Update app.py

app.py

@@ -4,14 +4,15 @@ import numpy as np
 import os
 import matplotlib.pyplot as plt
 import seaborn as sns
-from dotenv import load_dotenv
+from fuzzywuzzy import process
+from sklearn.cluster import KMeans
 from langchain_openai import ChatOpenAI
 from langchain.chains import LLMChain
 from langchain.prompts import PromptTemplate
+from dotenv import load_dotenv
 from pycaret.regression import setup as setup_reg, compare_models, predict_model
 from pandasai import SmartDatalake
 from pandasai.llm import OpenAI
-from crewai import Crew, Agent, Task
 
 # ✅ Load Environment Variables
 load_dotenv()
@@ -25,6 +26,21 @@ 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"),
+}
+
+# ✅ Load competitors data
+competitor_data = {}
+for mall, file in competitor_files.items():
+    if os.path.exists(file):
+        competitor_data[mall] = pd.read_csv(file, encoding="ISO-8859-1")
+
 # ✅ Store Categorization Dictionary
 store_categories = {
     "E-commerce": ["6th Street.com"],
@@ -49,64 +65,69 @@ store_categories = {
                "REN by Sushi Library", "Sushi Library", "Tim Hortons", "LM Exchange"]
 }
 
-# ✅ CrewAI Agents & Tasks with Backstories
-cleaning_agent = Agent(
-    role="Data Cleaning Expert",
-    goal="Ensure data quality by handling missing values and duplicates.",
-    backstory="A seasoned data scientist who specializes in cleaning and structuring large datasets.",
-    llm=ChatOpenAI(model="gpt-4"),
-    allow_delegation=True
-)
-
-competitor_agent = Agent(
-    role="Competitor Analyst",
-    goal="Fetch competitors for the selected mall or generate AI-based fallback.",
-    backstory="A retail market expert who analyzes competitors and provides key insights for store replacements.",
-    llm=ChatOpenAI(model="gpt-4"),
-    allow_delegation=True
-)
-
-replacement_agent = Agent(
-    role="Store Replacement Specialist",
-    goal="Find the best replacement stores matching the selected store’s category.",
-    backstory="An AI-powered retail strategist with expertise in category-based store recommendations.",
-    llm=ChatOpenAI(model="gpt-4"),
-    allow_delegation=True
-)
-
-prediction_agent = Agent(
-    role="AI KPI Forecaster",
-    goal="Use AutoML to predict sales, footfall, conversion rate, and margins.",
-    backstory="A machine learning expert trained in forecasting business metrics using the best models.",
-    llm=ChatOpenAI(model="gpt-4"),
-    allow_delegation=True
-)
-
-chatbot_agent = Agent(
-    role="Conversational Data Assistant",
-    goal="Answer business-related queries about the dataset.",
-    backstory="An AI-powered assistant trained to answer complex retail analytics questions based on data.",
-    llm=ChatOpenAI(model="gpt-4"),
-    allow_delegation=True
-)
-
-crew = Crew(agents=[cleaning_agent, competitor_agent, replacement_agent, prediction_agent, chatbot_agent])
-
 # ✅ Streamlit Page Configuration
 st.set_page_config(page_title="AI Store Replacement & Chatbot", layout="wide")
 
 # ✅ Title
-st.title("🔄 AI-Powered Store Replacement, AutoML Demand Prediction & CrewAI")
+st.title("🔄 AI-Powered Store Replacement, AutoML Demand Prediction & Smart Data Lake AI")
 
-# ✅ Data Preview (First 10 Rows)
+# ✅ Data Preview (Only First 10 Rows)
 st.subheader("📊 Sample Data Preview - KPI Sheet")
 st.dataframe(df.head(10))
 
+# ✅ Store Replacement System
+st.subheader("🏬 AI-Based Store Replacement")
+
+# Select Mall
+selected_mall = st.selectbox("🏬 Select a Mall:", df["Mall"].unique())
+
+if selected_mall:
+    # Get all stores in the selected mall
+    mall_data = df[df["Mall"] == selected_mall]
+    
+    # Select the store that will be replaced
+    store_to_replace = st.selectbox("🚪 Select the store that is closing:", mall_data["Store_Brand"].unique())
+
+    if store_to_replace:
+        llm = ChatOpenAI(model="gpt-4", api_key=openai_api_key)
+
+        # ✅ Fuzzy Matching to Find the Correct Category
+        all_store_names = [store for category in store_categories.values() for store in category]
+        best_match, score = process.extractOne(store_to_replace, all_store_names)
+
+        store_category = None
+        if score > 80:  # Ensure strong match confidence
+            for category, stores in store_categories.items():
+                if best_match in stores:
+                    store_category = category
+                    break
+        else:
+            store_category = "Others"
+
+        st.write(f"🔍 **Identified Category:** {store_category} (Matched: {best_match}, Score: {score})")
+
+        # ✅ Find Replacement Stores Based on Performance
+        replacement_candidates = df[(df["Category"] == store_category) & (df["Annual_Sales"] > df["Annual_Sales"].median())]
+        top_replacements = replacement_candidates["Store_Brand"].unique()[:3]
+
+        if top_replacements.any():
+            st.success("✅ **Recommended Replacement Stores:**")
+            for replacement_store in top_replacements:
+                store_data = df[df["Store_Brand"] == replacement_store]
+
+                # ✅ AutoML Model Selection
+                setup_reg(df, target="Annual_Sales", silent=True, verbose=False)
+                best_model = compare_models()
+                projected_sales = predict_model(best_model, data=store_data)["Label"].mean()
+
+                # ✅ Display KPIs in the Green Section
+                st.subheader(f"📊 Projected KPIs for {replacement_store}")
+                st.write(f"🏆 **Best Model Used:** {best_model}")
+                st.write(f"📊 **Projected Sales:** AED {projected_sales:,.2f}")
+
 # ✅ AI-Powered Conversational Smart Data Lake
 st.subheader("💬 AI-Powered Data Chatbot")
-
-llm_sdl = OpenAI(api_token=openai_api_key)
-sdl = SmartDatalake([df], config={"llm": llm_sdl, "enable_cache": False})
+sdl = SmartDatalake([df], config={"llm": OpenAI(api_token=openai_api_key), "enable_cache": False})
 
 user_query = st.text_input("🔍 Ask AI About Your Data:")
 if user_query:
@@ -116,9 +137,3 @@ if user_query:
             st.markdown(f"🤖 **AI:** {response}")
         except Exception as e:
             st.error(f"❌ AI Error: {str(e)}")
-
-# ✅ CrewAI Execution
-st.subheader("🚀 AI Task Execution with CrewAI")
-st.write("Executing all AI agents...")
-crew.kickoff()
-st.success("✅ CrewAI execution completed!")