config.py

import os
from dataclasses import dataclass, field
from typing import Dict, List

try:
    from data_loader import load_data_file

    DATA_LOADER_AVAILABLE = True
except ImportError:
    DATA_LOADER_AVAILABLE = False

    def load_data_file() -> str:
        return "data/optimized_data.xlsx"


def _get_data_file_path() -> str:
    if os.getenv("HF_TOKEN"):
        return load_data_file()
    elif os.getenv("DEMO_MODE", "false").lower() == "true":
        return "data/demo_data.xlsx"
    else:
        return "data/optimized_data.xlsx"


@dataclass
class DataConfig:
    FILE_PATH: str = field(default_factory=_get_data_file_path)

    SHEET_COURSES: str = "tabel1_data_matkul"
    SHEET_OFFERINGS: str = "tabel2_data_matkul_dibuka"
    SHEET_STUDENTS_YEARLY: str = "tabel3_data_mahasiswa_per_tahun"
    SHEET_STUDENTS_INDIVIDUAL: str = "tabel4_data_individu_mahasiswa"

    # Standardization
    OFFERINGS_RENAME: Dict[str, str] = field(
        default_factory=lambda: {"tahun": "thn", "semester": "smt"}
    )

    ELECTIVE_CATEGORY: str = "P"
    MANDATORY_CATEGORY: str = "W"

    VALID_CATEGORIES: List[str] = field(default_factory=lambda: ["P", "W"])


@dataclass
class ClassCapacityConfig:
    # Default maximum students per class
    DEFAULT_CLASS_CAPACITY: int = 50

    # Minimum students required to open a class
    MIN_STUDENTS_TO_OPEN_CLASS: int = 1

    # Threshold for opening additional classes
    ADDITIONAL_CLASS_THRESHOLD: float = 0.7

    # Always open at least 1 class if there's any historical enrollment
    OPEN_CLASS_IF_HAS_HISTORY: bool = True

    # Course-specific capacity overrides (kode_mk -> max_capacity)
    COURSE_CAPACITY_OVERRIDES: Dict[str, int] = field(default_factory=dict)

    # Warning threshold - if predicted > capacity * threshold, warn about capacity
    CAPACITY_WARNING_THRESHOLD: float = 0.8

    # Enable capacity-aware prediction
    # When True, predictions will be bounded by realistic capacity constraints
    ENABLE_CAPACITY_CONSTRAINTS: bool = True


@dataclass
class MultiYearForecastConfig:
    # How many years ahead to forecast
    FORECAST_YEARS_AHEAD: int = 3

    # Include trend analysis in output
    SHOW_TREND_ANALYSIS: bool = True

    # Confidence interval for forecasts (0-1)
    CONFIDENCE_INTERVAL: float = 0.95

    # Growth rate limits for sanity checking
    MAX_YEARLY_GROWTH_RATE: float = 0.5  # 50% max growth per year
    MIN_YEARLY_GROWTH_RATE: float = -0.3  # 30% max decline per year


@dataclass
class ModelConfig:
    # Prophet Hyperparameters
    GROWTH_MODE: str = "logistic"
    CHANGEPOINT_SCALE: float = 0.01
    SEASONALITY_MODE: str = "multiplicative"
    YEARLY_SEASONALITY: bool = True
    FALLBACK_DEFAULT: int = 20

    # Prediction safety limits
    # Maximum multiplier of historical max enrollment before flagging as unrealistic
    SANITY_CHECK_MAX_MULTIPLIER: float = 3.0
    # Minimum historical data points required for reliable prediction
    MIN_HISTORY_POINTS: int = 3

    # Use student population as regressor
    USE_POPULATION_REGRESSOR: bool = True
    # Use capacity as upper bound (cap in logistic growth)
    USE_CAPACITY_AS_CAP: bool = True


@dataclass
class PredictionConfig:
    """Business logic for predictions."""

    PREDICT_YEAR: int = 2025
    PREDICT_SEMESTER: int = 2

    BUFFER_PERCENT: float = 0.20
    MIN_QUOTA_OPEN: int = 25
    MIN_PREDICT_THRESHOLD: int = 15

    MAX_CAPACITY_MULTIPLIER: float = 2.0
    ABSOLUTE_MAX_STUDENTS: int = 400

    SEMESTER_TO_MONTH: Dict[int, str] = field(
        default_factory=lambda: {
            1: "09-01",
            2: "03-01",
        }
    )


@dataclass
class OutputConfig:
    OUTPUT_DIR: str = "output"
    LOG_LEVEL: str = "INFO"
    TOP_N_DISPLAY: int = 30


@dataclass
class BacktestConfig:
    START_YEAR: int = 2010
    END_YEAR: int = 2024
    VERBOSE: bool = True

    # Minimum elective enrollments required for backtesting
    MIN_ELECTIVE_ENROLLMENTS: int = 1
    # Minimum unique courses required for backtesting
    MIN_UNIQUE_COURSES: int = 1


class Config:
    def __init__(self):
        self.data: DataConfig = DataConfig()
        self.model: ModelConfig = ModelConfig()
        self.prediction: PredictionConfig = PredictionConfig()
        self.output: OutputConfig = OutputConfig()
        self.backtest: BacktestConfig = BacktestConfig()
        self.class_capacity: ClassCapacityConfig = ClassCapacityConfig()
        self.multi_year: MultiYearForecastConfig = MultiYearForecastConfig()

    def get_prediction_target_name(self) -> str:
        sem = "Ganjil" if self.prediction.PREDICT_SEMESTER == 1 else "Genap"
        return f"{self.prediction.PREDICT_YEAR} Semester {sem}"

    def get_elective_filter_description(self) -> str:
        return f"kategori_mk = '{self.data.ELECTIVE_CATEGORY}' in {self.data.SHEET_COURSES}"

    def get_class_capacity(self, course_code: str) -> int:
        if course_code in self.class_capacity.COURSE_CAPACITY_OVERRIDES:
            return self.class_capacity.COURSE_CAPACITY_OVERRIDES[course_code]
        return self.class_capacity.DEFAULT_CLASS_CAPACITY

    def calculate_classes_needed(
        self,
        predicted_enrollment: float,
        course_code: str,
        has_historical_data: bool = True,
    ) -> int:
        import math

        capacity = self.get_class_capacity(course_code)

        if predicted_enrollment <= 0:
            return 0

        if predicted_enrollment < 1 and has_historical_data:
            return 1

        classes = math.ceil(predicted_enrollment / capacity)

        return max(1, classes)

    def get_capacity_status(self, predicted_enrollment: float, course_code: str) -> str:
        capacity = self.get_class_capacity(course_code)
        classes_needed = self.calculate_classes_needed(
            predicted_enrollment, course_code
        )

        if classes_needed == 0:
            return "UNDER"

        total_capacity = classes_needed * capacity
        utilization = predicted_enrollment / total_capacity

        if utilization >= 1.0:
            return "OVER"
        elif utilization >= self.class_capacity.CAPACITY_WARNING_THRESHOLD:
            return "WARNING"
        else:
            return "NORMAL"


default_config = Config()