Update app.py

app.py

@@ -1,5 +1,5 @@
 import os
-from langchain_community.llms import HuggingFaceHub
+from langchain_huggingface import HuggingFaceEndpoint
 from langchain.chains.question_answering import load_qa_chain
 from langchain.vectorstores import FAISS
 from langchain.embeddings import HuggingFaceEmbeddings
@@ -7,7 +7,7 @@ from langchain.text_splitter import CharacterTextSplitter
 from langchain_community.document_loaders import PyPDFLoader, TextLoader
 import gradio as gr
 
-# Load documents (example using PDFs or TXT)
+# Load documents (PDF or TXT)
 def load_docs(file):
     if file.name.endswith(".pdf"):
         loader = PyPDFLoader(file.name)
@@ -26,15 +26,16 @@ def embed_docs(splits):
     embeddings = HuggingFaceEmbeddings()
     return FAISS.from_documents(splits, embeddings)
 
-# QA chain
+# QA chain using updated HuggingFaceEndpoint and invoke method
 def qa_chain(vectorstore, query):
-    llm = HuggingFaceHub(
+    llm = HuggingFaceEndpoint(
         repo_id="google/flan-t5-large",
-        model_kwargs={"temperature": 0.5, "max_length": 512}
+        huggingfacehub_api_token=os.environ["HUGGINGFACEHUB_API_TOKEN"],
+        model_kwargs={"temperature": 0.5, "max_new_tokens": 512}
     )
     chain = load_qa_chain(llm, chain_type="stuff")
     docs = vectorstore.similarity_search(query)
-    return chain.run(input_documents=docs, question=query)
+    return chain.invoke({"input_documents": docs, "question": query})
 
 # Gradio interface
 def analyze(file, question):