Update app.py

app.py

@@ -1,10 +1,16 @@
 
 
+# Load the trained model
+model = YOLO("best.pt")
+openai_api_key = os.environ.get("openai_api_key")
 
 import os
+import io
+import base64
 import streamlit as st
 import numpy as np
 import fitz  # PyMuPDF
+import tempfile
 from ultralytics import YOLO
 from sklearn.cluster import KMeans
 from sklearn.metrics.pairwise import cosine_similarity
@@ -13,27 +19,22 @@ from langchain_community.document_loaders import PyMuPDFLoader
 from langchain_openai import OpenAIEmbeddings
 from langchain_text_splitters import RecursiveCharacterTextSplitter
 from langchain_core.prompts import ChatPromptTemplate
-from sklearn.decomposition import PCA
 from langchain_openai import ChatOpenAI
-import string
 import re
-
+from PIL import Image
+from streamlit_chat import message
 
 # Load the trained model
-model = YOLO("best.pt")
-openai_api_key = os.environ.get("openai_api_key")
+model = YOLO("runs\\detect\\train7\\weights\\best.pt")
+openai_api_key = "sk-proj-J7kj0kbG1m0eIMPWMdjoT3BlbkFJqwZNqQeOYJ9UH6I0efPi"
 
 # Define the class indices for figures, tables, and text
-figure_class_index = 4  # class index for figures
-table_class_index = 3   # class index for tables
-
-# Global variables to store embeddings and contents
-global_embeddings = None
-global_split_contents = None
+figure_class_index = 4
+table_class_index = 3
 
+# Utility functions
 def clean_text(text):
-    text = re.sub(r'\s+', ' ', text).strip()
-    return text
+    return re.sub(r'\s+', ' ', text).strip()
 
 def remove_references(text):
     reference_patterns = [
@@ -47,9 +48,10 @@ def remove_references(text):
     return text
 
 def save_uploaded_file(uploaded_file):
-    with open(uploaded_file.name, 'wb') as f:
-        f.write(uploaded_file.getbuffer())
-    return uploaded_file.name
+    temp_file = tempfile.NamedTemporaryFile(delete=False)
+    temp_file.write(uploaded_file.getbuffer())
+    temp_file.close()
+    return temp_file.name
 
 def summarize_pdf(pdf_file_path, num_clusters=10):
     embeddings_model = OpenAIEmbeddings(model="text-embedding-3-small", api_key=openai_api_key)
@@ -70,66 +72,21 @@ def summarize_pdf(pdf_file_path, num_clusters=10):
     loader = PyMuPDFLoader(pdf_file_path)
     docs = loader.load()
     full_text = "\n".join(doc.page_content for doc in docs)
-    cleaned_full_text = remove_references(full_text)
-    cleaned_full_text = clean_text(cleaned_full_text)
+    cleaned_full_text = clean_text(remove_references(full_text))
 
-    text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0,separators=["\n\n", "\n",".", " "])
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=0, separators=["\n\n", "\n", ".", " "])
     split_contents = text_splitter.split_text(cleaned_full_text)
     embeddings = embeddings_model.embed_documents(split_contents)
 
-    X = np.array(embeddings)
     kmeans = KMeans(n_clusters=num_clusters, init='k-means++', random_state=0).fit(embeddings)
-    cluster_centers = kmeans.cluster_centers_
-
-    closest_point_indices = []
-    for center in cluster_centers:
-        distances = np.linalg.norm(embeddings - center, axis=1)
-        closest_point_indices.append(np.argmin(distances))
-    
+    closest_point_indices = [np.argmin(np.linalg.norm(embeddings - center, axis=1)) for center in kmeans.cluster_centers_]
     extracted_contents = [split_contents[idx] for idx in closest_point_indices]
-    results = chain.invoke({"topic": ' '.join(extracted_contents)})
-
-    summary_sentences = re.split(r'(?<!\w\.\w.)(?<![A-Z][a-z]\.)(?<=\.|\?)\s', results)
-    summary_embeddings = embeddings_model.embed_documents(summary_sentences)
-    extracted_embeddings = embeddings_model.embed_documents(extracted_contents)
-    similarity_matrix = cosine_similarity(summary_embeddings, extracted_embeddings)
 
-    cited_results = results
-    relevant_sources = []
-    source_mapping = {}
-    sentence_to_source = {}
-    similarity_threshold = 0.6
+    results = chain.invoke({"topic": ' '.join(extracted_contents)})
 
