AttributeError about self-supervised-depth-completion HOT 8 OPEN

@gautamjain1009 Thanks for the screenshot. It turns out that the problem was due to a version difference in the code and the previously saved results. I just pushed a minor update, and it should fix the issue. Let me know if it works for you now.

from self-supervised-depth-completion.

It seems that the file structure of KITTI Depth has changed recently. I will update my code accordingly. Please stay tuned.

from self-supervised-depth-completion.

Just updated the code and the README for the latest file structure.

from self-supervised-depth-completion.

Please provide more details.

from self-supervised-depth-completion.

I have the same issue! everything is working fine but when i run main.py with the pretrained model argument i got this similar error like down below.

I have the same folder structure as described by you in readme file.

Please help.

from self-supervised-depth-completion.

Thank you for your reply. @fangchangma
That error is gone but still not working properly,

under kitti_depth in the data folder, i am using the "train" and "val" which i got from-" Download projected raw LiDaR scans data set (2 GB) " at KITTI depth completion benchmark.

The argument is not able to find anything in the data, i have rechecked the structure and the path again.

from self-supervised-depth-completion.

By default we use the ../data/kitti_depth/val_selection_cropped folder for validation. This is the "manually selected validation and test data sets (5 GB)" on the KITTI website.

I am not entirely sure if the current version of the code still supports loading the uncropped ../data/kitti_depth/val folder, but feel free to give it a shot by adding --val full to your command.

from self-supervised-depth-completion.

I tried adding "--val full" while executing the main.py but still it gives the same issue, what should i do, to make it work.

from self-supervised-depth-completion.