Add frontend/streamlit_app.py

frontend/streamlit_app.py

@@ -0,0 +1,388 @@
+"""
+Streamlit Frontend - Agentic BI Dashboard
+==========================================
+Multi-page dashboard with:
+- KPI Overview
+- Agentic Chat Interface
+- Real-time Monitoring
+- Automated Reports
+"""
+
+import streamlit as st
+import pandas as pd
+import plotly.express as px
+import plotly.graph_objects as go
+from datetime import datetime, timedelta
+import os
+
+# ============================================================
+# PAGE CONFIG
+# ============================================================
+
+st.set_page_config(
+    page_title="🤖 Agentic BI - E-commerce Analytics",
+    page_icon="🤖",
+    layout="wide",
+    initial_sidebar_state="expanded"
+)
+
+# ============================================================
+# SIDEBAR
+# ============================================================
+
+with st.sidebar:
+    st.image("https://img.icons8.com/3d-fluency/94/robot-2.png", width=80)
+    st.title("Agentic BI")
+    st.markdown("---")
+    
+    page = st.radio(
+        "Navigation",
+        ["📊 KPI Dashboard", "💬 AI Analytics Chat", "⚡ Real-time Monitor", "📋 Reports"],
+        index=0
+    )
+    
+    st.markdown("---")
+    st.markdown("### System Status")
+    
+    col1, col2 = st.columns(2)
+    with col1:
+        st.metric("Kafka", "✅ Online")
+    with col2:
+        st.metric("PostgreSQL", "✅ Online")
+    
+    col1, col2 = st.columns(2)
+    with col1:
+        st.metric("ClickHouse", "✅ Online")
+    with col2:
+        st.metric("Agent", "✅ Ready")
+    
+    st.markdown("---")
+    st.caption("Agentic BI v1.0 | Olist E-commerce Dataset")
+    st.caption(f"Last updated: {datetime.now().strftime('%Y-%m-%d %H:%M')}")
+
+
+# ============================================================
+# PAGE: KPI DASHBOARD
+# ============================================================
+
+if page == "📊 KPI Dashboard":
+    st.title("📊 E-commerce KPI Dashboard")
+    st.markdown("Real-time business metrics from the Olist marketplace")
+    
+    # Top KPI Cards
+    col1, col2, col3, col4, col5 = st.columns(5)
+    
+    with col1:
+        st.metric(
+            "💰 Total GMV",
+            "R$ 15.4M",
+            "+12.3%",
+            help="Gross Merchandise Value = price + freight"
+        )
+    with col2:
+        st.metric(
+            "📦 Total Orders",
+            "99,441",
+            "+8.5%",
+        )
+    with col3:
+        st.metric(
+            "💵 Avg Order Value",
+            "R$ 154.78",
+            "+3.2%",
+        )
+    with col4:
+        st.metric(
+            "🚚 On-time Delivery",
+            "93.2%",
+            "-1.1%",
+            delta_color="inverse"
+        )
+    with col5:
+        st.metric(
+            "⭐ Avg Review Score",
+            "4.09",
+            "+0.05",
+        )
+    
+    st.markdown("---")
+    
+    # Charts Row 1
+    col1, col2 = st.columns(2)
+    
+    with col1:
+        st.subheader("📈 Monthly Revenue Trend")
+        # Sample data for demonstration
+        months = pd.date_range('2017-01', '2018-10', freq='M')
+        revenue = [380000, 450000, 520000, 490000, 560000, 610000, 
+                   580000, 640000, 750000, 820000, 1200000, 680000,
+                   720000, 780000, 850000, 820000, 910000, 980000,
+                   1050000, 1100000, 1350000, 900000]
+        
+        fig = px.line(
+            x=months, y=revenue,
+            labels={'x': 'Month', 'y': 'Revenue (BRL)'},
+            template='plotly_white'
+        )
+        fig.update_traces(line_color='#E53935', line_width=3)
+        fig.update_layout(height=350)
+        st.plotly_chart(fig, use_container_width=True)
+    
+    with col2:
+        st.subheader("🗺️ Revenue by State")
+        states = ['SP', 'RJ', 'MG', 'RS', 'PR', 'SC', 'BA', 'DF', 'ES', 'GO']
+        state_rev = [5200000, 2100000, 1800000, 1200000, 1100000, 
+                     800000, 700000, 600000, 500000, 400000]
+        
+        fig = px.bar(
+            x=states, y=state_rev,
+            labels={'x': 'State', 'y': 'Revenue (BRL)'},
+            template='plotly_white',
+            color=state_rev,
+            color_continuous_scale='Reds'
+        )
+        fig.update_layout(height=350, showlegend=False)
+        st.plotly_chart(fig, use_container_width=True)
+    
+    # Charts Row 2
+    col1, col2, col3 = st.columns(3)
+    
+    with col1:
