Not able to show result generated from model into the UI

LLMs and AI
1

st.header(":parrot::link: Chat with PDF ")

Upload multiple PDF files

uploaded_files = st.file_uploader(“Upload your PDFs”, type=‘pdf’, accept_multiple_files=True)
if st.button(“Process”):
if uploaded_files is not None:
docs =
for pdf in uploaded_files:
pdf_reader = PdfReader(pdf)

        text = ""
        for page in pdf_reader.pages:
            text += page.extract_text()

    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000,chunk_overlap=20,length_function=len)
    chunks = text_splitter.split_text(text=text)
    print(f"Chunks for {pdf.name}: {chunks}")
    print("0")
    # print('**************Implementing Similarity Search Using Vector Database***************************')
    # hugging face embeddings
    embeddings = HuggingFaceEmbeddings()

    persist_directory = 'chromaDB'

    # create vector db
    vectordb = Chroma.from_texts(texts=chunks,
                        embedding=embeddings,
                        persist_directory=persist_directory)
    # persist db to disk
    vectordb.persist()

    # using the persist chroma db
    new_db = Chroma(persist_directory=persist_directory,embedding_function=embeddings)
    # intantiate retreive object results from multiple indexes from vector database
    retriever = vectordb.as_retriever(search_kwargs={"k": 2})

    # intantiate LLM model
    llm=HuggingFaceHub(repo_id="declare-lab/flan-alpaca-large", model_kwargs={"temperature":0, "max_length":512})


    # creating chain to answer questions using vector database (chromadb) directly
    qa_chain = RetrievalQA.from_chain_type(llm=llm, 
                                    chain_type="stuff", 
                                    retriever=retriever, 
                                    input_key="question")
    
    st.header("Ask your data")
    user_q = st.text_area("Enter your questions here")
    
    if st.button("Get Response"):
        try:
            with st.spinner("Model is working on it..."):
                result = qa_chain({"question":user_q}, return_only_outputs=True)
                st.subheader('Your response:')
                st.write(result['answer'])
                st.subheader('Source pages:')
                st.write(result['sources'])
        except Exception as e:
            st.error(f"An error occurred: {e}")
            st.error('Oops, the GPT response resulted in an error :( Please try again with a different question.')
2

Hi @Pritesh_Patel

No error message was provided, nevertheless, I’d recommend to try to run the code segment by segment in order to verify that each segment is working as expected. This would help to figure out which segment is causing the issue.

Here’s a similar GH repo to a similar app for asking documents via LangChain that you can get ideas from:

Hope this helps!