Ubuntu20安装maskrcnn_benchmark（CUDA11.0.2）详细步骤

哈哈就不吃萝卜

于 2023-11-14 20:37:58 发布

阅读量544

点赞数

文章标签： ubuntu linux

本文链接：https://blog.csdn.net/qq_44750728/article/details/134406853

版权

一、环境说明

Ubuntu环境为20.04：

Linux version 5.15.0-88-generic (buildd@lcy02-amd64-011) (gcc (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0, GNU ld (GNU Binutils for Ubuntu) 2.34) #98~20.04.1-Ubuntu SMP Mon Oct 9 16:43:45 UTC 2023

gcc、g++为5.3.1版本

gcc version 5.3.1 20160413 (Ubuntu 5.3.1-14ubuntu2)
gcc version 5.3.1 20160413 (Ubuntu 5.3.1-14ubuntu2)

CUDA版本为11.0，CUDNN版本为8.9.2

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Thu_Jun_11_22:26:38_PDT_2020
Cuda compilation tools, release 11.0, V11.0.194
Build cuda_11.0_bu.TC445_37.28540450_0

#define CUDNN_MAJOR 8
#define CUDNN_MINOR 9
#define CUDNN_PATCHLEVEL 2
--
#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

显卡为3090

二、安装前的注意事项

官方最低环境推荐（不推荐、环境太老不适合30系显卡）：

PyTorch 1.0 from a nightly release. It will not work with 1.0 nor 1.0.1. Installation instructions can be found in https://pytorch.org/get-started/locally/
torchvision from master
cocoapi
yacs
matplotlib
GCC >= 4.9
OpenCV
CUDA >= 9.0

三、安装步骤

设置正确cuda路径（修改.bashrc，添加如下，保存刷新）

export CUDA_HOME="/usr/local/cuda-11.0"
export CUDA_PATH="$CUDA_HOME"
export PATH="$CUDA_HOME/bin:$PATH"
export LD_LIBRARY_PATH="$CUDA_HOME/lib64:$CUDA_HOME/lib:$CUDA_HOME/extras/CUPTI/lib64:$CUDA_HOME/efa/lib:$LD_LIBRARY_PATH"

创建conda环境

conda create --name maskrcnn_benchmark python==3.7 -y
conda activate maskrcnn_benchmark

为新python安装正确的pip和依赖项

conda install ipython pip

安装依赖（opencv建议安装低版本）：

pip install ninja yacs cython matplotlib tqdm -i https://pypi.tuna.tsinghua.edu.cn/simple
pip install opencv-python==3.4.10.35 -i https://pypi.tuna.tsinghua.edu.cn/simple

安装Pytorch：

conda install pytorch==1.7.0 torchvision==0.8.0 torchaudio==0.7.0 cudatoolkit=11.0 -c pytorch

正式安装（官方步骤）：

export INSTALL_DIR=$PWD

# install pycocotools
cd $INSTALL_DIR
git clone https://github.com/cocodataset/cocoapi.git
cd cocoapi/PythonAPI
python setup.py build_ext install

# install cityscapesScripts
cd $INSTALL_DIR
git clone https://github.com/mcordts/cityscapesScripts.git
cd cityscapesScripts/
python setup.py build_ext install

安装apex：

建议用git克隆完整的apex包，然后再退回老版本的apex

cd $INSTALL_DIR

git clone https://github.com/NVIDIA/apex.git #git克隆，不要用网页下载压缩包
cd apex
git checkout f3a960f80244cf9e80558ab30f7f7e8cbf03c0a0 # 回退到老版本
python setup.py install --cuda_ext --cpp_ext

安装maskrcnn：

cd $INSTALL_DIR
git clone https://github.com/facebookresearch/maskrcnn-benchmark.git
cd maskrcnn-benchmark
# 添加路径
cuda_dir="maskrcnn_benchmark/csrc/cuda"
perl -i -pe 's/AT_CHECK/TORCH_CHECK/' $cuda_dir/deform_pool_cuda.cu $cuda_dir/deform_conv_cuda.cu
# You can then run the regular setup command
# 编译
python setup.py build develop

