Fix

app.py

@@ -21,15 +21,7 @@ DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 
 @tool
 def advanced_web_search(query: str, num_results: int = 10) -> str:
-    """Advanced web search using multiple search engines with fallback
-    
-    Args:
-        query: The search query
-        num_results: Number of results to return (default 10)
-        
-    Returns:
-        Comprehensive search results as formatted string
-    """
+    """Advanced web search using multiple search engines with fallback"""
     try:
         # First try Serper API if available
         api_key = os.getenv("SERPER_API_KEY")
@@ -77,14 +69,7 @@ def advanced_web_search(query: str, num_results: int = 10) -> str:
 
 @tool
 def wikipedia_lookup(topic: str) -> str:
-    """Enhanced Wikipedia search and content extraction
-    
-    Args:
-        topic: Wikipedia topic to look up
-        
-    Returns:
-        Wikipedia content with structured information
-    """
+    """Enhanced Wikipedia search and content extraction"""
     try:
         # Clean the topic
         topic_clean = topic.replace(" ", "_").strip()
@@ -131,14 +116,7 @@ def wikipedia_lookup(topic: str) -> str:
 
 @tool
 def youtube_video_analyzer(url: str) -> str:
-    """Advanced YouTube video analysis with multiple extraction methods
-    
-    Args:
-        url: YouTube video URL
-        
-    Returns:
-        Comprehensive video information
-    """
+    """Advanced YouTube video analysis with multiple extraction methods"""
     try:
         # Extract video ID using multiple patterns
         video_id = None
@@ -223,15 +201,7 @@ def youtube_video_analyzer(url: str) -> str:
 
 @tool
 def text_manipulator(text: str, operation: str = "reverse") -> str:
-    """Advanced text manipulation and analysis tool
-    
-    Args:
-        text: Text to manipulate
-        operation: Operation type (reverse, analyze, extract_numbers, etc.)
-        
-    Returns:
-        Manipulated or analyzed text
-    """
+    """Advanced text manipulation and analysis tool"""
     try:
         if operation == "reverse":
             return text[::-1]
@@ -255,14 +225,7 @@ def text_manipulator(text: str, operation: str = "reverse") -> str:
 
 @tool
 def mathematical_solver(problem: str) -> str:
-    """Advanced mathematical problem solver with specific GAIA patterns
-    
-    Args:
-        problem: Mathematical problem description
-        
-    Returns:
-        Mathematical solution or analysis
-    """
+    """Advanced mathematical problem solver with specific GAIA patterns"""
     try:
         problem_lower = problem.lower()
         
@@ -303,71 +266,9 @@ STRATEGY: Systematically check each pair in the table"""
     except Exception as e:
         return f"Math solver error: {str(e)}"
 
-@tool
-def data_classifier(data_string: str, classification_type: str = "botanical") -> str:
-    """Advanced data classification tool for various categorization tasks
-    
-    Args:
-        data_string: String containing data to classify
-        classification_type: Type of classification (botanical, numerical, etc.)
-        
-    Returns:
-        Classified and sorted data
-    """
-    try:
-        if classification_type == "botanical" or "vegetable" in classification_type:
-            # Extract items from the string
-            items = []
-            
-            # Split by common delimiters
-            for delimiter in [',', ';', 'and', '&']:
-                if delimiter in data_string:
-                    items = [item.strip() for item in data_string.split(delimiter)]
-                    break
-            
-            if not items and ' ' in data_string:
-                items = data_string.split()
-            
-            # Classify as true botanical vegetables (not fruits used as vegetables)
-            true_vegetables = []
-            
-            # Botanical vegetable keywords (parts of plants that are not fruits/seeds)
-            vegetable_keywords = [
-                'basil', 'lettuce', 'celery', 'broccoli', 'cabbage', 'spinach',
-                'kale', 'chard', 'arugula', 'parsley', 'cilantro', 'dill',
-                'sweet potato', 'potato', 'carrot', 'beet', 'radish', 'turnip',
-                'onion', 'garlic', 'leek', 'scallion', 'asparagus', 'artichoke'
-            ]
-            
-            for item in items:
-                item_clean = item.lower().strip()
-                if any(veg in item_clean for veg in vegetable_keywords):
-                    true_vegetables.append(item.strip())
-            
-            # Sort alphabetically
-            true_vegetables.sort()
-            return ', '.join(true_vegetables)
-        
-        elif classification_type == "numerical":
-            numbers = re.findall(r'-?\d+\.?\d*', data_string)
-            return f"NUMBERS: {', '.join(numbers)}"
-        
-        return f"CLASSIFIED_DATA: {data_string[:100]}..."
-        
-    except Exception as e:
-        return f"Classification error: {str(e)}"
-
 @tool
 def specialized_lookup(query: str, domain: str = "general") -> str:
-    """Specialized lookup tool for domain-specific information
-    
-    Args:
-        query: Search query
-        domain: Domain to search in (olympics, music, sports, etc.)
-        
-    Returns:
-        Domain-specific information
-    """
+    """Specialized lookup tool for domain-specific information"""
     try:
         if domain == "olympics" or "olympics" in query.lower():
             # Enhanced Olympics search
@@ -419,7 +320,6 @@ class EnhancedGAIAAgent:
             youtube_video_analyzer,
             text_manipulator,
             mathematical_solver,
-            data_classifier,
             specialized_lookup
         ]
         
