About training loss

[Ltohka]

I followed readme to reproduce the project and when I do stage 2 training, the loss contains nan.

if torch.isnan(depth_pred).any():
                    logging.warning("model_pred contains NaN.")

I put this code after this line, and its output shows that this is a nan.

So Where did I go wrong?

[indu1ge]

Hi,

Regarding the nan loss issue, you can try commenting out Lines 325 and 326 in train_s2.py.

Additionally, we have fixed several other issues. Please refer to my recent commit: 63999f6 (click the hash to view the changes).

After updating the code, the training should be normal, and the loss should be around 0.006.

Hope this helps!