Upload app.py

app.py

@@ -815,31 +815,67 @@ def create_statistics_dashboard(df: pd.DataFrame) -> Tuple[str, go.Figure]:
 
     html_parts.append('</div>')
 
+    # Determine what to show in second chart
+    second_chart_title = "Category Distribution"
+    if 'sentence_category' in df.columns:
+        second_chart_title = "Sentence Category"
+    elif 'reasoning_pattern' in df.columns:
+        second_chart_title = "Reasoning Pattern"
+    elif 'task_type' in df.columns:
+        second_chart_title = "Task Type"
+    elif 'is_positive' in df.columns:
+        second_chart_title = "Positive vs Negative"
+
     # Create distribution charts
     fig = make_subplots(rows=1, cols=2,
-                        subplot_titles=("Probability Delta Distribution", "Category Distribution"))
+                        subplot_titles=("Probability Delta Distribution", second_chart_title))
 
-    if 'prob_delta' in df.columns:
+    # First chart: Probability Delta histogram
+    if 'prob_delta' in df.columns and len(df['prob_delta'].dropna()) > 0:
         fig.add_trace(
-            go.Histogram(x=df['prob_delta'], nbinsx=30, name="Prob Delta",
+            go.Histogram(x=df['prob_delta'].dropna(), nbinsx=30, name="Prob Delta",
                         marker_color='#6366f1'),
             row=1, col=1
         )
+    else:
+        # Fallback: show prob_before/prob_after if available
+        if 'prob_after' in df.columns:
+            fig.add_trace(
+                go.Histogram(x=df['prob_after'].dropna(), nbinsx=30, name="Prob After",
+                            marker_color='#6366f1'),
+                row=1, col=1
+            )
 
+    # Second chart: Categories, patterns, or task types
     if 'sentence_category' in df.columns:
         category_counts = df['sentence_category'].value_counts()
         fig.add_trace(
-            go.Bar(x=category_counts.index, y=category_counts.values, name="Categories",
+            go.Bar(x=category_counts.index.tolist(), y=category_counts.values.tolist(), name="Categories",
                   marker_color='#22c55e'),
             row=1, col=2
         )
     elif 'reasoning_pattern' in df.columns:
         pattern_counts = df['reasoning_pattern'].value_counts()
         fig.add_trace(
-            go.Bar(x=pattern_counts.index, y=pattern_counts.values, name="Patterns",
+            go.Bar(x=pattern_counts.index.tolist(), y=pattern_counts.values.tolist(), name="Patterns",
                   marker_color='#22c55e'),
             row=1, col=2
         )
+    elif 'task_type' in df.columns:
+        task_counts = df['task_type'].value_counts()
+        fig.add_trace(
+            go.Bar(x=task_counts.index.tolist(), y=task_counts.values.tolist(), name="Task Types",
+                  marker_color='#22c55e'),
+            row=1, col=2
+        )
+    elif 'is_positive' in df.columns:
+        pos_neg_counts = df['is_positive'].value_counts()
+        labels = ['Positive' if v else 'Negative' for v in pos_neg_counts.index.tolist()]
+        fig.add_trace(
+            go.Bar(x=labels, y=pos_neg_counts.values.tolist(), name="Impact",
+                  marker_color=['#22c55e' if l == 'Positive' else '#ef4444' for l in labels]),
+            row=1, col=2
+        )
 
     fig.update_layout(
         template="plotly_dark",