只留下拉選單跟滑桿

app.py

@@ -838,13 +838,13 @@ def step1_load_mseed_and_select_stations(event_name):
         )
 
         logger.info("[步驟 1] 完成 - mseed 已載入，測站已選擇")
-        return st, selected_stations, station_map, None, gr.update(interactive=False)
+        return st, selected_stations, station_map, None
 
     except Exception as e:
         logger.error(f"[步驟 1] 發生錯誤: {e}")
         import traceback
         traceback.print_exc()
-        return None, None, None, None, gr.update(interactive=False)
+        return None, None, None, None
 
 
 # ============ 步驟 2：提取波形（使用快取的 stream + stations）============
@@ -879,13 +879,13 @@ def step2_extract_and_plot_waveforms(cached_stream, cached_stations, event_name,
         waveform_plot = plot_waveform(cached_stream, cached_stations, first_pick, duration)
 
         logger.info(f"[步驟 2] 完成 - 已提取 {len(waveforms)} 個測站的波形")
-        return waveforms, station_info_list, waveform_plot, gr.update(interactive=True)
+        return waveforms, station_info_list, waveform_plot
 
     except Exception as e:
         logger.error(f"[步驟 2] 發生錯誤: {e}")
         import traceback
         traceback.print_exc()
-        return None, None, None, gr.update(interactive=False)
+        return None, None, None
 
 
 # ============ 步驟 3：執行模型推論（使用快取的波形）============
@@ -1015,12 +1015,11 @@ with gr.Blocks(title="TTSAM 震度預測系統", fill_height=True) as demo:
     with gr.Row():
         # 左上：使用步驟與狀態顯示
         with gr.Column(scale=1):
-            gr.Markdown("## 使用步驟")
+            gr.Markdown("## 使用說明")
             gr.Markdown(
                 """
-                1. 選擇地震事件（自動載入測站）
-                2. 調整 P 波後時間範圍（即時更新波形）
-                3. 點擊「執行預測」查看震度分布
+                - **選擇地震事件**：自動載入測站並執行預測
+                - **調整時間滑桿**：即時更新波形與震度預測
                 
                 系統會自動選擇距離震央最近的 25 個測站，並快取資料避免重複讀檔。
             """
@@ -1031,16 +1030,15 @@ with gr.Blocks(title="TTSAM 震度預測系統", fill_height=True) as demo:
                 value=list(earthquake_metadata.keys())[0],
                 label="選擇地震事件",
             )
+            duration_slider = gr.Slider(
+                2, 15, value=15, step=1, label="P 波後時間 (秒)"
+            )
     # ========== 中層：輸入測站地圖與波形圖 ==========
     with gr.Row():
         # 中左：輸入測站地圖
         with gr.Column(scale=1):
             gr.Markdown("## 輸入測站分布")
             input_station_map = gr.Plot(label="輸入測站地圖")
-            load_waveform_btn = gr.Button("📊 載入波形", variant="secondary", scale=1)
-            predict_btn = gr.Button(
-                "🔮 執行預測", variant="primary", scale=1, interactive=False
-            )
 
         # 中右：輸入波形
         with gr.Column(scale=1):
@@ -1048,10 +1046,7 @@ with gr.Blocks(title="TTSAM 震度預測系統", fill_height=True) as demo:
             waveform_plot = gr.Plot(
                 label="地震波形（選定的 25 個測站）",
             )
-            with gr.Row():
-                duration_slider = gr.Slider(
-                    2, 15, value=15, step=1, label="P 波後時間 (秒)"
-                )
+
 
     # ========== 下層：實際觀測 vs 預測結果 ==========
     with gr.Row():
@@ -1078,51 +1073,45 @@ with gr.Blocks(title="TTSAM 震度預測系統", fill_height=True) as demo:
 
     # ========== 事件綁定（使用鏈式觸發 + gr.State 快取）==========
 
-    # 【觸發點 1】事件切換：重新執行步驟 1 → 步驟 2
+    # 【觸發點 1】事件切換：自動執行完整流程 步驟 1 → 步驟 2 → 步驟 3
     event_dropdown.change(
         fn=step1_load_mseed_and_select_stations,
         inputs=[event_dropdown],
-        outputs=[cached_stream, cached_stations, input_station_map, waveform_plot, predict_btn]
+        outputs=[cached_stream, cached_stations, input_station_map, waveform_plot]
     ).then(  # 鏈式觸發步驟 2
         fn=step2_extract_and_plot_waveforms,
         inputs=[cached_stream, cached_stations, event_dropdown, duration_slider],
-        outputs=[cached_waveforms, cached_station_info, waveform_plot, predict_btn]
+        outputs=[cached_waveforms, cached_station_info, waveform_plot]
+    ).then(  # 鏈式觸發步驟 3
+        fn=step3_predict_intensity,
+        inputs=[cached_waveforms, cached_station_info, event_dropdown],
+        outputs=[observed_intensity_image, predicted_intensity_map]
     )
 
-    # 【觸發點 2】時間範圍調整：只重新執行步驟 2（快速，不重新讀檔）
+    # 【觸發點 2】時間範圍調整：自動執行步驟 2 → 步驟 3（不重新讀檔）
     duration_slider.change(
         fn=step2_extract_and_plot_waveforms,
         inputs=[cached_stream, cached_stations, event_dropdown, duration_slider],
-        outputs=[cached_waveforms, cached_station_info, waveform_plot, predict_btn]
-    )
-
-    # 【觸發點 3】載入波形按鈕：手動觸發步驟 1 → 步驟 2
-    load_waveform_btn.click(
-        fn=step1_load_mseed_and_select_stations,
-        inputs=[event_dropdown],
-        outputs=[cached_stream, cached_stations, input_station_map, waveform_plot, predict_btn]
-    ).then(
-        fn=step2_extract_and_plot_waveforms,
-        inputs=[cached_stream, cached_stations, event_dropdown, duration_slider],
-        outputs=[cached_waveforms, cached_station_info, waveform_plot, predict_btn]
-    )
-
-    # 【觸發點 4】執行預測：使用快取的波形執行步驟 3
-    predict_btn.click(
+        outputs=[cached_waveforms, cached_station_info, waveform_plot]
+    ).then(  # 鏈式觸發步驟 3
         fn=step3_predict_intensity,
         inputs=[cached_waveforms, cached_station_info, event_dropdown],
         outputs=[observed_intensity_image, predicted_intensity_map]
     )
 
-    # 【冷啟動】應用載入時自動執行步驟 1 → 步驟 2
+    # 【冷啟動】應用載入時自動執行完整流程 步驟 1 → 步驟 2 → 步驟 3
     demo.load(
         fn=step1_load_mseed_and_select_stations,
         inputs=[event_dropdown],
-        outputs=[cached_stream, cached_stations, input_station_map, waveform_plot, predict_btn]
+        outputs=[cached_stream, cached_stations, input_station_map, waveform_plot]
     ).then(
         fn=step2_extract_and_plot_waveforms,
         inputs=[cached_stream, cached_stations, event_dropdown, duration_slider],
-        outputs=[cached_waveforms, cached_station_info, waveform_plot, predict_btn]
+        outputs=[cached_waveforms, cached_station_info, waveform_plot]
+    ).then(
+        fn=step3_predict_intensity,
+        inputs=[cached_waveforms, cached_station_info, event_dropdown],
+        outputs=[observed_intensity_image, predicted_intensity_map]
     )
 
 demo.launch()