Can someone please confirm that the bug reported on Dec '17 by @Chris_Palmer was fixed? I still have exactly the same so it would help. Thank you!
Difference between sub sampling and bootstrapping
With reference to the following code discussed in lesson 5, the code is doing sampling with replacement. We may get a row which we have received for a previous decision tree. Why then it is not bootstrapping?
def create_tree(self):
rnd_idxs = np.random.permutation(len(self.y))[:self.sample_sz]
return DecisionTree(self.x.iloc[rnd_idxs], self.y[rnd_idxs], min_leaf=self.min_leaf)
Where I am going wrong in understanding the difference between sub-sampling and bootstrapping?
If your data set consists of N rows (examples), a bootstrap sample is a set of N samples with replacement, while a subsample is a set of fewer than N samples.