本家様 https://github.com/bionanopatterning/Ais

Fast and user-friendly annotation and segmentation of cryo-electron tomography data using convolutional neural networks
 
(deepL先生訳)
畳み込みニューラルネットワークを用いたクライオ電子断層撮影データの高速でユーザーフレンドリーなアノテーションとセグメンテーション

実行環境確認.

[root@rockylinux ~]# cat /etc/redhat-release
Rocky Linux release 8.8 (Green Obsidian)
 
[root@rockylinux ~]# cat /proc/driver/nvidia/version
NVRM version: NVIDIA UNIX x86_64 Kernel Module  535.161.07  Sat Feb 17 22:55:48 UTC 2024
GCC version:  gcc version 8.5.0 20210514 (Red Hat 8.5.0-18) (GCC)
 
[root@rockylinux ~]# ls -l /usr/local/cuda
ls: cannot access '/usr/local/cuda': No such file or directory
[root@rockylinux ~]#

Aisはpythonアプリです.
ここでは各自のホームディレクトリに各種pythonアプリの実行環境を用意するのではなく、rootで一元管理させるために「pyenv/conda」を使ってます.
pyenvでOS提供のpython,anaconda環境から離れ、そのpyenv上にconda環境を独自に敷いてます

git clone https://github.com/yyuu/pyenv.git /apps/pyenv
export PYENV_ROOT=/apps/pyenv
export PATH=$PYENV_ROOT/bin:$PATH
 
pyenv install --list
 
pyenv install anaconda3-2024.02-1
source /apps/pyenv/versions/anaconda3-2024.02-1/etc/profile.d/conda.sh
(あるいは)
pyenv install miniforge3-23.11.0-0
source /apps/pyenv/versions/miniforge3-23.11.0-0/etc/profile.d/conda.sh

もし既に環境があればそれを利用します

source /apps/pyenv/versions/anaconda3-2024.02-1/etc/profile.d/conda.sh
(あるいは)
source /apps/pyenv/versions/miniforge3-23.11.0-0/etc/profile.d/conda.sh

そうした上でアプリ構築.
本家様の記載にあるように

conda create --name ais
conda activate ais
conda install python==3.9
conda install pip
pip install ais-cryoet

で作れなくはないが、GPUが機能してくれないです.
別途 tensorflow-gpu 、nvidia-cudnn-cu11など入れる必要がありました.

なので下記で作ってみました.

別の方法で

[root@rockylinux ~]# conda env remove --name ais
 
[root@rockylinux ~]# conda create -n ais python=3.9 pip cudatoolkit=11  tensorflow=2.8=cuda* cudnn=8 -c conda-forge
 
[root@rockylinux ~]# conda activate ais
(ais) [root@rockylinux ~]# pip install ais-cryoet
 
(ais) [root@rockylinux ~]# python
Python 3.9.15 | packaged by conda-forge | (main, Nov 22 2022, 08:45:29)
[GCC 10.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>>
>>>
>>> import tensorflow as tf
>>>
>>> tf.test.is_gpu_available()
WARNING:tensorflow:From <stdin>:1: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.config.list_physical_devices('GPU')` instead.
2024-04-16 20:28:53.785728: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  SSE4.1 SSE4.2 AVX AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2024-04-16 20:28:53.796433: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2024-04-16 20:28:53.809461: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2024-04-16 20:28:53.809604: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2024-04-16 20:28:53.890853: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2024-04-16 20:28:53.890997: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2024-04-16 20:28:53.891095: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:936] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2024-04-16 20:28:53.891177: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1525] Created device /device:GPU:0 with 4286 MB memory:  -> device: 0, name: NVIDIA RTX A2000, pci bus id: 0000:13:00.0, compute capability: 8.6
True
>>>
(ais) [root@rockylinux ~]#

一応GPUは認識できたみたい.

environment-modules

[root@rockylinux ~]# vi /apps/modulefiles/ais
#%Module
set          root /apps/pyenv/versions/miniforge3-23.11.0-0/envs/ais
prepend-path PATH $root/bin
 
[root@rockylinux ~]#

使ってみる

[saber@rockylinux ~]$ module use /apps/modulefiles/
[saber@rockylinux ~]$
[saber@rockylinux ~]$ module load ais
 
[saber@rockylinux ~]$ ais

っでこんな画面が表示される
2024y04m16d_203411429.png

centos7では?

[root@centos7 ~]# cat /etc/redhat-release
CentOS Linux release 7.9.2009 (Core)
 
[root@centos7 ~]# cat /proc/driver/nvidia/version
NVRM version: NVIDIA UNIX x86_64 Kernel Module  535.154.05  Thu Dec 28 15:37:48 UTC 2023
GCC version:  gcc version 4.8.5 20150623 (Red Hat 4.8.5-44) (GCC)
 
[root@centos7 ~]# ls -l /usr/local/cuda
ls: cannot access /usr/local/cuda: No such file or directory
 
[root@centos7 ~]#

「pyenv-ana/conda」環境を作る

git clone https://github.com/yyuu/pyenv.git /apps/pyenv
export PYENV_ROOT=/apps/pyenv
export PATH=$PYENV_ROOT/bin:$PATH
pyenv install anaconda3-2024.02-1
source /apps/pyenv/versions/anaconda3-2024.02-1/etc/profile.d/conda.sh
 
(miniforge3でも構わないです)

