MouJiajun

Hi, I was confused about the smooth_l1_loss_layer.cu， that the diff_ in Forward_gpu has multiply the "inside" weights, why in Backward_gpu, multiply the "inside" weights again?

关于vfe

1

嗨，你好，我发现你的这个版本在group_pointcloud.py里去掉了vfe2之后的全连接层，直接进行了reduce_max。因此在vfe里，最后的操作也有一个reduce_max，然后在进行tile和concat。 那进行tile和concat之后，再进行reduce_max，得到的结果和vfe2里做的reduce_max的结果不就一样的么？那这样不就相当于做了重复操作？ 另外，为何要去掉那个全连接层呢？是对性能有影响么？

ERROR: cannot launch node of type [point_cloud_converter/point_cloud_converter]: point_cloud_converter

2

Hi: when I try to roslaunch ethzasl_icp_mapper ICRA14_dynamic_mapper.launch, I just get this error! so I download a package named point_cloud_converter, it still can run right. besides, in the package ,I...