = HuggingFacePipeline(pipeline=pipe)
 
     # 4. QA Chain with retrieval
     print("🔹 Building RetrievalQA...")
-    retriever = vectorstore.as_retriever(search_kwargs={"k": 3})
+    retriever = vectorstore.as_retriever()
     
 
     template = """