FutureWarning in Optuna during TabM hyperparameter tuning causes notebook failure after trials complete on Kaggle GPU

Question

I’m running Optuna to tune hyperparameters for a TabM regression model (10 trials) on Kaggle (GPU: Tesla P100) to minimize RMSE. The optimization runs fine — all trials complete — but right after finishing, the notebook fails with a FutureWarning from the traitlets package. After this error, no subsequent cells (like saving trial results to CSV) can run.

Here’s the main part of my Optuna code:

import optuna
from optuna.storages import RDBStorage
import warnings
warnings.filterwarnings("ignore", category=FutureWarning)
study_name = "tabm_tuning"
study = optuna.create_study(
 study_name=study_name,
 direction="minimize",
 storage=RDBStorage(url=f"sqlite:///{study_name}.db")
)
# enqueue one predefined parameter set
study.enqueue_trial(params_2)
# run optimization for 10 trials (example)
study.optimize(
 lambda trial: objective(trial, X_train, y_train, X_val, y_val),
 n_trials=10,
 timeout=41710
) # use of timeout to prevent runtime out error
study.best_params
df_trial=study.trials_dataframe()
df_trial.to_csv(f'{study_name}_df.csv',index=False)
df_trial

Everything runs smoothly until the end of the trials, where this warning appears and the notebook crashes:

[I 2025年10月27日 14:50:51,477] Trial 9 finished with value: 0.05611381708907078 and parameters: {'patience': 15, 'tabm_k': 112, 'gradient_clipping_norm': 0.915269632521597, 'lr': 0.0008291955423479689, 'weight_decay': 0.007870223162632283, 'n_blocks': 5, 'd_block': 210, 'dropout': 0.35379749466109023, 'd_embedding': 11}. Best is trial 5 with value: 0.056029200575024675.
20800.4s /usr/local/lib/python3.11/dist-packages/traitlets/traitlets.py:2915: FutureWarning: 
--Exporter.preprocessors=["remove_papermill_header.RemovePapermillHeader"] for containers 
is deprecated in traitlets 5.0. You can pass `--Exporter.preprocessors item` ... multiple times.
20803.4s /usr/local/lib/python3.11/dist-packages/traitlets/traitlets.py:2915: FutureWarning: 
--Exporter.preprocessors=["nbconvert.preprocessors.ExtractOutputPreprocessor"] for containers 
is deprecated in traitlets 5.0.
20804.2s Traceback (most recent call last):
20804.2s File "/usr/local/bin/jupyter-nbconvert", line 10, in <module>

After this, the notebook stops executing completely.

Question

How can I prevent or safely handle this FutureWarning from traitlets (triggered by nbconvert) so my Kaggle notebook doesn’t crash after Optuna finishes? Is there a reliable workaround or fix for this in Kaggle / Jupyter environments?

• colonel: Try to add these warning: warnings.filterwarnings("ignore", module="traitlets") warnings.filterwarnings("ignore", module="nbconvert") and run again. OR you can disable the autosave and checkpoint, but using %%javascript Jupyter.notebook.set_autosave_interval(0)