mit-han-lab / bevfusion

[ICRA'23] BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation
https://bevfusion.mit.edu
Apache License 2.0
2.26k stars 409 forks source link

'dict' object has no attribute 'rotate' #491

Closed quantumdotsss closed 4 months ago

quantumdotsss commented 1 year ago

image Hi guys, I would like to know how the radar data is used. image from the figure of the paper, I didn't see the usage of ladar image So I think I can just comment those lines. Am I right. This happens when I was running torchpack dist-run -np 1 python tools/train.py configs/nuscenes/seg/fusion-bev256d2-lss.yaml --model.encoders.camera.backbone.init_cfg.checkpoint pretrained/swint-nuimages-pretrained.pth

Charlie0804 commented 6 months ago

I'm also having the same issue, would like to ask if you solved it

zhijian-liu commented 4 months ago

Feel free to comment out the code related to radar if you are only working on lidar-only, camera-only, or camera-lidar fusion models.