-    for i, sentence in enumerate(summary_sentences):
-        if sentence in sentence_to_source:
-            continue
-        max_similarity = max(similarity_matrix[i])
-        if max_similarity >= similarity_threshold:
-            most_similar_idx = np.argmax(similarity_matrix[i])
-            if most_similar_idx not in source_mapping:
-                source_mapping[most_similar_idx] = len(relevant_sources) + 1
-                relevant_sources.append((most_similar_idx, extracted_contents[most_similar_idx]))
-            citation_idx = source_mapping[most_similar_idx]
-            citation = f"([Source {citation_idx}](#source-{citation_idx}))"
-            cited_sentence = re.sub(r'([.!?])$', f" {citation}\\1", sentence)
-            sentence_to_source[sentence] = citation_idx
-            cited_results = cited_results.replace(sentence, cited_sentence)
-    
-    sources_list = "\n\n## Sources:\n"
-    for idx, (original_idx, content) in enumerate(relevant_sources):
-        sources_list += f"""
-<details style="margin: 10px 0; padding: 10px; border: 1px solid #ccc; border-radius: 5px; background-color: #f9f9f9;">
-<summary style="font-weight: bold; cursor: pointer;">Source {idx + 1}</summary>
-<pre style="white-space: pre-wrap; word-wrap: break-word; margin-top: 10px;">{content}</pre>
-</details>
-"""
-    cited_results += sources_list
-    return cited_results
+    return generate_citations(results, extracted_contents)
 
 def qa_pdf(pdf_file_path, query, num_clusters=5, similarity_threshold=0.6):
-    global global_embeddings, global_split_contents
-
-    # Initialize models and embeddings
     embeddings_model = OpenAIEmbeddings(model="text-embedding-3-small", api_key=openai_api_key)
     llm = ChatOpenAI(model="gpt-3.5-turbo", api_key=openai_api_key, temperature=0.3)
     prompt = ChatPromptTemplate.from_template(
@@ -142,40 +99,37 @@ def qa_pdf(pdf_file_path, query, num_clusters=5, similarity_threshold=0.6):
     output_parser = StrOutputParser()
     chain = prompt | llm | output_parser
 
-    # Load and process the PDF if not already loaded
-    if global_embeddings is None or global_split_contents is None:
-        loader = PyMuPDFLoader(pdf_file_path)
-        docs = loader.load()
-        full_text = "\n".join(doc.page_content for doc in docs)
-        cleaned_full_text = remove_references(full_text)
-        cleaned_full_text = clean_text(cleaned_full_text)
+    loader = PyMuPDFLoader(pdf_file_path)
+    docs = loader.load()
+    full_text = "\n".join(doc.page_content for doc in docs)
+    cleaned_full_text = clean_text(remove_references(full_text))
 
-        text_splitter = RecursiveCharacterTextSplitter(chunk_size=800, chunk_overlap=0, separators=["\n\n", "\n", ".", " "])
-        global_split_contents = text_splitter.split_text(cleaned_full_text)
-        global_embeddings = embeddings_model.embed_documents(global_split_contents)
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=800, chunk_overlap=0, separators=["\n\n", "\n", ".", " "])
+    split_contents = text_splitter.split_text(cleaned_full_text)
+    embeddings = embeddings_model.embed_documents(split_contents)
 
-    # Embed the query and find the most relevant contexts
     query_embedding = embeddings_model.embed_query(query)
-    similarity_scores = cosine_similarity([query_embedding], global_embeddings)[0]
+    similarity_scores = cosine_similarity([query_embedding], embeddings)[0]
     top_indices = np.argsort(similarity_scores)[-num_clusters:]
-    relevant_contents = [global_split_contents[i] for i in top_indices]
+    relevant_contents = [split_contents[i] for i in top_indices]
 
-    # Generate the answer using the LLM chain
     results = chain.invoke({"question": query, "contexts": ' '.join(relevant_contents)})
 
-    # Split the answer into sentences and embed them
-    answer_sentences = re.split(r'(?<!\w\.\w.)(?<![A-Z][a-z]\.)(?<=\.|\?)\s', results)
-    answer_embeddings = embeddings_model.embed_documents(answer_sentences)
-    relevant_embeddings = embeddings_model.embed_documents(relevant_contents)
-    similarity_matrix = cosine_similarity(answer_embeddings, relevant_embeddings)
+    return generate_citations(results, relevant_contents, similarity_threshold)
 
