Deploy simplified Gradio app to HuggingFace

Documents/Claude/Projects/uber/lyft rider project/surge_pricing_analysis/app.py

@@ -1,214 +1,54 @@
-"""
-Gradio dashboard for Surge Pricing Impact Analysis.
-Causal inference analysis of marketplace dynamics.
-"""
-
 import gradio as gr
 import pandas as pd
 import numpy as np
-import plotly.graph_objects as go
-import plotly.express as px
-
-from src.data_simulator import MarketplaceSimulator
-from src.causal_inference import (
-    PropensityScoreMatching,
-    SyntheticControlMethod,
-    InterruptedTimeSeries,
-    compare_causal_methods
-)
-
-
-def generate_analysis(days, surge_mult, surge_freq, demand_elast, supply_elast, seed):
-    """Generate marketplace data and run causal analysis."""
-    try:
-        # Generate data
-        sim = MarketplaceSimulator(seed=seed)
-        surge_days = list(np.random.RandomState(seed).choice(
-            days, size=max(1, int(days * surge_freq)), replace=False
-        ))
-
-        df = sim.generate_daily_data(
-            days=days,
-            surge_multiplier=surge_mult,
-            surge_days=surge_days,
-            demand_elasticity=demand_elast,
-            supply_elasticity=supply_elast
-        )
-
-        # Summary stats
-        surge_group = df[df['is_surge'] == 1]
-        no_surge_group = df[df['is_surge'] == 0]
-
-        summary = f"""
-        **Simulation Summary**
-
-        Total Days: {len(df)}
-        Surge Days: {len(surge_group)} ({len(surge_group)/len(df)*100:.1f}%)
-
-        **Completion Rate**
-        - Surge Days: {surge_group['completion_rate'].mean():.2%}
-        - No Surge Days: {no_surge_group['completion_rate'].mean():.2%}
-        - Difference: {(surge_group['completion_rate'].mean() - no_surge_group['completion_rate'].mean()):.2%}
-
-        **Wait Time (minutes)**
-        - Surge Days: {surge_group['wait_time_minutes'].mean():.1f}
-        - No Surge Days: {no_surge_group['wait_time_minutes'].mean():.1f}
-        - Difference: {(surge_group['wait_time_minutes'].mean() - no_surge_group['wait_time_minutes'].mean()):.1f}
-
-        **Driver Earnings ($)**
-        - Surge Days: ${surge_group['driver_earnings'].mean():.2f}
-        - No Surge Days: ${no_surge_group['driver_earnings'].mean():.2f}
-        - Difference: ${(surge_group['driver_earnings'].mean() - no_surge_group['driver_earnings'].mean()):.2f}
-        """
-
-        # Time series plot
-        fig_ts = go.Figure()
-        fig_ts.add_trace(go.Scatter(
-            x=df['date'],
-            y=df['completion_rate'],
-            name='Completion Rate',
-            mode='lines',
-            line=dict(color='#4ECDC4', width=2)
-        ))
-
-        surge_periods = df[df['is_surge'] == 1]
-        fig_ts.add_trace(go.Scatter(
-            x=surge_periods['date'],
-            y=surge_periods['completion_rate'],
-            mode='markers',
-            name='Surge Period',
-            marker=dict(size=8, color='#FF6B6B')
-        ))
-
-        fig_ts.update_layout(
-            title='Completion Rate Over Time',
-            xaxis_title='Date',
-            yaxis_title='Completion Rate',
-            hovermode='x unified',
-            height=400
-        )
-
-        # Causal analysis
-        results = compare_causal_methods(df, outcome_col='completion_rate')
-
-        causal_text = "**Causal Inference Results (Completion Rate)**\n\n"
-        for method, result in results.items():
-            if 'error' in result:
-                causal_text += f"- **{method}:** Error - {result['error']}\n"
-            elif 'ate' in result:
-                ate = result['ate']
-                ci_lower = result.get('ci_lower', 0)
-                ci_upper = result.get('ci_upper', 0)
-                causal_text += f"- **{method}:** {ate:.4f} (95% CI: [{ci_lower:.4f}, {ci_upper:.4f}])\n"
-            elif 'level_change' in result:
-                causal_text += f"- **{method}:** Level Change = {result['level_change']:.4f}\n"
-
-        return summary, fig_ts, causal_text
-
-    except Exception as e:
-        return f"Error: {str(e)}", None, f"Error in causal analysis: {str(e)}"
-
-
-# Create Gradio interface
-with gr.Blocks(title="Surge Pricing Impact Analysis") as demo:
-    gr.Markdown("""
-    # 🚗 Surge Pricing Impact Analysis
-
-    **How does surge pricing truly affect rider demand, driver supply, and market equilibrium?**
-
-    This interactive dashboard applies causal inference to isolate the real incremental impact of surge pricing—moving beyond correlation to causation.
-    """)
-
-    with gr.Row():
-        with gr.Column(scale=1):
-            gr.Markdown("### Simulation Parameters")
-
-            days = gr.Slider(
-                label="Simulation Duration (days)",
-                minimum=30,
-                maximum=180,
-                value=90,
-                step=10
-            )
-
-            surge_multiplier = gr.Slider(
-                label="Surge Multiplier (fare multiplier)",
-                minimum=1.2,
-                maximum=3.0,
-                value=2.5,
-                step=0.1
-            )
-
-            surge_frequency = gr.Slider(
-                label="Surge Frequency (%)",
-                minimum=5,
-                maximum=50,
-                value=15,
-                step=5
-            )
-
-            demand_elasticity = gr.Slider(
-                label="Demand Elasticity",
-                minimum=-1.0,
-                maximum=-0.1,
-                value=-0.5,
-                step=0.1
-            )
-
-            supply_elasticity = gr.Slider(
-                label="Supply Elasticity",
-                minimum=0.2,
-                maximum=1.5,
-                value=0.65,
-                step=0.1
-            )
-
-            random_seed = gr.Number(
-                label="Random Seed",
-                value=42,
-                precision=0
-            )
-
-            analyze_btn = gr.Button("Analyze", scale=2)
-
-        with gr.Column(scale=2):
-            summary_output = gr.Textbox(
-                label="Summary Statistics",
-                lines=15,
-                interactive=False
-            )
 