+        st.subheader("💳 Payment Methods")
+        payment_data = {
+            'Type': ['Credit Card', 'Boleto', 'Voucher', 'Debit Card'],
+            'Count': [73886, 19784, 5775, 1529]
+        }
+        fig = px.pie(
+            payment_data, names='Type', values='Count',
+            color_discrete_sequence=['#E53935', '#FF5722', '#FF9800', '#FFC107'],
+            template='plotly_white'
+        )
+        fig.update_layout(height=300)
+        st.plotly_chart(fig, use_container_width=True)
+    
+    with col2:
+        st.subheader("⭐ Review Distribution")
+        scores = [1, 2, 3, 4, 5]
+        counts = [11424, 3151, 8179, 19142, 57328]
+        colors = ['#E53935', '#FF5722', '#FF9800', '#FFC107', '#4CAF50']
+        
+        fig = px.bar(
+            x=scores, y=counts,
+            labels={'x': 'Score', 'y': 'Count'},
+            template='plotly_white',
+            color=scores,
+            color_discrete_sequence=colors
+        )
+        fig.update_layout(height=300, showlegend=False)
+        st.plotly_chart(fig, use_container_width=True)
+    
+    with col3:
+        st.subheader("📦 Top Categories")
+        categories = ['Bed/Bath/Table', 'Health/Beauty', 'Sports/Leisure', 
+                      'Furniture', 'Computers']
+        cat_rev = [1850000, 1520000, 1180000, 980000, 870000]
+        
+        fig = px.barh(
+            x=cat_rev, y=categories,
+            labels={'x': 'Revenue (BRL)', 'y': ''},
+            template='plotly_white',
+            color=cat_rev,
+            color_continuous_scale='Reds'
+        )
+        fig.update_layout(height=300, showlegend=False, yaxis={'categoryorder': 'total ascending'})
+        st.plotly_chart(fig, use_container_width=True)
+
+
+# ============================================================
+# PAGE: AI ANALYTICS CHAT
+# ============================================================
+
+elif page == "💬 AI Analytics Chat":
+    st.title("💬 AI-Powered Analytics Chat")
+    st.markdown("Ask questions about the e-commerce data in natural language")
+    
+    # Chat interface
+    if "messages" not in st.session_state:
+        st.session_state.messages = [
+            {
+                "role": "assistant",
+                "content": (
+                    "🤖 Xin chào! Tôi là Agentic BI Assistant.\n\n"
+                    "Tôi có thể giúp bạn phân tích dữ liệu e-commerce Olist. "
+                    "Hãy hỏi bất kỳ câu hỏi nào, ví dụ:\n\n"
+                    "- *Doanh thu tháng này so với tháng trước?*\n"
+                    "- *Top 5 seller có hiệu suất cao nhất?*\n"
+                    "- *Tỷ lệ giao hàng trễ theo từng bang?*\n"
+                    "- *Phát hiện bất thường trong doanh thu tuần qua*\n"
+                    "- *Vẽ biểu đồ xu hướng đơn hàng theo tháng*"
+                )
+            }
+        ]
+    
+    # Display chat history
+    for message in st.session_state.messages:
+        with st.chat_message(message["role"]):
+            st.markdown(message["content"])
+    
+    # Chat input
+    if prompt := st.chat_input("Nhập câu hỏi của bạn..."):
+        st.session_state.messages.append({"role": "user", "content": prompt})
+        with st.chat_message("user"):
+            st.markdown(prompt)
+        
+        with st.chat_message("assistant"):
+            with st.spinner("🔄 Đang phân tích..."):
+                # Here we would call the orchestrator agent
+                # For demo, show a template response
+                response = (
+                    f"📊 **Phân tích cho câu hỏi:** *{prompt}*\n\n"
+                    "🔄 Agent đang xử lý...\n\n"
+                    "*(Kết nối với Orchestrator Agent để có kết quả thực tế)*"
+                )
+                st.markdown(response)
+                st.session_state.messages.append({"role": "assistant", "content": response})
+    
+    # Quick action buttons
+    st.markdown("---")
+    st.subheader("⚡ Quick Actions")
+    
+    col1, col2, col3, col4 = st.columns(4)
+    with col1:
+        if st.button("📈 Revenue Report"):
+            st.info("Generating revenue report...")
+    with col2:
+        if st.button("🔍 Anomaly Scan"):
+            st.info("Scanning for anomalies...")
+    with col3:
+        if st.button("🏪 Seller Analysis"):
+            st.info("Analyzing seller performance...")
+    with col4:
+        if st.button("👥 Customer Segments"):
+            st.info("Computing customer segments...")
