Update app.py

app.py

@@ -1,8 +1,10 @@
 import gradio as gr
 import pandas as pd
 from apscheduler.schedulers.background import BackgroundScheduler
+from datetime import datetime
 
-# --- Placeholder Imports / Definitions ---
+# --- Make sure these imports work relative to your file structure ---
+# Option 1: If src is a directory in the same folder as your script:
 try:
     from src.about import (
         CITATION_BUTTON_LABEL,
@@ -10,210 +12,846 @@ try:
         EVALUATION_QUEUE_TEXT,
         INTRODUCTION_TEXT,
         LLM_BENCHMARKS_TEXT,
-        TITLE,  # We will override TITLE below
+        TITLE,
     )
     from src.display.css_html_js import custom_css
     from src.envs import REPO_ID
     from src.submission.submit import add_new_eval
+    print("Successfully imported from src module.")
+# Option 2: If you don't have these files, define placeholders
 except ImportError:
+    print("Warning: Using placeholder values because src module imports failed.")
     CITATION_BUTTON_LABEL = "Citation"
     CITATION_BUTTON_TEXT = "Please cite us if you use this benchmark..."
     EVALUATION_QUEUE_TEXT = "Current evaluation queue:"
-    INTRODUCTION_TEXT = "Welcome to the MLE-Dojo Benchmark Leaderboard."
-    LLM_BENCHMARKS_TEXT = "Information about the benchmarks..."
+    INTRODUCTION_TEXT = """
+    # Welcome to the MLE-Dojo Benchmark Leaderboard
+    
+    This leaderboard tracks the performance of various AI models across multiple machine learning engineering domains.
+    Our comprehensive evaluation system uses ELO ratings to provide a fair comparison between different models.
+    
+    ## How to read this leaderboard
+    - Select a domain category to view specialized rankings
+    - Higher ELO scores indicate better performance
+    - Click on any model name to learn more about it
+    """
+    LLM_BENCHMARKS_TEXT = """
+    # About the MLE-Dojo Benchmark
+    
+    ## Evaluation Methodology
+    The MLE-Dojo benchmark evaluates models across various domains including:
+    
+    - **MLE-Lite**: Basic machine learning engineering tasks
+    - **Tabular**: Data manipulation, analysis, and modeling with structured data
+    - **NLP**: Natural language processing tasks including classification, generation, and understanding
+    - **CV**: Computer vision tasks including image classification, object detection, and generation
+    
+    Our evaluation uses a sophisticated ELO rating system that considers the relative performance of models against each other.
+    
+    ## Contact
+    For more information or to submit your model, please contact us at [email protected]
+    """
+    TITLE = "<h1>🏆 MLE-Dojo Benchmark Leaderboard</h1>"
     custom_css = ""
     REPO_ID = "your/space-id"
     def add_new_eval(*args): return "Submission placeholder."
 
