Typos in training guidelines?
noahcao opened this issue · 2 comments
noahcao commented
Hi Xingyi,
Thanks for the wonderful job. I tried to run the training on MOT17 following the guidelines. But I found some potential typos making that doable.
- Should we rename the MOT17 train to trainval, which is not explained in the prepare datasets doc?
- Should the datasets for training be ("mot17_halftrain","crowdhuman_train") instead of ("mot17_halftrain","crowdhuman_amodal_train") in the config file? the later one would raises an error of unregistered dataset:
xingyizhou commented
Thank you a lot for pointing out the issue! You are correct in both. I fixed them in the most recent commit.
Best,
Xingyi
noahcao commented
Thanks. I think the current version fixed the problem.