Update space

app.py

@@ -545,14 +545,15 @@ def update_leaderboard_data(submission_data):
     }
     
     df_to_update = split_to_df[submission_data['Split']]
+    submitted_dataset = submission_data['Dataset'].upper()
     
     # Prepare new row data
     new_row = {
-        'Method': submission_data['Method Name'],  # Only use method name in table
-        f'STARK-{submission_data["Dataset"].upper()}_Hit@1': submission_data['results']['hit@1'],
-        f'STARK-{submission_data["Dataset"].upper()}_Hit@5': submission_data['results']['hit@5'],
-        f'STARK-{submission_data["Dataset"].upper()}_R@20': submission_data['results']['recall@20'],
-        f'STARK-{submission_data["Dataset"].upper()}_MRR': submission_data['results']['mrr']
+        'Method': submission_data['Method Name'],
+        f'STARK-{submitted_dataset}_Hit@1': submission_data['results']['hit@1'],
+        f'STARK-{submitted_dataset}_Hit@5': submission_data['results']['hit@5'],
+        f'STARK-{submitted_dataset}_R@20': submission_data['results']['recall@20'],
+        f'STARK-{submitted_dataset}_MRR': submission_data['results']['mrr']
     }
     
     # Check if method already exists
@@ -562,8 +563,11 @@ def update_leaderboard_data(submission_data):
         for col in new_row:
             df_to_update.loc[method_mask, col] = new_row[col]
     else:
-        # Add new row
-        df_to_update.loc[len(df_to_update)] = new_row
+        # For new method, create row with NaN for other datasets
+        all_columns = df_to_update.columns
+        full_row = {col: None for col in all_columns}  # Initialize with NaN
+        full_row.update(new_row)  # Update with the submitted dataset's values
+        df_to_update.loc[len(df_to_update)] = full_row
 
 # Function to get emails from meta_data
 def get_emails_from_metadata(meta_data):
@@ -667,9 +671,12 @@ def validate_model_type(method_name, selected_type):
 
 def process_submission(
     method_name, team_name, dataset, split, contact_email,
-    code_repo, csv_file, model_description, hardware, paper_link, model_type
+    code_repo, csv_file, model_description, hardware, paper_link, model_type, honor_code
 ):
     """Process and validate submission"""
+    if not honor_code:
+        return "Error: Please accept the honor code to submit"
+        
     temp_files = []
     try:
         # Input validation
@@ -757,7 +764,7 @@ def process_submission(
             submission_data = {
                 **meta_data,
                 "results": processed_results,
-                "status": "approved",  # or "pending_review"
+                "status": "pending_review",  # or "approved"
                 "submission_date": datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
                 "csv_path": csv_path_in_repo
             }
@@ -779,7 +786,7 @@ def process_submission(
             # 3. Create or update latest.json
             latest_info = {
                 "latest_submission": timestamp,
-                "status": "approved",  # or "pending_review"
+                "status": "pending_review",  # or "approved"
                 "method_name": method_name,
                 "team_name": team_name
             }
@@ -854,10 +861,23 @@ def filter_by_model_type(df, selected_types):
     return df[df['Method'].isin(selected_models)]
 
 def format_dataframe(df, dataset):
+    """
+    Format DataFrame for display, removing rows with no data for the specified dataset.
+    """
+    # Get relevant columns
     columns = ['Method'] + [col for col in df.columns if dataset in col]
     filtered_df = df[columns].copy()
+    
+    # Remove rows where all metric columns are NaN
+    metric_columns = [col for col in filtered_df.columns if col != 'Method']
+    filtered_df = filtered_df.dropna(subset=metric_columns, how='all')
+    
+    # Rename columns to remove dataset prefix
     filtered_df.columns = [col.split('_')[-1] if '_' in col else col for col in filtered_df.columns]
+    
+    # Sort by MRR
     filtered_df = filtered_df.sort_values('MRR', ascending=False)
+    
     return filtered_df
 
 def update_tables(selected_types):
@@ -948,7 +968,7 @@ with gr.Blocks(css=css) as demo:
     """)
     
     with gr.Row():
-        with gr.Column():
+        with gr.Column(scale=1):
             method_name = gr.Textbox(
                 label="Method Name (max 25 chars)*",
                 placeholder="e.g., MyRetrievalModel-v1"
@@ -977,14 +997,14 @@ with gr.Blocks(css=css) as demo:
                 value="Others",
                 info="Select the appropriate category for your model"
             )
-            
-        
-        with gr.Column():
             model_description = gr.Textbox(
                 label="Model Description*",
                 lines=3,
                 placeholder="Briefly describe how your retriever model works..."
             )
+            
+        
+        with gr.Column(scale=1):
             code_repo = gr.Textbox(
                 label="Code Repository*",
                 placeholder="https://github.com/snap-stanford/stark-leaderboard"
@@ -993,6 +1013,11 @@ with gr.Blocks(css=css) as demo:
                 label="Hardware Specifications*",
                 placeholder="e.g., 4x NVIDIA A100 80GB"
             )
+            honor_code = gr.Checkbox(
+                label="Honor Code",
+                value=False,
+                info="By submitting these results, you confirm that they are truthful and reproducible, and you verify the integrity of your submission."
+            )
             csv_file = gr.File(
                 label="Prediction CSV*",
                 file_types=[".csv"],
@@ -1003,9 +1028,9 @@ with gr.Blocks(css=css) as demo:
                 placeholder="https://arxiv.org/abs/..."
             )
     
+    
     submit_btn = gr.Button("Submit", variant="primary")
     result = gr.Textbox(label="Submission Status", interactive=False)
-
     
     # Set up event handlers
     model_type_filter.change(
@@ -1013,13 +1038,18 @@ with gr.Blocks(css=css) as demo:
         inputs=[model_type_filter],
         outputs=all_dfs
     )
+    honor_code.change(
+        fn=lambda x: gr.Button.update(interactive=x),
+        inputs=[honor_code],
+        outputs=[submit_btn]
+    )
     
     # Event handler for submission button
     submit_btn.click(
         fn=process_submission,
         inputs=[
             method_name, team_name, dataset, split, contact_email,
-            code_repo, csv_file, model_description, hardware, paper_link, model_type
+            code_repo, csv_file, model_description, hardware, paper_link, model_type, honor_code
         ],
         outputs=result
     ).success(  # Add a success handler to update tables after successful submission