GitLab

* Reflect the changes of the default behavior of grid_sample and affine_grid. If PyTorch > 1.3.0 align_corners will be set to False by default. This is undesired behavior.

* Support grid_sample and affine_grid for PyTorch<1.3 as well.
Co-authored-by: Dusan Malic <dusan.malic@icg.tugraz.at>

Re-organize fix random seed code; Make saved checkpoint compatible among different torch versions (#986)

* fix TypeError in ROI subsampling function

* add comment

* bugfixed: stuck when training with dist_train.sh, support tcp_port

* bugfixed: solve the issue of blocking with dist_train.sh, automatically search tcp ports

* Fix missing __init__.py

* Add further missing __init__.py

* add cbgs_dyn_pp_centerpoint.yaml

* add cbgs_voxel01_res3d_centerpoint.yaml

* add support of centerpoint on Nuscenes in README

* add translation augmentation, re-implement the performance of CenterPoints with 0.075m voxels

* cover the original world translation augmentation

* modify random_translation_along_X

* fix a small bug in random_world_translation()

* add cbgs_dyn_pp_centerpoint.yaml

* add cbgs_voxel01_res3d_centerpoint.yaml

* add support of centerpoint on Nuscenes in README

* add translation augmentation, re-implement the performance of CenterPoints with 0.075m voxels

* cover the original world translation augmentation

* modify random_translation_along_X

* add dynamic pillar vfe

* make the upperbound unaccessible

* add place holder for voxel generation

* add DynPillarVFE

* add PFNLayerV2

* add try except for torch_scatter package

* add dynamic pillar in readme

* add the cfg file of centerpoint with dynamic pillar

* recover original mask_points_by_range

* masking points with points_coords out of grid_size

* add dynamic voxelization and config for voxel_rcnn

* add multi_scale_3d_strides in resbackbone

* add dynamic pillar vfe

* make the upperbound unaccessible

* add place holder for voxel generation

* add DynPillarVFE

* add PFNLayerV2

* add try except for torch_scatter package

* add dynamic pillar in readme

* add the cfg file of centerpoint with dynamic pillar

* recover original mask_points_by_range

* masking points with points_coords out of grid_size

* add dynamic voxelization and config for voxel_rcnn

* add dynamic pillar vfe

* make the upperbound unaccessible

* add place holder for voxel generation

* add DynPillarVFE

* add PFNLayerV2

* add try except for torch_scatter package

* add dynamic pillar in readme

* add the cfg file of centerpoint with dynamic pillar

* recover original mask_points_by_range

* masking points with points_coords out of grid_size

* add dynamic pillar vfe

* make the upperbound unaccessible

* add place holder for voxel generation

* add DynPillarVFE

* add PFNLayerV2

* add try except for torch_scatter package

* add dynamic pillar in readme

* add the cfg file of centerpoint with dynamic pillar

* add pyramid augmentation

* add pyramid augmentation

* format the modified file
Co-authored-by: Shaoshuai Shi <shaoshuaics@gmail.com>

* Fixes #718

* FIX: replaced wrong part

* #719 Fixes index error max_dt while inference: No gt_db_sampling -> MAX_SWEEPS-1 > # unique dt (e.g. first frame)

* move max_sweeps filter to PointFeatureEncoder with new config FILTER_SWEEPS

* update to lowercase
Co-authored-by: Christian Fruhwirth-Reisinger <christian.reisinger@student.tugraz.at>
Co-authored-by: Shaoshuai Shi <shaoshuaics@gmail.com>

update back to original version of torch.div and torch.meshgrid to also support lower version pytorch

* Fixes #718

* FIX: replaced wrong part
Co-authored-by: Christian Fruhwirth-Reisinger <christian.reisinger@student.tugraz.at>

Fix https://github.com/open-mmlab/OpenPCDet/issues/702

fixed a training bug

* support PointPillar (anchor-based, anchor-free) for WOD datasets

* update OpenPCDet to v0.5.1