Update benchmark_selection.py

benchmark_selection.py

@@ -78,13 +78,33 @@ class BenchmarkSelector:
             # Format results
             results = []
             for dataset in datasets:
+                # Handle cases where description might be missing
+                dataset_description = ""
+                if hasattr(dataset, 'description') and dataset.description:
+                    dataset_description = dataset.description[:200] + "..." if len(dataset.description) > 200 else dataset.description
+                
+                # Handle cases where tags might be missing
+                dataset_tags = []
+                if hasattr(dataset, 'tags'):
+                    dataset_tags = dataset.tags
+                
+                # Handle cases where downloads might be missing
+                dataset_downloads = 0
+                if hasattr(dataset, 'downloads'):
+                    dataset_downloads = dataset.downloads
+                
+                # Handle cases where author might be missing
+                dataset_author = ""
+                if hasattr(dataset, 'author'):
+                    dataset_author = dataset.author
+                
                 results.append({
                     "id": dataset.id,
                     "name": dataset.id.split("/")[-1],
-                    "author": dataset.author,
-                    "description": dataset.description[:200] + "..." if dataset.description and len(dataset.description) > 200 else dataset.description,
-                    "tags": dataset.tags,
-                    "downloads": dataset.downloads
+                    "author": dataset_author,
+                    "description": dataset_description,
+                    "tags": dataset_tags,
+                    "downloads": dataset_downloads
                 })
             
             return results
@@ -106,18 +126,43 @@ class BenchmarkSelector:
             dataset_info = self.hf_api.dataset_info(dataset_id)
             
             # Get available configurations
-            configs = get_dataset_config_names(dataset_id)
+            configs = []
+            try:
+                configs = get_dataset_config_names(dataset_id)
+            except Exception as e:
+                print(f"Error getting dataset configs: {e}")
+            
+            # Handle missing attributes safely
+            dataset_description = ""
+            if hasattr(dataset_info, 'description'):
+                dataset_description = dataset_info.description
+            
+            dataset_citation = ""
+            if hasattr(dataset_info, 'citation'):
+                dataset_citation = dataset_info.citation
+            
+            dataset_tags = []
+            if hasattr(dataset_info, 'tags'):
+                dataset_tags = dataset_info.tags
+            
+            dataset_downloads = 0
+            if hasattr(dataset_info, 'downloads'):
+                dataset_downloads = dataset_info.downloads
+            
+            dataset_author = ""
+            if hasattr(dataset_info, 'author'):
+                dataset_author = dataset_info.author
             
             # Format result
             result = {
                 "id": dataset_info.id,
                 "name": dataset_info.id.split("/")[-1],
-                "author": dataset_info.author,
-                "description": dataset_info.description,
-                "citation": dataset_info.citation,
+                "author": dataset_author,
+                "description": dataset_description,
+                "citation": dataset_citation,
                 "configs": configs,
-                "tags": dataset_info.tags,
-                "downloads": dataset_info.downloads
+                "tags": dataset_tags,
+                "downloads": dataset_downloads
             }
             
             return result
@@ -351,6 +396,9 @@ def create_benchmark_selection_ui(benchmark_selector, auth_manager):
                 
                 with gr.Row():
                     refresh_benchmarks_button = gr.Button("Refresh Benchmarks")
+                    reload_sample_benchmarks_button = gr.Button("Reload Sample Benchmarks", variant="secondary")
+                
+                reload_status = gr.Markdown("")
                 
                 benchmarks_container = gr.Column()
                 with benchmarks_container:
@@ -470,6 +518,14 @@ def create_benchmark_selection_ui(benchmark_selector, auth_manager):
             
             return gr.update(visible=False), gr.update(visible=True), formatted_benchmarks
         
+        def reload_sample_benchmarks_handler():
+            try:
+                from sample_benchmarks import add_sample_benchmarks
+                num_added = add_sample_benchmarks()
+                return f"✅ Successfully reloaded {num_added} sample benchmarks."
+            except Exception as e:
+                return f"❌ Error reloading benchmarks: {str(e)}"
+        
         # Connect event handlers
         search_button.click(
             fn=search_datasets_handler,
@@ -501,6 +557,12 @@ def create_benchmark_selection_ui(benchmark_selector, auth_manager):
             outputs=[no_benchmarks_message, my_benchmarks, my_benchmarks]
         )
         
+        reload_sample_benchmarks_button.click(
+            fn=reload_sample_benchmarks_handler,
+            inputs=[],
+            outputs=[reload_status]
+        )
+        
         # Initialize benchmarks on load
         benchmark_ui.load(
             fn=get_benchmarks_handler,
@@ -508,4 +570,4 @@ def create_benchmark_selection_ui(benchmark_selector, auth_manager):
             outputs=[no_benchmarks_message, my_benchmarks, my_benchmarks]
         )
     
-    return benchmark_ui
+    return benchmark_ui