NVIDIA-AI-IOT / CUDA-PointPillars

A project demonstrating how to use CUDA-PointPillars to deal with cloud points data from lidar.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

What should I do if I want more labels?

brillint opened this issue · comments

commented

when I have 8 labels,it returns:
onnx.onnx_cpp2py_export.shape_inference.InferenceError: [ShapeInferenceError] (op_type:Transpose, node name: Transpose_309): [ShapeInferenceError] Inferred shape and existing shape differ in dimension 3: (128) vs (32)

commented

You need to modify the shape of cls_preds , box_preds , dir_cls_preds according to the your grid size and number of categories, in the simplify_postprocess() func.