not understand "log probability in the center of the bin"
Beronx86 opened this issue · 2 comments
Beronx86 commented
Could anybody explain the default values to be used in extreme cases.
68 # log probability in the center of the bin, to be used in extreme cases
69 # (not actually used in our code)
70 log_pdf_mid = mid_in - log_scales - 2. * tf.nn.softplus(mid_in)
nagichan commented
xiaonengmiao commented
why this is used in extreme case? particularly:
inner_inner_cond = (cdf_delta > 1e-5).float()
inner_inner_out = inner_inner_cond * torch.log(torch.clamp(cdf_delta, min=1e-12)) + (1. - inner_inner_cond) * (log_pdf_mid - np.log(127.5))
inner_cond = (x > 0.999).float()
inner_out = inner_cond * log_one_minus_cdf_min + (1. - inner_cond) * inner_inner_out
cond = (x < -0.999).float()
log_probs = cond * log_cdf_plus + (1. - cond) * inner_out
log_probs = torch.sum(log_probs, dim=3) + log_prob_from_logits(logit_probs)
why is log_pdf_mid - np.log(127.5)
?