Rishabh Agrahari
Rishabh Agrahari
look into utils.py file, modify the functions accordingly.
Probably, your ground truth or predicted json file formats are not as expected.
So, it's because some of the classes in your ground truth file don't have any boxes. (Probably because of a bad validation set) On Sun 3 Nov, 2019, 5:15 PM...
> .. the above precision values at *each* IoU threshold So, it is indeed like running the same script with different thresholds.
Hey, what value are you using for `nms_range_list`? `nms_range_list = [-40.0, 40.0, 80.0]` should work, right?
Hey Lee, I'm not actively maintaining this repository as of now, please do the necessary debugging yourself and let me know if you need any help in doing so.
Hey guys, good to know that this issue is resolved now. Closing the issue.
Hey, thanks. What kind of segmentation do you want to visualize?
Oh okay, I'm not sure about that. On Fri 11 Oct, 2019 10:47 am ChengDahai, wrote: > For example, visualize the different classes for the segmentation result? > Or paint...
I set [0,0,0,0] as a box for those cases and it works pretty well. (All the priors come out to be negative, so no loc_loss for them and all of...