@@ -56,17 +56,19 @@ class Pipeline(pipeline.Pipeline):
5656 --------
5757
5858 >>> from collections import Counter
59- >>> from sklearn.datasets import fetch_mldata
59+ >>> from sklearn.datasets import make_classification
6060 >>> from sklearn.cross_validation import train_test_split as tts
6161 >>> from sklearn.decomposition import PCA
6262 >>> from sklearn.neighbors import KNeighborsClassifier as KNN
6363 >>> from sklearn.metrics import classification_report
6464 >>> from imblearn.over_sampling import SMOTE
6565 >>> from imblearn.pipeline import Pipeline
66- >>> pima = fetch_mldata('diabetes_scale')
67- >>> X, y = pima['data'], pima['target']
66+ >>> X, y = make_classification(n_classes=2, class_sep=2, weights=[0.1, 0.9],
67+ ... n_informative=3, n_redundant=1, flip_y=0,
68+ ... n_features=20, n_clusters_per_class=1,
69+ ... n_samples=1000, random_state=10)
6870 >>> print('Original dataset shape {}'.format(Counter(y)))
69- Original dataset shape Counter({1: 500, -1: 268 })
71+ Original dataset shape Counter({1: 900, 0: 100 })
7072 >>> pca = PCA()
7173 >>> smt = SMOTE(random_state=42)
7274 >>> knn = KNN()
@@ -81,10 +83,10 @@ class Pipeline(pipeline.Pipeline):
8183 >>> print(classification_report(y_test, y_hat))
8284 precision recall f1-score support
8385 <BLANKLINE>
84- -1 0.49 0.67 0.56 69
85- 1 0.77 0.61 0.68 123
86+ 0 0.71 1.00 0.83 24
87+ 1 1.00 0.96 0.98 226
8688 <BLANKLINE>
87- avg / total 0.67 0.63 0.64 192
89+ avg / total 0.97 0.96 0.96 250
8890 <BLANKLINE>
8991
9092 """
0 commit comments