Update app.py

app.py

@@ -20,6 +20,8 @@ import tempfile
 import csv
 from collections import defaultdict
 import filecmp
+from sklearn.manifold import TSNE
+from sklearn.cluster import DBSCAN
 
 # Command-line argument parsing
 parser = argparse.ArgumentParser(description='Typhoon Analysis Dashboard')
@@ -47,20 +49,20 @@ color_map = {
 
 # Classification standards with distinct colors for Matplotlib
 atlantic_standard = {
-    'C5 Super Typhoon': {'wind_speed': 137, 'color': 'Red', 'hex': '#FF0000'},         # Pure Red
-    'C4 Very Strong Typhoon': {'wind_speed': 113, 'color': 'Orange', 'hex': '#FFA500'}, # Bright Orange
-    'C3 Strong Typhoon': {'wind_speed': 96, 'color': 'Yellow', 'hex': '#FFFF00'},      # Pure Yellow
-    'C2 Typhoon': {'wind_speed': 83, 'color': 'Green', 'hex': '#00FF00'},             # Pure Green
-    'C1 Typhoon': {'wind_speed': 64, 'color': 'Cyan', 'hex': '#00FFFF'},              # Pure Cyan
-    'Tropical Storm': {'wind_speed': 34, 'color': 'Blue', 'hex': '#0000FF'},          # Pure Blue
-    'Tropical Depression': {'wind_speed': 0, 'color': 'Gray', 'hex': '#808080'}       # Medium Gray
+    'C5 Super Typhoon': {'wind_speed': 137, 'color': 'Red', 'hex': '#FF0000'},
+    'C4 Very Strong Typhoon': {'wind_speed': 113, 'color': 'Orange', 'hex': '#FFA500'},
+    'C3 Strong Typhoon': {'wind_speed': 96, 'color': 'Yellow', 'hex': '#FFFF00'},
+    'C2 Typhoon': {'wind_speed': 83, 'color': 'Green', 'hex': '#00FF00'},
+    'C1 Typhoon': {'wind_speed': 64, 'color': 'Cyan', 'hex': '#00FFFF'},
+    'Tropical Storm': {'wind_speed': 34, 'color': 'Blue', 'hex': '#0000FF'},
+    'Tropical Depression': {'wind_speed': 0, 'color': 'Gray', 'hex': '#808080'}
 }
 
 taiwan_standard = {
-    'Strong Typhoon': {'wind_speed': 51.0, 'color': 'Red', 'hex': '#FF0000'},         # Pure Red
-    'Medium Typhoon': {'wind_speed': 33.7, 'color': 'Orange', 'hex': '#FFA500'},      # Bright Orange
-    'Mild Typhoon': {'wind_speed': 17.2, 'color': 'Yellow', 'hex': '#FFFF00'},        # Pure Yellow
-    'Tropical Depression': {'wind_speed': 0, 'color': 'Gray', 'hex': '#808080'}       # Medium Gray
+    'Strong Typhoon': {'wind_speed': 51.0, 'color': 'Red', 'hex': '#FF0000'},
+    'Medium Typhoon': {'wind_speed': 33.7, 'color': 'Orange', 'hex': '#FFA500'},
+    'Mild Typhoon': {'wind_speed': 17.2, 'color': 'Yellow', 'hex': '#FFFF00'},
+    'Tropical Depression': {'wind_speed': 0, 'color': 'Gray', 'hex': '#808080'}
 }
 
 # Data loading and preprocessing functions
@@ -302,7 +304,7 @@ def generate_main_analysis(start_year, start_month, end_year, end_month, enso_ph
     
     return tracks_fig, wind_scatter, pressure_scatter, regression_fig, slopes_text
 
-# Video animation function with color examples, distinct colors, and dynamic sidebar
+# Video animation function with fixed sidebar
 def categorize_typhoon_by_standard(wind_speed, standard):
     if standard == 'taiwan':
         wind_speed_ms = wind_speed * 0.514444
@@ -341,9 +343,9 @@ def generate_track_video(year, typhoon, standard):
     lat_padding = max((max_lat - min_lat) * 0.3, 5)
     lon_padding = max((max_lon - min_lon) * 0.3, 5)
 
-    # Set up the figure with adjusted size to fit Gradio (900x700 pixels at 100 DPI)
-    fig = plt.figure(figsize=(9, 7), dpi=100)  # 9x7 inches = 900x700 pixels
-    ax = plt.axes([0.05, 0.15, 0.60, 0.80], projection=ccrs.PlateCarree())  # Map on left 60%
+    # Set up the figure (900x700 pixels at 100 DPI)
+    fig = plt.figure(figsize=(9, 7), dpi=100)
+    ax = plt.axes([0.05, 0.05, 0.65, 0.90], projection=ccrs.PlateCarree())  # Adjusted to leave space for sidebar
     ax.set_extent([min_lon - lon_padding, max_lon + lon_padding, min_lat - lat_padding, max_lat + lat_padding], crs=ccrs.PlateCarree())
 
     # Add world map features
@@ -360,18 +362,18 @@ def generate_track_video(year, typhoon, standard):
     point, = ax.plot([], [], 'o', markersize=8, transform=ccrs.PlateCarree())
     date_text = ax.text(0.02, 0.02, '', transform=ax.transAxes, fontsize=10, bbox=dict(facecolor='white', alpha=0.8))
 
-    # Add sidebar with typhoon details on the right
-    details_title = fig.text(0.65, 0.90, "Typhoon Details", fontsize=12, fontweight='bold', verticalalignment='top', horizontalalignment='left')
-    details_text = fig.text(0.65, 0.85, '', fontsize=10, verticalalignment='top', horizontalalignment='left',
+    # Add sidebar on the right
+    details_title = fig.text(0.75, 0.95, "Typhoon Details", fontsize=12, fontweight='bold', verticalalignment='top')
+    details_text = fig.text(0.75, 0.85, '', fontsize=10, verticalalignment='top',
                             bbox=dict(facecolor='white', alpha=0.8, boxstyle='round,pad=0.5'))
 
-    # Add color legend with colored markers below the details
+    # Add color legend
     standard_dict = atlantic_standard if standard == 'atlantic' else taiwan_standard
     legend_elements = [plt.Line2D([0], [0], marker='o', color='w', label=f"{cat}", 
-                                   markerfacecolor=details['hex'], markersize=10)
+                                  markerfacecolor=details['hex'], markersize=10)
                        for cat, details in standard_dict.items()]
-    fig.legend(handles=legend_elements, title="Color Legend", loc='center right', 
-               bbox_to_anchor=(0.95, 0.5), fontsize=10)
+    fig.legend(handles=legend_elements, title="Color Legend", loc='lower right', 
+               bbox_to_anchor=(0.95, 0.05), fontsize=10)
 
     def init():
         line.set_data([], [])
@@ -393,7 +395,6 @@ def generate_track_video(year, typhoon, standard):
         details_text.set_text(details)
         return line, point, date_text, details_text
 
-    # Create animation
     ani = animation.FuncAnimation(fig, update, init_func=init, frames=len(storm.time),
                                   interval=200, blit=True, repeat=True)
 
@@ -439,7 +440,118 @@ def perform_longitude_regression(start_year, start_month, end_year, end_month):
     beta_1, exp_beta_1, p_value = model.params['ONI'], np.exp(model.params['ONI']), model.pvalues['ONI']
     return f"Longitude Regression: β1={beta_1:.4f}, Odds Ratio={exp_beta_1:.4f}, P-value={p_value:.4f}"
 
