I noticed that np.var doesn’t compute the unbiased variance. If one wants to calculate the unbiased variance, np.var(args, ddof=1) must be used instead. This is not a big deal, but I think it is probably a good idea to mention it in the lecture.

You are abusing the word “unbiased”.

The definition of variance is divided by n instead of n-1. You set ddof=1 only when you are computing “the unbiased estimator of populatiion variance”.

Therefore, we shouldn’t change that. That said, if you spot some place where we should use unbiased estimator but we didn’t please let us know. Thanks!