added plots to separate tab

app.py

@@ -27,7 +27,7 @@ try:
     LEADERBOARD_DF = get_leaderboard_df_from_hf_dataset(EVAL_RESULTS_PATH)
     LEADERBOARD_DF = summarize_model_task_type_performance(LEADERBOARD_DF)
     LEADERBOARD_DF = prepare_leaderboard_df(LEADERBOARD_DF)
-    
+
     wrapped_task_plot = make_plot_wrapper(leaderboard_df=LEADERBOARD_DF, group_by="Model", filter_col="Task", orientation="h")
     wrapped_model_plot = make_plot_wrapper(leaderboard_df=LEADERBOARD_DF, group_by="Task", filter_col="Model", orientation="v")
 
@@ -74,39 +74,41 @@ with demo:
     gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
 
     with gr.Tabs(elem_classes="tab-buttons") as tabs:
-        with gr.TabItem("🏅 LLM Benchmark", elem_id="llm-benchmark-tab-table", id=0):
+        with gr.TabItem("🏅 LLM Leaderboard", elem_id="llm-benchmark-tab-table", id=0):
             leaderboard = init_leaderboard(LEADERBOARD_DF)
 
-            with gr.Accordion("🧬 Overview of Performance for a Specific Task", open=False):
-                gr.Markdown("Visualize model performance for a specific task and metric")
+        with gr.TabItem("📊 Performance Plots", elem_id="llm-benchmark-performance-plots", id=1):
+            with gr.TabItem("📊 Performance Plots", elem_id="llm-benchmark-tab-table", id=2):
+
+                with gr.Accordion("🧬 Overview of Performance for a Specific Task", open=False):
+                    gr.Markdown("Visualize model performance for a specific task and metric")
 
-                task_choices = sorted(LEADERBOARD_DF["Task"].dropna().unique())
+                    task_choices = sorted(LEADERBOARD_DF["Task"].dropna().unique())
 
-                with gr.Row():
-                    task_dropdown = gr.Dropdown(choices=task_choices, label="Select Task")
-                    metric_dropdown = gr.Dropdown(choices=METRICS_FOR_PLOTS, value="Accuracy", label="Select Metric")
-        
-                performance_plot = gr.Plot()
+                    with gr.Row():
+                        task_dropdown = gr.Dropdown(choices=task_choices, label="Select Task")
+                        metric_dropdown = gr.Dropdown(choices=METRICS_FOR_PLOTS, value="Accuracy", label="Select Metric")
 
-                task_dropdown.change(fn=wrapped_task_plot, inputs=[task_dropdown, metric_dropdown], outputs=performance_plot)
-                metric_dropdown.change(fn=wrapped_task_plot, inputs=[task_dropdown, metric_dropdown], outputs=performance_plot)
+                    performance_plot = gr.Plot()
 
+                    task_dropdown.change(fn=wrapped_task_plot, inputs=[task_dropdown, metric_dropdown], outputs=performance_plot)
+                    metric_dropdown.change(fn=wrapped_task_plot, inputs=[task_dropdown, metric_dropdown], outputs=performance_plot)
 
-            with gr.Accordion("🤖 Overview of Performance for a Specific Model", open=False):
-                gr.Markdown("Visualize model performance across tasks according to a specific metric")
+                with gr.Accordion("🤖 Overview of Performance for a Specific Model", open=False):
+                    gr.Markdown("Visualize model performance across tasks according to a specific metric")
 
-                model_choices = sorted(LEADERBOARD_DF["Model"].dropna().unique())
+                    model_choices = sorted(LEADERBOARD_DF["Model"].dropna().unique())
 
-                with gr.Row():
-                    model_dropdown = gr.Dropdown(choices=model_choices, label="Select Model")
-                    metric2_dropdown = gr.Dropdown(choices=METRICS_FOR_PLOTS, value="Accuracy", label="Select Metric")
+                    with gr.Row():
+                        model_dropdown = gr.Dropdown(choices=model_choices, label="Select Model")
+                        metric2_dropdown = gr.Dropdown(choices=METRICS_FOR_PLOTS, value="Accuracy", label="Select Metric")
 
-                model_plot = gr.Plot()
+                    model_plot = gr.Plot()
 
-                model_dropdown.change(fn=wrapped_model_plot, inputs=[model_dropdown, metric2_dropdown], outputs=model_plot)
-                metric2_dropdown.change(fn=wrapped_model_plot, inputs=[model_dropdown, metric2_dropdown], outputs=model_plot)
+                    model_dropdown.change(fn=wrapped_model_plot, inputs=[model_dropdown, metric2_dropdown], outputs=model_plot)
+                    metric2_dropdown.change(fn=wrapped_model_plot, inputs=[model_dropdown, metric2_dropdown], outputs=model_plot)
 
-        with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2):
+        with gr.TabItem("📝 About", elem_id="llm-benchmark-about", id=2):
             gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
 
     with gr.Row():