Björn Holtvogt

@achao-c How did you solve the testing for SECOND?

@JoeyforJoy What is wrong with SparseEncoder?

Thanks! Won't I need to change the `split` parameter to `'testing'` as well? Or what's the meaning behind this parameter? Does it affect the data preparation?

By doing so, I receive this after using the testing command: Command: ```shell python tools/test.py --eval mAP ``` Error: ``` [>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 31/31, 19.3 task/s, elapsed: 2s, ETA: 0s Converting prediction...

@24spiders Which point clouds are you displaying on which the bounding boxes look correctly placed? @Tai-Wang How's the status on this issue?

Is it possible to hotfix this by replacing the line in https://github.com/open-mmlab/mmdetection3d/blob/073f353cf21d31beabcbffa45bef10a6f81abff3/mmdet3d/core/visualizer/show_result.py#L106 with ```python if pred_labels is None or pred_labels.numel() == 0 ``` ?

My configuration: ```python model = dict( type='VoxelNet', voxel_layer=dict( max_num_points=5, point_cloud_range=[-1, -3, -2.5, 3, 0.5, 1], voxel_size=[0.05, 0.05, 0.1], max_voxels=(16000, 40000)), voxel_encoder=dict(type='HardSimpleVFE'), middle_encoder=dict( type='SparseEncoder', in_channels=4, sparse_shape=[36, 70, 80], order=('conv', 'norm', 'act')),...

Got this when changing to 128: ``` load checkpoint from local path: /home/ws/x/Dokumente/Repositories/dasad3d/checkpoints/itiv/second/latest.pth The model and loaded state dict do not match exactly size mismatch for backbone.blocks.0.0.weight: copying a param...

Can you give a more precise description of the actual problems/errors you're dealing with? For me though, the following setup worked (partially depends on your installed CUDA version): 1. Create...

These are not errors but user warnings. You should be good to go