api_examples.py

"""
API Usage Examples for Academic Evaluation System

This script demonstrates how to interact with the ML prediction API endpoints.
Run this after the Django server is started.
"""

import requests
import json

# Base URL - update if different
BASE_URL = "http://localhost:8000"
API_BASE = f"{BASE_URL}/api"


def predict_student_risk(student_data):
    """
    Example: Predict risk for student data
    """
    url = f"{API_BASE}/predict-risk/"
    
    try:
        response = requests.post(url, json=student_data)
        response.raise_for_status()
        
        result = response.json()
        print("✅ Prediction Success!")
        print(f"   Prediction: {result['prediction']}")
        print(f"   Confidence: {result['confidence']:.2%}")
        print(f"   Risk Score: {result['details']['risk_score']:.1f}%")
        print()
        return result
        
    except requests.exceptions.RequestException as e:
        print(f"❌ Error: {e}")
        return None


def get_risk_statistics():
    """
    Example: Get system-wide risk statistics
    """
    url = f"{API_BASE}/risk-statistics/"
    
    try:
        response = requests.get(url)
        response.raise_for_status()
        
        stats = response.json()
        print("📊 Risk Statistics:")
        print(f"   Total Students: {stats['overall']['total']}")
        print(f"   At Risk: {stats['overall']['at_risk']} ({stats['overall']['at_risk_percentage']}%)")
        print(f"   Safe: {stats['overall']['safe']}")
        print()
        return stats
        
    except requests.exceptions.RequestException as e:
        print(f"❌ Error: {e}")
        return None


def update_student_risk(student_id, csrf_token):
    """
    Example: Update specific student's risk status
    Note: Requires authentication and CSRF token
    """
    url = f"{API_BASE}/update-risk/{student_id}/"
    headers = {
        'X-CSRFToken': csrf_token,
        'Cookie': f'csrftoken={csrf_token}'
    }
    
    try:
        response = requests.post(url, headers=headers)
        response.raise_for_status()
        
        result = response.json()
        print(f"✅ Updated {student_id}")
        print(f"   Risk Status: {result['risk_status']}")
        print(f"   Confidence: {result['risk_confidence']:.2%}")
        print()
        return result
        
    except requests.exceptions.RequestException as e:
        print(f"❌ Error: {e}")
        return None


def bulk_predict_students(students_list):
    """
    Example: Predict risk for multiple students
    """
    results = []
    
    print(f"🔄 Processing {len(students_list)} students...")
    print()
    
    for i, student in enumerate(students_list, 1):
        print(f"Student {i}/{len(students_list)}: {student.get('name', 'Unknown')}")
        result = predict_student_risk(student['data'])
        if result:
            results.append({
                'name': student.get('name'),
                'prediction': result['prediction'],
                'confidence': result['confidence']
            })
    
    return results


# ============================================================================
# Example Usage
# ============================================================================

if __name__ == "__main__":
    print("=" * 70)
    print("Academic Evaluation System - API Examples")
    print("=" * 70)
    print()
    
    # Example 1: Predict for a high-performing student
    print("Example 1: High-Performing Student")
    print("-" * 70)
    high_performer = {
        "cgpa": 8.5,
        "attendance_percentage": 95.0,
        "assignment_completion": 98.0,
        "participation_score": 90.0,
        "assignments_late": 0,
        "avg_evaluation_score": 88.0,
        "semester": 6,
        "absence_count": 2
    }
    predict_student_risk(high_performer)
    
    # Example 2: Predict for an at-risk student
    print("Example 2: At-Risk Student")
    print("-" * 70)
    at_risk_student = {
        "cgpa": 4.2,
        "attendance_percentage": 55.0,
        "assignment_completion": 50.0,
        "participation_score": 45.0,
        "assignments_late": 7,
        "avg_evaluation_score": 48.0,
        "semester": 3,
        "absence_count": 22
    }
    predict_student_risk(at_risk_student)
    
