Update app.py

app.py

@@ -23,7 +23,7 @@ from sklearn.manifold import TSNE
 from sklearn.cluster import DBSCAN
 from scipy.interpolate import interp1d
 
-# Import tropycal for IBTrACS processing (for typhoon option updates)
+# Import tropycal for IBTrACS processing (for typhoon options)
 import tropycal.tracks as tracks
 
 # ------------------ Argument Parsing ------------------
@@ -189,6 +189,46 @@ def classify_enso_phases(oni_value):
     else:
         return 'Neutral'
 
+# ------------------ Regression Functions ------------------
+def perform_wind_regression(start_year, start_month, end_year, end_month):
+    start_date = datetime(start_year, start_month, 1)
+    end_date = datetime(end_year, end_month, 28)
+    data = merged_data[(merged_data['ISO_TIME'] >= start_date) & (merged_data['ISO_TIME'] <= end_date)].dropna(subset=['USA_WIND', 'ONI'])
+    data['severe_typhoon'] = (data['USA_WIND'] >= 64).astype(int)
+    X = sm.add_constant(data['ONI'])
+    y = data['severe_typhoon']
+    model = sm.Logit(y, X).fit(disp=0)
+    beta_1 = model.params['ONI']
+    exp_beta_1 = np.exp(beta_1)
+    p_value = model.pvalues['ONI']
+    return f"Wind Regression: β1={beta_1:.4f}, Odds Ratio={exp_beta_1:.4f}, P-value={p_value:.4f}"
+
+def perform_pressure_regression(start_year, start_month, end_year, end_month):
+    start_date = datetime(start_year, start_month, 1)
+    end_date = datetime(end_year, end_month, 28)
+    data = merged_data[(merged_data['ISO_TIME'] >= start_date) & (merged_data['ISO_TIME'] <= end_date)].dropna(subset=['USA_PRES', 'ONI'])
+    data['intense_typhoon'] = (data['USA_PRES'] <= 950).astype(int)
+    X = sm.add_constant(data['ONI'])
+    y = data['intense_typhoon']
+    model = sm.Logit(y, X).fit(disp=0)
+    beta_1 = model.params['ONI']
+    exp_beta_1 = np.exp(beta_1)
+    p_value = model.pvalues['ONI']
+    return f"Pressure Regression: β1={beta_1:.4f}, Odds Ratio={exp_beta_1:.4f}, P-value={p_value:.4f}"
+
+def perform_longitude_regression(start_year, start_month, end_year, end_month):
+    start_date = datetime(start_year, start_month, 1)
+    end_date = datetime(end_year, end_month, 28)
+    data = merged_data[(merged_data['ISO_TIME'] >= start_date) & (merged_data['ISO_TIME'] <= end_date)].dropna(subset=['LON', 'ONI'])
+    data['western_typhoon'] = (data['LON'] <= 140).astype(int)
+    X = sm.add_constant(data['ONI'])
+    y = data['western_typhoon']
+    model = sm.Logit(y, X).fit(disp=0)
+    beta_1 = model.params['ONI']
+    exp_beta_1 = np.exp(beta_1)
+    p_value = model.pvalues['ONI']
+    return f"Longitude Regression: β1={beta_1:.4f}, Odds Ratio={exp_beta_1:.4f}, P-value={p_value:.4f}"
+
 # ------------------ IBTrACS Data Loading (for typhoon options) ------------------
 def load_ibtracs_data():
     ibtracs_data = {}
@@ -403,7 +443,6 @@ def categorize_typhoon_by_standard(wind_speed, standard):
 
 # ------------------ Animation Functions Using Processed CSV ------------------
 def generate_track_video_from_csv(year, storm_id, standard):
-    # Filter processed CSV data for the storm ID
     storm_df = typhoon_data[typhoon_data['SID'] == storm_id].copy()
     if storm_df.empty:
         print("No data found for storm:", storm_id)
@@ -419,20 +458,17 @@ def generate_track_video_from_csv(year, storm_id, standard):
     storm_name = storm_df['NAME'].iloc[0]
     season = storm_df['SEASON'].iloc[0]
     
-    # Set up map boundaries
     min_lat, max_lat = np.min(lats), np.max(lats)
     min_lon, max_lon = np.min(lons), np.max(lons)
     lat_padding = max((max_lat - min_lat) * 0.3, 5)
     lon_padding = max((max_lon - min_lon) * 0.3, 5)
     
-    # Create a larger figure with custom central longitude for better regional focus
     fig = plt.figure(figsize=(12, 9), dpi=100)
     ax = plt.axes([0.05, 0.05, 0.60, 0.90],
                   projection=ccrs.PlateCarree(central_longitude=-25))
     ax.set_extent([min_lon - lon_padding, max_lon + lon_padding, min_lat - lat_padding, max_lat + lat_padding],
                   crs=ccrs.PlateCarree())
     
