config/optuna_config.yaml

# Optuna Hyperparameter Optimization Configuration
# Developer Salary Prediction - XGBoost Tuning

# Study settings
study:
  # Default number of optimization trials
  n_trials: 30

  # Optimization direction (minimize RMSE)
  direction: minimize

  # Cross-validation folds per trial
  cv_splits: 5

# Hyperparameter search space
# Each parameter specifies: type (int, float), low, high
# Optional: log (bool) for log-uniform distribution
search_space:
  max_depth:
    type: int
    low: 3
    high: 10

  learning_rate:
    type: float
    low: 0.005
    high: 0.3
    log: true

  min_child_weight:
    type: int
    low: 1
    high: 30

  subsample:
    type: float
    low: 0.5
    high: 1.0

  colsample_bytree:
    type: float
    low: 0.5
    high: 1.0

  reg_alpha:
    type: float
    low: 0.0001
    high: 10.0
    log: true

  reg_lambda:
    type: float
    low: 0.0001
    high: 10.0
    log: true

  gamma:
    type: float
    low: 0.0
    high: 5.0

# Fixed parameters (not tuned, passed through to every trial)
fixed:
  n_estimators: 5000
  early_stopping_rounds: 50
  n_jobs: -1
  random_state: 42