search

NVIDIA-AI-IOT / Lidar_AI_Solution

A project demonstrating Lidar related AI solutions, including three GPU accelerated Lidar/camera DL networks (PointPillars, CenterPoint, BEVFusion) and the related libs (cuPCL, 3D SparseConvolution, YUV2RGB, cuOSD,).
Other
1.26k stars 218 forks source link

export-scn.py Error: #243

Open Bonasdljlkj opened 5 months ago

Bonasdljlkj commented 5 months ago

(mitbevfusion) gss@Gss:~/Lidar_AI_Solution/CUDA-BEVFusion$ python qat/export-scn.py Tracing model inference...

Do inference... --> SparseConvolutionQunat0[subm] -> Input 0, Output 1 Traceback (most recent call last): File "qat/export-scn.py", line 73, in exptool.export_onnx(model, voxels, coors, batch_size, inverse_indices, args.save) File "/home/gss/Lidar_AI_Solution/CUDA-BEVFusion/qat/lean/exptool.py", line 401, in export_onnx y = model(voxels, coors, batch_size) File "/home/gss/anaconda3/envs/mitbevfusion/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl return forward_call(*input, *kwargs) File "/home/gss/Lidar_AI_Solution/CUDA-BEVFusion/qat/lean/exptool.py", line 275, in impl x = self.conv_input(input_sp_tensor) File "/home/gss/anaconda3/envs/mitbevfusion/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl return forward_call(input, *kwargs) File "/home/gss/Lidar_AI_Solution/CUDA-BEVFusion/qat/lean/exptool.py", line 62, in internal_forward bind_fn(self, ilayer, y, args) File "/home/gss/Lidar_AI_Solution/CUDA-BEVFusion/qat/lean/exptool.py", line 179, in symbolic_sparse_convolution_quant input_dynamic_range = self._input_quantizer.amax.cpu().item(), AttributeError: 'NoneType' object has no attribute 'cpu'

Felix-fz commented 1 month ago

I have the same problem. Did you solve it？