Update app.py

app.py

@@ -1,108 +1,98 @@
 import os
-import csv
 import streamlit as st
+import pandas as pd
 from langchain_community.vectorstores import FAISS
 from langchain_community.embeddings import HuggingFaceEmbeddings
 from langchain_huggingface import HuggingFaceEndpoint
 from langchain.prompts import PromptTemplate
-from langchain.chains import LLMChain, RetrievalQA
+from langchain.chains import LLMChain
 from huggingface_hub import login
+from langchain_community.document_loaders import TextLoader
+from langchain_text_splitters import CharacterTextSplitter
+from langchain_community.document_loaders import PyPDFLoader
+from langchain.chains import RetrievalQA
 
-# Login to Hugging Face
+# Authenticate with Hugging Face
 login(token=st.secrets["HF_TOKEN"])
 
-# Load FAISS index and ensure it only happens once
-if 'db' not in st.session_state:
-    st.session_state.db = FAISS.load_local(
-        "faiss_index", 
-        HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L12-v2'),
-        allow_dangerous_deserialization=True
-    )
-
-# Use session state for retriever
-retriever = st.session_state.db.as_retriever(
-    search_type="mmr",
-    search_kwargs={'k': 1}
-)
+# Load FAISS index
+db = FAISS.load_local("faiss_index", HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L12-v2'), allow_dangerous_deserialization=True)
+
+# Set up retriever
+retriever = db.as_retriever(search_type="mmr", search_kwargs={'k': 1})
 
-# Define prompt template
+# Prompt template for the LLM
 prompt_template = """
 ### [INST]
-Instruction: You are a Q&A assistant. Your goal is to answer questions as accurately as possible based on the instructions and context provided without using prior knowledge. You answer in FRENCH
-Analyse carefully the context and provide a direct answer based on the context. If the user said Bonjour or Hello your only answer will be Hi! comment puis-je vous aider?
-Answer in french only
+Instruction: You are a Q&A assistant. Your goal is to answer questions as accurately as possible based on the instructions and context provided without using prior knowledge. You answer in FRENCH.
+        Analyse carefully the context and provide a direct answer based on the context. If the user said Bonjour or Hello, your only answer will be Hi! comment puis-je vous aider?
+Answer in French only.
 
 {context}
 Vous devez répondre aux questions en français.
+
 ### QUESTION:
 {question}
 [/INST]
-Answer in french only
+Answer in French only.
 Vous devez répondre aux questions en français.
 """
 
+# Set up the LLM from Hugging Face
 repo_id = "mistralai/Mistral-7B-Instruct-v0.3"
 
-# Load the model only once
-if 'mistral_llm' not in st.session_state:
-    st.session_state.mistral_llm = HuggingFaceEndpoint(
-        repo_id=repo_id, 
-        max_length=2048, 
-        temperature=0.05, 
-        huggingfacehub_api_token=st.secrets["HF_TOKEN"]
-    )
+mistral_llm = HuggingFaceEndpoint(
+    repo_id=repo_id, max_length=2048, temperature=0.05, huggingfacehub_api_token=st.secrets["HF_TOKEN"]
+)
 
-# Create prompt and LLM chain
+# Create prompt from prompt template
 prompt = PromptTemplate(
     input_variables=["question"],
     template=prompt_template,
 )
-llm_chain = LLMChain(llm=st.session_state.mistral_llm, prompt=prompt)
 
-# Create QA chain
+# Create LLM chain
+llm_chain = LLMChain(llm=mistral_llm, prompt=prompt)
+
+# Set up RetrievalQA chain
+retriever.search_kwargs = {'k': 1}
 qa = RetrievalQA.from_chain_type(
-    llm=st.session_state.mistral_llm,
+    llm=mistral_llm,
     chain_type="stuff",
     retriever=retriever,
     chain_type_kwargs={"prompt": prompt},
 )
 
-import psycopg2
-import streamlit as st
-from datetime import datetime
-import os
-
-# PostgreSQL connection setup using secrets from Hugging Face Spaces
-def create_connection():
-    conn = psycopg2.connect(
-        host=os.getenv("DB_HOST"),
-        database=os.getenv("DB_NAME"),
-        user=os.getenv("DB_USER"),
-        password=os.getenv("DB_PASSWORD"),
-        port=os.getenv("DB_PORT")
-    )
-    return conn
-
-def create_table(conn):
-    with conn.cursor() as cur:
-        cur.execute('''
-        CREATE TABLE IF NOT EXISTS feedback (
-            id SERIAL PRIMARY KEY,
-            timestamp TIMESTAMP NOT NULL,
-            rating INTEGER NOT NULL,
-            comment TEXT NOT NULL
-        );
-        ''')
-        conn.commit()
-
-# Streamlit interface with improved aesthetics
+# Streamlit interface setup
 st.set_page_config(page_title="Alter-IA Chat", page_icon="🤖")
 
-# Define function to handle user input and display chatbot response
+# Function to handle user input and display chatbot response
 def chatbot_response(user_input):
     response = qa.run(user_input)
     return response
 