-# --- Elo Data ---
+# --- Elo Leaderboard Configuration ---
+# Enhanced data with Rank (placeholder), Organizer, License, and URL
 data = [
-    {'model_name': 'gpt-4o-mini',   'url': 'https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/', 'organizer': 'OpenAI',   'license': 'Proprietary', 'MLE-Lite_Elo': 753,  'Tabular_Elo': 839,  'NLP_Elo': 758,  'CV_Elo': 754,  'Overall': 778},
-    {'model_name': 'gpt-4o',        'url': 'https://openai.com/index/hello-gpt-4o/',                              'organizer': 'OpenAI',   'license': 'Proprietary', 'MLE-Lite_Elo': 830,  'Tabular_Elo': 861,  'NLP_Elo': 903,  'CV_Elo': 761,  'Overall': 841},
-    {'model_name': 'o3-mini',       'url': 'https://openai.com/index/openai-o3-mini/',                              'organizer': 'OpenAI',   'license': 'Proprietary', 'MLE-Lite_Elo': 1108, 'Tabular_Elo': 1019, 'NLP_Elo': 1056, 'CV_Elo': 1207, 'Overall': 1096},
-    {'model_name': 'deepseek-v3',   'url': 'https://api-docs.deepseek.com/news/news1226',                          'organizer': 'DeepSeek','license': 'DeepSeek',     'MLE-Lite_Elo': 1004, 'Tabular_Elo': 1015, 'NLP_Elo': 1028, 'CV_Elo': 1067, 'Overall': 1023},
-    {'model_name': 'deepseek-r1',   'url': 'https://api-docs.deepseek.com/news/news250120',                         'organizer': 'DeepSeek','license': 'DeepSeek',     'MLE-Lite_Elo': 1137, 'Tabular_Elo': 1053, 'NLP_Elo': 1103, 'CV_Elo': 1083, 'Overall': 1100},
-    {'model_name': 'gemini-2.0-flash','url': 'https://ai.google.dev/gemini-api/docs/models#gemini-2.0-flash',        'organizer': 'Google',   'license': 'Proprietary', 'MLE-Lite_Elo': 847,  'Tabular_Elo': 923,  'NLP_Elo': 860,  'CV_Elo': 978,  'Overall': 895},
-    {'model_name': 'gemini-2.0-pro', 'url': 'https://blog.google/technology/google-deepmind/gemini-model-updates-february-2025/', 'organizer': 'Google', 'license': 'Proprietary', 'MLE-Lite_Elo': 1064, 'Tabular_Elo': 1139, 'NLP_Elo': 1028, 'CV_Elo': 973,  'Overall': 1054},
-    {'model_name': 'gemini-2.5-pro', 'url': 'https://deepmind.google/technologies/gemini/pro/',               'organizer': 'Google',   'license': 'Proprietary', 'MLE-Lite_Elo': 1257, 'Tabular_Elo': 1150, 'NLP_Elo': 1266, 'CV_Elo': 1177, 'Overall': 1214},
+    {'model_name': 'gpt-4o-mini', 'url': 'https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/', 'organizer': 'OpenAI', 'license': 'Proprietary', 'MLE-Lite_Elo': 753, 'Tabular_Elo': 839, 'NLP_Elo': 758, 'CV_Elo': 754, 'Overall': 778},
+    {'model_name': 'gpt-4o', 'url': 'https://openai.com/index/hello-gpt-4o/', 'organizer': 'OpenAI', 'license': 'Proprietary', 'MLE-Lite_Elo': 830, 'Tabular_Elo': 861, 'NLP_Elo': 903, 'CV_Elo': 761, 'Overall': 841},
+    {'model_name': 'o3-mini', 'url': 'https://openai.com/index/openai-o3-mini/', 'organizer': 'OpenAI', 'license': 'Proprietary', 'MLE-Lite_Elo': 1108, 'Tabular_Elo': 1019, 'NLP_Elo': 1056, 'CV_Elo': 1207, 'Overall': 1096},
+    {'model_name': 'deepseek-v3', 'url': 'https://api-docs.deepseek.com/news/news1226', 'organizer': 'DeepSeek', 'license': 'DeepSeek', 'MLE-Lite_Elo': 1004, 'Tabular_Elo': 1015, 'NLP_Elo': 1028, 'CV_Elo': 1067, 'Overall': 1023},
+    {'model_name': 'deepseek-r1', 'url': 'https://api-docs.deepseek.com/news/news250120', 'organizer': 'DeepSeek', 'license': 'DeepSeek', 'MLE-Lite_Elo': 1137, 'Tabular_Elo': 1053, 'NLP_Elo': 1103, 'CV_Elo': 1083, 'Overall': 1100},
+    {'model_name': 'gemini-2.0-flash', 'url': 'https://ai.google.dev/gemini-api/docs/models#gemini-2.0-flash', 'organizer': 'Google', 'license': 'Proprietary', 'MLE-Lite_Elo': 847, 'Tabular_Elo': 923, 'NLP_Elo': 860, 'CV_Elo': 978, 'Overall': 895},
+    {'model_name': 'gemini-2.0-pro', 'url': 'https://blog.google/technology/google-deepmind/gemini-model-updates-february-2025/', 'organizer': 'Google', 'license': 'Proprietary', 'MLE-Lite_Elo': 1064, 'Tabular_Elo': 1139, 'NLP_Elo': 1028, 'CV_Elo': 973, 'Overall': 1054},
+    {'model_name': 'gemini-2.5-pro', 'url': 'https://deepmind.google/technologies/gemini/pro/', 'organizer': 'Google', 'license': 'Proprietary', 'MLE-Lite_Elo': 1257, 'Tabular_Elo': 1150, 'NLP_Elo': 1266, 'CV_Elo': 1177, 'Overall': 1214},
 ]
+
+# Add organization logos (for visual enhancement)
+org_logos = {
+    'OpenAI': '📱',  # You can replace these with actual icon URLs in production
+    'DeepSeek': '🔍',
+    'Google': '🌐',
+    'Default': '🤖'
+}
+
+# Create a master DataFrame
 master_df = pd.DataFrame(data)
 
-CATEGORIES = ["Overall", "MLE-Lite", "Tabular", "NLP", "CV"]
+# Add last updated timestamp
+last_updated = datetime.now().strftime("%B %d, %Y at %H:%M:%S")
+
+# Define categories with fancy icons
+CATEGORIES = [
+    ("🏆 Overall", "Overall"), 
+    ("💡 MLE-Lite", "MLE-Lite"), 
+    ("📊 Tabular", "Tabular"), 
+    ("📝 NLP", "NLP"), 
+    ("👁️ CV", "CV")
+]
 DEFAULT_CATEGORY = "Overall"
+
+# Map user-facing categories to DataFrame column names
 category_to_column = {
-    "Overall": "Overall",
     "MLE-Lite": "MLE-Lite_Elo",
     "Tabular": "Tabular_Elo",
     "NLP": "NLP_Elo",
     "CV": "CV_Elo",
+    "Overall": "Overall"
 }
 
