Question

使用sklearn的StratifiedKFold功能，有人可以帮我理解这里的错误吗？

我的猜测是它与我的输入标签数组有关，我注意到当我打印它们时（本例中的前16个）索引从0到15，但是在我之上打印了一个额外的0没有期待。也许我只是一个蟒蛇菜鸟，但这看起来很奇怪。

有人在这里看到了蠢事吗？

文档：http://scikit-learn.org...StratifiedKFold.html

代码：

import nltk
import sklearn

print('The nltk version is {}.'.format(nltk.__version__))
print('The scikit-learn version is {}.'.format(sklearn.__version__))

print type(skew_gendata_targets.values), skew_gendata_targets.values.shape
print skew_gendata_targets.head(16)

skew_sfold10 = cross_validation.StratifiedKFold(skew_gendata_targets.values, n_folds=10, shuffle=True, random_state=20160121)

结果

The nltk version is 3.1.
The scikit-learn version is 0.17.
<type 'numpy.ndarray'> (500L, 1L)
    0
0   0
1   0
2   0
3   0
4   0
5   0
6   0
7   0
8   0
9   0
10  0
11  0
12  0
13  0
14  1
15  0
---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
<ipython-input-373-653b6010b806> in <module>()
      8 print skew_gendata_targets.head(16)
      9 
---> 10 skew_sfold10 = cross_validation.StratifiedKFold(skew_gendata_targets.values, n_folds=10, shuffle=True, random_state=20160121)
     11 
     12 #print '\nSkewed Generated Dataset (', len(skew_gendata_data), ')'

d:\Program Files\Anaconda2\lib\site-packages\sklearn\cross_validation.pyc in __init__(self, y, n_folds, shuffle, random_state)
    531         for test_fold_idx, per_label_splits in enumerate(zip(*per_label_cvs)):
    532             for label, (_, test_split) in zip(unique_labels, per_label_splits):
--> 533                 label_test_folds = test_folds[y == label]
    534                 # the test split can be too big because we used
    535                 # KFold(max(c, self.n_folds), self.n_folds) instead of

IndexError: too many indices for array

Answer 1

检查skew_gendata_targets.values的形状。你会发现它不是像StratifiedKFold所期望的1d数组（形状（500，）），而是（500,1）数组。 SKlearn分开处理这些，而不是强迫它们相同。如果有帮助，请告诉我

StratifiedKFold：IndexError：数组的索引太多了

1 个答案: