TiagoCortinhal / SalsaNext

Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving
MIT License
417 stars 102 forks source link

Long type expected, Int provided #54

Closed Moldoteck closed 3 years ago

Moldoteck commented 3 years ago

At trainer.py, line 384> loss_m = criterion(torch.log(output.clamp(min=1e-8)), proj_labels) + self.ls(output, proj_labels.long()) should be replaced with loss_m = criterion(torch.log(output.clamp(min=1e-8)), proj_labels.long()) + self.ls(output, proj_labels.long())