ValueError: When passing a model not in PyCaret's model library, the custom_grid parameter must be provided.

# Comparing models and returning the top 4
models = exp.compare_models(include=models_to_ignore,n_select=4)

# Saving the models
from pathlib import Path
import os
model_dir = Path('../models')
for index, model in enumerate(models):
    model_file_name = models_metrics.index[index]
    print(f'Saving {model_file_name} model')
    save_model(model, os.path.join(model_dir, model_file_name))

#Loading and tuning the saved models
from pathlib import Path
from glob import glob
import os
model_dir = Path('../models')
for model in glob(os.path.join(model_dir, '*.pkl')):
    
    tuned_model_file = (os.path.basename(model))
    tuned_model_file = os.path.splitext(tuned_model_file)[0]
    
    model = os.path.splitext(model)[0]
    print(f'Started tuning {model} model')
    model = load_model(model)
    print(model)
    
    tuned_model = tune_model(
        model, 
        n_iter=20, 
        optimize='RMSLE',
        search_library="tune-sklearn",
        search_algorithm="optuna",
        early_stopping='asha',
        return_tuner=True,
        return_train_score=True
    )
    
    print(f'Saving Model to {tuned_model_file}')
    save_model(tuned_model, os.path.join(model_dir, tuned_file_name))

# Error output
Started tuning ../models/xgboost model
Transformation Pipeline and Model Successfully Loaded
Pipeline(memory=FastMemory(location=/var/folders/vc/262dx0q508d6dv26phhy52ph0000gn/T/joblib),
         steps=[('target_transformation',
                 TransformerWrapperWithInverse(transformer=TargetTransformer(estimator=PowerTransformer(standardize=False)))),
                ('numerical_imputer',
                 TransformerWrapper(include=['onpromotion', 'oil_price'],
                                    transformer=SimpleImputer())),
                ('categori...
                              feature_types=None, gamma=None, gpu_id=None,
                              grow_policy=None, importance_type=None,
                              interaction_constraints=None, learning_rate=None,
                              max_bin=None, max_cat_threshold=None,
                              max_cat_to_onehot=None, max_delta_step=None,
                              max_depth=None, max_leaves=None,
                              min_child_weight=None, missing=nan,
                              monotone_constraints=None, n_estimators=100,
                              n_jobs=2, num_parallel_tree=None, predictor=None,
                              random_state=4758, ...))])
Initiated	. . . . . . . . . . . . . . . . . .	16:12:40
Status	. . . . . . . . . . . . . . . . . .	Loading Dependencies
Estimator	. . . . . . . . . . . . . . . . . .	Compiling Library
Processing: 0%
0/7 [00:00<?, ?it/s]
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
/var/folders/vc/262dx0q508d6dv26phhy52ph0000gn/T/ipykernel_12938/1177041359.py in <cell line: 5>()
     13     print(model)
     14 
---> 15     tuned_model = tune_model(
     16         model
     17         # n_iter=20,

/opt/anaconda3/envs/store_sales/lib/python3.10/site-packages/pycaret/utils/generic.py in wrapper(*args, **kwargs)
    963                 if globals_d[name] is None:
    964                     raise ValueError(message)
--> 965             return func(*args, **kwargs)
    966 
    967         return wrapper

/opt/anaconda3/envs/store_sales/lib/python3.10/site-packages/pycaret/regression/functional.py in tune_model(estimator, fold, round, n_iter, custom_grid, optimize, custom_scorer, search_library, search_algorithm, early_stopping, early_stopping_max_iters, choose_better, fit_kwargs, groups, return_tuner, verbose, tuner_verbose, return_train_score, **kwargs)
   1203     """
   1204 
-> 1205     return _CURRENT_EXPERIMENT.tune_model(
   1206         estimator=estimator,
   1207         fold=fold,

/opt/anaconda3/envs/store_sales/lib/python3.10/site-packages/pycaret/regression/oop.py in tune_model(self, estimator, fold, round, n_iter, custom_grid, optimize, custom_scorer, search_library, search_algorithm, early_stopping, early_stopping_max_iters, choose_better, fit_kwargs, groups, return_tuner, verbose, tuner_verbose, return_train_score, **kwargs)
   1503         """
   1504 
-> 1505         return super().tune_model(
   1506             estimator=estimator,
   1507             fold=fold,

/opt/anaconda3/envs/store_sales/lib/python3.10/site-packages/pycaret/internal/pycaret_experiment/supervised_experiment.py in tune_model(self, estimator, fold, round, n_iter, custom_grid, optimize, custom_scorer, search_library, search_algorithm, early_stopping, early_stopping_max_iters, choose_better, fit_kwargs, groups, return_tuner, verbose, tuner_verbose, return_train_score, **kwargs)
   2238         if estimator_id is None:
   2239             if custom_grid is None:
-> 2240                 raise ValueError(
   2241                     "When passing a model not in PyCaret's model library, the custom_grid parameter must be provided."
   2242                 )

ValueError: When passing a model not in PyCaret's model library, the custom_grid parameter must be provided.