feat: updated location of files

app.py

@@ -94,56 +94,123 @@ def radar_chart(categories, values, amounts, title):
     
     return fig
 
+# Custom CSS for dark theme and styling
+st.markdown("""
+    <style>
+    body {
+        background-color: #121212;
+        color: white;
+    }
+    .stMetric {
+        background-color: #2E2E2E;
+        border-radius: 10px;
+        padding: 10px;
+    }
+    .stMetric > div {
+        color: white;
+    }
+    .stMetric > div > div {
+        color: white !important;
+    }
+    .custom-title {
+        color: white;
+        font-size: 36px;
+        margin-bottom: 20px;
+    }
+    .custom-metric-box {
+        background-color: #2E2E2E;
+        border-radius: 10px;
+        padding: 20px;
+        margin-bottom: 10px;
+        text-align: center;
+        color: white;
+    }
+    .up-arrow {
+        color: #00FF00;  /* Green */
+    }
+    .down-arrow {
+        color: #FF0000;  /* Red */
+    }
+    .container {
+        display: flex;
+        justify-content: space-between;
+    }
+    </style>
+""", unsafe_allow_html=True)
+
 # Navigation menu
 st.sidebar.title("DeepInsightz")
 page = st.sidebar.selectbox("Select the tool you want to use", ["Home", "Customer Analysis", "Articles Recommendations"])
 
-# Home Page
-# Home Page Design
 if page == "Home":
     st.title("Welcome to DeepInsightz")
     st.markdown("""
-        ### Personalized Sales Insights & Product Recommendations
-        Discover customer trends and boost your sales by understanding purchase patterns and getting tailored product suggestions.
+        ### Data-driven Customer Clustering
+        At DeepInsightz, we analyzed 4,000 customers, 400 suppliers, and over 800,000 sales lines to help businesses make smarter sales decisions.
+    """)
+
+    # 1. Quick Metrics for Data Preparation
+    st.subheader("Data Preparation Overview")
+    st.write("""
+        - **Customers Analyzed**: 4,000
+        - **Suppliers Analyzed**: 400
+        - **Invoice Lines Processed**: 800,000
     """)
 
-    # 1. Quick Metrics (Placeholder for now)
-    st.subheader("Quick Overview")
-    col1, col2, col3 = st.columns(3)
-    col1.metric("Total Customers", "1,532", "+3.8%")
-    col2.metric("Total Sales (2023)", "€2.9M", "+4.2%")
-    col3.metric("Top Cluster", "Cluster 3", "+1.5%")
+    st.info("The data was cleaned, duplicate entries were removed, and we standardized product codes and descriptions. We excluded customers with less than 12 purchases or total sales below €1,200 in 2024.")
 
-    # 2. 3D Cluster Model (Placeholder with random data)
-    st.subheader("Customer Segmentation (3D Clusters)")
-    st.write("Explore the different customer types based on purchase patterns.")
+    # 2. Customer Distribution
+    st.subheader("Customer Distribution by Purchase Volume")
+    
+    # Example data: distribution of customers by purchase volume
+    data = {
+        'Threshold (Items Purchased)': ['>1,000 items', '>5,000 items', '>10,000 items', '>20,000 items', '>50,000 items'],
+        'Number of Customers': [411, 180, 93, 47, 12],
+        'Percentage': [10, 4, 2, 1, 0.3]
+    }
+    df_customers = pd.DataFrame(data)
     
-    # Create random data for 3D scatter plot (replace this with your real data)
+    fig_customers = px.bar(df_customers, x='Threshold (Items Purchased)', y='Number of Customers',
+                           title="Distribution of Customers by Purchase Volume")
+    st.plotly_chart(fig_customers)
+
+    # 3. Interactive Cluster Visualization (placeholder)
+    st.subheader("3D Cluster Visualization")
+    st.write("Explore the customer clusters and their product purchase behavior.")
+
+    # Sample random data for a 3D cluster (replace with actual cluster data)
     np.random.seed(42)
-    df_3d = pd.DataFrame({
-        'Cluster': np.random.choice(['Cluster 1', 'Cluster 2', 'Cluster 3', 'Cluster 4', 'Cluster 5', 'Cluster 6', 'Cluster 7'], 100),
+    df_cluster = pd.DataFrame({
+        'Cluster': np.random.choice(['Cluster 1', 'Cluster 2', 'Cluster 3', 'Cluster 4', 'Cluster 5'], 100),
         'x': np.random.randn(100),
         'y': np.random.randn(100),
         'z': np.random.randn(100)
     })
 
-    fig_3d = px.scatter_3d(df_3d, x='x', y='y', z='z', color='Cluster', title="Customer Clusters (3D View)")
-    st.plotly_chart(fig_3d)
+    fig_cluster = px.scatter_3d(df_cluster, x='x', y='y', z='z', color='Cluster', title="Customer Clusters (3D View)")
+    st.plotly_chart(fig_cluster)
 
-    # 3. Sales by Cluster (Example of a simple bar chart)
-    st.subheader("Sales by Cluster")
-    sales_data = {
-        'Cluster': ['Cluster 1', 'Cluster 2', 'Cluster 3', 'Cluster 4', 'Cluster 5', 'Cluster 6', 'Cluster 7'],
-        'Sales': [500000, 400000, 600000, 350000, 450000, 550000, 300000]
+    # 4. Product Groups vs Manufacturer
+    st.subheader("Product Group vs Manufacturer Distribution")
+    st.write("See how different manufacturers align with various product groups purchased by customers.")
+    
+    # Example data for product groups vs manufacturers (replace with real data)
+    prod_data = {
+        'Product Group': ['Group A', 'Group B', 'Group C', 'Group D', 'Group E'],
+        'Manufacturer 1': [25, 40, 35, 30, 50],
+        'Manufacturer 2': [30, 35, 30, 45, 40],
+        'Manufacturer 3': [35, 25, 40, 25, 35]
     }
-    df_sales = pd.DataFrame(sales_data)
-    fig_sales = px.bar(df_sales, x='Cluster', y='Sales', title="Total Sales by Cluster (2023)")
-    st.plotly_chart(fig_sales)
+    df_prod_manu = pd.DataFrame(prod_data)
+    
+    fig_prod_manu = px.bar(df_prod_manu, x='Product Group', y=['Manufacturer 1', 'Manufacturer 2', 'Manufacturer 3'],
+                           title="Manufacturer Distribution Across Product Groups")
+    st.plotly_chart(fig_prod_manu)
 
-    # 4. Call to Action
+    # 5. Call to Action
     st.markdown("""
-        ## Ready to Dive Deeper?
-        - Use the tools in the sidebar to explore **Customer Analysis** or get **Articles Recommendations**.
+        ## Ready to Explore the Customer Clusters Further?
+        - Use the tools in the sidebar to dive deeper into **Customer Analysis** or get **Product Recommendations**.
     """)
 
 # Customer Analysis Page