Update app.py

app.py

@@ -5,9 +5,9 @@ import os
 import matplotlib.pyplot as plt
 import seaborn as sns
 from sklearn.cluster import KMeans
-from statsmodels.tsa.holtwinters import ExponentialSmoothing
 from langchain_openai import ChatOpenAI
-from pandasai import SmartDatalake
+from langchain.chains import LLMChain
+from langchain.prompts import PromptTemplate
 from dotenv import load_dotenv
 
 # ✅ Load Environment Variables
@@ -17,34 +17,18 @@ openai_api_key = os.getenv("OPENAI_API_KEY")
 if not openai_api_key:
     st.error("❌ OpenAI API Key is missing! Set it in your environment variables.")
 
-# ✅ Auto-Load All Files from the Folder
-folder_path = os.getcwd()  # Uses current folder
-
-# Load KPI Excel File (Main store dataset)
+# ✅ Load KPI Excel File
+folder_path = os.getcwd()
 data_file = os.path.join(folder_path, "kpi.xlsx")
 df = pd.read_excel(data_file, engine="openpyxl")
 
-# Load Competitor Mapping Files
-competitor_files = {
-    "Deira City Centre": os.path.join(folder_path, "Competitor Mapping Doc Final1(DEIRA CITY CENTRE).csv"),
-    "Arabian Centre": os.path.join(folder_path, "Competitor Mapping Doc Final1(ARABIAN CENTRE).csv"),
-    "Al Ghurair Centre": os.path.join(folder_path, "Competitor Mapping Doc Final1(AL GHURAIR CENTRE).csv"),
-    "Ibn Battuta Mall": os.path.join(folder_path, "Competitor Mapping Doc Final1(IBN BATTUTA MALL).csv"),
-    "Dubai Hills": os.path.join(folder_path, "Competitor Mapping Doc Final1(DUBAI HILLS).csv"),
-}
-
-# Store all competitor data in a dictionary
-competitor_data = {}
-for mall, file in competitor_files.items():
-    competitor_data[mall] = pd.read_csv(file, encoding="ISO-8859-1")
-
 # ✅ Streamlit Page Configuration
-st.set_page_config(page_title="AI Store Replacement, Demand Prediction & Chatbot", layout="wide")
+st.set_page_config(page_title="AI Store Replacement & Chatbot", layout="wide")
 
 # ✅ Title
-st.title("🔄 AI-Powered Store Replacement, Demand Prediction & Chatbot")
+st.title("🔄 AI-Powered Store Replacement & LangChain Chatbot")
 
-# ✅ Data Preview Section
+# ✅ Data Preview
 st.subheader("📊 Full Data Preview - KPI Sheet")
 st.dataframe(df)
 
@@ -65,81 +49,74 @@ if selected_mall:
         llm = ChatOpenAI(model="gpt-4", api_key=openai_api_key)
 
         # ✅ Identify Store Category
-        category_prompt = f"Given that {store_to_replace} is a retail brand, what type of store is it? Categorize it broadly (e.g., Fashion, Electronics, Footwear, Food, Home Decor, etc.)."
-        store_category = llm.invoke(category_prompt).content
+        category_prompt = PromptTemplate.from_template(
+            "Given that {store_name} is a retail brand, what type of store is it? Categorize it broadly (e.g., Fashion, Electronics, Footwear, Food, Home Decor, etc.)."
+        )
+        category_chain = LLMChain(llm=llm, prompt=category_prompt)
+        store_category = category_chain.run(store_name=store_to_replace)
         st.write(f"🔍 **Identified Category:** {store_category}")
 
-        # ✅ Find 5 Competitors in the Same Mall
-        if selected_mall in competitor_data:
-            mall_competitor_data = competitor_data[selected_mall]
-            competitor_list = mall_competitor_data["Competitors in mall"].dropna().unique()
-
-            if len(competitor_list) > 5:
-                competitor_list = np.random.choice(competitor_list, 5, replace=False)
-
-            st.subheader("🔎 Competitors in This Mall")
-            for competitor in competitor_list:
-                st.write(f"- {competitor}")
-
-        # ✅ AI-Based Store Clustering (Finding Best Replacements)
-        store_features = df[["Annual_Sales", "Footfall", "Gross_Margin"]].dropna()
-        kmeans = KMeans(n_clusters=5, random_state=42)  # 5 Clusters
-        df["Cluster"] = kmeans.fit_predict(store_features)
-
-        # ✅ Find a replacement store in the same category (From kpi.xlsx, Not in this mall)
+        # ✅ Find Replacement Stores from kpi.xlsx
         all_stores = set(df["Store_Brand"].unique())
         existing_stores = set(mall_data["Store_Brand"].unique())
         potential_replacements = list(all_stores - existing_stores)
 