-def update_leaderboard(category):
-    col = category_to_column.get(category, category_to_column[DEFAULT_CATEGORY])
-    df = master_df[['model_name','url','organizer','license',col]].copy()
-    df.sort_values(by=col, ascending=False, inplace=True)
+# --- Helper function to update leaderboard ---
+def update_leaderboard(category_label):
+    """
+    Enhanced function to update the leaderboard with visual improvements
+    """
+    # Extract the category value from the label if it's a tuple (icon, value)
+    if isinstance(category_label, tuple):
+        category = category_label[1]
+    else:
+        # For backward compatibility or direct values
+        category = category_label.split(" ")[-1] if " " in category_label else category_label
+    
+    score_column = category_to_column.get(category)
+    if score_column is None or score_column not in master_df.columns:
+        print(f"Warning: Invalid category '{category}' or column '{score_column}'. Falling back to default.")
+        score_column = category_to_column[DEFAULT_CATEGORY]
+        if score_column not in master_df.columns:
+            print(f"Error: Default column '{score_column}' also not found.")
+            return pd.DataFrame({
+                "Rank": [],
+                "Model": [],
+                "Organizer": [],
+                "License": [],
+                "Elo Score": []
+            })
+
+    # Select base columns + the score column for sorting
+    cols_to_select = ['model_name', 'url', 'organizer', 'license', score_column]
+    df = master_df[cols_to_select].copy()
+
+    # Sort by the selected 'Elo Score' descending
+    df.sort_values(by=score_column, ascending=False, inplace=True)
+
+    # Add Rank with medal emojis for top 3
     df.reset_index(drop=True, inplace=True)
-    df.insert(0, 'Rank', df.index+1)
+    
+    # Create fancy rank with medals for top positions
+    def get_rank_display(idx):
+        if idx == 0:
+            return "🥇 1"
+        elif idx == 1:
+            return "🥈 2"
+        elif idx == 2:
+            return "🥉 3"
+        else:
+            return f"{idx + 1}"
+    
+    df.insert(0, 'Rank', df.index.map(get_rank_display))
+    
+    # Add organization icons to model names
     df['Model'] = df.apply(
-        lambda r: f"<a href='{r['url']}' target='_blank'>{r['model_name']}</a>",
+        lambda row: f"""<div style="display: flex; align-items: center;">
+            <span style="font-size: 1.5em; margin-right: 10px;">{org_logos.get(row['organizer'], org_logos['Default'])}</span>
+            <a href='{row['url'] if pd.notna(row['url']) else '#'}' target='_blank' 
+               style='color: #0066cc; text-decoration: none; font-weight: 500; font-size: 1.05em;'>
+               {row['model_name']}
+            </a>
+        </div>""",
         axis=1
     )
-    df.rename(columns={col:'Elo Score','organizer':'Organizer','license':'License'}, inplace=True)
-    return df[['Rank','Model','Organizer','License','Elo Score']]
 
-# --- Dark Theme + Custom CSS ---
-custom_css += """
-@import url('https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&display=swap');
+    # Format Elo scores with visual indicators
+    df['Elo Display'] = df[score_column].apply(
+        lambda score: f"""<div style="display: flex; align-items: center;">
+            <span style="font-weight: bold; color: {'#1a5fb4' if score >= 1000 else '#2ec27e' if score >= 900 else '#e5a50a' if score >= 800 else '#ff7800'}">
+                {score}
+            </span>
+            <div style="margin-left: 10px; height: 12px; width: 60px; background-color: #eaeaea; border-radius: 6px; overflow: hidden;">
+                <div style="height: 100%; width: {min(100, max(5, (score-700)/7))}%; background-color: {'#1a5fb4' if score >= 1000 else '#2ec27e' if score >= 900 else '#e5a50a' if score >= 800 else '#ff7800'};"></div>
+            </div>
+        </div>"""
+    )
+
+    # Rename columns for display
+    df.rename(columns={score_column: 'Elo Score'}, inplace=True)
+    df.rename(columns={'organizer': 'Organizer', 'license': 'License'}, inplace=True)
+
+    # Select and reorder columns for final display
+    final_columns = ["Rank", "Model", "Organizer", "License", "Elo Display"]
+    df = df[final_columns]
+
+    # Rename for display
+    df.columns = ["Rank", "Model", "Organization", "License", f"Elo Score ({category})"]
+    
+    return df
+
+# --- Mock/Placeholder functions/data for other tabs ---
+print("Warning: Evaluation queue data fetching is disabled/mocked due to leaderboard changes.")
+finished_eval_queue_df = pd.DataFrame(columns=["Model", "Status", "Requested", "Started"])
+running_eval_queue_df = pd.DataFrame(columns=["Model", "Status", "Requested", "Started"])
+pending_eval_queue_df = pd.DataFrame(columns=["Model", "Status", "Requested", "Started"])
+EVAL_COLS = ["Model", "Status", "Requested", "Started"]
+EVAL_TYPES = ["str", "str", "str", "str"]
+
+# --- Keep restart function if relevant ---
+def restart_space():
+    print(f"Attempting to restart space: {REPO_ID}")
+    # Replace with your actual space restart mechanism if needed
 
-body {
-    font-family: 'Inter', sans-serif;
-    background-color: #1e1e2f !important;
-    color: #e0e0f0 !important;
+# --- Enhanced CSS for beauty and readability ---
+enhanced_css = """
+/* Base styling */
+:root {
+    --primary-color: #1a5fb4;
+    --secondary-color: #2ec27e;
+    --accent-color: #e5a50a;
+    --warning-color: #ff7800;
+    --text-color: #333333;
+    --background-color: #ffffff;
+    --card-background: #f9f9f9;
+    --border-color: #e0e0e0;
+    --shadow-color: rgba(0, 0, 0, 0.1);
 }
 
