__init __（）得到了意外的关键字参数'n_splits'错误

Question

I was going to try the code in this link :

I am getting error from the line which refers to StratifiedKFold(n_splits=60) . Can anybody tell me how i can solve this error?

Here is the Code:

import numpy as np
from scipy import interp
import matplotlib.pyplot as plt
from itertools import cycle

from sklearn import svm, datasets
from sklearn.metrics import roc_curve, auc
from sklearn.cross_validation import StratifiedKFold

iris = datasets.load_iris()
X = iris.data
y = iris.target
X, y = X[y != 2], y
X, y

cv = StratifiedKFold(n_splits=6)
classifier = svm.SVC(kernel='linear', probability=True,
                     random_state=random_state)

mean_tpr = 0.0
mean_fpr = np.linspace(0, 1, 100)

Here is the error:

TypeError                                 Traceback (most recent call last)
<ipython-input-227-2af2773f4987> in <module>()
----> 1 sklearn.cross_validation.StratifiedKFold(n_splits=6)
      2 #cv = StratifiedKFold(n_splits=6,  shuffle=True, random_state=1)
      3 classifier = svm.SVC(kernel='linear', probability=True,
      4                      random_state=random_state)
      5 

TypeError: __init__() got an unexpected keyword argument 'n_splits'

Answer 1

You are not getting any warnings while importing the sklearn.cross-validation module. This means that your installed version is less than 0.18.

If your scikit-learn version is < 0.18 , then change the below lines: (Taken from StratifiedKFold documentation for version 0.17 )

#Notice the extra parameter y and change of name for n_splits to n_folds
cv = StratifiedKFold(y, n_folds=6)

#Also note that the cv is called directly in for loop
for train_index, test_index in cv:
   print("TRAIN:", train_index, "TEST:", test_index)
   X_train, X_test = X[train_index], X[test_index]
   y_train, y_test = y[train_index], y[test_index]

If your scikit-learn version is >=0.18 , then only you can use the n_splits parameter for the cv : (Taken from StratifiedKFold current documentation , which is what I think you are referring to)

#Notice the extra parameter y is removed here
cv = StratifiedKFold(n_splits=6)

#Also note that the cv.split() is called here (opposed to cv in ver 0.17 above)
for train_index, test_index in cv.split(X, y):
   print("TRAIN:", train_index, "TEST:", test_index)
   X_train, X_test = X[train_index], X[test_index]
   y_train, y_test = y[train_index], y[test_index]

Recommendation :

Update your scikit-learn to the latest version 0.18. Because most documentation you will find by directly searching will be of this version and it will get you confused.

Edit:

I have already answered your similar question here: - Issue with Cross Validation

So next time, please mention the version of libraries you use in the question itself, and remember to access their relevant documentation, not the other ones.