K-fold cross validation

In the quest to find the best model, you can view the indices of cross-validation folds and see what data is in each fold.

Getting ready

Create a toy dataset that is very small:

import numpy as np
X = np.array([[1, 2], [3, 4], [5, 6], [7, 8],[1, 2], [3, 4], [5, 6], [7, 8]])
y = np.array([1, 2, 1, 2, 1, 2, 1, 2])

How to do it..

1. Import KFold and select the number of splits:

from sklearn.model_selection import KFold

kf= KFold(n_splits = 4)

2. You can iterate through the generator and print out the indices:

cc = 1
for train_index, test_index in kf.split(X):
      print "Round : ",cc,": ",
      print "Training indices :", train_index,
      print "Testing indices :", test_index
      cc += 1

Round 1 : Training indices : [2 3 4 5 6 7] Testing indices : [0 1]
Round 2 : Training indices : [0 1 4 5 6 7] Testing indices : [2 3]
Round 3 : Training indices : [0 1 2 3 6 7] Testing indices : [4 5]
Round 4 : Training indices : [0 1 2 3 4 5] Testing indices : [6 7]

You can see, for example...