search

davidtvs / PyTorch-ENet

PyTorch implementation of ENet
MIT License
389 stars 129 forks source link

Training with two classes #42

Closed PatrickNa closed 4 years ago

PatrickNa commented 4 years ago

When I try to train the model with only two classes, let's say 'road' and 'unlabeled' the IOU remains 1 all the time. Is there a bug or is it something that I need to configure differently?

PatrickNa commented 4 years ago

When I add a third class that as 'unlabeled' represents background the IoU changes over time. However this measure feels like a hack.

davidtvs commented 4 years ago

Have you tried using the command line option --with-unlabeled?

By default (without that option) the unlabeled class is ignored when computing metrics.

PatrickNa commented 4 years ago

This seems to do the trick. Thank you very much!