File size: 8,396 Bytes
a6fe9a8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
407931b
a6fe9a8
407931b
 
 
a6fe9a8
407931b
a6fe9a8
407931b
 
 
 
a6fe9a8
 
 
 
407931b
 
 
 
 
 
 
 
a6fe9a8
 
 
 
 
 
407931b
a6fe9a8
407931b
 
 
a6fe9a8
 
 
407931b
 
 
 
a6fe9a8
407931b
a6fe9a8
 
 
407931b
 
a6fe9a8
407931b
a6fe9a8
 
 
 
 
 
 
407931b
 
a6fe9a8
407931b
a6fe9a8
407931b
 
 
a6fe9a8
 
407931b
 
 
a6fe9a8
 
 
 
 
 
407931b
a6fe9a8
407931b
a6fe9a8
407931b
a6fe9a8
 
 
407931b
a6fe9a8
407931b
 
 
 
 
 
 
 
a6fe9a8
407931b
 
a6fe9a8
407931b
 
a6fe9a8
407931b
a6fe9a8
407931b
a6fe9a8
407931b
a6fe9a8
407931b
 
a6fe9a8
 
 
 
 
 
 
 
 
 
407931b
 
a6fe9a8
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
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
# import os

# import streamlit as st
# from dotenv import load_dotenv
# from PyPDF2 import PdfReader
# from langchain.text_splitter import CharacterTextSplitter
# from langchain_openai import OpenAIEmbeddings
# from langchain.vectorstores import FAISS
# # from langchain_community.vectorstores import FAISS
# from langchain.embeddings import HuggingFaceEmbeddings
# from langchain.memory import ConversationBufferMemory
# from langchain.chains import ConversationalRetrievalChain
# from langchain.chat_models import ChatOpenAI
# from htmlTemplates import css, bot_template, user_template
# from langchain.embeddings import HuggingFaceInstructEmbeddings
# from langchain.llms import HuggingFaceHub
# import os
# def get_pdf_text(pdf_doc):
#     text = ""
#     for pdf in pdf_doc:
#         pdf_reader = PdfReader(pdf)
#         for page in pdf_reader.pages:
#             text += page.extract_text()
#     return text


# def get_text_chunk(row_text):
#     text_splitter = CharacterTextSplitter(
#         separator="\n",
#         chunk_size = 1000,
#         chunk_overlap = 200,
#         length_function = len
#     )
#     chunk = text_splitter.split_text(row_text)
#     return chunk


# def get_vectorstore(text_chunk):
#     #embeddings = OpenAIEmbeddings(openai_api_key = os.getenv("OPENAI_API_KEY"))
#     embeddings = HuggingFaceInstructEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
#     vector = FAISS.from_texts(text_chunk,embeddings)
#     return vector


# def get_conversation_chain(vectorstores):
#     #llm = ChatOpenAI(openai_api_key = os.getenv("OPENAI_API_KEY"))
#     llm = HuggingFaceHub(repo_id="google/flan-t5-base", model_kwargs={"temperature":0.5, "max_length":512})
#     memory = ConversationBufferMemory(memory_key = "chat_history",return_messages = True)
#     conversation_chain = ConversationalRetrievalChain.from_llm(llm=llm,
#                                                                retriever=vectorstores.as_retriever(),
#                                                                memory=memory)
#     return conversation_chain


# def user_input(user_question):
#     response = st.session_state.conversation({"question":user_question})
#     st.session_state.chat_history = response["chat_history"]

#     for indx, msg in enumerate(st.session_state.chat_history):
#         if indx % 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():
#     # load secret key
#     load_dotenv()
    
#     # config the pg
#     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

#     st.header("Chat with multiple PDFs :books:")
#     user_question = st.text_input("Ask a question about your docs")
#     if user_question:
#         user_input(user_question)

#     # st.write(user_template.replace("{{MSG}}","Hello Robot"), unsafe_allow_html=True)
#     # st.write(bot_template.replace("{{MSG}}","Hello Human"), unsafe_allow_html=True)

