[BUG]: estimator gives a tuple ?

[celestinoxp]

pycaret version checks

• I have checked that this issue has not already been reported here.

• I have confirmed this bug exists on the latest version of pycaret.

• I have confirmed this bug exists on the master branch of pycaret (pip install -U git+https://github.com/pycaret/pycaret.git@master).

Issue Description

using this code

modelo_final = finalize_model(modelo_tunado)
modelo_treinado = save_model(modelo_final, './modelos/n1_maior_igual_9')
predict_model(modelo_treinado, data=dados_teste)

shows error about a tuple?

Reproducible Example

modelo_final = finalize_model(modelo_tunado)
modelo_treinado = save_model(modelo_final, './modelos/n1_maior_igual_9')
predict_model(modelo_treinado, data=dados_teste)

Expected Behavior

maybe a list?

Actual Results

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[83], line 2
      1 # Prever com o modelo
----> 2 predict_model(modelo_treinado, data=dados_teste)

File ~\anaconda3\Lib\site-packages\pycaret\classification\functional.py:2172, in predict_model(estimator, data, probability_threshold, encoded_labels, raw_score, round, verbose)
   2169 if experiment is None:
   2170     experiment = _EXPERIMENT_CLASS()
-> 2172 return experiment.predict_model(
   2173     estimator=estimator,
   2174     data=data,
   2175     probability_threshold=probability_threshold,
   2176     encoded_labels=encoded_labels,
   2177     raw_score=raw_score,
   2178     round=round,
   2179     verbose=verbose,
   2180 )

File ~\anaconda3\Lib\site-packages\pycaret\classification\oop.py:2824, in ClassificationExperiment.predict_model(self, estimator, data, probability_threshold, encoded_labels, raw_score, round, verbose)
   2752 def predict_model(
   2753     self,
   2754     estimator,
   (...)
   2760     verbose: bool = True,
   2761 ) -> pd.DataFrame:
   2762     """
   2763     This function predicts ``Label`` and ``Score`` (probability of predicted
   2764     class) using a trained model. When ``data`` is None, it predicts label and
   (...)
   2821 
   2822     """
-> 2824     return super().predict_model(
   2825         estimator=estimator,
   2826         data=data,
   2827         probability_threshold=probability_threshold,
   2828         encoded_labels=encoded_labels,
   2829         raw_score=raw_score,
   2830         round=round,
   2831         verbose=verbose,
   2832     )

File ~\anaconda3\Lib\site-packages\pycaret\internal\pycaret_experiment\supervised_experiment.py:4992, in _SupervisedExperiment.predict_model(self, estimator, data, probability_threshold, encoded_labels, raw_score, round, verbose, ml_usecase, preprocess)
   4989         probability_threshold = estimator.probability_threshold
   4990     estimator = get_estimator_from_meta_estimator(estimator)
-> 4992 pred = np.nan_to_num(estimator.predict(X_test_))
   4993 pred = pipeline.inverse_transform(pred)
   4994 # Need to convert labels back to numbers
   4995 # TODO optimize

AttributeError: 'tuple' object has no attribute 'predict'

Installed Versions

pycaret 3.3.2 python 3.11 windows 11

[celestinoxp]

more context:
model created (a tuple):

[celestinoxp]

solved! a package created conflicts... i don't know what package...