-# Gradio interface
+# t-SNE clustering functions
+def filter_west_pacific_coordinates(lons, lats):
+    mask = (lons >= 100) & (lons <= 180) & (lats >= 0) & (lats <= 50)
+    return lons[mask], lats[mask]
+
+def update_route_clusters(start_year, start_month, end_year, end_month, enso_value, season):
+    start_date = datetime(int(start_year), int(start_month), 1)
+    end_date = datetime(int(end_year), int(end_month), 28)
+    
+    all_storms_data = []
+    for year in range(int(start_year), int(end_year) + 1):
+        season_data = ibtracs.get_season(year)
+        for storm_id in season_data.summary()['id']:
+            storm = ibtracs.get_storm(storm_id)
+            if storm.time[0] >= start_date and storm.time[-1] <= end_date:
+                lons, lats = filter_west_pacific_coordinates(np.array(storm.lon), np.array(storm.lat))
+                if len(lons) > 1:
+                    all_storms_data.append((lons, lats, np.array(storm.vmax), np.array(storm.mslp), np.array(storm.time), storm.name))
+
+    if not all_storms_data:
+        return go.Figure(), go.Figure(), go.Figure(), "No storms found in the selected period."
+
+    # Prepare route vectors for t-SNE
+    max_length = max(len(st[0]) for st in all_storms_data)
+    route_vectors = []
+    for lons, lats, _, _, _, _ in all_storms_data:
+        interp_lons = np.interp(np.linspace(0, 1, max_length), np.linspace(0, 1, len(lons)), lons)
+        interp_lats = np.interp(np.linspace(0, 1, max_length), np.linspace(0, 1, len(lats)), lats)
+        route_vectors.append(np.column_stack((interp_lons, interp_lats)).flatten())
+    route_vectors = np.array(route_vectors)
+
+    # Perform t-SNE
+    tsne_results = TSNE(n_components=2, random_state=42, perplexity=min(30, len(route_vectors)-1)).fit_transform(route_vectors)
+
+    # Dynamic DBSCAN clustering
+    target_clusters = min(5, len(all_storms_data) // 3)
+    eps_range = np.arange(5.0, 50.0, 5.0)
+    min_samples = max(3, len(all_storms_data) // 20)
+    best_labels = None
+    best_eps = None
+    best_n_clusters = 0
+    best_noise_ratio = 1.0
+
+    for eps in eps_range:
+        dbscan = DBSCAN(eps=eps, min_samples=min_samples)
+        labels = dbscan.fit_predict(tsne_results)
+        n_clusters = len(set(labels)) - (1 if -1 in labels else 0)
+        noise_points = np.sum(labels == -1)
+        noise_ratio = noise_points / len(labels)
+        
+        if n_clusters >= target_clusters and noise_ratio < 0.3 and (n_clusters > best_n_clusters or (n_clusters == best_n_clusters and noise_ratio < best_noise_ratio)):
+            best_labels = labels
+            best_eps = eps
+            best_n_clusters = n_clusters
+            best_noise_ratio = noise_ratio
+
+    if best_labels is None:
+        dbscan = DBSCAN(eps=5.0, min_samples=min_samples)
+        best_labels = dbscan.fit_predict(tsne_results)
+        best_eps = 5.0
+        best_n_clusters = len(set(best_labels)) - (1 if -1 in best_labels else 0)
+
+    # t-SNE Scatter Plot
+    fig_tsne = go.Figure()
+    for cluster in set(best_labels):
+        mask = best_labels == cluster
+        name = "Noise" if cluster == -1 else f"Cluster {cluster}"
+        fig_tsne.add_trace(go.Scatter(
+            x=tsne_results[mask, 0], y=tsne_results[mask, 1], mode='markers',
+            name=name, text=[all_storms_data[i][5] for i in range(len(all_storms_data)) if mask[i]],
+            hoverinfo='text'
+        ))
+    fig_tsne.update_layout(title="t-SNE Clustering of Typhoon Routes", xaxis_title="t-SNE 1", yaxis_title="t-SNE 2")
+
+    # Typhoon Routes Plot
+    fig_routes = go.Figure()
+    for i, (lons, lats, _, _, _, name) in enumerate(all_storms_data):
+        cluster = best_labels[i]
+        color = 'gray' if cluster == -1 else px.colors.qualitative.Plotly[cluster % len(px.colors.qualitative.Plotly)]
+        fig_routes.add_trace(go.Scattergeo(
+            lon=lons, lat=lats, mode='lines+markers', name=name,
+            line=dict(color=color), marker=dict(size=4), hoverinfo='text', text=name
+        ))
+    fig_routes.update_layout(
+        title="Typhoon Routes by Cluster",
+        geo=dict(scope='asia', projection_type='mercator', showland=True, landcolor='lightgray')
+    )
+
+    # Cluster Statistics Plot
+    cluster_stats = []
+    for cluster in set(best_labels) - {-1}:
+        mask = best_labels == cluster
+        winds = [all_storms_data[i][2].max() for i in range(len(all_storms_data)) if mask[i]]
+        pressures = [all_storms_data[i][3].min() for i in range(len(all_storms_data)) if mask[i]]
+        cluster_stats.append({
+            'Cluster': cluster,
+            'Count': np.sum(mask),
+            'Mean Wind': np.mean(winds),
+            'Mean Pressure': np.mean(pressures)
+        })
+    stats_df = pd.DataFrame(cluster_stats)
+    fig_stats = px.bar(stats_df, x='Cluster', y=['Mean Wind', 'Mean Pressure'], barmode='group',
+                       title="Cluster Statistics (Mean Wind Speed and Pressure)")
+
+    # Cluster Information
+    cluster_info = f"Number of Clusters: {best_n_clusters}\nBest EPS: {best_eps}\nNoise Points: {best_noise_ratio*100:.1f}%"
+    for stat in cluster_stats:
+        cluster_info += f"\nCluster {stat['Cluster']}: {stat['Count']} storms, Mean Wind: {stat['Mean Wind']:.1f} kt, Mean Pressure: {stat['Mean Pressure']:.1f} hPa"
+
+    return fig_tsne, fig_routes, fig_stats, cluster_info
+
+# Gradio Interface
 with gr.Blocks(title="Typhoon Analysis Dashboard") as demo:
     gr.Markdown("# Typhoon Analysis Dashboard")
     