#     # create side bar
#     with st.sidebar:
#         st.subheader("Your Documents")
#         pdf_doc = st.file_uploader(label="Upload your documents",accept_multiple_files=True)
#         if st.button("Process"):
#             with st.spinner(text="Processing"):

#             # get pdf text
#                 row_text = get_pdf_text(pdf_doc)
#             # get the text chunk
#                 text_chunk = get_text_chunk(row_text)
#                 # st.write(text_chunk)
#             # create vecor store
#                 vectorstores = get_vectorstore(text_chunk)
#                 # st.write(vectorstores)
#             # create conversation chain
#                 st.session_state.conversation = get_conversation_chain(vectorstores)


# if __name__ == "__main__":
#     main()

import os
import streamlit as st
from dotenv import load_dotenv
from PyPDF2 import PdfReader
from pdf2image import convert_from_path
from langchain.text_splitter import CharacterTextSplitter
from sentence_transformers import SentenceTransformer
from langchain.vectorstores import FAISS
from langchain.memory import ConversationBufferMemory
from langchain.chains import ConversationalRetrievalChain
from langchain.llms import HuggingFaceHub
from htmlTemplates import css, bot_template, user_template
from transformers import pipeline

# Function to extract text from PDF
def get_pdf_text(pdf_doc):
    text = ""
    for pdf in pdf_doc:
        pdf_reader = PdfReader(pdf)
        for page in pdf_reader.pages:
            text += page.extract_text()
    return text

# Function to extract images from PDF
def get_pdf_images(pdf_doc):
    images = []
    for pdf in pdf_doc:
        images.extend(convert_from_path(pdf))
    return images

# Function to split text into chunks
def get_text_chunk(row_text):
    text_splitter = CharacterTextSplitter(
        separator="\n",
        chunk_size=1000,
        chunk_overlap=200,
        length_function=len
    )
    chunk = text_splitter.split_text(row_text)
    return chunk

# Function to create vector store
def get_vectorstore(text_chunk):
    model = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')
    embeddings = model.encode(text_chunk)
    vector = FAISS.from_embeddings(embeddings)
    return vector

# Function to create conversation chain
def get_conversation_chain(vectorstores):
    llm = HuggingFaceHub(repo_id="google/flan-t5-base", model_kwargs={"temperature": 0.5, "max_length": 512})
    memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
    conversation_chain = ConversationalRetrievalChain.from_llm(
        llm=llm,
        retriever=vectorstores.as_retriever(),
        memory=memory
    )
    return conversation_chain

# Function to handle user input
def user_input(user_question):
    response = st.session_state.conversation({"question": user_question})
    st.session_state.chat_history = response["chat_history"]

    for indx, msg in enumerate(st.session_state.chat_history):
        if indx % 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)

# Function to generate images from text using a DALL-E model
def generate_image_from_text(prompt):
    # Ensure you have a DALL-E or similar model for text-to-image generation
    generator = pipeline("text-to-image", model="dalle-mini/dalle-mini")
    images = generator(prompt)
    return images

# Main function
def main():
    # Load secret key
    load_dotenv()

    # Config the page
    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

    st.header("Chat with multiple PDFs :books:")
    user_question = st.text_input("Ask a question about your docs")
    if user_question:
        user_input(user_question)

    # Create side bar
    with st.sidebar:
        st.subheader("Your Documents")
        pdf_doc = st.file_uploader(label="Upload your documents", accept_multiple_files=True, type=["pdf"])
        if st.button("Process"):
            with st.spinner(text="Processing"):
                # Get PDF text
                row_text = get_pdf_text(pdf_doc)
                # Get the text chunk
                text_chunk = get_text_chunk(row_text)
                # Create vector store
                vectorstores = get_vectorstore(text_chunk)
                # Create conversation chain
                st.session_state.conversation = get_conversation_chain(vectorstores)

                # Extract and display images from PDFs
                images = get_pdf_images(pdf_doc)
                for img in images:
                    st.image(img)

                # Generate and display images from text using DALL-E
                if user_question:
                    generated_images = generate_image_from_text(user_question)
                    for gen_img in generated_images:
                        st.image(gen_img)

if __name__ == "__main__":
    main()