jwyang / faster-rcnn.pytorch

A faster pytorch implementation of faster r-cnn

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

failure of Compilation with command "python setup.py build develop"

ThomaswellY opened this issue · comments

Dear Jian Wei
I've tried to build this project in conda envs with python3.6 and pytorch1.0.0 . But encountered failure when compiling the cuda dependencies using commands "python setup.py build develop" in lib folder.

  • Envs:

      OS: Manjaro Linux 21.2.6
      GCC version: gcc 12.1.0
      Gpu version: GeForce RTX 2050
      Pytorch version: pytorch 1.0.0 (py3.6_cuda9.0.176_cudnn7.4.1_1)
    
  • Configuration path:

      PATH=$PATH:/opt/cuda/bin:~/anaconda3/bin
      LD_LIBRARY_PATH=/opt/cuda/lib64:$LD_LIBRARY_PATH
      CUDA_HOME=/opt/cuda
    
  • Traceback:

required from here
/home/yang_hao/anaconda3/envs/faster-rcnn/lib/python3.6/site-packages/torch/lib/include/c10/util/ArrayRef.h:96:93: 警告:initializing ‘c10::ArrayRef::Data’ from ‘std::initializer_list::begin’ does not extend the lifetime of the underlying array [-Winit-list-lifetime]
error: command '/opt/cuda/bin/nvcc' failed with exit status 1

Same error!