Which of the following is NOT correct about the Occam$$s Razor principle?

Question $33$ options:

It implies that one should factor model complexity when evaluating a model.

All of the other choices are correct statements about the Occam's Razor principle

When there is overfitting, training error no longer provides a good estimate of how well the tree will perform on previously unseen records.

It cautions that for complex models, there is a greater chance that it was fitted accidentally by errors or noise in data.

It is a principle of model parsimony which suggests that given two models of similar generalization errors, one should prefer the simpler model over the more complex model.