hyperparameter_tuning_summary.param_tuning_summary

hyperparameter_tuning_summary.param_tuning_summary(param_search_cv)

Create a summary of the hyperparameter tuning results and extract the best estimator

Parameters

Name	Type	Description	Default
param_search_cv			required

Returns

Name	Type	Description
df_summary	pandas.DataFrame	a Dataframe containing the best set of parameters’ names, values and scores.
best_estimator	the best model with the best set of parameters

Examples

>>> from sklearn.svm import SVC
>>> from sklearn.model_selection import GridSearchCV
>>> from sklearn.datasets import load_iris
>>> 
>>> X, y = load_iris(return_X_y=True)
>>> param_grid = {'C': [0.1, 1, 10], 'kernel': ['rbf', 'linear']}
>>> grid_search = GridSearchCV(SVC(), param_grid, cv=5)
>>> grid_search.fit(X, y)
GridSearchCV(...)
>>> 
>>> summary_df, best_estimator = param_tuning_summary(grid_search)
>>> print(summary_df)
   Parameter  Value  Best_Score
0          C    1.0    0.980000
1     kernel    rbf    0.980000
>>> print(type(best_estimator))
<class 'sklearn.svm._classes.SVC'>