Yuan Lijiang
Yuan Lijiang
Hi, I would like to know how this comparison experiment is done and how to go about comparing performance gains by distance? Is it compared by a certain distance within...
作者您好,目前在复现您的工作,我用的不是kitti和nyu数据集,用的是清华大学发布的dair数据集,里面也有激光雷达点云图和rgb图片,图片的分辨率大小是在1920*1080,但由于本人目前计算资源不够,所以全尺寸跑的话batch_size只能设为2,训练过程中大约100次左右的iteration后就loss就不变了大约在1点几左右,于是我将图片resize了一下,现在尺寸为960*448,batch_size为8,但loss就变成了30多。想向您请教俩个问题 1.您之前训练kitti的时候有对深度图进行resize吗? 2.对于我这种情况您是建议寻找更好的计算资源还是去调整loss
I have my own dataset and want to train this model on my own dataset but i didn‘t see the detail of how to train