"RuntimeError: CUDA error: device-side assert triggered CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1."

[rockywind]

Hi, thank you for sharing you greak work! I met the error below. I ran the model pretrained weights on the Mapillary Vistas Dataset. This is my infer script.

python demo/demo_my.py \
--config-file /rockywin.wang/NeRF/Mask2Former/configs/mapillary-vistas/panoptic-segmentation/swin/maskformer2_swin_large_IN21k_384_bs16_300k.yaml \
--input /rockywin.wang/NeRF/RoMe/RoMe_data/nuscenes/samples/CAM_FRONT/ \
--output /rockywin.wang/NeRF/suds/data/kitti/20230201-15-13-45_obj_nerf_norm_0_100/seg \ 
--predictions /rockywin.wang/NeRF/suds/data/kitti/20230201-15-13-45_obj_nerf_norm_0_100/seg \
--opts MODEL.WEIGHTS /SHFP12/02_bevdet/SceneRF/01_thrd_data/model_final_132c71_map_pano.pkl

[RYHSmmc]

same error