File size: 3,480 Bytes
74748ba
3be2bfb
2d2e179
3657970
2d2e179
3be2bfb
 
2d2e179
 
 
74748ba
2d2e179
 
 
9fa297f
74748ba
4fe0e78
3be2bfb
 
2d2e179
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
74748ba
2d2e179
bfc97c8
2d2e179
 
 
 
 
74748ba
3be2bfb
2d2e179
 
3be2bfb
2d2e179
3be2bfb
 
 
 
 
2d2e179
3be2bfb
2d2e179
74748ba
3be2bfb
 
 
 
 
 
678e471
3be2bfb
 
 
 
 
 
 
2d2e179
 
 
 
3be2bfb
 
 
 
 
 
 
2d2e179
3be2bfb
 
2d2e179
8ff6f4e
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
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
import gradio as gr
import openai, os
import tqdm
import time
from langchain.vectorstores import Chroma
from PyPDF2 import PdfReader
from langchain.text_splitter import CharacterTextSplitter, RecursiveCharacterTextSplitter
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain import VectorDBQA
from langchain.llms import AzureOpenAI

os.environ["OPENAI_API_TYPE"] = openai.api_type = "azure"
os.environ["OPENAI_API_VERSION"] = openai.api_version = "2022-12-01"
os.environ["OPENAI_API_BASE"] = openai.api_base = "https://openai-endpoint.openai.azure.com/"
openai.api_key =os.environ["OPENAI_API_KEY"]


def upload_pdf(file, pdf_text, embeddings, vectorstore, azure_embeddings, qa, progress = gr.Progress(track_tqdm=True)):
    reader = PdfReader(file)
    number_of_pages = len(reader.pages)
    pdf_text = ""
    for page_number in range(number_of_pages):
        page = reader.pages[page_number]
        pdf_text += page.extract_text()
    text_splitter = RecursiveCharacterTextSplitter(        
    chunk_size = 1000,
    chunk_overlap  = 200,
    length_function = len,)
    texts = text_splitter.split_text(pdf_text)
    for text in tqdm.tqdm(texts):
        try:
            response = openai.Embedding.create(
            input=text,
            engine="text-embedding-ada-002")
            emb = response['data'][0]['embedding']
            embeddings.append(emb)
        except Exception as e:
            print(e)
            time.sleep(8)
            response = openai.Embedding.create(
            input=text,
            engine="text-embedding-ada-002")
            emb = response['data'][0]['embedding']
            embeddings.append(emb)
    

    azure_embeddings = OpenAIEmbeddings(document_model_name="text-embedding-ada-002",query_model_name="text-embedding-ada-002")
    vectorstore = Chroma("collection", embedding_function=azure_embeddings)

    vectorstore._collection.add(
        ids= [f"doc_{i}" for i in range(len(texts))],
        documents=texts,
        embeddings=embeddings,
        metadatas=[{"source": "source"} for text in texts])
    qa = VectorDBQA.from_chain_type(llm= AzureOpenAI(deployment_name="davinci003", model_name="text-davinci-003"), chain_type="stuff", vectorstore=vectorstore)

    return pdf_text, pdf_text, embeddings, vectorstore, azure_embeddings, qa, gr.update(visible=True), gr.update(visible=True), gr.update(visible=False)


def add_text(chatstate, query, qa):
    # chain.run(input_documents=docs, question=query)
    chatstate = chatstate + [(query, qa.run(query))]

    return chatstate, chatstate, qa

with gr.Blocks(css="footer {visibility: hidden}") as demo:
    qa = pdf_text = embeddings = vectorstore = azure_embeddings = gr.State([])
    with gr.Row(visible=False) as chat_row:
        chatbot = gr.Chatbot()
    with gr.Row(visible=False) as submit_row:
        text = gr.Textbox(show_label=False, placeholder="Enter text and press enter").style(container=False)
    chatstate = gr.State([])
    text.submit(add_text, [chatstate, text, qa], [chatbot, chatstate, qa])

    


    # set state
    with gr.Column() as upload_column:

        file = gr.File()
        upload_btn = gr.Button("Upload")
        output_text = gr.TextArea()
        upload_btn.click(upload_pdf, inputs=[file, pdf_text, embeddings, vectorstore, azure_embeddings, qa], outputs=[output_text, pdf_text, embeddings, vectorstore, azure_embeddings, qa, chat_row, submit_row, upload_column])




demo.launch(enable_queue=True)