How is the structure of the Tartan datasets arranged during training?

[Tianci-Wen]

Thank you for open sourcing your excellent work. I encountered some problems during training. For one of the sequences, my path is as follows： ～/datasets/TartanAir/abandonedfactory/Easy/depth_left/abandonedfactory/EasyP000/depth_left/xxxx.npy ~/datasets/TartanAir/abandonedfactory/Easy/image_left/abandonedfactory/Easy/P000/image_left/xxxx.png

How can I rearrange the file structure? Is there any preprocessing after downloading the data set? Thank you very much and look forward to your answer.

[lahavlipson]

The data should be stored as DPVO/datasets/TartanAir/abandonedfactory/abandonedfactory/Easy/P001/image_left/xxxx.png

Sorry for the confusion. I believe this is how the tartanair_tools github repo unpacks it

[Tianci-Wen]

The data should be stored as DPVO/datasets/TartanAir/abandonedfactory/abandonedfactory/Easy/P001/image_left/xxxx.png

Sorry for the confusion. I believe this is how the tartanair_tools github repo unpacks it

Thank you very much for your patient reply! I found this in the tartanair_tools github repo, but it still doesn't work when testing driod-slam. I tested the first abandonedfactory on DPVO and it works. I will test and train DPVO after all sequences have been downloaded.