centerpoint_pillar pretrained model error

[iloveai8086]

Hello.
when i am using the following command:
python demo/pcd_demo.py --pcd demo/data/nuscenes/1.bin --config configs/centerpoint/centerpoint_02pillar_second_secfpn_circlenms_4x8_cyclic_20e_nus.py --checkpoint checkpoints/centerpoint_02pillar_second_secfpn_circlenms_4x8_cyclic_20e_nus_20210816_064624-0f3299c0.pth --show

and the model is downloaded from the model link in configs/centerpoint

Run the above command and it will get the following results：
load checkpoint from local path: checkpoints/centerpoint_02pillar_second_secfpn_circlenms_4x8_cyclic_20e_nus_20210816_064624-0f3299c0.pth
The model and loaded state dict do not match exactly
size mismatch for pts_voxel_encoder.pfn_layers.0.linear.weight: copying a param with shape torch.Size([64, 10]) from checkpoint, the shape in current model is torch.Size([64, 11])

and my env list:
mmcls 0.23.1
mmcv-full 1.5.3
mmdeploy 0.5.0
mmdet 2.25.0
mmdet3d 1.0.0rc3
mmsegmentation 0.25.0

Is this a bug, or the model of centerpoint pillar is not updated?

[Tai-Wang]

@ZCMax Please check this problem. Maybe it is another problem related to the previous KITTI performance upgrading of PointPillars.

[hlhzau]

also encountered this problem. What caused this problem? Has it been solved?

[rkotimi]

@ZCMax Hello! When can you solve this problem? I need the pretrained model.

[ZCMax]

@ZCMax Hello! When can you solve this problem? I need the pretrained model.

This checkpoint will be provided in one or two days, and other centerpoint checkpoints will be provided in this week.

[ZCMax]

@ZCMax Hello! When can you solve this problem? I need the pretrained model.

Sorry for the inconvenience caused by the model checkpoint problem

[rkotimi]

@ZCMax Hello! When can you solve this problem? I need the pretrained model.

Sorry for the inconvenience caused by the model checkpoint problem

Thank you for your reply. I will wait for the new checkpoint.

[rkotimi]

@ZCMax Hello! When can you solve this problem? I need the pretrained model.

This checkpoint will be provided in one or two days, and other centerpoint checkpoints will be provided in this week.

@ZCMax Hello! I wonder if the new checkpoints are available. It seems the checkpoint links are not updated in this page

[hlhzau]

@ZCMax I wonder if the new checkpoints are available?

[ZCMax]

I've prepared a pretrained model of centerpoint_02pillar_second_secfpn_4x8_cyclic_20e_nus for temporal link:

link: https://pan.baidu.com/s/1u1dS6XPbzhvrMNfquuoeuA?pwd=g5qi
password: g5qi

[rahuja123]

Is there a non baidu link?

[jaan1729]

I tried to use pretrained model from the documentation for centerpoint_pillar02_second_secfpn_8xb4-cyclic-20e_nus-3d.py to test Centerpoint with the following command.

python tools/test.py --task lidar_det configs/centerpoint/centerpoint_pillar02_second_secfpn_8xb4-cyclic-20e_nus-3d.py checkpoints/centerpoint/centerpoint_01voxel_second_secfpn_circlenms_4x8_cyclic_20e_nus_20220810_030004-9061688e.pth --show --show-dir ./data/centerpoint/show_results

I'm facing weights mismatch issue with the following log....

Loads checkpoint by local backend from path: checkpoints/centerpoint/centerpoint_01voxel_second_secfpn_circlenms_4x8_cyclic_20e_nus_20220810_030004-9061688e.pth

