second.pytorch
second.pytorch copied to clipboard
traveller59
SECOND for KITTI/NuScenes object detection
Hi, I am running create_data.py for trainval database for training and I got this error. Anyone know how to solve it? ====== Loading NuScenes tables for version v1.0-trainval... 23 category,...
Hi, I am running the `create_data.py` script to create the trainval database for training. I get the following error: `ValueError: cannot reshape array of size 39424 into shape (5)` for...
Hello,everyone! I successfully train my model and evaluate KITTI results. And when I began to use kittiviewer to visualize my results in bev/rgb/3dpointcloud area, I came across a problem that...
/local/lib/python3.8/dist-packages/numba/core/object_mode_passes.py:177: NumbaWarning: Function "points_in_convex_polygon_jit" was compiled in object mode without forceobj=True, but has lifted loops. File "core/geometry.py", line 283: def points_in_convex_polygon_jit(points, polygon, clockwise=True): # first convert polygon to directed lines...
I am using Anaconda base environment for second.pytorch of ubuntu 16.04 after executing the following: nvidia-docker run -it --rm -v /media/yy/960evo/datasets/:/root/data -v $HOME/pretrained_models:/root/model --ipc=host second-pytorch:latest the next thing I excute...
I run the creation of the training data for Nuscene, and it displays that it need 48 hour to process this training data. I don't know other people need how...
numba should be version 0.40.0 `conda install numba==0.40.0` and set the variable in ~/.bashrc export NUMBAPRO_CUDA_DRIVER=/usr/lib/x86_64-linux-gnu/libcuda.so export NUMBAPRO_NVVM=/usr/local/cuda/nvvm/lib64/libnvvm.so export NUMBAPRO_LIBDEVICE=/usr/local/cuda/nvvm/libdevice
@dashidhy @traveller59 how can we set voxelization values for multiclass objects on custom dataset, i have a sparser point cloud in my data Thanks in advance
After ran the whole process of this project, including prep dataset, train, eval, and kitti viewer on my own model, i tried to import some point cloud data to the...
