Tangtao

Problems of the dimension: @Tai-Wang in the Loss function of CenterPointBBoxHead: the loss_reg : bbox_pred dim : [batch_size * **num_bbox_sampled**, bbox_code_size] bbox_targets dim: [batch_size * **num_pos_bboxes_sampled**, bbox_code_size]

So what's the mapping between old and new filenames? Why not just keep consistent with the original valset

> > Is it possible not to run `torchpack` in the root mode or in a container? > > I run the following code in the root folder: > >...

> Hi @luoyuchenmlcv @kentang-mit, I have a problem about where to place the data which are generated by the 'create_data.py' file. When the data in the directory of mmdet3d, I...

> after run code: > > ``` > cd ranking_nats > python get_model_score_nats.py > ``` > > I got: > > ``` > kendall tau begin > BossNAS: KendalltauResult(correlation=-0.534180602248828, pvalue=0.0)...

Are they depth_loss and sigma_loss in the code, i.e., DS-NeRF&MSE and DS-NeRF&KL divergence in the paper respectively?

Looking forward to your models and training code.

Yes, I have pre-trained a 2D backbone on Waymo first. The config and log as follows: [waymo-2d-log.txt](https://github.com/XuyangBai/TransFusion/files/8678971/waymo-2d-log.txt). And I fixed the backbone of image and lidar for training. I don't...

Thanks for the kind reply, but if I directly changed for labels in _for labels in frame.projected_lidar_labels_ to _for labels in frame.camera_labels_ , it will not work. Maybe the same...

Hello, thanks for your feedback. I will double-check the results on mvl, but this may take several days since I have been busy with some work recently. I provide the...