Danila Rukhovich
Danila Rukhovich
Hi @Cindy0725 , But we use mmcv of 1.2.7 not 1.7.0. Can you please try with versions from our Dockerfile? You'll probably need to downgrade cuda and pytorch :(
The thing is you don't need to install mmdet3d :( You should install `imvoxelnet/mmdet3d` of version 0.8.0 by `pip install --no-cache-dir -e .` And it for sure will not work...
Great! May be you can pull request your fixes to mmdetection3d codebase? I think you need to load these .json files and dump them as .pkl in exactly the same...
Hi @Geralt-of-winterfall , Can you please share the full output log. Otherwise it's hard for me to understand `they didn't work`.
Unfotunately, we didn't fix the random seed while sampling 100 images per scene, so you can try random ones.
As I remember we provide the confidence interval in our paper. The randomness is less then 1%.
Hi @gyhandy , You can just set `n_images` to 1 [here](https://github.com/SamsungLabs/imvoxelnet/blob/master/configs/imvoxelnet/imvoxelnet_scannet_fast.py#L66).
Setting `n_images` to 1 in training [pipeline](https://github.com/SamsungLabs/imvoxelnet/blob/master/configs/imvoxelnet/imvoxelnet_scannet_fast.py#L52) should also be ok.
Yes, you need it.
Just a random one for each train or test iteration.