-    # Add map features
     ax.add_feature(cfeature.LAND, facecolor='lightgray')
     ax.add_feature(cfeature.OCEAN, facecolor='lightblue')
     ax.add_feature(cfeature.COASTLINE, edgecolor='black')
@@ -440,18 +476,17 @@ def generate_track_video_from_csv(year, storm_id, standard):
     ax.gridlines(draw_labels=True, linestyle='--', color='gray', alpha=0.5)
     ax.set_title(f"{year} {storm_name} - {season}", fontsize=16)
     
-    # Plot track and state marker
     line, = ax.plot([], [], 'b-', linewidth=2, transform=ccrs.PlateCarree())
     point, = ax.plot([], [], 'o', markersize=10, transform=ccrs.PlateCarree())
-    
-    # Dynamic text elements
-    date_text = ax.text(0.02, 0.02, '', transform=ax.transAxes, fontsize=12, bbox=dict(facecolor='white', alpha=0.8))
+    date_text = ax.text(0.02, 0.02, '', transform=ax.transAxes, fontsize=12,
+                         bbox=dict(facecolor='white', alpha=0.8))
+    # Dynamic state display at right side
     state_text = fig.text(0.70, 0.60, '', fontsize=14, verticalalignment='top',
-                           bbox=dict(facecolor='white', alpha=0.8, boxstyle='round,pad=0.5'))
+                          bbox=dict(facecolor='white', alpha=0.8, boxstyle='round,pad=0.5'))
     
-    # Persistent legend for color mapping (placed on right)
+    # Persistent legend for category colors
     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 (atlantic_standard if standard=='atlantic' else taiwan_standard).items()]
     ax.legend(handles=legend_elements, title="Storm Categories", loc='upper right', fontsize=10)
     
@@ -463,16 +498,13 @@ def generate_track_video_from_csv(year, storm_id, standard):
         return line, point, date_text, state_text
 
     def update(frame):
-        # Update the track line
         line.set_data(lons[:frame+1], lats[:frame+1])
-        # Update the marker position using lists to avoid deprecation warning
         point.set_data([lons[frame]], [lats[frame]])
         wind_speed = winds[frame] if frame < len(winds) else np.nan
         category, color = categorize_typhoon_by_standard(wind_speed, standard)
         point.set_color(color)
         dt_str = pd.to_datetime(times[frame]).strftime('%Y-%m-%d %H:%M')
         date_text.set_text(dt_str)
-        # Update state information dynamically in the sidebar
         state_info = (f"Name: {storm_name}\n"
                       f"Date: {dt_str}\n"
                       f"Wind: {wind_speed:.1f} kt\n"
@@ -542,7 +574,7 @@ def update_typhoon_options(year, basin):
 
 def update_typhoon_options_anim(year, basin):
     try:
-        # For animation, use the processed CSV data for all storms in the given year
+        # For animation, use the processed CSV data (without filtering by a specific prefix)
         data = typhoon_data.copy()
         data['Year'] = pd.to_datetime(data['ISO_TIME']).dt.year
         season_data = data[data['Year'] == int(year)]
@@ -625,8 +657,8 @@ def update_route_clusters(start_year, start_month, end_year, end_month, enso_val
             name=f"Cluster {label}"
         ))
     fig_tsne.update_layout(title="t-SNE of WP Storm Routes")
-    fig_routes = go.Figure()
-    fig_stats = make_subplots(rows=2, cols=1)
+    fig_routes = go.Figure()  # Placeholder
+    fig_stats = make_subplots(rows=2, cols=1)  # Placeholder
     info = "TSNE clustering complete."
     return fig_tsne, fig_routes, fig_stats, info
 
@@ -645,7 +677,7 @@ with gr.Blocks(title="Typhoon Analysis Dashboard") as demo:
         - **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 tropical cyclone paths with dynamic state display and a legend (using processed CSV data)
+        - **Path Animation**: Watch animated tropical cyclone paths with a dynamic state display and persistent legend (using processed CSV data)
         - **TSNE Cluster**: Perform t-SNE clustering on WP storm routes with mean routes and region analysis
         
         Select a tab above to begin your analysis.
@@ -715,11 +747,11 @@ with gr.Blocks(title="Typhoon Analysis Dashboard") as demo:
         path_video = gr.Video(label="Tropical Cyclone Path Animation", format="mp4", interactive=False, elem_id="path_video")
         animation_info = gr.Markdown("""
         ### Animation Instructions
-        1. Select a year and basin from the dropdowns. (This animation uses the processed CSV data.)
+        1. Select a year and basin from the dropdowns. (This animation uses processed CSV data.)
         2. Choose a tropical cyclone from the populated list.
         3. Select a classification standard (Atlantic or Taiwan).
         4. Click "Generate Animation".
-        5. The animation displays the storm track along with a dynamic sidebar that updates the state (name, date, wind, category) and includes a persistent legend.
+        5. The animation displays the storm track along with a dynamic sidebar that shows the current state (name, date, wind, category) and a persistent legend for colors.
         """)
         year_dropdown.change(fn=update_typhoon_options_anim, inputs=[year_dropdown, basin_dropdown], outputs=typhoon_dropdown)
         basin_dropdown.change(fn=update_typhoon_options_anim, inputs=[year_dropdown, basin_dropdown], outputs=typhoon_dropdown)