The model and loaded state dict do not match exactly

size mismatch for pts_backbone.blocks.0.0.weight: copying a param with shape torch.Size([128, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for pts_backbone.blocks.0.1.weight: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.1.bias: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.1.running_mean: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.1.running_var: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.3.weight: copying a param with shape torch.Size([128, 128, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for pts_backbone.blocks.0.4.weight: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.4.bias: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.4.running_mean: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.4.running_var: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.6.weight: copying a param with shape torch.Size([128, 128, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for pts_backbone.blocks.0.7.weight: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.7.bias: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.7.running_mean: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.7.running_var: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.9.weight: copying a param with shape torch.Size([128, 128, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 64, 3, 3]).
size mismatch for pts_backbone.blocks.0.10.weight: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.10.bias: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.10.running_mean: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.0.10.running_var: copying a param with shape torch.Size([128]) from checkpoint, the shape in current model is torch.Size([64]).
size mismatch for pts_backbone.blocks.1.0.weight: copying a param with shape torch.Size([256, 128, 3, 3]) from checkpoint, the shape in current model is torch.Size([128, 64, 3, 3]).
size mismatch for pts_backbone.blocks.1.1.weight: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.1.bias: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.1.running_mean: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.1.running_var: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.3.weight: copying a param with shape torch.Size([256, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([128, 128, 3, 3]).
size mismatch for pts_backbone.blocks.1.4.weight: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.4.bias: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.4.running_mean: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.4.running_var: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.6.weight: copying a param with shape torch.Size([256, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([128, 128, 3, 3]).
size mismatch for pts_backbone.blocks.1.7.weight: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.7.bias: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.7.running_mean: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.7.running_var: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.9.weight: copying a param with shape torch.Size([256, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([128, 128, 3, 3]).
size mismatch for pts_backbone.blocks.1.10.weight: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.10.bias: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.10.running_mean: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.10.running_var: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.12.weight: copying a param with shape torch.Size([256, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([128, 128, 3, 3]).
size mismatch for pts_backbone.blocks.1.13.weight: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.13.bias: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.13.running_mean: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.13.running_var: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.15.weight: copying a param with shape torch.Size([256, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([128, 128, 3, 3]).
size mismatch for pts_backbone.blocks.1.16.weight: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.16.bias: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.16.running_mean: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_backbone.blocks.1.16.running_var: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_neck.deblocks.0.0.weight: copying a param with shape torch.Size([256, 128, 1, 1]) from checkpoint, the shape in current model is torch.Size([128, 64, 2, 2]).
size mismatch for pts_neck.deblocks.0.1.weight: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_neck.deblocks.0.1.bias: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_neck.deblocks.0.1.running_mean: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_neck.deblocks.0.1.running_var: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_neck.deblocks.1.0.weight: copying a param with shape torch.Size([256, 256, 2, 2]) from checkpoint, the shape in current model is torch.Size([128, 128, 1, 1]).
size mismatch for pts_neck.deblocks.1.1.weight: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_neck.deblocks.1.1.bias: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_neck.deblocks.1.1.running_mean: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_neck.deblocks.1.1.running_var: copying a param with shape torch.Size([256]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for pts_bbox_head.shared_conv.conv.weight: copying a param with shape torch.Size([64, 512, 3, 3]) from checkpoint, the shape in current model is torch.Size([64, 384, 3, 3]).
unexpected key in source state_dict: pts_middle_encoder.conv_input.0.weight, pts_middle_encoder.conv_input.1.weight, pts_middle_encoder.conv_input.1.bias, pts_middle_encoder.conv_input.1.running_mean, pts_middle_encoder.conv_input.1.running_var, pts_middle_encoder.conv_input.1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer1.0.conv1.weight, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn1.weight, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn1.bias, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn1.running_var, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer1.0.conv2.weight, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn2.weight, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn2.bias, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn2.running_mean, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn2.running_var, pts_middle_encoder.encoder_layers.encoder_layer1.0.bn2.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer1.1.conv1.weight, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn1.weight, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn1.bias, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn1.running_var, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer1.1.conv2.weight, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn2.weight, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn2.bias, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn2.running_mean, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn2.running_var, pts_middle_encoder.encoder_layers.encoder_layer1.1.bn2.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer1.2.0.weight, pts_middle_encoder.encoder_layers.encoder_layer1.2.1.weight, pts_middle_encoder.encoder_layers.encoder_layer1.2.1.bias, pts_middle_encoder.encoder_layers.encoder_layer1.2.1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer1.2.1.running_var, pts_middle_encoder.encoder_layers.encoder_layer1.2.1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer2.0.conv1.weight, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn1.weight, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn1.bias, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn1.running_var, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer2.0.conv2.weight, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn2.weight, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn2.bias, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn2.running_mean, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn2.running_var, pts_middle_encoder.encoder_layers.encoder_layer2.0.bn2.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer2.1.conv1.weight, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn1.weight, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn1.bias, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn1.running_var, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer2.1.conv2.weight, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn2.weight, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn2.bias, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn2.running_mean, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn2.running_var, pts_middle_encoder.encoder_layers.encoder_layer2.1.bn2.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer2.2.0.weight, pts_middle_encoder.encoder_layers.encoder_layer2.2.1.weight, pts_middle_encoder.encoder_layers.encoder_layer2.2.1.bias, pts_middle_encoder.encoder_layers.encoder_layer2.2.1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer2.2.1.running_var, pts_middle_encoder.encoder_layers.encoder_layer2.2.1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer3.0.conv1.weight, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn1.weight, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn1.bias, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn1.running_var, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer3.0.conv2.weight, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn2.weight, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn2.bias, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn2.running_mean, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn2.running_var, pts_middle_encoder.encoder_layers.encoder_layer3.0.bn2.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer3.1.conv1.weight, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn1.weight, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn1.bias, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn1.running_var, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer3.1.conv2.weight, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn2.weight, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn2.bias, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn2.running_mean, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn2.running_var, pts_middle_encoder.encoder_layers.encoder_layer3.1.bn2.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer3.2.0.weight, pts_middle_encoder.encoder_layers.encoder_layer3.2.1.weight, pts_middle_encoder.encoder_layers.encoder_layer3.2.1.bias, pts_middle_encoder.encoder_layers.encoder_layer3.2.1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer3.2.1.running_var, pts_middle_encoder.encoder_layers.encoder_layer3.2.1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer4.0.conv1.weight, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn1.weight, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn1.bias, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn1.running_var, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer4.0.conv2.weight, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn2.weight, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn2.bias, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn2.running_mean, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn2.running_var, pts_middle_encoder.encoder_layers.encoder_layer4.0.bn2.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer4.1.conv1.weight, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn1.weight, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn1.bias, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn1.running_mean, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn1.running_var, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn1.num_batches_tracked, pts_middle_encoder.encoder_layers.encoder_layer4.1.conv2.weight, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn2.weight, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn2.bias, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn2.running_mean, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn2.running_var, pts_middle_encoder.encoder_layers.encoder_layer4.1.bn2.num_batches_tracked, pts_middle_encoder.conv_out.0.weight, pts_middle_encoder.conv_out.1.weight, pts_middle_encoder.conv_out.1.bias, pts_middle_encoder.conv_out.1.running_mean, pts_middle_encoder.conv_out.1.running_var, pts_middle_encoder.conv_out.1.num_batches_tracked, pts_backbone.blocks.0.12.weight, pts_backbone.blocks.0.13.weight, pts_backbone.blocks.0.13.bias, pts_backbone.blocks.0.13.running_mean, pts_backbone.blocks.0.13.running_var, pts_backbone.blocks.0.13.num_batches_tracked, pts_backbone.blocks.0.15.weight, pts_backbone.blocks.0.16.weight, pts_backbone.blocks.0.16.bias, pts_backbone.blocks.0.16.running_mean, pts_backbone.blocks.0.16.running_var, pts_backbone.blocks.0.16.num_batches_tracked