-/* Hero Section */
-.hero-section {
-    background: linear-gradient(135deg, #6c63ff, #8f94fb);
-    color: #fff;
-    padding: 2rem 1rem;
-    border-radius: .75rem;
-    margin-bottom: 1.5rem;
-    text-align: center;
-    box-shadow: 0 4px 10px rgba(0,0,0,0.3);
+/* Typography */
+body, .gradio-container {
+    font-family: 'Inter', 'Segoe UI', Roboto, -apple-system, BlinkMacSystemFont, system-ui, sans-serif !important;
+    font-size: 16px !important;
+    line-height: 1.6 !important;
+    color: var(--text-color) !important;
+    background-color: var(--background-color) !important;
 }
-.hero-section h1 {
-    margin: 0;
+
+/* Headings */
+h1 {
     font-size: 2.5rem !important;
     font-weight: 700 !important;
+    margin-bottom: 1.5rem !important;
+    color: var(--primary-color) !important;
+    text-align: center !important;
+    letter-spacing: -0.02em !important;
+    line-height: 1.2 !important;
+}
+
+h2 {
+    font-size: 1.8rem !important;
+    font-weight: 600 !important;
+    margin-top: 1.5rem !important;
+    margin-bottom: 1rem !important;
+    color: var(--primary-color) !important;
+    letter-spacing: -0.01em !important;
+}
+
+h3 {
+    font-size: 1.4rem !important;
+    font-weight: 600 !important;
+    margin-top: 1.2rem !important;
+    margin-bottom: 0.8rem !important;
+    color: var(--text-color) !important;
+}
+
+/* Tabs styling */
+.tabs {
+    margin-top: 1rem !important;
+    border-radius: 12px !important;
+    overflow: hidden !important;
+    box-shadow: 0 4px 12px var(--shadow-color) !important;
+}
+
+.tab-nav button {
+    font-size: 1.1rem !important;
+    font-weight: 500 !important;
+    padding: 0.8rem 1.5rem !important;
+    border-radius: 0 !important;
+    transition: all 0.2s ease !important;
+}
+
+.tab-nav button.selected {
+    background-color: var(--primary-color) !important;
+    color: white !important;
+    font-weight: 600 !important;
+}
+
+/* Card styling */
+.gradio-container .gr-box, .gradio-container .gr-panel {
+    border-radius: 12px !important;
+    border: 1px solid var(--border-color) !important;
+    box-shadow: 0 4px 12px var(--shadow-color) !important;
+    overflow: hidden !important;
+}
+
+/* Table styling */
+table {
+    width: 100% !important;
+    border-collapse: separate !important;
+    border-spacing: 0 !important;
+    margin: 1.5rem 0 !important;
+    border-radius: 8px !important;
+    overflow: hidden !important;
+    box-shadow: 0 4px 12px var(--shadow-color) !important;
 }
-.hero-section h2 {
-    margin: .5rem 0 0 !important;
-    font-size: 1.25rem !important;
-    font-weight: 400 !important;
-    opacity: 0.9;
+
+th {
+    background-color: #f0f5ff !important;
+    color: var(--primary-color) !important;
+    font-weight: 600 !important;
+    padding: 1rem !important;
+    font-size: 1.1rem !important;
+    text-align: left !important;
+    border-bottom: 2px solid var(--primary-color) !important;
 }
 
-/* Tab Buttons */
-.tab-buttons button {
-    border-radius: 20px !important;
-    padding: 0.5rem 1rem !important;
-    margin-right: 0.5rem !important;
-    background: #3a3a4c !important;
-    color: #e0e0f0 !important;
+td {
+    padding: 1rem !important;
+    border-bottom: 1px solid var(--border-color) !important;
+    font-size: 1rem !important;
+    vertical-align: middle !important;
+}
+
+tr:nth-child(even) {
+    background-color: #f8fafd !important;
+}
+
+tr:hover {
+    background-color: #edf2fb !important;
+}
+
+tr:first-child td {
+    border-top: none !important;
+}
+
+/* Button styling */
+button.primary, .gr-button.primary {
+    background-color: var(--primary-color) !important;
+    color: white !important;
+    font-weight: 500 !important;
+    padding: 0.8rem 1.5rem !important;
+    border-radius: 8px !important;
     border: none !important;
-    transition: background 0.3s !important;
+    cursor: pointer !important;
+    transition: all 0.2s ease !important;
+    box-shadow: 0 2px 5px rgba(0, 0, 0, 0.1) !important;
+}
+
+button.primary:hover, .gr-button.primary:hover {
+    background-color: #0b4a9e !important;
+    box-shadow: 0 4px 8px rgba(0, 0, 0, 0.15) !important;
+    transform: translateY(-1px) !important;
+}
+
+/* Radio buttons */
+.gr-radio {
+    display: flex !important;
+    flex-wrap: wrap !important;
+    gap: 10px !important;
+    margin: 1rem 0 !important;
+}
+
+.gr-radio label {
+    background-color: #f5f7fa !important;
+    border: 1px solid var(--border-color) !important;
+    border-radius: 8px !important;
+    padding: 0.7rem 1.2rem !important;
+    font-size: 1rem !important;
     font-weight: 500 !important;
+    cursor: pointer !important;
+    transition: all 0.2s ease !important;
+    display: flex !important;
+    align-items: center !important;
+    gap: 8px !important;
+}
+
+.gr-radio label:hover {
+    background-color: #eaeef3 !important;
+    border-color: #c0c9d6 !important;
 }
-.tab-buttons button:hover {
-    background: #4a4a6f !important;
+
+.gr-radio label.selected {
+    background-color: #e0e9f7 !important;
+    border-color: var(--primary-color) !important;
+    color: var(--primary-color) !important;
+    font-weight: 600 !important;
 }
-.tab-buttons button[aria-selected="true"] {
-    background: #6c63ff !important;
-    color: #fff !important;
+
+/* Input fields */
+input, textarea, select {
+    font-size: 1rem !important;
+    padding: 0.8rem !important;
+    border-radius: 8px !important;
+    border: 1px solid var(--border-color) !important;
+    transition: all 0.2s ease !important;
+}
+
+input:focus, textarea:focus, select:focus {
+    border-color: var(--primary-color) !important;
+    box-shadow: 0 0 0 2px rgba(26, 95, 180, 0.2) !important;
+    outline: none !important;
 }
 
