File size: 5,253 Bytes
5102822
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
# Importing dependencies
from dotenv import load_dotenv
import streamlit as st
from PyPDF2 import PdfReader
from langchain.text_splitter import CharacterTextSplitter
from langchain.embeddings import HuggingFaceEmbeddings
from langchain.vectorstores import FAISS
from langchain.prompts import PromptTemplate
from langchain.memory import ConversationBufferMemory
from transformers import pipeline
from htmlTemplates import css, bot_template, user_template

# Load environment variables
load_dotenv()

# Creating custom template to guide LLM model
custom_template = """Given the following conversation and a follow-up question, rephrase the follow-up question to be a standalone question, in its original language.
Chat History:
{chat_history}
Follow Up Input: {question}
Standalone question:"""

CUSTOM_QUESTION_PROMPT = PromptTemplate.from_template(custom_template)

# Extracting text from PDF
def get_pdf_text(docs):
    text = ""
    for pdf in docs:
        pdf_reader = PdfReader(pdf)
        for page in pdf_reader.pages:
            text += page.extract_text()
    return text

# Converting text to chunks
def get_chunks(raw_text):
    text_splitter = CharacterTextSplitter(
        separator="\n",
        chunk_size=1000,
        chunk_overlap=200,
        length_function=len
    )
    chunks = text_splitter.split_text(raw_text)
    return chunks

# Using Hugging Face embeddings model and FAISS to create vectorstore
def get_vectorstore(chunks):
    embeddings = HuggingFaceEmbeddings(
        model_name="sentence-transformers/all-MiniLM-L6-v2",
        model_kwargs={'device': 'cpu'}
    )
    vectorstore = FAISS.from_texts(texts=chunks, embedding=embeddings)
    return vectorstore

# Generating conversation chain with improved out-of-scope handling
def get_conversationchain(vectorstore):
    # Use a Hugging Face model for question-answering
    model_name = "distilbert-base-uncased-distilled-squad"  # Pretrained QA model
    qa_pipeline = pipeline("question-answering", model=model_name, tokenizer=model_name)

    def qa_function(question, context):
        response = qa_pipeline(question=question, context=context)
        return response['answer'], response['score']

    memory = ConversationBufferMemory(
        memory_key='chat_history',
        return_messages=True,
        output_key='answer'
    )

    def conversation_chain(inputs):
        question = inputs['question']
        # Extract text content from Document objects
        documents = vectorstore.similarity_search(question, k=5)
        
        # If no similar documents are found or similarity is too low
        if not documents:
            answer = "Sorry, I couldn't find relevant information in the document. Please ask a question related to the document."
            memory.save_context({"user_input": question}, {"answer": answer})
            return {"chat_history": memory.chat_memory.messages, "answer": answer}

        context = "\n".join([doc.page_content for doc in documents])  # Extract `page_content` from each Document
        answer, score = qa_function(question, context)

        # Define a threshold for confidence (e.g., 0.5)
        if score < 0.5:
            answer = "Sorry, I couldn't find relevant information in the document. Please ask a question related to the document."
        
        memory.save_context({"user_input": question}, {"answer": answer})
        return {"chat_history": memory.chat_memory.messages, "answer": answer}

    return conversation_chain

# Generating response from user queries and displaying them accordingly
def handle_question(question):
    response = st.session_state.conversation({'question': question})
    st.session_state.chat_history = response["chat_history"]
    for i, msg in enumerate(st.session_state.chat_history):
        if i % 2 == 0:
            st.write(user_template.replace("{{MSG}}", msg.content), unsafe_allow_html=True)
        else:
            st.write(bot_template.replace("{{MSG}}", msg.content), unsafe_allow_html=True)

def main():
    st.set_page_config(page_title="Chat with multiple PDFs", page_icon=":books:")
    st.write(css, unsafe_allow_html=True)

    if "conversation" not in st.session_state:
        st.session_state.conversation = None

    if "chat_history" not in st.session_state:
        st.session_state.chat_history = None
    
    st.header("CSS Edge - Intelligent Document Chatbot :books:")
    question = st.text_input("Ask a question from your document:")
    if question:
        handle_question(question)
    
    with st.sidebar:
        st.subheader("Your documents")
        docs = st.file_uploader("Upload your PDF here and click on 'Process'", accept_multiple_files=True)
        if st.button("Process"):
            with st.spinner("Processing..."):
                # Get the PDF text
                raw_text = get_pdf_text(docs)
                
                # Get the text chunks
                text_chunks = get_chunks(raw_text)
                
                # Create vectorstore
                vectorstore = get_vectorstore(text_chunks)
                
                # Create conversation chain
                st.session_state.conversation = get_conversationchain(vectorstore)

if __name__ == '__main__':
    main()