+# Function to save user feedback
+def save_feedback(user_input, bot_response, rating, comment):
+    feedback = {
+        "user_input": user_input,
+        "bot_response": bot_response,
+        "rating": rating,
+        "comment": comment
+    }
+    
+    # Check if the feedback file exists
+    feedback_file = "feedback.csv"
+    if os.path.exists(feedback_file):
+        # Load existing feedback and append new feedback
+        feedback_df = pd.read_csv(feedback_file)
+        feedback_df = feedback_df.append(feedback, ignore_index=True)
+    else:
+        # Create a new dataframe for the feedback
+        feedback_df = pd.DataFrame([feedback])
+    
+    # Save feedback to CSV
+    feedback_df.to_csv(feedback_file, index=False)
+
 # Create columns for logos
 col1, col2, col3 = st.columns([2, 3, 2])
 
@@ -112,40 +102,13 @@ with col1:
 with col3:
     st.image("Altereo logo 2023 original - eau et territoires durables.png", width=150, use_column_width=True)
 
-# CSS for styling
+# Adding centered header and subtitle
 st.markdown("""
     <style>
-    .centered-text {
-        text-align: center;
-    }
-    .centered-orange-text {
-        text-align: center;
-        color: darkorange;
-    }
-    .star-rating {
-        display: flex;
-        flex-direction: row-reverse;
-        justify-content: center;
-        cursor: pointer;
-    }
-    .star-rating input[type="radio"] {
-        display: none;
-    }
-    .star-rating label {
-        font-size: 2em;
-        color: #ddd;
-        padding: 0 5px;
-        transition: color 0.3s;
-    }
-    .star-rating input[type="radio"]:checked ~ label {
-        color: gold;
-    }
-    .star-rating input[type="radio"]:hover ~ label {
-        color: gold;
-    }
+    .centered-text { text-align: center; }
+    .centered-orange-text { text-align: center; color: darkorange; }
     </style>
-    """, unsafe_allow_html=True)
-
+""", unsafe_allow_html=True)
 st.markdown('<h3 class="centered-text">🤖 AlteriaChat 🤖 </h3>', unsafe_allow_html=True)
 st.markdown('<p class="centered-orange-text">"Votre Réponse à Chaque Défi Méthodologique "</p>', unsafe_allow_html=True)
 
@@ -153,54 +116,32 @@ st.markdown('<p class="centered-orange-text">"Votre Réponse à Chaque Défi Mé
 user_input = st.text_input("You:", "")
 submit_button = st.button("Ask 📨")
 
-if submit_button:
-    if user_input.strip() != "":
-        bot_response = chatbot_response(user_input)
-        st.markdown("### Bot:")
-        st.text_area("", value=bot_response, height=300)
-
-        # Add rating and comment section
-        st.markdown("---")
-        st.markdown("#### Rate the Response:")
-        
-        # Custom star rating HTML
-        rating_html = """
-        <div class="star-rating">
-            <input type="radio" id="5-stars" name="rating" value="5"><label for="5-stars">★</label>
-            <input type="radio" id="4-stars" name="rating" value="4"><label for="4-stars">★</label>
-            <input type="radio" id="3-stars" name="rating" value="3" checked><label for="3-stars">★</label>
-            <input type="radio" id="2-stars" name="rating" value="2"><label for="2-stars">★</label>
-            <input type="radio" id="1-star" name="rating" value="1"><label for="1-star">★</label>
-        </div>
-        """
-        st.markdown(rating_html, unsafe_allow_html=True)
-        
-        # Get the selected rating via JavaScript
-        rating = st.text_input("Selected Rating:", value="3", key="rating_input", label_visibility="hidden")
-
-        comment = st.text_area("Your Comment:")
-
-        # Submit feedback
-        feedback_button = st.button("Submit Feedback")
-
-        if feedback_button:
-            if comment.strip() == "":
-                st.warning("⚠ Please provide a comment.")
-            else:
-                st.success("Thank you for your feedback!")
-                
-                # Store feedback in PostgreSQL
-                conn = create_connection()
-                with conn.cursor() as cur:
-                    timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
-                    cur.execute('INSERT INTO feedback (timestamp, rating, comment) VALUES (%s, %s, %s)', 
-                                (timestamp, int(rating), comment))
-                    conn.commit()
-                conn.close()
-
-    else:
-        st.warning("⚠ Please enter a message.")
+# Handle user input and display response
+if submit_button and user_input.strip():
+    bot_response = chatbot_response(user_input)
+    st.markdown("### Bot:")
+    st.text_area("", value=bot_response, height=300)
+    
+    # Star rating system
+    st.markdown("### How would you rate the response?")
+    rating = st.slider("Rate from 1 star to 5 stars", min_value=1, max_value=5, value=3)
+    
+    # Comment section
+    comment = st.text_area("Any comments or suggestions for improvement?", "")
+    
+    # Save feedback when the user submits a rating and comment
+    if st.button("Submit Feedback"):
+        save_feedback(user_input, bot_response, rating, comment)
+        st.success("Thank you for your feedback!")
 
 # Motivational quote at the bottom
 st.markdown("---")
-st.markdown("La collaboration est la clé du succès. Chaque question trouve sa réponse, chaque défi devient une opportunité.")
+st.markdown("La collaboration est la clé du succès. Chaque question trouve sa réponse, chaque défi devient une opportunité.")
+
+# Section for the developer to review feedback
+if st.checkbox("Show Feedback (Developer Only)"):
+    if os.path.exists("feedback.csv"):
+        feedback_df = pd.read_csv("feedback.csv")
+        st.dataframe(feedback_df)
+    else:
+        st.warning("No feedback available yet.")