Jiangxb

Why can't I use the torchex.connected_components to get the same result as scipy.parse.csgraph.connected_components? Sometimes it's the same, but sometimes it's different, and the initial estimate is 50-50。 the code is...

About intra-modal spatial perception module parameters update problem

1

Hello author, in the SimIPU paper, I find that the inter-modal module loss is gradient truncation in the point cloud features. When gradient backpropagation is considered, this step does not...

Why not use 200, Was it a clerical error?

1

https://github.com/LiWentomng/BoxInstSeg/blob/main/configs/boxinst/boxinst_r50_fpn_1x_coco.py#L88 max_per_img = 2000, but in other framework configurations it's 200.

[Bug] 'CocoDataset is not in the mmdet3d::dataset registry.

2

### Prerequisite - [X] I have searched [Issues](https://github.com/open-mmlab/mmdetection3d/issues) and [Discussions](https://github.com/open-mmlab/mmdetection3d/discussions) but cannot get the expected help. - [X] I have read the [FAQ documentation](https://mmdetection3d.readthedocs.io/en/latest/notes/faq.html) but cannot get the expected help....

The render result on the KITTI dataset

7

Hello author, do you have the rendering result of 3dgs on kitti? My own rendering result was very poor, and both sides of the picture could not be rendered.

How do semantic labels align with multiple sweeps in nuScenes dataset?

1

When training on the nuScenes dataset, the semantic segmentation label of nuScenes is only the current frame, which is 10 sweeps input during training. How does it match the corresponding...

mumu12，雷电模拟器，每次重启游戏都会突然返回主界面

3

### 在提问之前... - [x] 我已经搜索了现有的 issues - [x] 我在提问题之前至少花费了 5 分钟来思考和准备 - [x] 我已经阅读了 Wiki 中的 常见问题(FAQ) - [x] 我正在使用最新版的 Alas ### 描述你的问题 每次使用alas重启模拟器，进入游戏后都会突然跳到模拟器的主界面 ### 如何复现 1. 前往 alas 2....