-/* Category Selector Pills */
-#category-selector input[type="radio"] { display: none; }
-#category-selector label {
+/* Accordion styling */
+.gr-accordion {
+    border-radius: 8px !important;
+    overflow: hidden !important;
+    margin: 1rem 0 !important;
+    border: 1px solid var(--border-color) !important;
+}
+
+.gr-accordion-header {
+    padding: 1rem !important;
+    background-color: #f5f7fa !important;
+    font-weight: 600 !important;
+    font-size: 1.1rem !important;
+    color: var(--text-color) !important;
+}
+
+.gr-accordion-content {
+    padding: 1rem !important;
+    background-color: white !important;
+}
+
+/* Markdown text improvements */
+.markdown-text {
+    font-size: 1.05rem !important;
+    line-height: 1.7 !important;
+}
+
+.markdown-text p {
+    margin-bottom: 1rem !important;
+}
+
+.markdown-text ul, .markdown-text ol {
+    margin-left: 1.5rem !important;
+    margin-bottom: 1rem !important;
+}
+
+.markdown-text li {
+    margin-bottom: 0.5rem !important;
+}
+
+.markdown-text strong {
+    font-weight: 600 !important;
+    color: #333 !important;
+}
+
+/* Status indicators */
+.status-badge {
     display: inline-block;
-    padding: 0.5rem 1rem;
-    margin-right: 0.5rem;
-    border-radius: 999px;
-    background: #3a3a4c;
-    color: #e0e0f0;
-    cursor: pointer;
-    transition: background 0.3s, color 0.3s;
+    padding: 0.3rem 0.7rem;
+    border-radius: 99px;
+    font-size: 0.85rem;
     font-weight: 500;
+    text-align: center;
 }
-#category-selector input[type="radio"]:checked + label {
-    background: #6c63ff;
-    color: #fff;
+
+.status-pending {
+    background-color: #fff8e0;
+    color: #b58a00;
+    border: 1px solid #ffd74d;
 }
 
-/* Table Styling */
-table {
-    width: 100%;
-    border: none;
-    border-radius: .5rem;
-    overflow: hidden;
-    box-shadow: 0 2px 4px rgba(0,0,0,0.3);
-    margin: 1rem 0;
+.status-running {
+    background-color: #e0f2ff;
+    color: #0066cc;
+    border: 1px solid #66b3ff;
 }
-th {
-    background: #6c63ff !important;
-    color: #fff !important;
+
+.status-completed {
+    background-color: #e6f7ef;
+    color: #00875a;
+    border: 1px solid #57d9a3;
 }
-td, th {
-    padding: 0.75rem 1rem;
-    background: #1e1e2f;
-    color: #e0e0f0;
+
+/* Footer */
+.footer {
+    margin-top: 2rem;
+    padding: 1rem;
+    text-align: center;
+    font-size: 0.9rem;
+    color: #666;
+    border-top: 1px solid var(--border-color);
 }
-tr:nth-child(even) td {
-    background: #2a2a3c;
+
+/* Enhanced leaderboard title */
+.leaderboard-header {
+    display: flex;
+    align-items: center;
+    justify-content: space-between;
+    margin-bottom: 1.5rem;
+    padding-bottom: 1rem;
+    border-bottom: 2px solid var(--border-color);
 }
-tr:hover td {
-    background: #3c3b52;
+
+.leaderboard-title {
+    font-size: 2.2rem;
+    font-weight: 700;
+    color: var(--primary-color);
+    margin: 0;
+    display: flex;
+    align-items: center;
+    gap: 0.5rem;
 }
-td a {
-    color: #9afeff;
-    text-decoration: none;
+
+.leaderboard-subtitle {
+    font-size: 1.1rem;
+    color: #666;
+    margin-top: 0.5rem;
 }
-td a:hover {
-    text-decoration: underline;
+
+.timestamp {
+    font-size: 0.85rem;
+    color: #666;
+    font-style: italic;
+}
+
+/* Category selector buttons */
+.category-buttons {
+    display: flex;
+    flex-wrap: wrap;
+    gap: 10px;
+    margin-bottom: 1.5rem;
+}
+
+.category-button {
+    padding: 0.7rem 1.2rem;
+    background-color: #f0f5ff;
+    border: 1px solid #d0e0ff;
+    border-radius: 8px;
+    font-weight: 500;
+    cursor: pointer;
+    transition: all 0.2s ease;
+    display: flex;
+    align-items: center;
+    gap: 8px;
+}
+
+.category-button:hover {
+    background-color: #e0ebff;
+    border-color: #b0d0ff;
+}
+
+.category-button.active {
+    background-color: var(--primary-color);
+    color: white;
+    border-color: var(--primary-color);
 }
-"""
 