+
+
+# ============================================================
+# PAGE: REAL-TIME MONITOR
+# ============================================================
+
+elif page == "⚡ Real-time Monitor":
+    st.title("⚡ Real-time Streaming Monitor")
+    st.markdown("Live metrics from Apache Kafka → Flink → ClickHouse pipeline")
+    
+    # Auto-refresh
+    auto_refresh = st.checkbox("Auto-refresh (5s)", value=False)
+    
+    col1, col2, col3, col4 = st.columns(4)
+    with col1:
+        st.metric("📨 Events/sec", "1,247", "+156")
+    with col2:
+        st.metric("📦 Orders (5min)", "43", "+5")
+    with col3:
+        st.metric("💰 Revenue (5min)", "R$ 6,720", "+R$ 890")
+    with col4:
+        st.metric("🔔 Active Alerts", "2", "-1", delta_color="inverse")
+    
+    st.markdown("---")
+    
+    col1, col2 = st.columns(2)
+    
+    with col1:
+        st.subheader("📊 Kafka Topic Throughput")
+        topics = ['orders.created', 'orders.items', 'orders.payments', 
+                  'orders.status', 'orders.delivered', 'reviews.submitted']
+        throughput = [245, 312, 198, 156, 89, 67]
+        
+        fig = px.bar(
+            x=topics, y=throughput,
+            labels={'x': 'Topic', 'y': 'Messages/min'},
+            template='plotly_white',
+            color=throughput,
+            color_continuous_scale='RdYlGn'
+        )
+        fig.update_layout(height=300, showlegend=False)
+        st.plotly_chart(fig, use_container_width=True)
+    
+    with col2:
+        st.subheader("🔔 Recent Anomaly Alerts")
+        alerts = pd.DataFrame({
+            'Time': ['14:05', '13:42', '12:15', '11:30', '10:45'],
+            'Type': ['Revenue Drop', 'Late Delivery Spike', 'Payment Failure',
+                     'Review Score Drop', 'Order Surge'],
+            'Severity': ['🔴 Critical', '🟡 Warning', '🟡 Warning', 
+                         '🟢 Info', '🟢 Info'],
+            'Status': ['Investigating', 'Resolved', 'Resolved', 
+                       'Monitoring', 'Resolved']
+        })
+        st.dataframe(alerts, use_container_width=True, hide_index=True)
+    
+    st.markdown("---")
+    st.subheader("📈 Pipeline Health")
+    
+    col1, col2, col3 = st.columns(3)
+    with col1:
+        st.markdown("**Kafka Consumer Lag**")
+        st.progress(15, text="15 messages behind")
+    with col2:
+        st.markdown("**Flink Job Status**")
+        st.progress(100, text="All 4 jobs running ✅")
+    with col3:
+        st.markdown("**ClickHouse Ingestion**")
+        st.progress(98, text="98% write success rate")
+
+
+# ============================================================
+# PAGE: REPORTS
+# ============================================================
+
+elif page == "📋 Reports":
+    st.title("📋 Automated Reports")
+    st.markdown("AI-generated business intelligence reports")
+    
+    st.subheader("📅 Scheduled Reports")
+    
+    reports = pd.DataFrame({
+        'Report': ['Daily Revenue Summary', 'Weekly Seller Performance', 
+                   'Monthly Customer Analysis', 'Anomaly Investigation Report',
+                   'Delivery SLA Report'],
+        'Schedule': ['Daily 8:00 AM', 'Weekly Monday', 'Monthly 1st',
+                     'On anomaly detection', 'Daily 6:00 AM'],
+        'Last Run': ['Today 08:00', 'Mon 08:00', 'Apr 01', 
+                     'Today 14:05', 'Today 06:00'],
+        'Status': ['✅ Generated', '✅ Generated', '✅ Generated',
+                   '🔄 Generating', '✅ Generated']
+    })
+    
+    st.dataframe(reports, use_container_width=True, hide_index=True)
+    
+    st.markdown("---")
+    
+    st.subheader("📄 Latest Report: Daily Revenue Summary")
+    
+    with st.expander("View Report", expanded=True):
+        st.markdown("""
+        ### 📊 Daily Revenue Summary - April 27, 2026
+        
+        **Generated by: Insight Agent (automatic)**
+        
+        #### Key Findings:
+        
+        1. **Revenue**: R$ 52,340 (+8.3% vs yesterday, +12.1% vs same day last week)
+        2. **Orders**: 342 orders processed (+15 vs yesterday)
+        3. **AOV**: R$ 153.04 (-2.1% vs yesterday — more small orders)
+        4. **Top Category**: Bed/Bath/Table (R$ 8,920, 17.0% of daily revenue)
+        5. **Delivery SLA**: 94.2% on-time (+1.1% improvement)
+        
+        #### 🔔 Alerts:
+        - ⚠️ Payment failures spiked at 14:00-14:05 (12 failures vs baseline 2)
+        - ℹ️ Seller #a1b2c3 had 0 orders today (usually averages 5/day)
+        
+        #### 📈 Recommendations:
+        - Investigate payment gateway issue at 14:00
+        - Contact seller #a1b2c3 to check inventory/status
+        - Consider promotion for Sports/Leisure category (declining 3 weeks)
+        """)
+    
+    if st.button("🤖 Generate New Report"):
+        with st.spinner("Agent generating report..."):
+            st.success("Report generation triggered! Check back in 1-2 minutes.")