You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello author, what trick did you use when training the neural network? In the forest scene, I used the train.py file directly, but the trained model could never achieve the same effect as your model. can you help me? Thanks.
The text was updated successfully, but these errors were encountered:
We trained on multiple scenes and a large number of environments. The quickest way to do this is to download the data from the associated link and train on our previously collected dataset. After a finetuning round at the speed you need to go, the network would reach approximately (accounting for the randomness in the simulator) the same performance of our checkpoint.
Hello author, what trick did you use when training the neural network? In the forest scene, I used the train.py file directly, but the trained model could never achieve the same effect as your model. can you help me? Thanks.
The text was updated successfully, but these errors were encountered: