Update app.py

app.py

@@ -38,7 +38,7 @@ def make_link(mname):
     display_name = parts[1] if len(parts) > 1 else mname
     return f'[{display_name}](https://huggingface.co/{mname})'
 
-# --- Plot Functions (Axes swapped) ---
+# --- Plot Functions (Bar Chart) ---
 
 def get_plots(task):
     df = pd.read_csv('data/energy/' + task)
@@ -53,8 +53,8 @@ def get_plots(task):
     # Use the energy score to control color
     color_map = {"1": "red", "2": "orange", "3": "yellow", "4": "lightgreen", "5": "green"}
 
-    # Now plot with the model name on the X axis and GPU Energy on the Y axis.
-    fig = px.scatter(
+    # Now plot as a bar chart
+    fig = px.bar(
         df,
         x="Display Model",
         y="total_gpu_energy",
@@ -75,9 +75,9 @@ def get_plots(task):
     fig.update_layout(
         xaxis_title="Model",
         yaxis_title="GPU Energy (Wh)",
-        yaxis_tickformat=".4f", # Add this line to format y-axis ticks
+        yaxis_tickformat=".4f", # Add this line to format y-axis ticks - might not be needed for bar chart
         yaxis = dict(
-            tickformat=".4f" # Ensure tickformat is set within yaxis dict as well
+            tickformat=".4f" # Ensure tickformat is set within yaxis dict as well - might not be needed for bar chart
         )
     )
     return fig
@@ -96,7 +96,7 @@ def get_all_plots():
 
     color_map = {"1": "red", "2": "orange", "3": "yellow", "4": "lightgreen", "5": "green"}
 
-    fig = px.scatter(
+    fig = px.bar(
         all_df,
         x="Display Model",
         y="total_gpu_energy",
@@ -116,43 +116,14 @@ def get_all_plots():
     fig.update_layout(
         xaxis_title="Model",
         yaxis_title="GPU Energy (Wh)",
-        yaxis_tickformat=".4f", # Add this line to format y-axis ticks
+        yaxis_tickformat=".4f", # Add this line to format y-axis ticks - might not be needed for bar chart
         yaxis = dict(
-            tickformat=".4f" # Ensure tickformat is set within yaxis dict as well
+            tickformat=".4f" # Ensure tickformat is set within yaxis dict as well - might not be needed for bar chart
         )
     )
     return fig
 
-# --- Leaderboard Table Functions (unchanged except stars) ---
-
-def get_model_names(task):
-    df = pd.read_csv('data/energy/' + task)
-    if df.columns[0].startswith("Unnamed:"):
-        df = df.iloc[:, 1:]
-    df['energy_score'] = df['energy_score'].astype(int)
-    # For leaderboard display, format GPU Energy to 4 decimals
-    df['GPU Energy (Wh)'] = pd.to_numeric(df['total_gpu_energy'], errors='raise').apply(lambda x: f"{x:.4f}")
-    df['Model'] = df['model'].apply(make_link)
-    df['Score'] = df['energy_score'].apply(format_stars)
-    # Remove any Class column if it exists
-    df = df[['Model', 'GPU Energy (Wh)', 'Score']]
-    df = df.sort_values(by='GPU Energy (Wh)')
-    return df
-
-def get_all_model_names():
-    all_df = pd.DataFrame()
-    for task in tasks:
-        df = pd.read_csv('data/energy/' + task)
-        df['energy_score'] = df['energy_score'].astype(int)
-        df['GPU Energy (Wh)'] = pd.to_numeric(df['total_gpu_energy'], errors='raise').apply(lambda x: f"{x:.4f}")
-        df['Model'] = df['model'].apply(make_link)
-        df['Score'] = df['energy_score'].apply(format_stars)
-        all_df = pd.concat([all_df, df], ignore_index=True)
-    all_df = all_df.drop_duplicates(subset=['model'])
-    all_df = all_df.sort_values(by='GPU Energy (Wh)')
-    return all_df[['Model', 'GPU Energy (Wh)', 'Score']]
-
-# --- New functions for Text Generation filtering by model class (with swapped axes) ---
+# --- New functions for Text Generation filtering by model class (with Bar Chart) ---
 
 def get_text_generation_plots(model_class):
     df = pd.read_csv('data/energy/text_generation.csv')
@@ -167,7 +138,7 @@ def get_text_generation_plots(model_class):
 
     color_map = {"1": "red", "2": "orange", "3": "yellow", "4": "lightgreen", "5": "green"}
 
-    fig = px.scatter(
+    fig = px.bar(
         df,
         x="Display Model",
         y="total_gpu_energy",
@@ -177,7 +148,6 @@ def get_text_generation_plots(model_class):
         width=800,
         color_discrete_map=color_map
     )
-    # Update hover text to show the model and GPU Energy (with 4 decimals)
     fig.update_traces(
         hovertemplate="<br>".join([
             "Model: %{x}",
@@ -188,13 +158,45 @@ def get_text_generation_plots(model_class):
     fig.update_layout(
         xaxis_title="Model",
         yaxis_title="GPU Energy (Wh)",
-        yaxis_tickformat=".4f", # Add this line to format y-axis ticks
+        yaxis_tickformat=".4f", # Add this line to format y-axis ticks - might not be needed for bar chart
         yaxis = dict(
-            tickformat=".4f" # Ensure tickformat is set within yaxis dict as well
+            tickformat=".4f" # Ensure tickformat is set within yaxis dict as well - might not be needed for bar chart
         )
     )
     return fig
 
+
+# --- Leaderboard Table Functions and Gradio Interface are unchanged ---
+# (Keep the rest of the code same as previous response)
+
+def get_model_names(task):
+    df = pd.read_csv('data/energy/' + task)
+    if df.columns[0].startswith("Unnamed:"):
+        df = df.iloc[:, 1:]
+    df['energy_score'] = df['energy_score'].astype(int)
+    # For leaderboard display, format GPU Energy to 4 decimals
+    df['GPU Energy (Wh)'] = pd.to_numeric(df['total_gpu_energy'], errors='raise').apply(lambda x: f"{x:.4f}")
+    df['Model'] = df['model'].apply(make_link)
+    df['Score'] = df['energy_score'].apply(format_stars)
+    # Remove any Class column if it exists
+    df = df[['Model', 'GPU Energy (Wh)', 'Score']]
+    df = df.sort_values(by='GPU Energy (Wh)')
+    return df
+
+def get_all_model_names():
+    all_df = pd.DataFrame()
+    for task in tasks:
+        df = pd.read_csv('data/energy/' + task)
+        df['energy_score'] = df['energy_score'].astype(int)
+        df['GPU Energy (Wh)'] = pd.to_numeric(df['total_gpu_energy'], errors='raise').apply(lambda x: f"{x:.4f}")
+        df['Model'] = df['model'].apply(make_link)
+        df['Score'] = df['energy_score'].apply(format_stars)
+        all_df = pd.concat([all_df, df], ignore_index=True)
+    all_df = all_df.drop_duplicates(subset=['model'])
+    all_df = all_df.sort_values(by='GPU Energy (Wh)')
+    return all_df[['Model', 'GPU Energy (Wh)', 'Score']]
+
+
 def get_text_generation_model_names(model_class):
     df = pd.read_csv('data/energy/text_generation.csv')
     if df.columns[0].startswith("Unnamed:"):
@@ -243,12 +245,12 @@ Select different tasks to see scored models. Submit open models for testing and
             # Dropdown moved above the plot and leaderboard
             model_class_dropdown = gr.Dropdown(choices=["A", "B", "C"],
                                                  label="Select Model Class",
-                                                 value="C") # Default to C for testing
+                                                 value="A")
             with gr.Row():
                 with gr.Column(scale=1.3):
-                    tg_plot = gr.Plot(get_text_generation_plots("C")) # Default to C for testing
+                    tg_plot = gr.Plot(get_text_generation_plots("A"))
                 with gr.Column(scale=1):
-                    tg_table = gr.Dataframe(get_text_generation_model_names("C"), datatype="markdown")
+                    tg_table = gr.Dataframe(get_text_generation_model_names("A"), datatype="markdown")
             # Update plot and table when the dropdown value changes
             model_class_dropdown.change(fn=update_text_generation,
                                         inputs=model_class_dropdown,