-
The evaluation log of 3DSSD for kitti with default config is below.
> Car AP@0.70, 0.70, 0.70:
bbox AP:96.6093, 89.8885, 89.2153
bev AP:90.2018, 87.8845, 86.5672
3d AP:88.7361, 78.3761, 77.18…
-
My environment:
fatal: not a git repository (or any parent up to mount point /)
Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).
sys.platform: linux
Python: 3.7.10 (defau…
-
Trying to training model with the all_pp_nu.config, the model got from server face some problem when inference:
1. this part code in rpn.py will occur an error said the ups[2] has the shape of [1, …
-
想问一下关于apollo里面pointpillars模型是怎么转成onnx的?apollo里面有转换脚本么?谢谢
-
Hi,
I successfully built the environment from your official Dockfile
```
ARG PYTORCH="1.6.0"
ARG CUDA="10.1"
ARG CUDNN="7"
FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel
…
-
**Notice**
There are several common situations in the reimplementation issues as below
1. Reimplement a model in the model zoo using the provided configs
2. Reimplement a model in the model zoo o…
-
I'm confused about the anchor settings in fpn neck when using nuScenes dataset, could you tell me why setting anchor sizes like this?
https://github.com/open-mmlab/mmdetection3d/blob/440e44a0e03d575b…
-
I noticed that in recent update #13189 , you remove all onnx models and use several libtorch models instead. Does this improve the speed? And why use several separate libtorch models rather than using…
-
I also meet the same bugs with [#21](https://github.com/open-mmlab/mmdetection3d/issues/21)
-
I want to train the DeepRoute Open Dataset. Here is the website: https://gas.graviti.cn/dataset/graviti-open-dataset/DeepRoute
There is no camera data in the dataset, only lidar data. It is differe…