File size: 1,449 Bytes
166b8dd 561086d ac2d7ba 561086d 166b8dd ac2d7ba 166b8dd 986b0f9 166b8dd ba3683e 166b8dd |
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 |
import os
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.text_splitter import CharacterTextSplitter
from langchain.vectorstores import Qdrant
from langchain.document_loaders import TextLoader
from langchain.chains.question_answering import load_qa_chain
from langchain.llms import OpenAI
import gradio as gr
from qdrant_client import QdrantClient
from langchain.document_loaders import PagedPDFSplitter
#load data
loader = PagedPDFSplitter("Philippine National Formulary 8th Edition.pdf")
docs = loader.load_and_split()
#declare constants here
OPEN_API_KEY = os.environ["OPENAI_API_KEY"]
host = os.environ["QDRANT_HOST"]
api_key = os.environ["QDRANT_API_KEY"]
embeddings = OpenAIEmbeddings()
#initialize vectorstore
qdrant = Qdrant.from_documents(docs, embeddings, host=host, prefer_grpc=True, api_key=api_key)
#query pipeline
def question_answering(question):
chain = load_qa_chain(OpenAI(temperature=0), chain_type="stuff")
query = question
docs = qdrant.similarity_search(query)
answer = chain.run(input_documents=docs, question=query)
return answer
with gr.Blocks() as demo:
gr.Markdown("Start typing below and then click **Run** to see the output.")
with gr.Row():
inp = gr.Textbox(placeholder="Ask question here?")
out = gr.Textbox()
btn = gr.Button("Run", api_name="search")
btn.click(fn=question_answering, inputs=inp, outputs=out)
demo.launch(debug=True)
|