Huang Hongqian comments

Results 9 comments of


                                            Huang Hongqian

using pytorch1.11, I still can not compile DCNv2 because of #include <ATen/ceil_div.h>

try python3 setup.py build develop without sudo...

using pytorch1.11, I still can not compile DCNv2 because of #include <ATen/ceil_div.h>

Try to use the previous version. i.e. git checkout fa9b2fd740ced2a22e0e7e913c3bf3934fd08098 Please read the README carefully.

using pytorch1.11, I still can not compile DCNv2 because of #include <ATen/ceil_div.h>

What I mean is that the lateset version of DCNv2 only support pytorch 1.11.+. On the other hand, when you use pytorch 1.8.2 you need to use previous version.

Error when compiling: 'TORCH_CHECK_ARG' was not declared in this scope

According to the latest README.md. Try to checkout the last commit. e.g ``` git checkout fa9b2fd740ced2a22e0e7e913c3bf3934fd08098

Linking Problems While Building

Do you download and set the libtorch library refering to the proposed README?

Linking Problems While Building

Thanks for your feedback. I agree with you. I am afraid that a deployment on the Jetson Platform is beyond my experience. Best wishes for your solution.

how to export onnx from lidar-bonnetal?

https://github.com/Natsu-Akatsuki/RangeNetTrt8?tab=readme-ov-file#install 参考步骤 2

how to export onnx from lidar-bonnetal?

1. ![image](https://github.com/user-attachments/assets/40405eff-8317-4d6b-86ee-60316491f702) 2. 我这边直接用的是原仓库（https://github.com/PRBonn/lidar-bonnetal ）提供的 onnx 文件，他怎么从 pytorch 模型转换为 onnx 模型不在当前仓库讨论范围

生成trt时报错

只适用直接使用原仓库的 onnx 模型。如果您这边 onnx 模型跟原仓库 onnx 模型的网络结构有差别，则可能需要修改当前仓库的网络层的相关代码。