std / sqrt(len)

[jingweiz]

Hi,
maybe I'm missing something obvious here, but could you inform me as why the std is divided again by sqrt(len) as this should already be taken care of in the std calculation?

attention-learn-to-route/eval.py

Line 77 in 6dbad47

print("Average cost: {} +- {}".format(np.mean(costs), 2 * np.std(costs) / np.sqrt(len(costs))))

Thanks!

[wouterkool]

Hi! This is to compute the std of the mean, see https://math.stackexchange.com/questions/504288/what-situation-calls-for-dividing-the-standard-deviation-by-sqrt-n.

…

Op 24 jun. 2021 om 09:14 heeft Jingwei Zhang ***@***.***> het volgende geschreven:  Hi, maybe I'm missing something obvious here, but could you inform me as why the std is divided again by sqrt(len) as this should already be taken care of in the std calculation? https://github.com/wouterkool/attention-learn-to-route/blob/6dbad47a415a87b5048df8802a74081a193fe240/eval.py#L77 Thanks! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[jingweiz]

Oh thanks for the quick reply!
I understand that e.g. when integrating the plots of multiple runs the standard error should be calculated and plotted, however is there a particular reason here to give the se instead of the std of the sample? Thanks in advance!