Inconsistency between X and Y loss

nachiket92 / conv-social-pooling

Code for model proposed in: Nachiket Deo and Mohan M. Trivedi,"Convolutional Social Pooling for Vehicle Trajectory Prediction." CVPRW, 2018

MIT License

336 stars 102 forks source link

Inconsistency between X and Y loss #26

Closed AmitaiPerlstein closed 3 years ago

AmitaiPerlstein commented 4 years ago

utils lines 170, 172 (for feet, meters respectively) have 0.5torch.pow(sigY, 2)torch.pow(y-muY, 2) for Y but torch.pow(sigX, 2) * torch.pow(x - muX, 2) for X

nachiket92 commented 4 years ago

Both terms have the 0.5 multiplicative factor.

Edit: Ah yes, you're right there was still another typo. The 0.5 factor was outside the bracket for both terms, but was repeated inside for the Y term. This has been fixed. Thanks!