っで構築

[root@centos7 ~]# conda env remove --name ais
[root@centos7 ~]# conda create -n ais python=3.9 pip cudatoolkit=11  tensorflow=2.8=cuda* cudnn=8 -c conda-forge
[root@centos7 ~]# conda activate ais
(ais) [root@centos7 ~]# pip install ais-cryoet

テスト

(ais) [root@centos7 ~]# python
 
>>> import tensorflow as tf
>>> tf.test.is_gpu_available()
True
>>>

environment-modules

[root@centos7 ~]# vi /apps/modulefiles/ais
#%Module
set          root /apps/pyenv/versions/anaconda3-2024.02-1/envs/ais
prepend-path PATH $root/bin
 
[root@centos7 ~]#

テスト

[illya@centos7 ~]$ module use /apps/modulefiles/
[illya@centos7 ~]$ module load ais
[illya@centos7 ~]$ ais
Traceback (most recent call last):
  File "/apps/pyenv/versions/anaconda3-2024.02-1/envs/ais/bin/ais", line 5, in <module>
    from Ais.main import main
  File "/apps/pyenv/versions/anaconda3-2024.02-1/envs/ais/lib/python3.9/site-packages/Ais/main.py", line 4, in <module>
    from Ais.core.window import *
  File "/apps/pyenv/versions/anaconda3-2024.02-1/envs/ais/lib/python3.9/site-packages/Ais/core/window.py", line 3, in <module>
    from OpenGL.GL import *
  File "/apps/pyenv/versions/anaconda3-2024.02-1/envs/ais/lib/python3.9/site-packages/OpenGL/GL/__init__.py", line 4, in <module>
    from OpenGL.GL.VERSION.GL_1_1 import *
  File "/apps/pyenv/versions/anaconda3-2024.02-1/envs/ais/lib/python3.9/site-packages/OpenGL/GL/VERSION/GL_1_1.py", line 14, in <module>
    from OpenGL.raw.GL.VERSION.GL_1_1 import *
  File "/apps/pyenv/versions/anaconda3-2024.02-1/envs/ais/lib/python3.9/site-packages/OpenGL/raw/GL/VERSION/GL_1_1.py", line 7, in <module>
    from OpenGL.raw.GL import _errors
  File "/apps/pyenv/versions/anaconda3-2024.02-1/envs/ais/lib/python3.9/site-packages/OpenGL/raw/GL/_errors.py", line 4, in <module>
    _error_checker = _ErrorChecker( _p, _p.GL.glGetError )
AttributeError: 'NoneType' object has no attribute 'glGetError'
[illya@centos7 ~]$

aisの中身を確認

[root@centos7 ~]# source /apps/pyenv/versions/anaconda3-2024.02-1/etc/profile.d/conda.sh
[root@centos7 ~]# conda activate ais
(ais) [root@centos7 ~]# conda list
 :
pyopengl                  3.1.7                    pypi_0    pypi
 :
(ais) [root@centos7 ~]#

っで

(ais) [root@centos7 ~]# conda search pyopengl
Loading channels: done
# Name                       Version           Build  Channel
pyopengl                     3.1.1a1          py27_0  pkgs/main
pyopengl                     3.1.1a1 py310h06a4308_0  pkgs/main
pyopengl                     3.1.1a1 py311h06a4308_0  pkgs/main
pyopengl                     3.1.1a1 py312h06a4308_0  pkgs/main
pyopengl                     3.1.1a1          py36_0  pkgs/main
pyopengl                     3.1.1a1  py36h06a4308_0  pkgs/main
pyopengl                     3.1.1a1          py37_0  pkgs/main
pyopengl                     3.1.1a1  py37h06a4308_0  pkgs/main
pyopengl                     3.1.1a1          py38_0  pkgs/main
pyopengl                     3.1.1a1  py38h06a4308_0  pkgs/main
pyopengl                     3.1.1a1  py39h06a4308_0  pkgs/main
(ais) [root@centos7 ~]#
(ais) [root@centos7 ~]# conda install pyopengl -c anaconda
 
(pipなら「pip install PyOpenGL==3.1.6」でも有効みたい)

これで pyopengl は 3.1.1a1 になると思うのだが、、表示的には変わらなかった

(ais) [root@centos7 ~]# conda list
 :
pyopengl                  3.1.7                    pypi_0    pypi
 :
(ais) [root@centos7 ~]#

けどこれで起動しました.
2024y04m25d_235309326.png

最新の60件
2024-05-21 2024-05-16 2024-04-26
  • Ais
2024-04-19 2024-04-18 2024-04-15 2024-04-11 2024-04-06 2024-04-05 2024-03-30 2024-03-29 2024-03-19 2024-03-17 2024-03-15 2024-03-06 2024-03-05 2024-03-03 2024-02-23 2024-02-22 2024-02-21 2024-02-20 2024-02-19 2024-02-18 2024-02-17 2024-02-13 2024-02-12 2024-02-09 2024-02-04 2024-02-03 2024-01-31 2024-01-30 2024-01-29 2024-01-28 2024-01-22 2024-01-16 2024-01-13 2024-01-09 2024-01-08 2024-01-07 2024-01-03 2023-12-22 2023-12-17

edit


トップ   編集 差分 履歴 添付 複製 名前変更 リロード   新規 一覧 検索 最終更新   ヘルプ   最終更新のRSS
Last-modified: 2024-04-26 (金) 00:06:02