[python] ML Model Parameter Optimization : GridSearchCV

 

from sklearn.model_selection import GridSearchCV
from sklearn.neighbors import KNeighborsClassifier
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import MinMaxScaler
knn_pipe = Pipeline([('mms', MinMaxScaler()),
                     ('knn', KNeighborsClassifier())])
                     
params = [{'knn__n_neighbors': [3, 5, 7, 9],
         'knn__weights': ['uniform', 'distance'],
         'knn__leaf_size': [15, 20]}]
         
gs_knn = GridSearchCV(knn_pipe,
                      param_grid=params,
                      scoring='accuracy',
                      cv=5)
gs_knn.fit(X_train, y_train)
gs_knn.best_params_

#Output:
#{'knn__leaf_size': 15, 'knn__n_neighbors': 5, 'knn__weights': 'distance'}
# find best model score

gs_knn.score(X_train, y_train)
#Output: 0.84035137