@@ -451,9 +563,11 @@ with gr.Blocks(title="Typhoon Analysis Dashboard") as demo:
         
         ### Features:
         - **Track Visualization**: View typhoon tracks by time period and ENSO phase
-        - **Statistical Analysis**: Examine relationships between ONI values and typhoon characteristics
-        - **Path Animation**: Watch an animated typhoon path with video controls, world map, color legend with examples, and dynamic sidebar
-        - **Regression Analysis**: Perform statistical regression on typhoon data
+        - **Wind Analysis**: Examine wind speed vs ONI relationships
+        - **Pressure Analysis**: Analyze pressure vs ONI relationships
+        - **Longitude Analysis**: Study typhoon generation longitude vs ONI
+        - **Path Animation**: Watch animated typhoon paths with a sidebar
+        - **TSNE Cluster**: Perform t-SNE clustering on typhoon routes
         
         Select a tab above to begin your analysis.
         """)
@@ -623,7 +737,7 @@ with gr.Blocks(title="Typhoon Analysis Dashboard") as demo:
         5. The animation shows the typhoon track growing over a world map, with:
            - Date on the bottom left
            - Sidebar on the right showing typhoon details (name, date, wind speed, category) as it moves
-           - Color legend with colored markers (e.g., Red: C5 Super Typhoon) on the right
+           - Color legend with colored markers on the bottom right
         """)
         
         def update_typhoon_options(year):
@@ -638,6 +752,26 @@ with gr.Blocks(title="Typhoon Analysis Dashboard") as demo:
             inputs=[year_dropdown, typhoon_dropdown, standard_dropdown],
             outputs=path_video
         )
+    
+    with gr.Tab("TSNE Cluster"):
+        with gr.Row():
+            tsne_start_year = gr.Number(label="Start Year", value=2000, minimum=1900, maximum=2024, step=1)
+            tsne_start_month = gr.Dropdown(label="Start Month", choices=list(range(1, 13)), value=1)
+            tsne_end_year = gr.Number(label="End Year", value=2024, minimum=1900, maximum=2024, step=1)
+            tsne_end_month = gr.Dropdown(label="End Month", choices=list(range(1, 13)), value=12)
+            tsne_enso_phase = gr.Dropdown(label="ENSO Phase", choices=['all', 'El Nino', 'La Nina', 'Neutral'], value='all')
+            tsne_season = gr.Dropdown(label="Season", choices=['all', 'summer', 'winter'], value='all')
+        tsne_analyze_btn = gr.Button("Analyze")
+        tsne_plot = gr.Plot(label="t-SNE Clusters")
+        routes_plot = gr.Plot(label="Typhoon Routes")
+        stats_plot = gr.Plot(label="Cluster Statistics")
+        cluster_info = gr.Textbox(label="Cluster Information", lines=10)
+        
+        tsne_analyze_btn.click(
+            fn=update_route_clusters,
+            inputs=[tsne_start_year, tsne_start_month, tsne_end_year, tsne_end_month, tsne_enso_phase, tsne_season],
+            outputs=[tsne_plot, routes_plot, stats_plot, cluster_info]
+        )
 
     # Custom CSS for better visibility
     gr.HTML("""