NirmitSachde/Valorant-Match-Predictor

🎮 Valorant Match Predictor API

A comprehensive ML-powered prediction API for VALORANT esports matches.

🚀 Features

• —Team Matchup Predictions: Head-to-head analysis with historical data
• —Player Synergy Calculations: Team chemistry and compatibility scores
• —Map Preference Analysis: Team performance on different maps
• —Role Composition Optimization: Best team compositions for each map
• —Agent Selection Recommendations: Optimal agent picks for players
• —Comprehensive Match Predictions: Multi-factor analysis with confidence scores

📡 API Endpoints

Data Endpoints (GET)

• —/api/health - Health check
• —/api/data/teams - Get all teams with ratings
• —/api/data/players - Get all players with stats
• —/api/data/maps - Get available maps
• —/api/data/agents - Get agents by role
• —/api/data/rankings - Get current team rankings
• —/api/stats/team/{team_name} - Get detailed team statistics
• —/api/stats/player/{player_name} - Get detailed player statistics

Prediction Endpoints (POST)

• —/api/predict/matchup - Predict team vs team outcome
• —/api/predict/synergy - Calculate player synergy scores
• —/api/predict/map-preference - Get map advantages for teams
• —/api/predict/role-composition - Get optimal role distributions
• —/api/predict/agent-selection - Get agent recommendations
• —/api/predict/final - Generate comprehensive match prediction

🔧 Usage Example

// Predict a match
fetch('https://your-space.hf.space/api/predict/matchup', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    team_a: 'Sentinels',
    team_b: 'Gen.G'
  })
})
.then(res => res.json())
.then(data => console.log(data));

📊 Response Format

{
  "win_probability": 0.52,
  "confidence": 0.85,
  "components": {
    "team_matchup": 0.48,
    "player_synergy": 0.55,
    "map_advantage": 0.51,
    "role_composition": 0.53,
    "agent_selection": 0.50,
    "individual_form": 0.54
  }
}

🛠️ Technical Stack

• —Framework: FastAPI
• —ML Models: Scikit-learn (pickle format)
• —Data Processing: Pandas, NumPy
• —Server: Uvicorn
• —Deployment: Docker on Hugging Face Spaces

📈 Model Components

1. 1.Team Network Model: Analyzes team relationships and historical performance
2. 2.Player Synergy Model: Evaluates player compatibility and team chemistry
3. 3.Map Selection Model: Predicts map-specific advantages
4. 4.Role Composition Model: Optimizes team role distributions
5. 5.Agent Selection Model: Recommends optimal agent picks

🔗 Frontend Integration

This API is designed to work with the Valorant Predictor React application.

CORS is enabled for all origins. For production, update the CORS settings in app.py:

app.add_middleware(
    CORSMiddleware,
    allow_origins=["https://your-frontend-domain.com"],
    allow_credentials=True,
    allow_methods=["*"],
    allow_headers=["*"],
)

📝 Environment Variables

• —HF_TOKEN: Hugging Face token for private model access (optional)
• —PORT: Server port (default: 7860)

🚦 Status

• —✅ API Status: Active
• —✅ Models Loaded: 5/5
• —✅ Data Files: 12/12
• —✅ Health Check: Passing

📧 Contact

For issues or questions, please open an issue on the repository.

Built with ❤️ for the VALORANT esports community