-    # Map sentences to sources and create citations
-    cited_results = results
+def generate_citations(text, contents, similarity_threshold=0.6):
+    embeddings_model = OpenAIEmbeddings(model="text-embedding-3-small", api_key=openai_api_key)
+    text_sentences = re.split(r'(?<!\w\.\w.)(?<![A-Z][a-z]\.)(?<=\.|\?)\s', text)
+    text_embeddings = embeddings_model.embed_documents(text_sentences)
+    content_embeddings = embeddings_model.embed_documents(contents)
+    similarity_matrix = cosine_similarity(text_embeddings, content_embeddings)
+
+    cited_text = text
     relevant_sources = []
     source_mapping = {}
     sentence_to_source = {}
 
-    for i, sentence in enumerate(answer_sentences):
+    for i, sentence in enumerate(text_sentences):
         if sentence in sentence_to_source:
             continue
         max_similarity = max(similarity_matrix[i])
@@ -183,34 +137,42 @@ def qa_pdf(pdf_file_path, query, num_clusters=5, similarity_threshold=0.6):
             most_similar_idx = np.argmax(similarity_matrix[i])
             if most_similar_idx not in source_mapping:
                 source_mapping[most_similar_idx] = len(relevant_sources) + 1
-                relevant_sources.append((most_similar_idx, relevant_contents[most_similar_idx]))
+                relevant_sources.append((most_similar_idx, contents[most_similar_idx]))
             citation_idx = source_mapping[most_similar_idx]
-            citation = f"<strong style='color:blue;'>[Source {citation_idx}]</strong>"            
+            citation = f"([Source {citation_idx}](#source-{citation_idx}))"
             cited_sentence = re.sub(r'([.!?])$', f" {citation}\\1", sentence)
             sentence_to_source[sentence] = citation_idx
-            cited_results = cited_results.replace(sentence, cited_sentence)
+            cited_text = cited_text.replace(sentence, cited_sentence)
 
-    # Format the sources for markdown rendering
     sources_list = "\n\n## Sources:\n"
     for idx, (original_idx, content) in enumerate(relevant_sources):
-        sources_list += f"""
-<details style="margin: 10px 0; padding: 10px; border: 1px solid #ccc; border-radius: 5px; background-color: #f9f9f9;">
-<summary style="font-weight: bold; cursor: pointer;">Source {idx + 1}</summary>
-<pre style="white-space: pre-wrap; word-wrap: break-word; margin-top: 10px;">{content}</pre>
+        sources_list +=  f"""
+<details style="margin: 1px 0; padding: 5px; border: 1px solid #ccc; border-radius: 8px; background-color: #f9f9f9; transition: all 0.3s ease;">
+  <summary style="font-weight: bold; cursor: pointer; outline: none; padding: 5px 0; transition: color 0.3s ease;">Source {idx + 1}</summary>
+  <pre style="white-space: pre-wrap; word-wrap: break-word; margin: 1px 0; padding: 10px; background-color: #fff; border-radius: 5px; border: 1px solid #ddd; box-shadow: 0 2px 5px rgba(0, 0, 0, 0.1);">{content}</pre>
 </details>
 """
-    cited_results += sources_list
-    return cited_results
 
+    # Add dummy blanks after the last source
+    dummy_blanks = """
+<div style="margin: 20px 0;"></div>
+<div style="margin: 20px 0;"></div>
+<div style="margin: 20px 0;"></div>
+<div style="margin: 20px 0;"></div>
+<div style="margin: 20px 0;"></div>
 
-def infer_image_and_get_boxes(image, confidence_threshold=0.6):
+"""
+
+    cited_text += sources_list + dummy_blanks
+    return cited_text
+
+def infer_image_and_get_boxes(image, confidence_threshold=0.8):
     results = model.predict(image)
-    boxes = [
+    return [
         (int(box.xyxy[0][0]), int(box.xyxy[0][1]), int(box.xyxy[0][2]), int(box.xyxy[0][3]), int(box.cls[0]))
         for result in results for box in result.boxes
         if int(box.cls[0]) in {figure_class_index, table_class_index} and box.conf[0] > confidence_threshold
     ]
-    return boxes
 
 def crop_images_from_boxes(image, boxes, scale_factor):
     figures = []
@@ -223,7 +185,6 @@ def crop_images_from_boxes(image, boxes, scale_factor):
             tables.append(cropped_img)
     return figures, tables
 
