为啥导入 cugraph 在 kaggle 中不起作用?
Posted
技术标签:
【中文标题】为啥导入 cugraph 在 kaggle 中不起作用?【英文标题】:why import cugraph not working in kaggle?为什么导入 cugraph 在 kaggle 中不起作用? 【发布时间】:2022-01-05 01:45:19 【问题描述】:我需要在 kaggle 中使用 cugraph,因为我使用以下代码安装了 rapids:
import sys
!cp ../input/rapids/rapids.21.06 /opt/conda/envs/rapids.tar.gz
!cd /opt/conda/envs/ && tar -xzvf rapids.tar.gz > /dev/null
sys.path = ["/opt/conda/envs/rapids/lib/python3.7/site-packages"] + sys.path
sys.path = ["/opt/conda/envs/rapids/lib/python3.7"] + sys.path
sys.path = ["/opt/conda/envs/rapids/lib"] + sys.path
!cp /opt/conda/envs/rapids/lib/libxgboost.so /opt/conda/lib/
那么,
conda install -c nvidia -c rapidsai -c numba -c conda-forge -c defaults cugraph cudatoolkit=10.2
但没有安装cugraph,通常我有这个错误:
ModuleNotFoundError Traceback (most recent call last) /tmp/ipykernel_241/370751810.py in <module>
----> 1 import cugraph
ModuleNotFoundError: No module named 'cugraph'
当我尝试在安装中使用 !pip 时,我收到此错误:
Collecting cugraph
Using cached cugraph-0.6.1.post1.tar.gz (1.1 kB)
Building wheels for collected packages: cugraph
Building wheel for cugraph (setup.py) ... error
ERROR: Command errored out with exit status 1:
command: /opt/conda/bin/python3.7 -u -c 'import io, os, sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-vayp36a0/cugraph_6683f309da3e4f3da55486b6a34ad47d/setup.py'"'"'; __file__='"'"'/tmp/pip-install-vayp36a0/cugraph_6683f309da3e4f3da55486b6a34ad47d/setup.py'"'"';f = getattr(tokenize, '"'"'open'"'"', open)(__file__) if os.path.exists(__file__) else io.StringIO('"'"'from setuptools import setup; setup()'"'"');code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' bdist_wheel -d /tmp/pip-wheel-nar691rc
cwd: /tmp/pip-install-vayp36a0/cugraph_6683f309da3e4f3da55486b6a34ad47d/
Complete output (25 lines):
running bdist_wheel
running build
installing to build/bdist.linux-x86_64/wheel
running install
Traceback (most recent call last):
File "<string>", line 1, in <module>
File "/tmp/pip-install-vayp36a0/cugraph_6683f309da3e4f3da55486b6a34ad47d/setup.py", line 35, in <module>
cmdclass='install': InstallWrapper,
File "/opt/conda/lib/python3.7/site-packages/setuptools/__init__.py", line 153, in setup
return distutils.core.setup(**attrs)
File "/opt/conda/lib/python3.7/distutils/core.py", line 148, in setup
dist.run_commands()
File "/opt/conda/lib/python3.7/distutils/dist.py", line 966, in run_commands
self.run_command(cmd)
File "/opt/conda/lib/python3.7/distutils/dist.py", line 985, in run_command
cmd_obj.run()
File "/opt/conda/lib/python3.7/site-packages/wheel/bdist_wheel.py", line 335, in run
self.run_command('install')
File "/opt/conda/lib/python3.7/distutils/cmd.py", line 313, in run_command
self.distribution.run_command(command)
File "/opt/conda/lib/python3.7/distutils/dist.py", line 985, in run_command
cmd_obj.run()
File "/tmp/pip-install-vayp36a0/cugraph_6683f309da3e4f3da55486b6a34ad47d/setup.py", line 15, in run
raise Exception(long_description)
Exception: Please install cugraph via the rapidsai conda channel. See https://rapids.ai/start.html for instructions.
----------------------------------------
ERROR: Failed building wheel for cugraph
【问题讨论】:
为什么不!pip install ...
?
不工作!pip 安装
你有什么错误吗?始终将有问题的完整错误消息(从“Traceback”一词开始)(不在 cmets 中)作为文本(不是屏幕截图,不链接到外部门户)。还有其他有用的信息。
@furas,我编辑帖子并在尝试使用 !pip install cugraph 时输入错误
你能确认你的jupyter内核实际上是在使用环境的python吗?在单元格中运行print(sys.executable())
并检查它是否指向/opt/conda/envs/rapids
【参考方案1】:
为了将来的读者参考,完整的答案在这里详述:
https://github.com/rapidsai/cugraph/issues/1972#issuecomment-982925312
用户需要指定 RPAIDS 版本、python 版本和 CUDA 工具包版本。在撰写本文时,Kaggle 的工作 conda install 命令是:
!conda install -y -c rapidsai -c nvidia -c conda-forge cugraph=21.08 python=3.7 cudatoolkit=11.0
【讨论】:
以上是关于为啥导入 cugraph 在 kaggle 中不起作用?的主要内容,如果未能解决你的问题,请参考以下文章
为啥 IOCP 在 BeginExecuteReader 中不起作用
为啥 setVisibility() 在 onResumeFragments() 中不起作用?