[Fix] plot axes and visibility

app.py

@@ -5,31 +5,35 @@ from src.dataloading import get_leaderboard_models_cached, get_leaderboard_datas
 
 def create_heatmap(selected_models, selected_dataset):
     if not selected_models or not selected_dataset:
-        return None
+        return None, gr.update(visible=False)  # Hide if no selection
     
     # Generate random similarity matrix
     size = len(selected_models)
     similarities = np.random.rand(size, size)
     similarities = (similarities + similarities.T) / 2  # Make symmetric
     
-    # ✅ Explicitly create a Figure object before adding Heatmap
-    fig = go.Figure()
-    fig.add_trace(go.Heatmap(
+    # Create plot 
+    fig = go.Figure(data=go.Heatmap(
         z=similarities,
         x=selected_models,
         y=selected_models,
-        colorscale='Viridis'
+        colorscale='Viridis',
+        hoverongaps=False
     ))
-
+    
+    # Improve axis readability
     fig.update_layout(
         title=f"Similarity Matrix for {selected_dataset}",
-        width=800,
-        height=800,
         xaxis_title="Models",
-        yaxis_title="Models"
+        yaxis_title="Models",
+        xaxis=dict(tickangle=45, automargin=True),
+        yaxis=dict(automargin=True),
+        width=800 + 20*len(selected_models),
+        height=800 + 20*len(selected_models),
+        margin=dict(b=100, l=100)  # Add bottom/left margin for labels
     )
     
-    return fig
+    return fig, gr.update(visible=True)  
 
 def validate_inputs(selected_models, selected_dataset):
     if not selected_models:
@@ -69,7 +73,7 @@ with gr.Blocks(title="LLM Similarity Analyzer") as demo:
     ).then(
         fn=create_heatmap,
         inputs=[model_dropdown, dataset_dropdown],
-        outputs=[heatmap]
+        outputs=[heatmap, heatmap.visible]
     )
     
     # Clear button