Create app.py

app.py

@@ -0,0 +1,110 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import plotly.express as px
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.model_selection import train_test_split
+
+# ----------------------
+# Data Generation (Synthetic)
+# ----------------------
+def generate_data(num_devices=100):
+    np.random.seed(42)
+    timestamps = pd.date_range("2024-01-01", periods=30, freq="D")
+    data = []
+    for device_id in range(num_devices):
+        for timestamp in timestamps:
+            energy_use = np.random.normal(200, 50)  # Base energy (kWh)
+            latency = np.random.normal(100, 20)     # Latency (ms)
+            traffic = np.random.randint(0, 100)      # Network traffic (%)
+            # Simulate "inefficient" devices (20% of data)
+            if np.random.random() < 0.2:
+                energy_use *= 1.5
+            data.append([device_id, timestamp, energy_use, latency, traffic])
+    df = pd.DataFrame(data, columns=["device_id", "timestamp", "energy_kwh", "latency_ms", "traffic_pct"])
+    return df
+
+# ----------------------
+# AI Model (Energy Optimization)
+# ----------------------
+def train_model(df):
+    # Feature engineering
+    df["hour"] = df["timestamp"].dt.hour
+    df["day_of_week"] = df["timestamp"].dt.dayofweek
+    X = df[["hour", "day_of_week", "latency_ms", "traffic_pct"]]
+    y = df["energy_kwh"]
+    
+    # Train/test split
+    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2)
+    
+    # Model training
+    model = RandomForestRegressor(n_estimators=10)
+    model.fit(X_train, y_train)
+    df["predicted_energy"] = model.predict(X)
+    
+    # Calculate savings potential
+    df["savings_kwh"] = df["energy_kwh"] - df["predicted_energy"]
+    return df, model.feature_importances_
+
+# ----------------------
+# Streamlit App
+# ----------------------
+st.set_page_config(page_title="Energy Optimizer", layout="wide")
+st.title("🌱 AI for Energy-Efficient Public Networks")
+
+# Load data
+df = generate_data()
+df, feature_importances = train_model(df)
+
+# ----------------------
+# Dashboard Sections
+# ----------------------
+tab1, tab2, tab3 = st.tabs(["📈 Analysis", "🔧 Recommendations", "About"])
+
+with tab1:
+    st.subheader("Energy Consumption Trends")
+    selected_device = st.selectbox("Select Device", df["device_id"].unique())
+    device_df = df[df["device_id"] == selected_device]
+    
+    # Plot energy vs. predictions
+    fig = px.line(device_df, x="timestamp", y=["energy_kwh", "predicted_energy"],
+                  labels={"value": "Energy (kWh)"}, title="Actual vs. Optimal Energy Use")
+    st.plotly_chart(fig, use_container_width=True)
+    
+    # Feature importance
+    st.subheader("What Impacts Energy Use?")
+    features = ["Hour", "Day of Week", "Latency", "Traffic"]
+    fig = px.bar(x=features, y=feature_importances, labels={"x": "Factor", "y": "Importance"})
+    st.plotly_chart(fig, use_container_width=True)
+
+with tab2:
+    st.subheader("Optimization Recommendations")
+    
+    # Top inefficient devices
+    st.write("### Top Energy-Wasting Devices")
+    inefficient_devices = df.groupby("device_id")["savings_kwh"].mean().nlargest(5).reset_index()
+    st.dataframe(inefficient_devices, hide_index=True)
+    
+    # Savings calculator
+    st.write("### Potential Savings")
+    total_savings = df["savings_kwh"].sum()
+    co2_reduction = total_savings * 0.5  # Assume 0.5 kg CO2 per kWh
+    st.metric("Total Monthly Energy Savings", f"{total_savings:,.0f} kWh")
+    st.metric("Estimated CO₂ Reduction", f"{co2_reduction:,.0f} kg")
+    
+    # Recommendations
+    st.write("### Actionable Steps")
+    st.markdown("""
+    - **Schedule Off-Peak Maintenance**: Reduce energy use during low-traffic hours.
+    - **Upgrade Inefficient Devices**: Prioritize devices with the highest savings potential.
+    - **Adjust Traffic Routing**: Use latency data to balance load efficiently.
+    """)
+
+with tab3:
+    st.write("## About")
+    st.markdown("""
+    **Open-Source Energy Optimizer for Public Sector Networks**  
+    Built for the AI for Connectivity Hackathon II.  
+    - Code: [GitHub](https://github.com/)  
+    - Data: Synthetic dataset (replace with real telemetry).  
+    """)