missing keys in source state_dict: pts_voxel_encoder.pfn_layers.0.norm.weight, pts_voxel_encoder.pfn_layers.0.norm.bias, pts_voxel_encoder.pfn_layers.0.norm.running_mean, pts_voxel_encoder.pfn_layers.0.norm.running_var, pts_voxel_encoder.pfn_layers.0.linear.weight, pts_backbone.blocks.2.0.weight, pts_backbone.blocks.2.1.weight, pts_backbone.blocks.2.1.bias, pts_backbone.blocks.2.1.running_mean, pts_backbone.blocks.2.1.running_var, pts_backbone.blocks.2.3.weight, pts_backbone.blocks.2.4.weight, pts_backbone.blocks.2.4.bias, pts_backbone.blocks.2.4.running_mean, pts_backbone.blocks.2.4.running_var, pts_backbone.blocks.2.6.weight, pts_backbone.blocks.2.7.weight, pts_backbone.blocks.2.7.bias, pts_backbone.blocks.2.7.running_mean, pts_backbone.blocks.2.7.running_var, pts_backbone.blocks.2.9.weight, pts_backbone.blocks.2.10.weight, pts_backbone.blocks.2.10.bias, pts_backbone.blocks.2.10.running_mean, pts_backbone.blocks.2.10.running_var, pts_backbone.blocks.2.12.weight, pts_backbone.blocks.2.13.weight, pts_backbone.blocks.2.13.bias, pts_backbone.blocks.2.13.running_mean, pts_backbone.blocks.2.13.running_var, pts_backbone.blocks.2.15.weight, pts_backbone.blocks.2.16.weight, pts_backbone.blocks.2.16.bias, pts_backbone.blocks.2.16.running_mean, pts_backbone.blocks.2.16.running_var, pts_neck.deblocks.2.0.weight, pts_neck.deblocks.2.1.weight, pts_neck.deblocks.2.1.bias, pts_neck.deblocks.2.1.running_mean, pts_neck.deblocks.2.1.running_var

@ZCMax , is there any reason for not making the model available in Baidu as official?