Thanks for your great work! I have run the training code for depth estimation and foun

Some problems of nusc-depth training code about occnerf HOT 2 OPEN

linshan-bin commented on July 28, 2024

Some problems of nusc-depth training code

from occnerf.

Comments (2)

LinShan-Bin commented on July 28, 2024

Thanks for your feedback! Since we didn't observe this particular issue in our initial experiments, could you kindly provide us with additional details (the training log)? We are actively working to replicate this error and are committed to enhancing the stability of the fp16 training process.
It's a normal phenomenon when using the photometric loss. But the network is actually learning and you can wait for the result.

from occnerf.

eliliu2233 commented on July 28, 2024

Thanks for your feedback! Since we didn't observe this particular issue in our initial experiments, could you kindly provide us with additional details (the training log)? We are actively working to replicate this error and are committed to enhancing the stability of the fp16 training process.

It's a normal phenomenon when using the photometric loss. But the network is actually learning and you can wait for the result.

Thanks for your reply. I found that nan always happen while calculating rendering weights, so I force the tensor type in weights calculation to fp32 and solve this problem.

from occnerf.

Recommend Projects