RuntimeError: points must be a contiguous tensor

[Hitjss]

(sam6d) js@js-UbtSer:~/Documents/SAM-6D/SAM-6D-main/SAM-6D/Pose_Estimation_Model$ python run_inference_custom.py --output_dir $OUTPUT_DIR --cad_path $CAD_PATH --rgb_path $RGB_PATH --depth_path $DEPTH_PATH --cam_path $CAMERA_PATH --seg_path $SEG_PATH set CUDA_VISIBLE_DEVICES as 0 => creating model ... load pre-trained checkpoint from: checkpoints/mae_pretrain_vit_base.pth => extracting templates ... Traceback (most recent call last): File "/home/js/Documents/SAM-6D/SAM-6D-main/SAM-6D/Pose_Estimation_Model/run_inference_custom.py", line 277, in all_tem_pts, all_tem_feat = model.feature_extraction.get_obj_feats(all_tem, all_tem_pts, all_tem_choose) File "/home/js/Documents/SAM-6D/SAM-6D-main/SAM-6D/Pose_Estimation_Model/../Pose_Estimation_Model/model/feature_extraction.py", line 181, in get_obj_feats return sample_pts_feats(tem_pts, tem_feat, npoint) File "/home/js/Documents/SAM-6D/SAM-6D-main/SAM-6D/Pose_Estimation_Model/../Pose_Estimation_Model/utils/model_utils.py", line 58, in sample_pts_feats sample_idx = furthest_point_sample(pts, npoint) File "/home/js/anaconda3/envs/sam6d/lib/python3.9/site-packages/torch/autograd/function.py", line 598, in apply return super().apply(*args, **kwargs) # type: ignore[misc] File "/home/js/Documents/SAM-6D/SAM-6D-main/SAM-6D/Pose_Estimation_Model/model/pointnet2/pointnet2_utils.py", line 71, in forward fps_inds = _ext.furthest_point_sampling(xyz.contiguous(), npoint) RuntimeError: points must be a contiguous tensor