The isolationforest from elena-sharova

isolationforest's Issues

TypeError: cannot convert the series to <class 'int'> when trying GridSearchCV

I have a problem regarding section 4 of notebook (4. Setting Parameters of Isolation Forest - n_estimators) I want to print the optimized number of trees which is n_estimators as follows:

#import libraries
from sklearn import datasets
from sklearn import preprocessing
from sklearn.model_selection import train_test_split
from sklearn.metrics import confusion_matrix
from sklearn.metrics import recall_score, silhouette_score, roc_curve, roc_auc_score, f1_score, precision_recall_curve, auc
from sklearn.model_selection import GridSearchCV

# The part of notebook related to section 4 (4. Setting Parameters of Isolation Forest - n_estimators)
scoring = {'AUC': 'roc_auc', 'Recall': make_scorer(recall_score, pos_label=-1)}

gs = GridSearchCV(IsolationForest(max_samples=0.25, random_state=11, contamination = 0.15, n_jobs=-1),
                  param_grid={'n_estimators': range(20, 230, 30)},
                  scoring=scoring, refit='Recall', return_train_score=True, cv=3, verbose=1, n_jobs=-1)
gs.fit(X_train_sf, y_train_sf)
results = gs.cv_results_

#Frame the results via pandas
pd.DataFrame(results)

#print the best n_estimators based on recall metric
max_recall_n_estimators = int(pd.DataFrame(results)[["param_n_estimators", "mean_train_Recall"]].loc[pd.DataFrame(results)["mean_train_Recall"] == pd.DataFrame(results)["mean_train_Recall"].max()]["param_n_estimators"])
max_recall_n_estimators

visually I need to access the marked part on plot which is the optimum value on train-set and test-set based on recall metric:

and I get the error which its Traceback is following:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-61-95626cf7130a> in <module>
----> 1 max_recall_n_estimators = int(pd.DataFrame(results)[["param_n_estimators", "mean_train_Recall"]].loc[pd.DataFrame(results)["mean_train_Recall"] == pd.DataFrame(results)["mean_train_Recall"].max()]["param_n_estimators"])
      2 max_recall_n_estimators = max_recall_n_estimators.astype(int)

/opt/anaconda3/envs/.../lib/python3.8/site-packages/pandas/core/series.py in wrapper(self)
    137         if len(self) == 1:
    138             return converter(self.iloc[0])
--> 139         raise TypeError(f"cannot convert the series to {converter}")
    140 
    141     wrapper.__name__ = f"__{converter.__name__}__"

TypeError: cannot convert the series to <class 'int'>

I already checked this answer1 & answer2 via .astype(int) but I couldn't figure it out how I can return/print max_recall_n_estimators from results of GridSearchCV elegantly.

Recommend Projects