Dear Robert, I agree this is definitely not worth changing anything major. I was mostly suggesting this for convenience and was envisioning simply taking p = weights / sum(weights) internally if weights is supplied. I will note that if you are working with weights that happen to all be integers, which they often are, you have to take some care when converting them to probabilities to use in the random.choice() function (i.e., you cannot use "/="). However, you are right in that this is a pretty minor thing that can be written in-line.
Thank you for your time. _______________________________________________ NumPy-Discussion mailing list -- [email protected] To unsubscribe send an email to [email protected] https://mail.python.org/mailman3//lists/numpy-discussion.python.org Member address: [email protected]
