Update app.py

app.py

@@ -6,7 +6,6 @@ from langchain import OpenAI
 from langchain.chains import RetrievalQAWithSourcesChain
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 from langchain.document_loaders import UnstructuredURLLoader
-# from langchain.embeddings import OpenAIEmbeddings
 from langchain.embeddings import FakeEmbeddings
 from langchain.llms import HuggingFaceHub
 from langchain.chains import LLMChain
@@ -16,67 +15,64 @@ from dotenv import load_dotenv
 load_dotenv()  # take environment variables from .env (especially openai api key)
 os.environ["HUGGINGFACEHUB_API_TOKEN"] = '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'
 
-st.title("RockyBot: News Research Tool 📈")
-st.sidebar.title("News Article URLs")
 
-urls = []
-for i in range(3):
-    url = st.sidebar.text_input(f"URL {i+1}")
-    urls.append(url)
+class RockyBot:
+    def __init__(self, llm):
+        self.llm = llm
+        self.vectorstore = None
 
-process_url_clicked = st.sidebar.button("Process URLs")
-file_path = "faiss_store_openai.pkl"
+    def process_urls(self, urls):
+        """Processes the given URLs and saves the FAISS index to a pickle file."""
 
-main_placeholder = st.empty()
-llm = HuggingFaceHub( repo_id="google/flan-t5-xxl", model_kwargs={"temperature": 0.5, "max_length": 64} )
+        # load data
+        loader = UnstructuredURLLoader(urls=urls)
 
-@st.cache
-def process_urls(urls):
-    """Processes the given URLs and saves the FAISS index to a pickle file."""
+        # split data
+        text_splitter = RecursiveCharacterTextSplitter(
+            separators=['\n\n', '\n', '.', ','],
+            chunk_size=1000
+        )
+        docs = text_splitter.split_documents(loader.load())
 
-    # load data
-    loader = UnstructuredURLLoader(urls=urls)
+        # create embeddings and save it to FAISS index
+        embeddings = FakeEmbeddings(size=1352)
+        self.vectorstore = FAISS.from_documents(docs, embeddings)
 
-    # split data
-    text_splitter = RecursiveCharacterTextSplitter(
-        separators=['\n\n', '\n', '.', ','],
-        chunk_size=1000
-    )
-    docs = text_splitter.split_documents(loader.load())
+        # Save the FAISS index to a pickle file
+        with open("faiss_store_openai.pkl", "wb") as f:
+            pickle.dump(self.vectorstore, f)
 
-    # create embeddings and save it to FAISS index
-    embeddings = FakeEmbeddings(size=1352)
-    vectorstore_openai = FAISS.from_documents(docs, embeddings)
+    def answer_question(self, question):
+        """Answers the given question using the LLM and retriever."""
 
-    # Save the FAISS index to a pickle file
-    with open(file_path, "wb") as f:
-        pickle.dump(vectorstore_openai, f)
+        chain = RetrievalQAWithSourcesChain.from_llm(llm=self.llm, retriever=self.vectorstore.as_retriever())
+        result = chain({"question": question}, return_only_outputs=True)
 
-if process_url_clicked:
-    with st.progress(0.0):
-        process_urls(urls)
+        return result["answer"], result.get("sources", "")
+
+
+if __name__ == '__main__':
+    llm = HuggingFaceHub(repo_id="google/flan-t5-xxl", model_kwargs={"temperature": 0.5, "max_length": 64})
+    rockybot = RockyBot(llm)
+
+    # Process URLs if the button is clicked
+    if st.sidebar.button("Process URLs"):
+        rockybot.process_urls(st.sidebar.text_input("URL 1"), st.sidebar.text_input("URL 2"), st.sidebar.text_input("URL 3"))
         st.progress(100.0)
 
-query = main_placeholder.text_input("Question: ")
-if query:
-    try:
-        with open(file_path, "rb") as f:
-            vectorstore = pickle.load(f)
-            chain = RetrievalQAWithSourcesChain.from_llm(llm=llm, retriever=vectorstore.as_retriever())
-            result = chain({"question": query}, return_only_outputs=True)
-            # result will be a dictionary of this format --> {"answer": "", "sources": [] }
-            st.header("Answer")
-            st.write(result["answer"])
-
-            # Display sources, if available
-            sources = result.get("sources", "")
-            if sources:
-                st.subheader("Sources:")
-                sources_list = sources.split("\n")  # Split the sources by newline
-                for source in sources_list:
-                    st.write(source)
-    except Exception as e:
-        st.error(e)
+    # Answer the question if it is not empty
+    query = st.text_input("Question: ")
+    if query:
+        answer, sources = rockybot.answer_question(query)
+
+        st.header("Answer")
+        st.write(answer)
+
+        # Display sources, if available
+        if sources:
+            st.subheader("Sources:")
+            for source in sources.split("\n"):
+                st.write(source)
 
 if __name__ == '__main__':
-    st.main()
+    st.main()