-
 def process_pdf(pdf_file_path):
     doc = fitz.open(pdf_file_path)
     all_figures = []
@@ -246,213 +207,173 @@ def process_pdf(pdf_file_path):
     
     return all_figures, all_tables
 
-# Set the page configuration for a modern look
+def image_to_base64(img):
+    buffered = io.BytesIO()
+    img = Image.fromarray(img)
+    img.save(buffered, format="PNG")
+    return base64.b64encode(buffered.getvalue()).decode()
+
+def on_btn_click():
+    del st.session_state.chat_history[:]
 
-# Set the page configuration for a modern look
-# Set the page configuration for a modern look
-st.set_page_config(page_title="PDF Reading Assistant", page_icon="📄", layout="wide")
+# Streamlit interface
+
+# Custom CSS for the file uploader
+uploadercss='''
+<style>
+    [data-testid='stFileUploader'] {
+        width: max-content;
+    }
+    [data-testid='stFileUploader'] section {
+        padding: 0;
+        float: left;
+    }
+    [data-testid='stFileUploader'] section > input + div {
+        display: none;
+    }
+    [data-testid='stFileUploader'] section + div {
+        float: right;
+        padding-top: 0;
+    }
+
+</style>
+'''
+
+st.set_page_config(page_title="PDF Reading Assistant", page_icon="📄")
+
+# Initialize chat history in session state if not already present
+if 'chat_history' not in st.session_state:
+    st.session_state.chat_history = []
+
+st.title("📄 PDF Reading Assistant")
+st.markdown("### Extract tables, figures, summaries, and answers from your PDF files easily.")
+chat_placeholder = st.empty()
+
+# File uploader for PDF
+uploaded_file = st.file_uploader("Upload a PDF", type="pdf")
+st.markdown(uploadercss, unsafe_allow_html=True)
+if uploaded_file:
+    file_path = save_uploaded_file(uploaded_file)
 