    # Example 3: Borderline student
    print("Example 3: Borderline Student")
    print("-" * 70)
    borderline_student = {
        "cgpa": 6.5,
        "attendance_percentage": 72.0,
        "assignment_completion": 75.0,
        "participation_score": 65.0,
        "assignments_late": 3,
        "avg_evaluation_score": 68.0,
        "semester": 5,
        "absence_count": 9
    }
    predict_student_risk(borderline_student)
    
    # Example 4: Get statistics (requires authentication)
    print("Example 4: System Statistics")
    print("-" * 70)
    # Note: This might fail if not authenticated
    # get_risk_statistics()
    print("⚠️  Requires authentication - run from authenticated session")
    print()
    
    # Example 5: Batch prediction
    print("Example 5: Batch Prediction")
    print("-" * 70)
    students_batch = [
        {
            'name': 'Alice Johnson',
            'data': {
                "cgpa": 7.8,
                "attendance_percentage": 88.0,
                "assignment_completion": 92.0,
                "participation_score": 85.0,
                "assignments_late": 1,
                "avg_evaluation_score": 82.0,
                "semester": 4,
                "absence_count": 4
            }
        },
        {
            'name': 'Bob Smith',
            'data': {
                "cgpa": 5.5,
                "attendance_percentage": 68.0,
                "assignment_completion": 65.0,
                "participation_score": 60.0,
                "assignments_late": 5,
                "avg_evaluation_score": 58.0,
                "semester": 4,
                "absence_count": 15
            }
        },
        {
            'name': 'Carol Davis',
            'data': {
                "cgpa": 9.2,
                "attendance_percentage": 97.0,
                "assignment_completion": 100.0,
                "participation_score": 95.0,
                "assignments_late": 0,
                "avg_evaluation_score": 92.0,
                "semester": 7,
                "absence_count": 1
            }
        }
    ]
    
    batch_results = bulk_predict_students(students_batch)
    
    # Summary
    print("=" * 70)
    print("Batch Prediction Summary")
    print("=" * 70)
    for result in batch_results:
        status_icon = "🟢" if result['prediction'] == 'Safe' else "🔴"
        print(f"{status_icon} {result['name']:<20} {result['prediction']:<10} ({result['confidence']:.1%} confidence)")
    print()
    
    print("✅ Examples completed!")
    print()
    print("Note: For authenticated endpoints (update, bulk-update), you need to:")
    print("1. Login to the web interface")
    print("2. Extract CSRF token from cookies")
    print("3. Pass token in request headers")
    print()
    print("Or use Django's test client for authenticated requests.")


# ============================================================================
# Integration Example with Custom Application
# ============================================================================

class RiskPredictionClient:
    """
    Client class for easy integration with other applications
    """
    
    def __init__(self, base_url="http://localhost:8000"):
        self.base_url = base_url
        self.api_base = f"{base_url}/api"
    
    def predict(self, cgpa, attendance, assignments, participation, 
                late_assignments=0, avg_eval=0, semester=1, absences=0):
        """
        Simplified prediction method
        """
        data = {
            "cgpa": cgpa,
            "attendance_percentage": attendance,
            "assignment_completion": assignments,
            "participation_score": participation,
            "assignments_late": late_assignments,
            "avg_evaluation_score": avg_eval,
            "semester": semester,
            "absence_count": absences
        }
        
        url = f"{self.api_base}/predict-risk/"
        response = requests.post(url, json=data)
        
        if response.status_code == 200:
            return response.json()
        else:
            raise Exception(f"API Error: {response.status_code}")
    
    def is_at_risk(self, **kwargs):
        """
        Returns True if student is at risk, False otherwise
        """
        result = self.predict(**kwargs)
        return result['prediction'] == 'At Risk'


# Usage:
# client = RiskPredictionClient()
# at_risk = client.is_at_risk(cgpa=5.0, attendance=60, assignments=55, participation=50)
# if at_risk:
#     print("⚠️ Student needs intervention!")