Update app.py

app.py

@@ -1,50 +1,53 @@
 import os
 import gradio as gr
+
+# ✅ New LangChain Hugging Face imports
+from langchain_huggingface import HuggingFaceEndpoint
 from langchain.text_splitter import CharacterTextSplitter
 from langchain_community.embeddings import HuggingFaceEmbeddings
 from langchain_community.vectorstores import FAISS
 from langchain.chains import ConversationalRetrievalChain
-from langchain_community.llms import HuggingFaceHub
 from langchain_community.document_loaders import PyPDFLoader
 
-# Load PDF
+# --- 1️⃣ Load your PDF ---
 loader = PyPDFLoader("chimera.pdf")
 documents = loader.load()
 
-# Split documents
+# --- 2️⃣ Split into chunks ---
 text_splitter = CharacterTextSplitter(chunk_size=800, chunk_overlap=100)
 texts = text_splitter.split_documents(documents)
 
-# Embeddings
+# --- 3️⃣ Create embeddings + FAISS vector store ---
 embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
 db = FAISS.from_documents(texts, embeddings)
-
 retriever = db.as_retriever(search_kwargs={"k": 3})
 
-# Hugging Face Hub LLM
+# --- 4️⃣ Hugging Face Endpoint setup ---
 hf_token = os.getenv("HUGGINGFACEHUB_API_TOKEN")
 if hf_token is None:
-    raise ValueError("HUGGINGFACEHUB_API_TOKEN not set in Space Secrets!")
+    raise ValueError(
+        "HUGGINGFACEHUB_API_TOKEN not found! Add it in Space Settings → Secrets."
+    )
 
-llm = HuggingFaceHub(
-    repo_id="google/flan-t5-base",
-    model_kwargs={"temperature": 0},
-    huggingfacehub_api_token=hf_token
+llm = HuggingFaceEndpoint(
+    endpoint_url="https://api-inference.huggingface.co/models/google/flan-t5-base",
+    headers={"Authorization": f"Bearer {hf_token}"},
+    model_kwargs={"temperature": 0}
 )
 
-qa = ConversationalRetrievalChain.from_llm(
-    llm,
-    retriever=retriever
-)
+# --- 5️⃣ Build conversational chain ---
+qa = ConversationalRetrievalChain.from_llm(llm, retriever=retriever)
 
 chat_history = []
 
+# --- 6️⃣ Respond function ---
 def respond(message, history):
-    history = history[-6:]
+    history = history[-6:]  # keep last 3 exchanges
     result = qa({"question": message, "chat_history": history})
     history.append((message, result["answer"]))
     return history, history
 
+# --- 7️⃣ Gradio UI ---
 with gr.Blocks() as demo:
     with gr.Column():
         warning_text = gr.HTML(
@@ -58,12 +61,12 @@ with gr.Blocks() as demo:
 
     def enter_case():
         return (
-            gr.update(visible=True),
-            gr.update(visible=True),
-            gr.update(visible=True),
-            gr.update(value=""),
-            gr.update(visible=False),
-            gr.update(visible=False)
+            gr.update(visible=True),  # chatbot
+            gr.update(visible=True),  # user_input
+            gr.update(visible=True),  # submit_btn
+            gr.update(value=""),      # hide warning
+            gr.update(visible=False), # hide enter_btn
+            gr.update(visible=False)  # hide exit_btn
         )
 
     def exit_case():