-# Add some custom CSS for a modern look
+    # Chat container where all messages will be displayed
+    chat_container = st.container()
+    user_input = st.chat_input("Ask a question about the pdf......", key="user_input")
+    with chat_container:
+        # Scrollable chat messages
+        for idx, chat in enumerate(st.session_state.chat_history):
+            if chat.get("user"):
+                message(chat["user"], is_user=True, allow_html=True, key=f"user_{idx}", avatar_style="initials", seed="user")
+            if chat.get("bot"):
+                message(chat["bot"], is_user=False, allow_html=True, key=f"bot_{idx}",seed="bot")
+
+        # Input area and buttons for user interaction
+        with st.form(key="chat_form", clear_on_submit=True,border=False):
+
+            col1, col2, col3 = st.columns([1, 1, 1])
+            with col1:
+                summary_button = st.form_submit_button("Generate Summary")
+            with col2:
+                extract_button = st.form_submit_button("Extract Tables and Figures")
+            with col3:
+                st.form_submit_button("Clear message", on_click=on_btn_click)
+
+            # Handle responses based on user input and button presses
+            if summary_button:
+                with st.spinner("Generating summary..."):
+                    summary = summarize_pdf(file_path)
+                st.session_state.chat_history.append({"user": "Generate Summary", "bot": summary})
+                st.rerun()
+
+            if extract_button:
+                with st.spinner("Extracting tables and figures..."):
+                    figures, tables = process_pdf(file_path)
+                    if figures:
+                        st.session_state.chat_history.append({"user": "Figures"})
+
+                        for idx, figure in enumerate(figures):
+                            figure_base64 = image_to_base64(figure)
+                            result_html = f'<img src="data:image/png;base64,{figure_base64}" style="width:100%; display:block;" alt="Figure {idx+1}"/>'
+                            st.session_state.chat_history.append({"bot": f"Figure {idx+1} {result_html}"})
+                    if tables:
+                        st.session_state.chat_history.append({"user": "Tables"})
+                        for idx, table in enumerate(tables):
+                            table_base64 = image_to_base64(table)
+                            result_html = f'<img src="data:image/png;base64,{table_base64}" style="width:100%; display:block;" alt="Table {idx+1}"/>'
+                            st.session_state.chat_history.append({"bot": f"Table {idx+1} {result_html}"})
+                st.rerun()
+
+            if user_input:
+                st.session_state.chat_history.append({"user": user_input, "bot": None})
+                with st.spinner("Processing..."):
+                    answer = qa_pdf(file_path, user_input)
+                st.session_state.chat_history[-1]["bot"] = answer
+                st.rerun()
+
+# Additional CSS and JavaScript to ensure the chat container is scrollable and scrolls to the bottom
 st.markdown("""
     <style>
-        /* Main background and padding */
-        .main {
-            background-color: #f8f9fa;
-            padding: 2rem;
-            font-family: 'Arial', sans-serif;
+        #chat-container {
+            max-height: 500px;
+            overflow-y: auto;
+            padding: 1rem;
+            border: 1px solid #ddd;
+            border-radius: 8px;
+            background-color: #fefefe;
+            box-shadow: 0 2px 4px rgba(0, 0, 0, 0.1);
+            transition: background-color 0.3s ease;
         }
-        
-        /* Section headers */
-        .section-header {
-            font-size: 2rem;
-            font-weight: bold;
-            color: #343a40;
-            margin-top: 2rem;
-            margin-bottom: 1rem;
-            text-align: center;
-            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+        #chat-container:hover {
+            background-color: #f9f9f9;
         }
-
-        /* Containers */
-        .uploaded-file-container, .chat-container, .summary-container, .extract-container {
-            padding: 2rem;
-            background-color: #ffffff;
-            border-radius: 10px;
-            margin-bottom: 2rem;
-            box-shadow: 0 4px 12px rgba(0, 0, 0, 0.1);
+        .stChatMessage {
+            padding: 0.75rem;
+            margin: 0.75rem 0;
+            border-radius: 8px;
+            box-shadow: 0 1px 3px rgba(0, 0, 0, 0.1);
+            transition: background-color 0.3s ease;
         }
-
-        /* Buttons */
-        .stButton>button {
-            background-color: #007bff;
-            color: white;
-            padding: 0.6rem 1.2rem;
-            border-radius: 5px;
-            border: none;
-            cursor: pointer;
-            font-size: 1rem;
-            transition: background-color 0.3s ease, transform 0.3s ease;
+        .stChatMessage--user {
+            background-color: #E3F2FD;
         }
-        .stButton>button:hover {
-            background-color: #0056b3;
-            transform: translateY(-2px);
+        .stChatMessage--user:hover {
+            background-color: #BBDEFB;
         }
-
-        /* Chat messages */
-        .chat-message {
-            padding: 1rem;
-            border-radius: 10px;
-            margin-bottom: 1rem;
-            font-size: 1rem;
-            transition: all 0.3s ease;
-            box-shadow: 0 2px 8px rgba(0, 0, 0, 0.1);
-        }
-        .chat-message.user {
-            background-color: #e6f7ff;
-            border-left: 5px solid #007bff;
-            text-align: left;
+        .stChatMessage--bot {
+            background-color: #EDE7F6;
         }
-        .chat-message.bot {
-            background-color: #fff0f1;
-            border-left: 5px solid #dc3545;
-            text-align: left;
-        }
-
-        /* Input area */
-        .input-container {
-            display: flex;
-            align-items: center;
-            gap: 10px;
-            margin-top: 1rem;
+        .stChatMessage--bot:hover {
+            background-color: #D1C4E9;
         }
