Spaces:
Running
Running
File size: 2,299 Bytes
31a77e6 3f4785b 31a77e6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 |
import gradio as gr
from langchain.document_loaders import PyPDFLoader # for loading the pdf
from langchain.embeddings import OpenAIEmbeddings # for creating embeddings
from langchain.vectorstores import Chroma # for the vectorization part
from langchain.chains import ChatVectorDBChain # for chatting with the pdf
from langchain.llms import OpenAI # the LLM model we'll use (CHatGPT)
class Chat:
def __init__(self, pdf, api_input):
self.api = api_input
loader = PyPDFLoader(pdf)
pages = loader.load_and_split()
embeddings = OpenAIEmbeddings(openai_api_key=self.api)
vectordb = Chroma.from_documents(pages, embedding=embeddings, persist_directory=".")
vectordb.persist()
self.pdf_qa = ChatVectorDBChain.from_llm(OpenAI(temperature=0.9, model_name="gpt-3.5-turbo",
openai_api_key=self.api),
vectordb, return_source_documents=True)
def question(self, query):
result = self.pdf_qa({"question": "请使用中文回答" + query, "chat_history": ""})
print("Answer:")
print(result["answer"])
return result["answer"]
def analyse(pdf_file, api_input):
print(pdf_file.name)
session = Chat(pdf_file.name, api_input)
return session, "文章分析完成"
def ask_question(data, question):
if data == "":
return "Please upload PDF file first!"
return data.question(question)
with gr.Blocks() as demo:
gr.Markdown(
"""
# ChatPDF based on Langchain
""")
data = gr.State()
with gr.Tab("Upload PDF File"):
pdf_input = gr.File(label="PDF File")
api_input = gr.Textbox(label="OpenAI API Key")
result = gr.Textbox()
upload_button = gr.Button("Start Analyse")
question_input = gr.Textbox(label="Your Question", placeholder="Authors of this paper?")
answer = gr.Textbox(label="Answer")
ask_button = gr.Button("Ask")
upload_button.click(fn=analyse, inputs=[pdf_input, api_input], outputs=[data, result])
ask_button.click(ask_question, inputs=[data, question_input], outputs=answer)
if __name__ == "__main__":
demo.title = "ChatPDF Based on Langchain"
demo.launch()
|