总体环境及其版本如下:

pip list

Package            Version         Editable project location
------------------ --------------- ----------------------------
apex               0.1
appdirs            1.4.4
backcall           0.2.0
certifi            2022.12.7
charset-normalizer 3.3.2
cityscapesScripts  2.2.2
coloredlogs        15.0.1
cPython            0.0.6
cycler             0.11.0
Cython             3.0.5
decorator          5.1.1
dnspython          2.3.0
flit_core          3.6.0
fonttools          4.38.0
humanfriendly      10.0
idna               3.4
ipython            7.31.1
jedi               0.18.1
kiwisolver         1.4.5
maskrcnn-benchmark 0.1             /home/ztt/maskrcnn_benchmark
matplotlib         3.5.3
matplotlib-inline  0.1.6
mkl-fft            1.3.1
mkl-random         1.2.2
mkl-service        2.4.0
ninja              1.11.1.1
numpy              1.21.6
opencv-python      3.4.10.35
packaging          23.2
parso              0.8.3
pexpect            4.8.0
pickleshare        0.7.5
Pillow             9.4.0
pip                22.3.1
prompt-toolkit     3.0.36
ptyprocess         0.7.0
pycocotools        2.0
Pygments           2.11.2
pymongo            4.6.0
pyparsing          3.1.1
pyquaternion       0.9.9
python-dateutil    2.8.2
PyYAML             6.0.1
requests           2.31.0
setuptools         65.6.3
six                1.16.0
torch              1.7.0
torchaudio         0.7.0a0+ac17b64
torchvision        0.8.0
tqdm               4.66.1
traitlets          5.7.1
typing             3.10.0.0
typing_extensions  4.7.1
urllib3            2.0.7
wcwidth            0.2.5
wheel              0.38.4
yacs               0.1.8

