app.py

import gradio as gr
import os

from langchain.chains.question_answering import load_qa_chain
from langchain.document_loaders import UnstructuredURLLoader
from langchain import OpenAI
from langchain import HuggingFaceHub
os.environ[
    "HUGGINGFACEHUB_API_TOKEN"] = "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"
os.environ["LANGCHAIN_TRACING_V2"] = "true"
os.environ["LANGCHAIN_ENDPOINT"] = "https://api.smith.langchain.com"
os.environ["LANGCHAIN_API_KEY"] = "ls__ae9b316f4ee9475b84f66c616344d713"
os.environ["LANGCHAIN_PROJECT"] = "Sequential-Chain"


def main():
    with gr.Blocks() as demo:
        with gr.Tab(label="tab1", id="tab1"):  #标签页1
          input_url1 = gr.inputs.Textbox(label="URL", lines=1)
          outputs1 = gr.outputs.Textbox(label="输出")
          text_button = gr.Button("提交")
          # gr.Interface(fn=my_inference_function,inputs=input_url1,outputs=outputs1)
        with gr.Tab(label="tab2", id="tab2"):  #标签页2
          input_api_key = gr.inputs.Textbox(label="ChatGPT API Key", lines=1)
          input_api_base = gr.inputs.Textbox(label="ChatGPT API 地址（默认无地址）", lines=1)
          input_url2 = gr.inputs.Textbox(label="URL", lines=1)
          outputs2 = gr.outputs.Textbox(label="输出")
          vid_button = gr.Button("提交")
          # gr.Interface(fn=my_chatgpt_function,inputs=[input_api_key, input_api_base, input_url2],outputs=outputs2)
    demo.launch()


def my_chatgpt_function(api_key, api_base, url):
    os.environ["OPENAI_API_KEY"] = api_key
    os.environ['OPENAI_API_BASE'] = api_base
    llm = OpenAI(temperature=0.7, model_name="gpt-3.5-turbo", max_tokens=1024)
    loader = UnstructuredURLLoader(urls=[url])
    data = loader.load()
    chain = load_qa_chain(llm=llm, chain_type="stuff")
    response = chain.run(input_documents=data,
                         question="""请用中文总结文章的内容，并以下面模版给出结果：
        《文章标题》摘要如下：
        ## 一句话描述
        文章摘要内容
        ## 文章略读
        文章要点""")
    return response


def my_inference_function(url):
    llm = HuggingFaceHub(repo_id="declare-lab/flan-alpaca-large",
                         model_kwargs={
                             "temperature": 0.1,
                             "max_length": 512
                         })
    loader = UnstructuredURLLoader(urls=[url])
    data = loader.load()
    chain = load_qa_chain(llm=llm, chain_type="stuff")
    response = chain.run(input_documents=data,
                         question="Summarize this article in one paragraph")
    return response


if __name__ == '__main__':
    main()