save-me-ctrl-c
commented
1 week ago 老哥解决了吗，我也是报这个错，有的人说是要在gcc5下面重新下cuda，但是我改了之后还是不行，网上也没有可以参考的
Open QilongWu2 opened 3 months ago
save-me-ctrl-c
commented
1 week ago 老哥解决了吗，我也是报这个错，有的人说是要在gcc5下面重新下cuda，但是我改了之后还是不行，网上也没有可以参考的
Prerequisite
Task
I'm using the official example scripts/configs for the officially supported tasks/models/datasets.
Branch
main branch https://github.com/open-mmlab/mmdetection3d
Environment
/mmdetection3d$ python3 mmdet3d/utils/collect_env.py No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda-11.3' Traceback (most recent call last): File "mmdet3d/utils/collect_env.py", line 5, in
import mmdet3d
ModuleNotFoundError: No module named 'mmdet3d'
由于还没编译，所以collect_env.py运行不了。另外这个是我的安装步骤：
conda create -n env1 python=3.8 -y
conda activate env1
conda install pytorch==1.9.0 torchvision==0.10.0 cudatoolkit=11.3 -c pytorch -c conda-forge
pip install mmcv-full==1.3.12 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.9.0/index.html
pip install mmdet==2.14.0
pip install mmsegmentation==0.14.1
pip uninstall numpy -y
pip install numpy==1.19.5
pip install cython==0.29.36
pip install mmpycocotools
pip install pycocotools==2.0.1
git clone https://github.com/open-mmlab/mmdetection3d.git
cd mmdetection3d
git checkout v0.16.0
pip install -v -e .
在最后一步会出错。另外这是我的环境变量： echo $CUDA_HOME /usr/local/cuda-11.3 echo $PATH /usr/local/anaconda3/envs/radiant/bin:/usr/local/cuda-11.3/bin:/opt/ros/melodic/bin:/home/wql/anaconda3/bin:/home/wql/anaconda3/condabin:/home/wql/anaconda3/bin:/usr/local/cuda-11.3/bin:/home/wql/.local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin echo $LD_LIBRARY_PATH /usr/local/cuda-11.3/lib64:/home/wql/dai_ws/devel/lib:/home/wql/SG-SLAM-old/devel/lib:/home/wql/car_root1/devel/lib:/home/wql/catkin_ws/devel/lib:/opt/ros/melodic/lib:/usr/local/cuda-11.3/lib64
Reproduces the problem - code sample
这个是setup.py文件： from setuptools import find_packages, setup
import os import shutil import sys import torch import warnings from os import path as osp from torch.utils.cpp_extension import (BuildExtension, CppExtension, CUDAExtension)
def readme(): with open('README.md', encoding='utf-8') as f: content = f.read() return content
version_file = 'mmdet3d/version.py'
def get_version(): with open(version_file, 'r') as f: exec(compile(f.read(), version_file, 'exec')) import sys
def make_cuda_ext(name, module, sources, sources_cuda=[], extra_args=[], extra_include_path=[]):
def parse_requirements(fname='requirements.txt', with_version=True): """Parse the package dependencies listed in a requirements file but strips specific versioning information.
def add_mim_extention(): """Add extra files that are required to support MIM into the package.
if name == 'main': add_mim_extention() setup( name='mmdet3d', version=get_version(), description=("OpenMMLab's next-generation platform" 'for general 3D object detection.'), long_description=readme(), long_description_content_type='text/markdown', author='OpenMMLab', author_email='zwwdev@gmail.com', keywords='computer vision, 3D object detection', url='https://github.com/open-mmlab/mmdetection3d', packages=find_packages(), include_package_data=True, package_data={'mmdet3d.ops': ['/.so']}, classifiers=[ 'Development Status :: 4 - Beta', 'License :: OSI Approved :: Apache Software License', 'Operating System :: OS Independent', 'Programming Language :: Python :: 3', 'Programming Language :: Python :: 3.6', 'Programming Language :: Python :: 3.7', ], license='Apache License 2.0', setup_requires=parse_requirements('requirements/build.txt'), tests_require=parse_requirements('requirements/tests.txt'), install_requires=parse_requirements('requirements/runtime.txt'), extras_require={ 'all': parse_requirements('requirements.txt'), 'tests': parse_requirements('requirements/tests.txt'), 'build': parse_requirements('requirements/build.txt'), 'optional': parse_requirements('requirements/optional.txt'), }, ext_modules=[ make_cuda_ext( name='sparse_conv_ext', module='mmdet3d.ops.spconv', extra_include_path=[
PyTorch 1.5 uses ninjia, which requires absolute path
Reproduces the problem - command or script
conda create -n env1 python=3.8 -y
conda activate env1
conda install pytorch==1.9.0 torchvision==0.10.0 cudatoolkit=11.3 -c pytorch -c conda-forge
pip install mmcv-full==1.3.12 -f https://download.openmmlab.com/mmcv/dist/cu111/torch1.9.0/index.html
pip install mmdet==2.14.0
pip install mmsegmentation==0.14.1
pip uninstall numpy -y
pip install numpy==1.19.5
pip install cython==0.29.36
pip install mmpycocotools
pip install pycocotools==2.0.1
git clone https://github.com/open-mmlab/mmdetection3d.git
cd mmdetection3d
git checkout v0.16.0
pip install -v -e .
主要是上面这些命令
Reproduces the problem - error message
这个是报错片段： Running setup.py develop for mmdet3d Running command python setup.py develop No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda-11.3' Compiling sparse_conv_ext without CUDA Compiling iou3d_cuda without CUDA Compiling voxel_layer without CUDA Compiling roiaware_pool3d_ext without CUDA Compiling ball_query_ext without CUDA Compiling knn_ext without CUDA Compiling assign_score_withk_ext without CUDA Compiling group_points_ext without CUDA Compiling interpolate_ext without CUDA Compiling furthest_point_sample_ext without CUDA Compiling gather_points_ext without CUDA running develop running egg_info writing mmdet3d.egg-info/PKG-INFO writing dependency_links to mmdet3d.egg-info/dependency_links.txt writing requirements to mmdet3d.egg-info/requires.txt writing top-level names to mmdet3d.egg-info/top_level.txt reading manifest file 'mmdet3d.egg-info/SOURCES.txt' reading manifest template 'MANIFEST.in' /usr/local/anaconda3/envs/radiant/lib/python3.8/site-packages/setuptools/command/develop.py:40: EasyInstallDeprecationWarning: easy_install command is deprecated. !!
error: subprocess-exited-with-error
× python setup.py develop did not run successfully. │ exit code: 1 ╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip.
Additional information
我真的不知道该怎么办了，这个问题卡了很久，由于我想复现一个基于mmdet的论文（radiant），它的装的版本就是v0.16.0的