UNQ_C11 Block - Event times can only be 1-dimensional:

Course Q&A AI For Medicine AI For Medical Treatment
1

Hi!

My implementations passed all the tests so far, but when I run this cell:

treatment_model, best_hyperparam_treat  = holdout_grid_search(RandomForestClassifier,
                                      X_treat_train, y_treat_train,
                                      X_treat_val, y_treat_val, hyperparams)

I get the following error:

dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
dict_keys(['n_estimators', 'max_depth', 'min_samples_leaf', 'random_state'])
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-85-ebddece39f65> in <module>()
      2 treatment_model, best_hyperparam_treat  = holdout_grid_search(RandomForestClassifier,
      3                                       X_treat_train, y_treat_train,
----> 4                                       X_treat_val, y_treat_val, hyperparams)

<ipython-input-83-04fc2f5d43af> in holdout_grid_search(clf, X_train_hp, y_train_hp, X_val_hp, y_val_hp, hyperparam, verbose)
     67 
     68         # Evaluate the model's performance using the regular concordance index
---> 69         estimator_score = concordance_index(y_val_hp, preds)
     70 
     71         # if the model's c-index is better than the previous best:

/opt/conda/lib/python3.6/site-packages/lifelines/utils/concordance.py in concordance_index(event_times, predicted_scores, event_observed)
     52     """
     53     event_times, predicted_scores, event_observed = _preprocess_scoring_data(
---> 54         event_times, predicted_scores, event_observed
     55     )
     56     num_correct, num_tied, num_pairs = _concordance_summary_statistics(event_times, predicted_scores, event_observed)

/opt/conda/lib/python3.6/site-packages/lifelines/utils/concordance.py in _preprocess_scoring_data(event_times, predicted_scores, event_observed)
    255         raise ValueError("Event times and predictions must have the same shape")
    256     if event_times.ndim != 1:
--> 257         raise ValueError("Event times can only be 1-dimensional: (n,)")
    258 
    259     if event_observed is None:

ValueError: Event times can only be 1-dimensional: (n,)

Any tips on how to solve this?

Thanks!

2

Check your estimator_Score, it is basically mentioning your event_times and predicted score should be 1-D

1 Like
3

Magno,

Is your issue resolved?

4

Yes. There were some issues in previous functions. I made a submission and saw which functions were incorrect.

Thanks!