Negative loss
Closed this issue · 2 comments
shawrby commented
I am currently training the bev model with my dataset.
However, after enough steps, the total loss becomes negative as shown below.
After debugging, when calculating uncertainty_loss(ce_uncertainty_loss, center_uncertainty_loss, offset_uncertainty_loss),
weight(model.module.ce_weight, model.module.center_weight, model.module.offset_weight) is negative.
Are there any possible causes?
aharley commented
Negative loss is OK here. For context you can check these:
https://github.com/yaringal/multi-task-learning-example
https://arxiv.org/abs/1705.07115
shawrby commented
Negative loss is OK here. For context you can check these:
https://github.com/yaringal/multi-task-learning-example
https://arxiv.org/abs/1705.07115
Thank you so much!