-        .input-container textarea {
-            border: 2px solid #ccc;
-            border-radius: 10px;
-            padding: 10px;
+        textarea {
             width: 100%;
-            background-color: #fff;
-            transition: border-color 0.3s ease;
-            margin: 0;
-            font-size: 1rem;
+            padding: 1rem;
+            border: 1px solid #ddd;
+            border-radius: 8px;
+            box-shadow: inset 0 1px 3px rgba(0, 0, 0, 0.1);
+            transition: border-color 0.3s ease, box-shadow 0.3s ease;
         }
-        .input-container textarea:focus {
-            border-color: #007bff;
-            outline: none;
+        textarea:focus {
+            border-color: #4CAF50;
+            box-shadow: 0 0 5px rgba(76, 175, 80, 0.5);
         }
-        .input-container button {
-            background-color: #007bff;
+        .stButton > button {
+            width: 100%;
+            background-color: #4CAF50;
             color: white;
-            padding: 0.6rem 1.2rem;
-            border-radius: 5px;
-            border: none;
-            cursor: pointer;
-            font-size: 1rem;
-            transition: background-color 0.3s ease, transform 0.3s ease;
-        }
-        .input-container button:hover {
-            background-color: #0056b3;
-            transform: translateY(-2px);
-        }
-
-        /* Expander */
-        .st-expander {
             border: none;
-            box-shadow: 0 4px 12px rgba(0, 0, 0, 0.1);
-            margin-bottom: 2rem;
+            border-radius: 8px;
+            padding: 0.75rem;
+            font-size: 16px;
+            box-shadow: 0 2px 4px rgba(0, 0, 0, 0.1);
+            transition: background-color 0.3s ease, box-shadow 0.3s ease;
         }
-
-        /* Markdown elements */
-        .stMarkdown {
-            font-size: 1rem;
-            color: #343a40;
-            line-height: 1.6;
-        }
-        
-        /* Titles and subtitles */
-        .stTitle {
-            color: #343a40;
-            text-align: center;
-            margin-bottom: 1rem;
-            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
-        }
-        .stSubtitle {
-            color: #6c757d;
-            text-align: center;
-            margin-bottom: 1rem;
-            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+        .stButton > button:hover {
+            background-color: #45A049;
+            box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
         }
     </style>
+    <script>
+        const chatContainer = document.getElementById('chat-container');
+        chatContainer.scrollTop = chatContainer.scrollHeight;
+    </script>
 """, unsafe_allow_html=True)
-
-# Streamlit interface
-# Streamlit interface
-st.title("📄 PDF Reading Assistant")
-st.markdown("### Extract tables, figures, summaries, and answers from your PDF files easily.")
-
-uploaded_file = st.file_uploader("Upload a PDF", type="pdf")
-if uploaded_file:
-    file_path = save_uploaded_file(uploaded_file)
-
-    if 'figures' not in st.session_state:
-        st.session_state['figures'] = None
-    if 'tables' not in st.session_state:
-        st.session_state['tables'] = None
-    if 'summary' not in st.session_state:
-        st.session_state['summary'] = None
-
-    with st.container():
-        st.markdown("<div class='section-header'>Extract Tables and Figures</div>", unsafe_allow_html=True)
-        with st.expander("Click to Extract Tables and Figures", expanded=True):
-            with st.container():
-                extract_button = st.button("Extract")
-                if extract_button:
-                    figures, tables = process_pdf(file_path)
-                    st.session_state['figures'] = figures
-                    st.session_state['tables'] = tables
-
-                if st.session_state['figures']:
-                    col1, col2 = st.columns(2)
-                    with col1:
-                        st.write("### Figures")
-                        for figure in st.session_state['figures']:
-                            st.image(figure, use_column_width=True)
-                    with col2:
-                        st.write("### Tables")
-                        for table in st.session_state['tables']:
-                            st.image(table, use_column_width=True)
-                else:
-                    st.write("No figures or tables found.")
-    
-    with st.container():
-        st.markdown("<div class='section-header'>Get Summary</div>", unsafe_allow_html=True)
-        with st.expander("Click to Generate Summary", expanded=True):
-            with st.container():
-                summary_button = st.button("Generate Summary")
-                if summary_button:
-                    summary = summarize_pdf(file_path)
-                    st.session_state['summary'] = summary
-
-                if st.session_state['summary']:
-                    st.markdown(st.session_state['summary'], unsafe_allow_html=True)
-
-    with st.container():
-        st.markdown("<div class='section-header'>Chat with your PDF</div>", unsafe_allow_html=True)
-        st.write("### Chat with your PDF")
-        if 'chat_history' not in st.session_state:
-            st.session_state['chat_history'] = []
-
-        for chat in st.session_state['chat_history']:
-            chat_user_class = "user" if chat["user"] else ""
-            chat_bot_class = "bot" if chat["bot"] else ""
-            st.markdown(f"<div class='chat-message {chat_user_class}'>{chat['user']}</div>", unsafe_allow_html=True)
-            st.markdown(f"<div class='chat-message {chat_bot_class}'>{chat['bot']}</div>", unsafe_allow_html=True)
-
-        with st.form(key="chat_form", clear_on_submit=True):
-            user_input = st.text_area("Ask a question about the PDF:", key="user_input")
-            submit_button = st.form_submit_button(label="Send")
-
-            if submit_button and user_input:
-                st.session_state['chat_history'].append({"user": user_input, "bot": None})
-                answer = qa_pdf(file_path, user_input)
-                st.session_state['chat_history'][-1]["bot"] = answer
-                st.rerun()