+        # ✅ Debugging: Print available replacements
+        st.write(f"🔍 **Available Replacement Stores:** {potential_replacements}")
+
         if potential_replacements:
-            replacement_prompt = f"Based on the category {store_category}, suggest 3 stores from the list below that are not already in {selected_mall}. The stores should match the mall’s customer base. Store List: {potential_replacements}"
-            replacement_response = llm.invoke(replacement_prompt).content
-            replacement_stores = replacement_response.split(", ")
-
-            st.success(f"🔄 **Recommended Replacement Stores:**")
-            for replacement_store in replacement_stores:
-                store_data = df[df["Store_Brand"] == replacement_store]
-                
-                if not store_data.empty:
-                    sales_history = store_data["Annual_Sales"]
-                    if len(sales_history) > 3:
-                        # ✅ Time-Series Forecasting for Demand Prediction
-                        model = ExponentialSmoothing(sales_history, trend="add", seasonal="add", seasonal_periods=3)
-                        fit = model.fit()
-                        forecast_sales = fit.forecast(1).values[0]  # Predict next year's sales
-
-                        expected_footfall = store_data["Footfall"].mean()
-                        expected_conversion = store_data["Conversion_Rate"].mean()
-                        expected_margin = store_data["Gross_Margin"].mean()
+            replacement_prompt = PromptTemplate.from_template(
+                "Based on the category {store_category}, suggest 3 stores from the list below that are not already in {mall_name}. The stores should match the mall’s customer base. Store List: {store_list}"
+            )
+            replacement_chain = LLMChain(llm=llm, prompt=replacement_prompt)
+            replacement_response = replacement_chain.run(
+                store_category=store_category, mall_name=selected_mall, store_list=potential_replacements
+            )
+
+            # ✅ Debugging: Print AI Response
+            st.write(f"🧠 **AI Raw Response:** {replacement_response}")
+
+            # Ensure AI response is formatted correctly
+            replacement_stores = [store.strip() for store in replacement_response.split("\n") if store.strip()]
+
+            if replacement_stores:
+                st.success(f"🔄 **Recommended Replacement Stores:**")
+                for replacement_store in replacement_stores[:3]:
+                    store_data = df[df["Store_Brand"] == replacement_store]
+
+                    if not store_data.empty:
+                        projected_sales = store_data["Annual_Sales"].mean()
+                        projected_footfall = store_data["Footfall"].mean()
+                        projected_margin = store_data["Gross_Margin"].mean()
 
                         st.write(f"**{replacement_store}**")
-                        st.write(f"📊 **Projected Sales:** AED {forecast_sales:,.2f}")
-                        st.write(f"👣 **Projected Footfall:** {expected_footfall:,.0f} visitors")
-                        st.write(f"💰 **Projected Margin:** AED {expected_margin:,.2f}")
-                        st.write(f"🔄 **Projected Conversion Rate:** {expected_conversion:.2f}%")
+                        st.write(f"📊 **Projected Sales:** AED {projected_sales:,.2f}")
+                        st.write(f"👣 **Projected Footfall:** {projected_footfall:,.0f} visitors")
+                        st.write(f"💰 **Projected Margin:** AED {projected_margin:,.2f}")
 
                         # ✅ Justification
-                        justification_prompt = f"Explain why {replacement_store} is a good replacement for {store_to_replace} in {selected_mall}. Include numeric KPIs: expected sales, footfall, conversion rate, and profit margins."
-                        justification = llm.invoke(justification_prompt).content
+                        justification_prompt = PromptTemplate.from_template(
+                            "Explain why {replacement_store} is a good replacement for {store_to_replace} in {mall_name}. "
+                            "Include numeric KPIs: expected sales, footfall, conversion rate, and profit margins."
+                        )
+                        justification_chain = LLMChain(llm=llm, prompt=justification_prompt)
+                        justification = justification_chain.run(
+                            replacement_store=replacement_store, store_to_replace=store_to_replace, mall_name=selected_mall
+                        )
+
                         st.write(f"🤖 **AI Justification:** {justification}")
 
-# ✅ AI Chatbot for Dynamic Insights
+# ✅ AI Chatbot using LangChain
 st.subheader("💬 AI-Powered Chatbot")
 
-# AI Chatbot using PandasAI
-llm_pandas = ChatOpenAI(model="gpt-3.5-turbo", api_key=openai_api_key)
-sdl = SmartDatalake([df], config={"llm": llm_pandas, "enable_cache": False})
-
 user_query = st.text_input("🔍 Ask AI About Your Data:")
 if user_query:
-    with st.spinner("🤖 AI is thinking..."):
-        try:
-            response = sdl.chat(user_query)
-            st.markdown(f"🤖 **AI:** {response}")
-
-        except Exception as e:
-            st.error(f"❌ AI Error: {str(e)}")
+    chat_prompt = PromptTemplate.from_template("Answer this retail business question based on the dataset: {question}")
+    chat_chain = LLMChain(llm=llm, prompt=chat_prompt)
+    chat_response = chat_chain.run(question=user_query)
+    
+    st.markdown(f"🤖 **AI Response:** {chat_response}")
 
 # ✅ Footer
 st.markdown("---")
-st.markdown("💡 Built with ❤️ using **AI-Powered Store Replacement & Demand Prediction**")
+st.markdown("💡 Built with ❤️ using **LangChain AI for Store Replacement & Demand Prediction**")