-# --- Override Title with Hero ---
-TITLE = """
-<div class="hero-section">
-  <h1>🏆 MLE-Dojo Benchmark Leaderboard</h1>
-  <h2>Improving LLM Agents for Machine Learning Engineering</h2>
-</div>
+/* Logo and brand styling */
+.logo {
+    font-size: 2.5em;
+    margin-right: 0.5rem;
+}
+
+/* Medal styling for top ranks */
+.rank-1 {
+    color: #ffd700;
+    font-weight: bold;
+}
+
+.rank-2 {
+    color: #c0c0c0;
+    font-weight: bold;
+}
+
+.rank-3 {
+    color: #cd7f32;
+    font-weight: bold;
+}
 """
 
-# --- Build App ---
-demo = gr.Blocks(css=custom_css, theme=gr.themes.SoftDark())
+# Combine with any existing CSS
+custom_css = enhanced_css + custom_css
+
+# --- Gradio App Definition ---
+demo = gr.Blocks(css=custom_css, theme=gr.themes.Soft())
 
 with demo:
-    gr.HTML(TITLE)
+    # Enhanced header with timestamp
+    gr.HTML(f"""
+    <div class="leaderboard-header">
+        <div>
+            <div class="leaderboard-title">
+                <span class="logo">🏆</span> MLE-Dojo Benchmark Leaderboard
+            </div>
+            <div class="leaderboard-subtitle">
+                Comprehensive evaluation of AI models across multiple domains
+            </div>
+        </div>
+        <div class="timestamp">
+            Last updated: {last_updated}
+        </div>
+    </div>
+    """)
+
+    # Introduction with enhanced styling
     gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text")
 
