import gradio as gr from langchain.document_loaders import WebBaseLoader from langchain.text_splitter import RecursiveCharacterTextSplitter text_splitter = RecursiveCharacterTextSplitter(chunk_size=350, chunk_overlap=10) from langchain.llms import HuggingFaceHub model_id = HuggingFaceHub(repo_id="HuggingFaceH4/zephyr-7b-beta", model_kwargs={"temperature":0.1, "max_new_tokens":300}) from langchain.embeddings import HuggingFaceHubEmbeddings embeddings = HuggingFaceHubEmbeddings() from langchain.vectorstores import Chroma from langchain.chains import RetrievalQA web_links = ["https://www.databricks.com/"] loader = WebBaseLoader(web_links) documents = loader.load() texts = text_splitter.split_documents(documents) db = Chroma.from_documents(texts, embeddings) retriever = db.as_retriever() global qa qa = RetrievalQA.from_chain_type(llm=model_id, chain_type="stuff", retriever=retriever, return_source_documents=True) def add_text(history, text): history = history + [(text, None)] return history, "" def bot(history): response = infer(history[-1][0]) history[-1][1] = response['result'] return history def infer(question): query = question result = qa({"query": query}) return result css=""" #col-container {max-width: 700px; margin-left: auto; margin-right: auto;} """ title = """

Chat with PDF

Upload a .PDF from your computer, click the "Load PDF to LangChain" button,
when everything is ready, you can start asking questions about the pdf ;)

""" with gr.Blocks(css=css) as demo: with gr.Column(elem_id="col-container"): gr.HTML(title) chatbot = gr.Chatbot([], elem_id="chatbot").style(height=350) with gr.Row(): question = gr.Textbox(label="Question", placeholder="Type your question and hit Enter ") question.submit(add_text, [chatbot, question], [chatbot, question]).then( bot, chatbot, chatbot ) demo.launch()