Closed moskewcz closed 6 years ago
see also issue #10 -- i think that issue was meant to apply to both r_map and p_map, but the fix was only applied to r_map. i dunno why the creator of that issue didn't comment further, though, so again maybe i'm just confused.
Thanks, fix in d24c3ee.
i'm having trouble understanding why the p_map layer doesn't have 'activation=False':
https://github.com/jeasinema/VoxelNet-tensorflow/blob/master/model/rpn.py#L98
it's fed into a sigmoid, so if the activations are ReLU'd to be non-negative, it seems the objectness outputs:
https://github.com/jeasinema/VoxelNet-tensorflow/blob/master/model/rpn.py#L104
will be limited to [.5,1]. maybe even if that's a bit wrong-ish, it doesn't much matter somehow?