     with gr.Tabs(elem_classes="tab-buttons") as tabs:
-        with gr.TabItem("📋 Leaderboard"):
-            gr.Markdown("## Model Elo Rankings by Category")
-            category_selector = gr.Radio(
-                choices=CATEGORIES,
-                value=DEFAULT_CATEGORY,
-                interactive=True,
-                elem_id="category-selector"
-            )
-            leaderboard_df = gr.Dataframe(
-                value=update_leaderboard(DEFAULT_CATEGORY),
-                headers=["Rank","Model","Organizer","License","Elo Score"],
-                datatype=["number","html","str","str","number"],
-                interactive=False,
-                row_count=(len(master_df),"fixed"),
-                col_count=(5,"fixed"),
-                wrap=True,
-                elem_id="leaderboard-table"
-            )
-            category_selector.change(
-                fn=update_leaderboard,
-                inputs=category_selector,
-                outputs=leaderboard_df
-            )
-
-        with gr.TabItem("ℹ️ About"):
+        with gr.TabItem("📊 Leaderboard", elem_id="llm-benchmark-tab-table", id=0):
+            with gr.Column():
+                gr.HTML("""
+                <h2 style="display: flex; align-items: center; gap: 10px;">
+                    <span style="font-size: 1.3em;">📈</span> Model Performance Rankings
+                </h2>
+                <p class="leaderboard-subtitle">Select a category to view specialized performance metrics</p>
+                """)
+                
+                # Enhanced category selector
+                category_selector = gr.Radio(
+                    choices=[x[0] for x in CATEGORIES],
+                    label="Select Performance Domain:",
+                    value="🏆 Overall",
+                    interactive=True,
+                    elem_classes="fancy-radio"
+                )
+                
+                # Visual separator
+                gr.HTML('<div style="height: 1px; background-color: #e0e0e0; margin: 20px 0;"></div>')
+                
+                # Enhanced leaderboard table
+                leaderboard_df_component = gr.Dataframe(
+                    value=update_leaderboard(DEFAULT_CATEGORY),
+                    headers=["Rank", "Model", "Organization", "License", f"Elo Score ({DEFAULT_CATEGORY})"],
+                    datatype=["html", "html", "str", "str", "html"],
+                    interactive=False,
+                    row_count=(len(master_df), "fixed"),
+                    col_count=(5, "fixed"),
+                    wrap=True,
+                    elem_id="leaderboard-table",
+                )
+                
+                # Stats cards (visual enhancement)
+                with gr.Row():
+                    with gr.Column(scale=1):
+                        gr.HTML(f"""
+                        <div style="background-color: #f0f5ff; padding: 20px; border-radius: 12px; text-align: center;">
+                            <div style="font-size: 2em;">🔍</div>
+                            <div style="font-size: 2em; font-weight: bold; color: #1a5fb4;">{len(master_df)}</div>
+                            <div style="font-size: 1.1em; color: #666;">Models Evaluated</div>
+                        </div>
+                        """)
+                    with gr.Column(scale=1):
+                        gr.HTML(f"""
+                        <div style="background-color: #e6f7ef; padding: 20px; border-radius: 12px; text-align: center;">
+                            <div style="font-size: 2em;">🌐</div>
+                            <div style="font-size: 2em; font-weight: bold; color: #00875a;">{master_df['organizer'].nunique()}</div>
+                            <div style="font-size: 1.1em; color: #666;">Organizations</div>
+                        </div>
+                        """)
+                    with gr.Column(scale=1):
+                        gr.HTML(f"""
+                        <div style="background-color: #fff8e0; padding: 20px; border-radius: 12px; text-align: center;">
+                            <div style="font-size: 2em;">🏅</div>
+                            <div style="font-size: 2em; font-weight: bold; color: #b58a00;">{len(CATEGORIES)}</div>
+                            <div style="font-size: 1.1em; color: #666;">Performance Domains</div>
+                        </div>
+                        """)
+                
+                # Link the radio button change to the update function
+                category_selector.change(
+                    fn=update_leaderboard,
+                    inputs=category_selector,
+                    outputs=leaderboard_df_component
+                )
+
+        with gr.TabItem("📚 About", elem_id="llm-benchmark-tab-about", id=1):
+            # Enhanced about section
+            gr.HTML("""
+            <div class="about-header" style="display: flex; align-items: center; gap: 20px; margin-bottom: 20px;">
+                <div style="font-size: 4em;">🧪</div>
+                <div>
+                    <h2 style="margin: 0;">About the MLE-Dojo Benchmark</h2>
+                    <p style="margin: 5px 0 0 0; color: #666;">A comprehensive evaluation framework for AI models</p>
+                </div>
+            </div>
+            """)
+            
+            # Use the LLM_BENCHMARKS_TEXT variable
             gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text")
+            
+            # Add methodology cards for visual enhancement
+            with gr.Row():
+                with gr.Column():
+                    gr.HTML("""
+                    <div style="background-color: #f5f7fa; padding: 20px; border-radius: 12px; height: 100%;">
+                        <div style="font-size: 2em; text-align: center; margin-bottom: 15px;">💡</div>
+                        <h3 style="text-align: center; margin-top: 0;">MLE-Lite</h3>
+                        <p>Evaluates a model's ability to handle basic machine learning engineering tasks including 
+                        data preprocessing, feature engineering, model selection, and basic deployment.</p>
+                    </div>
+                    """)
+                with gr.Column():
+                    gr.HTML("""
+                    <div style="background-color: #f5f7fa; padding: 20px; border-radius: 12px; height: 100%;">
+                        <div style="font-size: 2em; text-align: center; margin-bottom: 15px;">📊</div>
+                        <h3 style="text-align: center; margin-top: 0;">Tabular</h3>
+                        <p>Tests a model's ability to process, analyze and model structured data, including 
+                        statistical analysis,statistical analysis, predictive modeling, and data visualization with tabular datasets.</p>
+                    </div>
+                    """)
+            
+            with gr.Row():
+                with gr.Column():
+                    gr.HTML("""
+                    <div style="background-color: #f5f7fa; padding: 20px; border-radius: 12px; height: 100%;">
+                        <div style="font-size: 2em; text-align: center; margin-bottom: 15px;">📝</div>
+                        <h3 style="text-align: center; margin-top: 0;">NLP</h3>
+                        <p>Evaluates natural language processing capabilities including text classification, 
+                        sentiment analysis, entity recognition, text generation, and language understanding.</p>
+                    </div>
+                    """)
+                with gr.Column():
+                    gr.HTML("""
+                    <div style="background-color: #f5f7fa; padding: 20px; border-radius: 12px; height: 100%;">
+                        <div style="font-size: 2em; text-align: center; margin-bottom: 15px;">👁️</div>
+                        <h3 style="text-align: center; margin-top: 0;">CV</h3>
+                        <p>Tests computer vision capabilities including image classification, object detection, 
+                        image generation, and visual understanding tasks across various domains.</p>
+                    </div>
+                    """)
 
