feat: expose fold option to hyperparameter optimization

xgboostlss/model.py

@@ -319,6 +319,7 @@ def hyper_opt(
         dtrain: DMatrix,
         num_boost_round=500,
         nfold=10,
+        folds=None,
         early_stopping_rounds=20,
         max_minutes=10,
         n_trials=None,
@@ -340,6 +341,13 @@ def hyper_opt(
             Number of boosting iterations.
         nfold: int
             Number of folds in CV.
+        folds: a KFold or StratifiedKFold instance or list of fold indices
+            Sklearn KFolds or StratifiedKFolds object.
+            Alternatively may explicitly pass sample indices for each fold.
+            For ``n`` folds, **folds** should be a length ``n`` list of tuples.
+            Each tuple is ``(in,out)`` where ``in`` is a list of indices to be used
+            as the training samples for the ``n`` th fold and ``out`` is a list of
+            indices to be used as the testing samples for the ``n`` th fold.
         early_stopping_rounds: int
             Activates early stopping. Cross-Validation metric (average of validation
             metric computed over CV folds) needs to improve at least once in
@@ -417,7 +425,8 @@ def objective(trial):
                                           early_stopping_rounds=early_stopping_rounds,
                                           callbacks=[pruning_callback],
                                           seed=seed,
-                                          verbose_eval=False
+                                          verbose_eval=False,
+                                          folds=folds,
                                           )
 
             # Add the optimal number of rounds