@@ -456,8 +356,6 @@ class EnhancedGAIAAgent:
             return "youtube"
         elif "ecnetnes siht dnatsrednu uoy fi" in question_lower or any(reversed_word in question_lower for reversed_word in ["fi", "dnif", "eht"]):
             return "reversed_text"
-        elif "botanical" in question_lower and "vegetable" in question_lower:
-            return "botanical_classification"
         elif any(math_term in question_lower for math_term in ["commutative", "operation", "chess", "checkmate"]):
             return "mathematical"
         elif any(olympics_term in question_lower for olympics_term in ["olympics", "olympic", "1928", "amsterdam"]):
@@ -513,13 +411,6 @@ class EnhancedGAIAAgent:
                     
                     return result
             
-            elif question_type == "botanical_classification":
-                # Extract the grocery list
-                food_items = re.search(r'milk.*?peanuts', question, re.IGNORECASE)
-                if food_items:
-                    item_list = food_items.group(0)
-                    return data_classifier(item_list, "botanical")
-            
             elif question_type == "mathematical":
                 return mathematical_solver(question)
             
@@ -715,103 +606,27 @@ Processing Summary:
         print(error_status)
         return error_status, pd.DataFrame(results_log)
 
-# --- Enhanced Gradio Interface ---
+# --- Simplified Gradio Interface ---
 with gr.Blocks(title="Enhanced GAIA Agent") as demo:
-    gr.Markdown("# 🚀 Enhanced GAIA Benchmark Agent")
-    gr.Markdown("""
-    **Advanced Multi-Tool Agent for GAIA Benchmark**
-    
-    **🛠️ Enhanced Capabilities:**
-    - **Advanced Web Search**: Multi-engine search with Serper API + DuckDuckGo fallback
-    - **Wikipedia Integration**: Comprehensive Wikipedia lookup and content extraction  
-    - **YouTube Analysis**: Deep video content analysis and metadata extraction
-    - **Text Processing**: Reverse text decoding, pattern recognition, number extraction
-    - **Mathematical Solver**: Group theory, chess analysis, number theory problems
-    - **Data Classification**: Botanical classification, categorical data sorting
-    - **Domain Specialists**: Olympics, music, sports, scientific information lookup
-    
-    **🎯 Target: 35%+ Accuracy**
-    
-    **📋 Instructions:**
-    1. Login to your Hugging Face account using the button below
-    2. Click 'Run Enhanced Evaluation' to start the benchmark
-    3. The agent will automatically process all questions using optimal strategies
-    4. Results will be submitted and displayed with detailed analytics
-    
-    **⏱️ Processing Time:** ~5-10 minutes depending on question complexity
-    """)
+    gr.Markdown("# Enhanced GAIA Benchmark Agent")
+    gr.Markdown("Advanced Multi-Tool Agent with Web Search, Wikipedia, YouTube Analysis, and Domain Specialists")
 
     gr.LoginButton()
-
-    with gr.Row():
-        run_button = gr.Button(
-            "🚀 Run Enhanced Evaluation & Submit All Answers", 
-            variant="primary",
-            size="lg"
-        )
-
-    status_output = gr.Textbox(
-        label="📊 Evaluation Status & Results", 
-        lines=15, 
-        interactive=False,
-        placeholder="Results will appear here after evaluation..."
-    )
     
-    results_table = gr.DataFrame(
-        label="📋 Detailed Question Analysis", 
-        wrap=True,
-        interactive=False
-    )
+    run_button = gr.Button("Run Enhanced Evaluation & Submit All Answers", variant="primary")
+    
+    status_output = gr.Textbox(label="Status & Results", lines=10, interactive=False)
+    results_table = gr.DataFrame(label="Question Analysis", wrap=True, interactive=False)
 
-    run_button.click(
-        fn=run_and_submit_all,
-        outputs=[status_output, results_table]
-    )
+    run_button.click(fn=run_and_submit_all, outputs=[status_output, results_table])
 
 if __name__ == "__main__":
-    print("\n" + "="*60)
-    print("🚀 ENHANCED GAIA AGENT STARTING")
-    print("="*60)
+    print("Enhanced GAIA Agent Starting...")
     
     # Environment check
-    env_status = []
-    required_vars = [
-        ("SPACE_HOST", "Space hosting"),
-        ("SPACE_ID", "Space identification"), 
-        ("SERPER_API_KEY", "Advanced web search"),
-        ("HUGGINGFACE_INFERENCE_TOKEN", "Model access")
-    ]
-    
-    for var_name, description in required_vars:
-        if os.getenv(var_name):
-            env_status.append(f"✅ {var_name}: Ready")
-        else:
-            env_status.append(f"❌ {var_name}: Missing ({description})")
+    env_vars = ["SPACE_HOST", "SPACE_ID", "SERPER_API_KEY", "HUGGINGFACE_INFERENCE_TOKEN"]
+    for var in env_vars:
+        status = "✅" if os.getenv(var) else "❌" 
+        print(f"{status} {var}")
     
-    print("\n📋 Environment Status:")
-    for status in env_status:
-        print(f"  {status}")
-    
-    print(f"\n🎯 Target Accuracy: 35%")
-    print(f"🔧 Enhanced Tools: 7 specialized tools loaded")
-    print(f"🌐 Web Search: Serper API + DuckDuckGo fallback")
-    print(f"📚 Knowledge: Wikipedia + Domain specialists")
-    print("\n" + "="*60)
-    
-    # Launch the interface
-    try:
-        demo.launch(
-            server_name="0.0.0.0",
-            server_port=7860,
-            share=False,
-            show_error=True,
-            quiet=False
-        )
-    except Exception as e:
-        print(f"❌ Error launching Gradio interface: {e}")
-        print("Attempting fallback launch...")
-        try:
-            demo.launch()
-        except Exception as fallback_error:
-            print(f"❌ Fallback launch failed: {fallback_error}")
-            exit(1)
+    demo.launch(server_name="0.0.0.0", server_port=7860, share=False)