visdom.py reports an error when running the pre trained dimp50

[NKdryer]

The command I run is python pytracking/run_tracker.py dimp dimp50 --dataset_name got10k_val --debug 1 --threads 0 and then rise an error in visdom.py line 332 ValueError: only one element tensors can be converted to Python scalars how can I solve this problem, thanks!

[2006pmach]

Thanks for pointing this out. I think something was changed when we added RTS such that it is no longer running for other trackers. @mattpfr could you have a look? Until this is fixed I recommend working with the repo at this commit 174728a95c74802cd99aa019f9d71f5dfa22b44f

[mattpfr]

@NKdryer @2006pmach thanks for pointing that out. I am looking into the problem, you can expect a fix soon!

[mattpfr]

@NKdryer this should be fix now, please let me know if you still have issues

[NKdryer]

@mattpfr Thanks for your attention! But another error occurred in evaluation/ trackers.py when I ran the above instruction changing the got10k dataset to otb and assigning a sequence. When the visualize method in tracker.py reads the bbox, an exception occurs that is IndexError: invalid index to scalar variable.

[mattpfr]

@NKdryer I am sorry but I cannot reproduce your issue, I can run python run_tracker.py dimp dimp50 --dataset_name otb --sequence Basketball --debug 1 --threads 0 without problem. Can you verify that you don't have local changes on your side, and if not, give me the exact command that you run ?