Briefly explain why the estimated true error from k-fold cross validation is biased. Why might k=10 be a reasonable choice for k? code example

Example 1: cross validation error

ModuleNotFoundError: No module named 'sklearn.cross_validation'

Example 2: cross validation error

cross validation error