PaddlePaddle / Paddle3D

A 3D computer vision development toolkit based on PaddlePaddle. It supports point-cloud object detection, segmentation, and monocular 3D object detection models.
Apache License 2.0
567 stars 141 forks source link

执行命令:“python ./deploy/smoke/python/infer.py --model_file ./output_kitti_inference/centerpoint.pdmodel --params_file ./output_kitti_inference/centerpoint.pdiparams --image ./datasets/KITTI/training/images_2/000425.png > ./output_kitti/pred.txt ”进行推理,报错:“I0603 07:01:28.572183 1454 analysis_predictor.cc:1314] ======= optimize end ======= I0603 07:01:28.572906 1454 naive_executor.cc:110] --- skip [feed], feed -> data Traceback (most recent call last): File "./deploy/smoke/python/infer.py", line 141, in <module> pred = init_predictor(args) File "./deploy/smoke/python/infer.py", line 106, in init_predictor predictor = create_predictor(config) RuntimeError: (NotFound) Operator (hard_voxelize) is not registered. [Hint: op_info_ptr should not be null.] (at /paddle/paddle/fluid/framework/op_info.h:156) ” #468

Open domainsense opened 4 months ago