Skip to content

Latest commit

 

History

History
40 lines (22 loc) · 1.69 KB

model_zoo.md

File metadata and controls

40 lines (22 loc) · 1.69 KB
Raw

Model Zoo

Common settings

  • We use distributed training.
  • For fair comparison with other codebases, we report the GPU memory as the maximum value of torch.cuda.max_memory_allocated() for all 8 GPUs. Note that this value is usually less than what nvidia-smi shows.
  • We report the inference time as the total time of network forwarding and post-processing, excluding the data loading time. Results are obtained with the script benchmark.py which computes the average time on 2000 images.

Baselines

SECOND

Please refer to SECOND for details.

PointPillars

Please refer to PointPillars for details.

Part-A2

Please refer to Part-A2 for details.

VoteNet

Please refer to VoteNet for details.

Dynamic Voxelization

Please refer to Dynamic Voxelization for details.

MVXNet

Please refer to MVXNet for details.

RegNetX

Please refer to RegNet for details.

nuImages

We also support baseline models on nuImages dataset. Please refer to nuImages for details.