+def analyze_surge(days, surge_mult, surge_freq):
+    """Simple surge pricing analysis."""
+    np.random.seed(42)
+    dates = pd.date_range('2024-01-01', periods=days, freq='D')
+    
+    base_demand = np.random.normal(500, 50, days)
+    base_supply = np.random.normal(300, 30, days)
+    
+    surge_days = np.random.choice(days, int(days * surge_freq), replace=False)
+    surge_multiplier = np.ones(days)
+    surge_multiplier[surge_days] = surge_mult
+    
+    demand = base_demand * (1 - (surge_mult - 1) * 0.3)
+    supply = base_supply * (1 + (surge_mult - 1) * 0.5)
+    
+    completion_rate = np.minimum(supply / demand, 1.0)
+    wait_time = 10 * (1 - completion_rate)
+    
+    summary = f"""**Surge Pricing Analysis**
+
+Days: {days}
+Surge Events: {len(surge_days)} ({len(surge_days)/days*100:.1f}%)
+Multiplier: {surge_mult}x
+
+**Completion Rate**
+Overall: {completion_rate.mean():.2%}
+Surge Days: {completion_rate[surge_days].mean():.2%}
+Normal Days: {completion_rate[~np.isin(range(days), surge_days)].mean():.2%}
+
+**Wait Time (minutes)**
+Overall: {wait_time.mean():.1f}
+Surge Days: {wait_time[surge_days].mean():.1f}
+Normal Days: {wait_time[~np.isin(range(days), surge_days)].mean():.1f}"""
+    
+    return summary
+
+with gr.Blocks(title="Surge Pricing Analysis") as demo:
+    gr.Markdown("# 🚗 Surge Pricing Impact Analysis")
+    
     with gr.Row():
-        timeseries_plot = gr.Plot(label="Time Series: Completion Rate")
-
-    with gr.Row():
-        causal_output = gr.Textbox(
-            label="Causal Inference Results",
-            lines=10,
-            interactive=False
-        )
-
-    # Connect button
-    analyze_btn.click(
-        fn=generate_analysis,
-        inputs=[days, surge_multiplier, surge_frequency, demand_elasticity, supply_elasticity, random_seed],
-        outputs=[summary_output, timeseries_plot, causal_output]
-    )
-
-    gr.Markdown("""
-    ---
-
-    ## Methodology
-
-    **Propensity Score Matching (PSM):** Matches surge days to no-surge days based on observable confounders.
-
-    **Synthetic Control Method:** Builds a synthetic "no surge" counterfactual using pre-treatment periods.
-
-    **Interrupted Time Series (ITS):** Estimates level and slope changes at the intervention point.
-
-    [GitHub Repository](https://github.com/data-geek-astronomy/surge_pricing_analysis)
-    """)
-
+        days = gr.Slider(30, 180, 90, step=10, label="Days")
+        surge_mult = gr.Slider(1.2, 3.0, 2.5, step=0.1, label="Multiplier")
+        surge_freq = gr.Slider(0.05, 0.5, 0.15, step=0.05, label="Frequency")
+    
+    output = gr.Textbox(label="Results", lines=12)
+    analyze_btn = gr.Button("Analyze")
+    analyze_btn.click(analyze_surge, [days, surge_mult, surge_freq], output)
 
 if __name__ == "__main__":
     demo.launch()