added traceback

app.py

@@ -1,3 +1,4 @@
+import traceback
 import gradio as gr
 import os
 import torch
@@ -14,18 +15,23 @@ OUT_PATH ="./savedPredictions/results.csv"
 
 def dataProcessing(file, timestamp_column: str = None):
     global chronos2
-    if os.path.exists(OUT_PATH):
-        os.remove(OUT_PATH)
-
+    
     # Mostra messaggio di processing
     yield (
-        gr.update(visible=True, value="⏳ Processing file, please wait..."),  # processing_msg
-        gr.update(visible=False),   # plot_output
-        gr.update(visible=False),   # download_output
-        gr.update(visible=False, value="")  # errorHandler
+        gr.update(visible=True),  # processing_msg
+        gr.update(visible=False, value=None),   # plot_output
+        gr.update(visible=False, value=None),   # download_output
+        gr.update(visible=False)  # errorHandler
     )
 
     try:
+        if os.path.exists(OUT_PATH):
+            os.remove(OUT_PATH)
+        
+        # Converti stringa vuota in None
+        if timestamp_column == "":
+            timestamp_column = None
+            
         validateData(file, timestamp_column)
         preProcessedData, timestamp_old, target_cols = preProcessData(file, timestamp_column)
         predictions, indexes = predictData(chronos2, preProcessedData, target_cols)
@@ -37,30 +43,38 @@ def dataProcessing(file, timestamp_column: str = None):
         df.to_csv(OUT_PATH, index=False)
 
         # Nascondi processing, mostra risultati
-        yield (
+        if fig is not None:
+            yield (
                 gr.update(visible=False),  # processing_msg
-                gr.update(visible=True, value=fig) if fig is not None else gr.update(visible=False),  # plot_output
+                gr.update(visible=True, value=fig),  # plot_output
                 gr.update(visible=True, value=OUT_PATH),  # download_output
-                gr.update(visible=False, value="")  # errorHandler
-        )
+                gr.update(visible=False)  # errorHandler
+            )
+        else:
+            yield (
+                gr.update(visible=False),  # processing_msg
+                gr.update(visible=False, value=None),  # plot_output
+                gr.update(visible=True, value=OUT_PATH),  # download_output
+                gr.update(visible=False)  # errorHandler
+            )
+            
     except Exception as e:
         # Mostra errore
         yield (
             gr.update(visible=False),  # processing_msg
-            gr.update(visible=False),  # plot_output
-            gr.update(visible=False),  # download_output
-            gr.update(visible=True, value=f"Error: {e}")  # errorHandler
+            gr.update(visible=False, value=None),  # plot_output
+            gr.update(visible=False, value=None),  # download_output
+            gr.update(visible=True, value=f"❌ **Error:** {str(e)}\n\n```\n{traceback.format_exc()}\n```")  # errorHandler
         )
 
+
 if os.path.exists(OUT_PATH):
     os.remove(OUT_PATH)
 
-
 os.makedirs(os.path.dirname(OUT_PATH), exist_ok=True)
 chronos2 = Chronos2Pipeline.from_pretrained("amazon/chronos-2", device_map=device)
 
 
-
 with gr.Blocks(title="Time series anomaly detection with Chronos2") as demo:
     gr.Markdown(
         """
@@ -88,39 +102,42 @@ with gr.Blocks(title="Time series anomaly detection with Chronos2") as demo:
 
     with gr.Row():
         with gr.Column(scale=1):
-            file_input = gr.File(label="Upload Time Series Data (CSV)", file_types=[".csv"], file_count="single", )
+            file_input = gr.File(label="Upload Time Series Data (CSV)", file_types=[".csv"], file_count="single")
             timestamp_question = gr.Radio(
                 label="Does your data contain a timestamp column?",
                 choices=["Yes", "No"],
                 value="No",
                 interactive=True,
-
             )
-            timestamp_column_input = gr.Textbox(label="Please specify the column name of the timestamp:", visible=False, value="", interactive=False)
+            timestamp_column_input = gr.Textbox(
+                label="Please specify the column name of the timestamp:", 
+                visible=False, 
+                value="", 
+                interactive=False
+            )
 
             timestamp_question.change(
-                lambda x: gr.update(visible=(x == "Yes"), interactive=(x == "Yes"), value="timestamp" if x == "Yes" else ""),
+                lambda x: gr.update(
+                    visible=(x == "Yes"), 
+                    interactive=(x == "Yes"), 
+                    value="timestamp" if x == "Yes" else ""
+                ),
                 inputs=timestamp_question,
                 outputs=timestamp_column_input
             )
             
-            detect_button = gr.Button("Detect Anomalies")
+            detect_button = gr.Button("Detect Anomalies", variant="primary")
 
         with gr.Column(scale=4):
-            plot_output = gr.Plot(label="Time Series with Detected Anomalies", visible=False)
-            download_output = gr.File(label="Download Anomaly Detection Results (CSV)", visible=False, interactive=False)
-            errorHandler = gr.Markdown(label="Error Messages", visible=False)
-
             processing_msg = gr.Markdown("⏳ Processing file, please wait...", visible=False)
+            errorHandler = gr.Markdown(visible=False)
+            plot_output = gr.Plot(label="Time Series with Detected Anomalies", visible=False)
+            download_output = gr.File(label="Download Anomaly Detection Results (CSV)", visible=False)
 
-            detect_button.click(
-                lambda file, timestamp_question, timestamp_column:
-                    dataProcessing(
-                        file.name if file is not None else None,
-                        timestamp_column if timestamp_question == "Yes" else None
-                    ),
-                inputs=[file_input, timestamp_question, timestamp_column_input],
-                outputs=[processing_msg, plot_output, download_output, errorHandler]
-            )
+    detect_button.click(
+        fn=dataProcessing,
+        inputs=[file_input, timestamp_column_input],
+        outputs=[processing_msg, plot_output, download_output, errorHandler]
+    )
 
 demo.launch(share=True)