Fix: Remove nonexistent routing model, restore OpenAI-based routing

app.py

@@ -7,10 +7,10 @@ from langchain_community.embeddings import HuggingFaceEmbeddings
 from langchain_community.vectorstores import FAISS
 from langchain.chains import RetrievalQA
 from langchain_community.llms import HuggingFacePipeline
+from langchain_openai import ChatOpenAI
 from transformers import pipeline
 
-# ✅ Hugging Face pipelines
-routing_agent = pipeline("text-classification", model="BivinSadler/statistics-routing-agent")
+# ✅ Load writer model and wrap it for LangChain
 writer_model = pipeline("text2text-generation", model="BivinSadler/llama3-finetuned-Statistics")
 writer_llm = HuggingFacePipeline(pipeline=writer_model)
 
@@ -25,19 +25,37 @@ def build_rag_agent(pdf_path):
     retriever = vectorstore.as_retriever()
     return RetrievalQA.from_chain_type(llm=writer_llm, retriever=retriever, chain_type="stuff")
 
-# ✅ Load agents
+# ✅ Load RAG agents
 stat6371_agent = build_rag_agent("PDFs/DS 6371 Syllabus Ver 6.pdf")
 ds7333_agent = build_rag_agent("PDFs/ds-7333_syllabus.pdf")
 
-# ✅ Routing
+# ✅ Load OpenAI LLM for routing
+openai_key = os.environ.get("OPENAI_API_KEY")
+llm = ChatOpenAI(api_key=openai_key, model="gpt-3.5-turbo", temperature=0)
+
+# ✅ Routing logic
 def route_question(question):
-    label = routing_agent(question)[0]["label"]
-    return {
-        "LABEL_0": "stat6371",
-        "LABEL_1": "ds7333"
-    }.get(label, "general")
+    routing_prompt = f"""
+You are a classification agent that helps route questions to the appropriate expert.
+
+There are three possible categories:
+A. Stat 6371 (Theoretical statistics course)
+B. DS 7333 (Decision Analytics Course)
+C. General statistics (any other statistics question)
+
+Classify the following question into one of those three categories by answering only with a single letter: A, B, or C.
+
+Question: "{question}"
+Answer:"""
+    response = llm.invoke(routing_prompt).content.strip().upper()
+    if response.startswith("A"):
+        return "stat6371"
+    elif response.startswith("B"):
+        return "ds7333"
+    else:
+        return "general"
 
-# ✅ Writing
+# ✅ Explanation agent
 def writer_agent(raw_answer, audience="high school students"):
     prompt = f"""
 You are a talented science communicator. Your job is to explain the following answer in a way that is clear, short, and engaging for {audience}.
@@ -50,7 +68,7 @@ Write your response in 2–3 sentences. Avoid technical jargon.
     result = writer_model(prompt, max_new_tokens=200, do_sample=False)
     return result[0]['generated_text']
 
-# ✅ Core Logic
+# ✅ Main logic
 def multiagent_system(question):
     print(f"🧭 Routing: {question}")
     route = route_question(question)
@@ -69,7 +87,7 @@ def multiagent_system(question):
     print("✍️ Simplifying...")
     return writer_agent(raw_answer)
 
-# ✅ Gradio
+# ✅ Gradio UI
 iface = gr.Interface(
     fn=multiagent_system,
     inputs=gr.Textbox(lines=2, label="Ask a statistics question"),