File size: 2,761 Bytes
63337f5
7d0a6ff
11ef280
63337f5
 
 
 
6f9cc9b
 
7de3632
078b2c2
7de3632
 
078b2c2
6f9cc9b
 
63337f5
 
6f9cc9b
 
7de3632
63337f5
 
7de3632
 
63337f5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
90
91
import streamlit as st 
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain_community.document_loaders import PyPDFLoader
from transformers import T5Tokenizer, T5ForConditionalGeneration
from transformers import pipeline
import torch
import base64
#import os
#from dotenv import load_dotenv
#from huggingface_hub import HfApi

#api = HfApi()
#token = api.retrieve_token("secret_token")  # Replace with your secret name

#load_dotenv()
#token = os.environ.get("HF_TOKEN")




checkpoint = "MBZUAI/LaMini-Flan-T5-248M"
#model and tokenizer loading

tokenizer = T5Tokenizer.from_pretrained(checkpoint)
base_model = T5ForConditionalGeneration.from_pretrained(checkpoint, device_map='auto', torch_dtype=torch.float32, offload_folder='D:/project/offload')

#file loader and preprocessing
def file_preprocessing(file):
    loader =  PyPDFLoader(file)
    pages = loader.load_and_split()
    text_splitter = RecursiveCharacterTextSplitter(chunk_size=200, chunk_overlap=50)
    texts = text_splitter.split_documents(pages)
    final_texts = ""
    for text in texts:
        print(text)
        final_texts = final_texts + text.page_content
    return final_texts

#LLM pipeline
def llm_pipeline(filepath):
    pipe_sum = pipeline(
        'summarization',
        model = base_model,
        tokenizer = tokenizer,
        max_length = 500, 
        min_length = 50)
    input_text = file_preprocessing(filepath)
    result = pipe_sum(input_text)
    result = result[0]['summary_text']
    return result

@st.cache_data
#function to display the PDF of a given file 
def displayPDF(file):
    # Opening file from file path
    with open(file, "rb") as f:
        base64_pdf = base64.b64encode(f.read()).decode('utf-8')

    # Embedding PDF in HTML
    pdf_display = F'<iframe src="data:application/pdf;base64,{base64_pdf}" width="100%" height="600" type="application/pdf"></iframe>'

    # Displaying File
    st.markdown(pdf_display, unsafe_allow_html=True)

#streamlit code 
st.set_page_config(layout="wide")

def main():
    st.title("Document Summarization App using Language Model")

    uploaded_file = st.file_uploader("Upload your PDF file", type=['pdf'])

    if uploaded_file is not None:
        if st.button("Summarize"):
            col1, col2 = st.columns(2)
            filepath = "data/"+uploaded_file.name
            with open(filepath, "wb") as temp_file:
                temp_file.write(uploaded_file.read())
            with col1:
                st.info("Uploaded File")
                pdf_view = displayPDF(filepath)

            with col2:
                summary = llm_pipeline(filepath)
                st.info("Summarization Complete")
                st.success(summary)



if __name__ == "__main__":
    main()