CPP error in grid_search

[acrofales]

Problem: catboost grid_search throws the following error: CatBoostError: library/cpp/json/writer/json_value.cpp:457: Not a string
catboost version: 0.25.1
Operating System: linux (Google Compute Engine)
CPU: 4 vCPUs type Intel Haswell

GPU: N/A

Here is the full stack trace, and the initial call:

params = {'depth': [4, 7, 10],
         'per_float_feature_quantization':[None, [f'{n25_index}:border_count=1024'], [f'{n75_index}:border_count=1024'],[f'{n25_index}:border_count=1024', f'{n75_index}:border_count=1024']],
         'l2_leaf_reg': [1, 3, 5, 7, 9],
         'iterations': [2000],
         'learning_rate':[0.3],
         'verbose': [100]
         }
        

cbr = CatBoostRegressor()
cb_model = cbr.grid_search(params,
            trainpool,
            cv=3,
            partition_random_seed=42,
            calc_cv_statistics=True,
            search_by_train_test_split=True,
            refit=True,
            shuffle=True,
            stratified=None,
            train_size=0.8, 
            verbose=5,
            plot=True)


---------------------------------------------------------------------------
CatBoostError                             Traceback (most recent call last)
<ipython-input-141-be7bb715851e> in <module>
     24             train_size=0.8,
     25             verbose=5,
---> 26             plot=True)

/opt/conda/lib/python3.7/site-packages/catboost/core.py in grid_search(self, param_grid, X, y, cv, partition_random_seed, calc_cv_statistics, search_by_train_test_split, refit, shuffle, stratified, train_size, verbose, plot)
   3618             partition_random_seed=partition_random_seed, calc_cv_statistics=calc_cv_statistics,
   3619             search_by_train_test_split=search_by_train_test_split, refit=refit, shuffle=shuffle,
-> 3620             stratified=stratified, train_size=train_size, verbose=verbose, plot=plot
   3621         )
   3622 

/opt/conda/lib/python3.7/site-packages/catboost/core.py in _tune_hyperparams(self, param_grid, X, y, cv, n_iter, partition_random_seed, calc_cv_statistics, search_by_train_test_split, refit, shuffle, stratified, train_size, verbose, plot)
   3526             assert not self.is_fitted()
   3527             self.set_params(**cv_result['params'])
-> 3528             self.fit(X, y, silent=True)
   3529         return cv_result
   3530 

/opt/conda/lib/python3.7/site-packages/catboost/core.py in fit(self, X, y, cat_features, sample_weight, baseline, use_best_model, eval_set, verbose, logging_level, plot, column_description, verbose_eval, metric_period, silent, early_stopping_rounds, save_snapshot, snapshot_file, snapshot_interval, init_model)
   5087                          use_best_model, eval_set, verbose, logging_level, plot, column_description,
   5088                          verbose_eval, metric_period, silent, early_stopping_rounds,
-> 5089                          save_snapshot, snapshot_file, snapshot_interval, init_model)
   5090 
   5091     def predict(self, data, prediction_type=None, ntree_start=0, ntree_end=0, thread_count=-1, verbose=None):

/opt/conda/lib/python3.7/site-packages/catboost/core.py in _fit(self, X, y, cat_features, text_features, embedding_features, pairs, sample_weight, group_id, group_weight, subgroup_id, pairs_weight, baseline, use_best_model, eval_set, verbose, logging_level, plot, column_description, verbose_eval, metric_period, silent, early_stopping_rounds, save_snapshot, snapshot_file, snapshot_interval, init_model)
   1909             column_description=column_description, verbose_eval=verbose_eval, metric_period=metric_period,
   1910             silent=silent, early_stopping_rounds=early_stopping_rounds, save_snapshot=save_snapshot,
-> 1911             snapshot_file=snapshot_file, snapshot_interval=snapshot_interval, init_model=init_model
   1912         )
   1913         params = train_params["params"]

/opt/conda/lib/python3.7/site-packages/catboost/core.py in _prepare_train_params(self, X, y, cat_features, text_features, embedding_features, pairs, sample_weight, group_id, group_weight, subgroup_id, pairs_weight, baseline, use_best_model, eval_set, verbose, logging_level, plot, column_description, verbose_eval, metric_period, silent, early_stopping_rounds, save_snapshot, snapshot_file, snapshot_interval, init_model)
   1834         _check_param_types(params)
   1835         params = _params_type_cast(params)
-> 1836         _check_train_params(params)
   1837 
   1838         eval_set_list = eval_set if isinstance(eval_set, list) else [eval_set]

_catboost.pyx in _catboost._check_train_params()

_catboost.pyx in _catboost._check_train_params()

CatBoostError: library/cpp/json/writer/json_value.cpp:457: Not a string

This effectively breaks the grid_search method shipped with catboost. I'm happy to use sklearn's gridsearch, but it would have been nice to know before starting the hyperparameter tuning...