Closed cloudy-toast closed 2 years ago
91 is approximately the maximum "error" in terms of euclidean distance in heat map you may get (i.e. prediction at [0, 0] while ground truth is [63, 63]
Yes, in fact 91 = 64*square(2). We need normalization here.
Hi Walid,
Thanks for sharing the code! Could you please tell me why you used 91 in the function ' _compute_err(self, u, v)'? I understand you want to calculate the distance btw two locations and rescale it to [0,1], but why not divide it by 64?
Best regards, Ruobing