Note that the optional watermark extension is a small IPython notebook plugin that I developed to make the code reproducible. You can just skip the following line(s).
%load_ext watermark
%watermark -a 'Sebastian Raschka' -u -d -v -p numpy,pandas,matplotlib,scikit-learn
Sebastian Raschka Last updated: 01/20/2016 CPython 3.5.1 IPython 4.0.1 numpy 1.10.1 pandas 0.17.1 matplotlib 1.5.0 scikit-learn 0.17
from sklearn.grid_search import GridSearchCV
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.svm import SVC
from sklearn.datasets import load_iris
from sklearn.cross_validation import train_test_split
# load and split data
iris = load_iris()
X, y = iris.data, iris.target
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.20, random_state=42)
# pipeline setup
cls = SVC(C=10.0,
kernel='rbf',
gamma=0.1,
decision_function_shape='ovr')
kernel_svm = Pipeline([('std', StandardScaler()),
('svc', cls)])
# gridsearch setup
param_grid = [
{'svc__C': [1, 10, 100, 1000],
'svc__gamma': [0.001, 0.0001],
'svc__kernel': ['rbf']},
]
gs = GridSearchCV(estimator=kernel_svm,
param_grid=param_grid,
scoring='accuracy',
n_jobs=-1,
cv=5,
verbose=1,
refit=True,
pre_dispatch='2*n_jobs')
# run gridearch
gs.fit(X_train, y_train)
Fitting 5 folds for each of 8 candidates, totalling 40 fits
[Parallel(n_jobs=-1)]: Done 40 out of 40 | elapsed: 0.2s finished
GridSearchCV(cv=5, error_score='raise', estimator=Pipeline(steps=[('std', StandardScaler(copy=True, with_mean=True, with_std=True)), ('svc', SVC(C=10.0, cache_size=200, class_weight=None, coef0=0.0, decision_function_shape='ovr', degree=3, gamma=0.1, kernel='rbf', max_iter=-1, probability=False, random_state=None, shrinking=True, tol=0.001, verbose=False))]), fit_params={}, iid=True, n_jobs=-1, param_grid=[{'svc__kernel': ['rbf'], 'svc__C': [1, 10, 100, 1000], 'svc__gamma': [0.001, 0.0001]}], pre_dispatch='2*n_jobs', refit=True, scoring='accuracy', verbose=1)
print('Best GS Score %.2f' % gs.best_score_)
print('best GS Params %s' % gs.best_params_)
# prediction on the training set
y_pred = gs.predict(X_train)
train_acc = (y_train == y_pred).sum()/len(y_train)
print('\nTrain Accuracy: %.2f' % (train_acc))
# evaluation on the test set
y_pred = gs.predict(X_test)
test_acc = (y_test == y_pred).sum()/len(y_test)
print('\nTest Accuracy: %.2f' % (test_acc))
Best GS Score 0.96 best GS Params {'svc__kernel': 'rbf', 'svc__C': 100, 'svc__gamma': 0.001} Train Accuracy: 0.97 Test Accuracy: 0.97
GridSearchCV
's best_score_
attribute¶Please note that gs.best_score_
is the average k-fold cross-validation score. I.e., if we have a GridSearchCV
object with 5-fold cross-validation (like the one above), the best_score_
attribute returns the average score over the 5-folds of the best model. To illustrate this with an example:
from sklearn.cross_validation import StratifiedKFold, cross_val_score
from sklearn.linear_model import LogisticRegression
import numpy as np
np.random.seed(0)
np.set_printoptions(precision=6)
y = [np.random.randint(3) for i in range(25)]
X = (y + np.random.randn(25)).reshape(-1, 1)
cv5_idx = list(StratifiedKFold(y, n_folds=5, shuffle=False, random_state=0))
cross_val_score(LogisticRegression(random_state=123), X, y, cv=cv5_idx)
array([ 0.6, 0.4, 0.6, 0.2, 0.6])
By executing the code above, we created a simple data set of random integers that shall represent our class labels. Next, we fed the indices of 5 cross-validation folds (cv3_idx
) to the cross_val_score
scorer, which returned 5 accuracy scores -- these are the 5 accuracy values for the 5 test folds.
Next, let us use the GridSearchCV
object and feed it the same 5 cross-validation sets (via the pre-generated cv3_idx
indices):
from sklearn.grid_search import GridSearchCV
gs = GridSearchCV(LogisticRegression(), {}, cv=cv5_idx, verbose=3).fit(X, y)
Fitting 5 folds for each of 1 candidates, totalling 5 fits [CV] ................................................................ [CV] ....................................... , score=0.600000 - 0.0s [CV] ................................................................ [CV] ....................................... , score=0.400000 - 0.0s [CV] ................................................................ [CV] ....................................... , score=0.600000 - 0.0s [CV] ................................................................ [CV] ....................................... , score=0.200000 - 0.0s [CV] ................................................................ [CV] ....................................... , score=0.600000 - 0.0s
[Parallel(n_jobs=1)]: Done 5 out of 5 | elapsed: 0.0s finished
As we can see, the scores for the 5 folds are exactly the same as the ones from cross_val_score
earlier.
Now, the best_score_ attribute of the GridSearchCV
object, which becomes available after fit
ting, returns the average accuracy score of the best model:
gs.best_score_
0.47999999999999998
As we can see, the result above is consistent with the average score computed the cross_val_score
.
cross_val_score(LogisticRegression(), X, y, cv=cv5_idx).mean()
0.47999999999999998