Loss function converges

[guoees]

hello,The loss function converges relatively high, and I want to know how much the loss function converges to a better model.

[Arthur151]

Hi, @guoees .
Yes, the training process needs quite some time.
Which model are you triaining, ROMP or BEV?
For ROMP, the overall loss is supposed to converge to ~ 120.
For BEV, the overall loss is supposed to converge to ~ 350.
Training more time would be very benefical to the performance.