What is a reasonable training time on kitti dataset? about delora HOT 6 CLOSED

Hi @g1y5x3 ,
It is hard to say what the right duration is, as it highly depends on the chosen learning-rate.
But for sure I never trained it for more than 200 epochs (you will also see that the loss kinda converges after that amount of time).
Best,

from delora.

Thank you for your clarification. So the 10,000 epochs that was in deployer/trainer.py was just a generic configuration but not necessarily the one used for reproducing the results right?

from delora.

Actually, would you mind share the training parameters that you used in the paper?

from delora.

yeah, usually I deployed it on servers and wanted to run it as long as possible rather than being killed by an internal epoch limit.

from delora.

Which parameters do you mean? model parameters (i.e. weights)?

from delora.

There is an example checkpoint for kitti provided here: https://github.com/leggedrobotics/DeLORA/tree/main/checkpoints

from delora.