conda list

# packages in environment at /home/ztt/anaconda3/envs/maskrcnn:
#
# Name                    Version                   Build  Channel
_libgcc_mutex             0.1                        main    defaults
_openmp_mutex             5.1                       1_gnu    defaults
apex                      0.1                      pypi_0    pypi
appdirs                   1.4.4                    pypi_0    pypi
backcall                  0.2.0              pyhd3eb1b0_0    defaults
blas                      1.0                         mkl    defaults
ca-certificates           2023.08.22           h06a4308_0    defaults
certifi                   2022.12.7        py37h06a4308_0    defaults
charset-normalizer        3.3.2                    pypi_0    pypi
cityscapesscripts         2.2.2                    pypi_0    pypi
coloredlogs               15.0.1                   pypi_0    pypi
cpython                   0.0.6                    pypi_0    pypi
cudatoolkit               11.0.221             h6bb024c_0    defaults
cycler                    0.11.0                   pypi_0    pypi
cython                    3.0.5                    pypi_0    pypi
decorator                 5.1.1              pyhd3eb1b0_0    defaults
dnspython                 2.3.0                    pypi_0    pypi
flit-core                 3.6.0              pyhd3eb1b0_0    defaults
fonttools                 4.38.0                   pypi_0    pypi
freetype                  2.12.1               h4a9f257_0    defaults
giflib                    5.2.1                h5eee18b_3    defaults
humanfriendly             10.0                     pypi_0    pypi
idna                      3.4                      pypi_0    pypi
intel-openmp              2021.4.0          h06a4308_3561    defaults
ipython                   7.31.1           py37h06a4308_1    defaults
jedi                      0.18.1           py37h06a4308_1    defaults
jpeg                      9e                   h5eee18b_1    defaults
kiwisolver                1.4.5                    pypi_0    pypi
lcms2                     2.12                 h3be6417_0    defaults
ld_impl_linux-64          2.38                 h1181459_1    defaults
lerc                      3.0                  h295c915_0    defaults
libdeflate                1.17                 h5eee18b_1    defaults
libedit                   3.1.20221030         h5eee18b_0    defaults
libffi                    3.2.1             hf484d3e_1007    defaults
libgcc-ng                 11.2.0               h1234567_1    defaults
libgomp                   11.2.0               h1234567_1    defaults
libpng                    1.6.39               h5eee18b_0    defaults
libstdcxx-ng              11.2.0               h1234567_1    defaults
libtiff                   4.5.1                h6a678d5_0    defaults
libuv                     1.44.2               h5eee18b_0    defaults
libwebp                   1.2.4                h11a3e52_1    defaults
libwebp-base              1.2.4                h5eee18b_1    defaults
lz4-c                     1.9.4                h6a678d5_0    defaults
maskrcnn-benchmark        0.1                       dev_0    <develop>
matplotlib                3.5.3                    pypi_0    pypi
matplotlib-inline         0.1.6            py37h06a4308_0    defaults
mkl                       2021.4.0           h06a4308_640    defaults
mkl-service               2.4.0            py37h7f8727e_0    defaults
mkl_fft                   1.3.1            py37hd3c417c_0    defaults
mkl_random                1.2.2            py37h51133e4_0    defaults
ncurses                   6.4                  h6a678d5_0    defaults
ninja                     1.11.1.1                 pypi_0    pypi
ninja-base                1.10.2               hd09550d_5    defaults
numpy                     1.21.6                   pypi_0    pypi
numpy-base                1.21.5           py37ha15fc14_3    defaults
opencv-python             3.4.10.35                pypi_0    pypi
openssl                   1.1.1w               h7f8727e_0    defaults
packaging                 23.2                     pypi_0    pypi
parso                     0.8.3              pyhd3eb1b0_0    defaults
pexpect                   4.8.0              pyhd3eb1b0_3    defaults
pickleshare               0.7.5           pyhd3eb1b0_1003    defaults
pillow                    6.2.0                    pypi_0    pypi
pip                       22.3.1           py37h06a4308_0    defaults
prompt-toolkit            3.0.36           py37h06a4308_0    defaults
ptyprocess                0.7.0              pyhd3eb1b0_2    defaults
pycocotools               2.0                      pypi_0    pypi
pygments                  2.11.2             pyhd3eb1b0_0    defaults
pymongo                   4.6.0                    pypi_0    pypi
pyparsing                 3.1.1                    pypi_0    pypi
pyquaternion              0.9.9                    pypi_0    pypi
python                    3.7.7           hcf32534_0_cpython    defaults
python-dateutil           2.8.2                    pypi_0    pypi
pytorch                   1.7.0           py3.7_cuda11.0.221_cudnn8.0.3_0    pytorch
pyyaml                    6.0.1                    pypi_0    pypi
readline                  8.2                  h5eee18b_0    defaults
requests                  2.31.0                   pypi_0    pypi
setuptools                65.6.3           py37h06a4308_0    defaults
six                       1.16.0             pyhd3eb1b0_1    defaults
sqlite                    3.33.0               h62c20be_0    defaults
tk                        8.6.12               h1ccaba5_0    defaults
torchaudio                0.7.0                      py37    pytorch
torchvision               0.8.0                py37_cu110    pytorch
tqdm                      4.66.1                   pypi_0    pypi
traitlets                 5.7.1            py37h06a4308_0    defaults
typing                    3.10.0.0                 pypi_0    pypi
typing-extensions         4.7.1                    pypi_0    pypi
typing_extensions         4.4.0            py37h06a4308_0    defaults
urllib3                   2.0.7                    pypi_0    pypi
wcwidth                   0.2.5              pyhd3eb1b0_0    defaults
wheel                     0.38.4           py37h06a4308_0    defaults
xz                        5.4.2                h5eee18b_0    defaults
yacs                      0.1.8                    pypi_0    pypi
zlib                      1.2.13               h5eee18b_0    defaults
zstd                      1.5.5                hc292b87_0    defaults