-    with gr.Accordion("📙 Citation", open=False):
-        gr.Textbox(
+        # Optional: Uncomment if you want to re-enable the Submit tab
+        # with gr.TabItem("🚀 Submit Model", elem_id="llm-benchmark-tab-submit", id=2):
+        #     with gr.Column():
+        #         gr.HTML("""
+        #         <div class="about-header" style="display: flex; align-items: center; gap: 20px; margin-bottom: 20px;">
+        #             <div style="font-size: 4em;">🚀</div>
+        #             <div>
+        #                 <h2 style="margin: 0;">Submit Your Model for Evaluation</h2>
+        #                 <p style="margin: 5px 0 0 0; color: #666;">Add your model to the MLE-Dojo leaderboard</p>
+        #             </div>
+        #         </div>
+        #         """)
+        #
+        #         with gr.Row():
+        #             gr.Markdown(EVALUATION_QUEUE_TEXT, elem_classes="markdown-text")
+        #         
+        #         with gr.Column():
+        #             with gr.Accordion(f"✅ Finished Evaluations ({len(finished_eval_queue_df)})", open=False):
+        #                  finished_eval_table = gr.components.Dataframe(
+        #                      value=finished_eval_queue_df, headers=EVAL_COLS, datatype=EVAL_TYPES, row_count=5,
+        #                 )
+        #             with gr.Accordion(f"🔄 Running Evaluation Queue ({len(running_eval_queue_df)})", open=False):
+        #                  running_eval_table = gr.components.Dataframe(
+        #                      value=running_eval_queue_df, headers=EVAL_COLS, datatype=EVAL_TYPES, row_count=5,
+        #                 )
+        #             with gr.Accordion(f"⏳ Pending Evaluation Queue ({len(pending_eval_queue_df)})", open=False):
+        #                 pending_eval_table = gr.components.Dataframe(
+        #                     value=pending_eval_queue_df, headers=EVAL_COLS, datatype=EVAL_TYPES, row_count=5,
+        #                 )
+        #         
+        #         gr.HTML('<div style="height: 1px; background-color: #e0e0e0; margin: 20px 0;"></div>')
+        #         
+        #         gr.HTML("""
+        #         <h2 style="display: flex; align-items: center; gap: 10px;">
+        #             <span style="font-size: 1.3em;">📝</span> Model Submission Form
+        #         </h2>
+        #         """)
+        #         
+        #         with gr.Row():
+        #             with gr.Column():
+        #                 model_name_textbox = gr.Textbox(
+        #                     label="Model Name (on Hugging Face Hub)",
+        #                     placeholder="Enter your model name...",
+        #                     elem_classes="enhanced-input"
+        #                 )
+        #                 revision_name_textbox = gr.Textbox(
+        #                     label="Revision / Commit Hash",
+        #                     placeholder="main",
+        #                     elem_classes="enhanced-input"
+        #                 )
+        #                 model_type = gr.Dropdown(
+        #                     choices=["Type A", "Type B", "Type C"],
+        #                     label="Model Type",
+        #                     multiselect=False,
+        #                     value=None,
+        #                     interactive=True,
+        #                     elem_classes="enhanced-dropdown"
+        #                 )
+        #             with gr.Column():
+        #                 precision = gr.Dropdown(
+        #                     choices=["float16", "bfloat16", "float32", "int8", "auto"],
+        #                     label="Precision",
+        #                     multiselect=False,
+        #                     value="auto",
+        #                     interactive=True,
+        #                     elem_classes="enhanced-dropdown"
+        #                 )
+        #                 weight_type = gr.Dropdown(
+        #                     choices=["Original", "Adapter", "Delta"],
+        #                     label="Weights Type",
+        #                     multiselect=False,
+        #                     value="Original",
+        #                     interactive=True,
+        #                     elem_classes="enhanced-dropdown"
+        #                 )
+        #                 base_model_name_textbox = gr.Textbox(
+        #                     label="Base Model (for delta or adapter weights)",
+        #                     placeholder="Only needed for adapter/delta weights",
+        #                     elem_classes="enhanced-input"
+        #                 )
+        #         
+        #         submit_button = gr.Button(
+        #             "Submit for Evaluation",
+        #             elem_classes="primary-button"
+        #         )
+        #         submission_result = gr.Markdown()
+        #         submit_button.click(
+        #             add_new_eval,
+        #             [model_name_textbox, base_model_name_textbox, revision_name_textbox, precision, weight_type, model_type],
+        #             submission_result,
+        #         )
+
+    # Enhanced citation section
+    with gr.Accordion("📄 Citation", open=False, elem_classes="citation-accordion"):
+        gr.HTML("""
+        <div style="display: flex; align-items: center; gap: 20px; margin-bottom: 15px;">
+            <div style="font-size: 2.5em;">📄</div>
+            <div>
+                <h3 style="margin: 0;">How to Cite This Benchmark</h3>
+                <p style="margin: 5px 0 0 0; color: #666;">Please use the following citation if you use this benchmark in your research</p>
+            </div>
+        </div>
+        """)
+        
+        citation_button = gr.Textbox(
             value=CITATION_BUTTON_TEXT,
             label=CITATION_BUTTON_LABEL,
             lines=10,
             elem_id="citation-button",
-            show_copy_button=True
+            show_copy_button=True,
         )
+    
+    # Footer
+    gr.HTML("""
+    <div class="footer">
+        <p>© 2025 MLE-Dojo Benchmark. All rights reserved.</p>
+        <p style="margin-top: 5px; display: flex; justify-content: center; gap: 20px;">
+            <a href="#" style="color: #1a5fb4; text-decoration: none;">Privacy Policy</a>
+            <a href="#" style="color: #1a5fb4; text-decoration: none;">Terms of Service</a>
+            <a href="#" style="color: #1a5fb4; text-decoration: none;">Contact Us</a>
+        </p>
+    </div>
+    """)
+
+# --- Keep scheduler if relevant ---
+if __name__ == "__main__":
+    try:
+        scheduler = BackgroundScheduler()
+        if callable(restart_space):
+             if REPO_ID and REPO_ID != "your/space-id":
+                 scheduler.add_job(restart_space, "interval", seconds=1800)  # Restart every 30 mins
+                 scheduler.start()
+             else:
+                 print("Warning: REPO_ID not set or is placeholder; space restart job not scheduled.")
+        else:
+             print("Warning: restart_space function not available; space restart job not scheduled.")
+    except Exception as e:
+        print(f"Failed to initialize or start scheduler: {e}")
 
+# --- Launch the app ---
 if __name__ == "__main__":
-    demo.launch()
+    print("Launching Enhanced Gradio App...")
+    demo.launch()