Open LandaChina opened 1 year ago
LandaChina
commented
1 year ago Additional Info: Next, conda uninstall pytorch-scatter, successfully, the error 'OSError:[WinError 127]' happened again. Maybe the wrong is from the torch-sparse wheel installed. The torch_sparse is found from https://[pytorch-geometric.com/whl/torch-2.0.1%2Bcpu.html](https://pytorch-geometric.com/whl/torch-2.0.1%2Bcpu.html).
general502570
commented
1 year ago 谢谢您对AutoGL的关注。pytorch-sparse 的依赖应该是ogb数据集所需要的。您可以试一试能不能直接import ogb库。如果不能的话，请在ogb库的讨论区寻找相关解答。如果可以的话，请再回复给我们。
LandaChina
commented
1 year ago 谢谢您对AutoGL的关注。pytorch-sparse 的依赖应该是ogb数据集所需要的。您可以试一试能不能直接import ogb库。如果不能的话，请在ogb库的讨论区寻找相关解答。如果可以的话，请再回复给我们。
成功import ogb！
(autogl) C:\Users\llzhang>python Python 3.8.17 (default, Jul 5 2023, 20:44:21) [MSC v.1916 64 bit (AMD64)] :: Anaconda, Inc. on win32 Type "help", "copyright", "credits" or "license" for more information.
import ogb import autogl Traceback (most recent call last): File "
", line 1, in File "C:\Users\All Users\anaconda3\envs\autogl\lib\site-packages\autogl__init.py", line 1, in from . import ( File "C:\Users\All Users\anaconda3\envs\autogl\lib\site-packages\autogl\datasets__init.py", line 14, in init.py", line 373, in init__ self._handle = _dlopen(self._name, mode) OSError: [WinError 127] 找不到指定的程序。from ._ogb import ( File "C:\Users\All Users\anaconda3\envs\autogl\lib\site-packages\autogl\datasets_ogb.py", line 8, in init__.py", line 18, infrom torch_sparse import SparseTensor File "C:\Users\llzhang\AppData\Roaming\Python\Python38\site-packages\torch_sparse\ torch.ops.load_library(spec.origin) File "C:\Users\All Users\anaconda3\envs\autogl\lib\site-packages\torch_ops.py", line 255, in load_library ctypes.CDLL(path) File "C:\Users\All Users\anaconda3\envs\autogl\lib\ctypes\
但是import autogl还是存在以上问题。
ZW-ZHANG
commented
1 year ago 感谢您的回复，该问题看起来是由于torch_sparse库引起的。您可以先试一下直接import torch_sparse能否成功，如过不能的话，请查看torch_sparse寻找相关原因（比如是否为python/torch/gpu版本与torch_sparse库不匹配等）
LandaChina
commented
1 year ago pytorch_sparse 是pytorch的1.6后版本使用，torch-sparse是1.6之前的版本，同时pytorch_sparse没有Windows版本，conda install找不到，从而得出的结论是autogl对操作系统有要求。
ZW-ZHANG
commented
1 year ago 您好，可能之前没有说清，我说的是https://github.com/rusty1s/pytorch_sparse 中的torch_sparse，它是pytorch geometric库的一个依赖（虽然您用的是DGL backend，但好像因为ogb或者别的库的依赖，也需要安装torch_sparse）。我个人尝试按照上面网页的说明安装torch_sparse后，是可以在windows下成功运行autogl
Describe the bug 安装成功dgl和autogl后，import验证时报错，pytorch-sparse安装后python.exe无法找到入口，卸载pytorch-sparse又会提示不存在模块，问题在torch_sparse循环。
To Reproduce
Expected behavior 成功import autogl，如何更好的安装torch_sparse，使之成功安装并且不报错。
Screenshots If applicable, add screenshots to help explain your problem.
Environment (please complete the following information):
Name Version Build Channel
appdirs 1.4.4 pyhd3eb1b0_0 defaults astor 0.8.1 pypi_0 pypi autogl 0.4.0 pypi_0 pypi bayesian-optimization 1.4.3 pypi_0 pypi blas 1.0 mkl defaults brotlipy 0.7.0 py38h2bbff1b_1003 defaults ca-certificates 2023.05.30 haa95532_0 defaults certifi 2023.7.22 py38haa95532_0 defaults cffi 1.15.1 py38h2bbff1b_3 defaults charset-normalizer 2.0.4 pyhd3eb1b0_0 defaults chocolate 0.0.2 pypi_0 pypi cloudpickle 2.2.1 pypi_0 pypi colorama 0.4.6 py38haa95532_0 defaults contextlib2 21.6.0 pypi_0 pypi contourpy 1.1.0 pypi_0 pypi cryptography 41.0.2 py38hac1b9e3_0 defaults cycler 0.11.0 pypi_0 pypi cython 0.29.14 pypi_0 pypi deeprobust 0.2.8 pypi_0 pypi dgl 1.1.1 py38_0 dglteam dill 0.3.7 pypi_0 pypi dnspython 2.4.2 pypi_0 pypi filelock 3.12.2 pypi_0 pypi fonttools 4.42.0 pypi_0 pypi gensim 3.8.3 pypi_0 pypi hyperopt 0.1.2 pypi_0 pypi icc_rt 2022.1.0 h6049295_2 defaults idna 3.4 py38haa95532_0 defaults imageio 2.31.1 pypi_0 pypi importlib-metadata 6.8.0 pypi_0 pypi importlib-resources 6.0.1 pypi_0 pypi intel-openmp 2023.1.0 h59b6b97_46319 defaults jinja2 3.1.2 py38haa95532_0 defaults joblib 1.3.2 pypi_0 pypi json-tricks 3.17.2 pypi_0 pypi kiwisolver 1.4.4 pypi_0 pypi lazy-loader 0.3 pypi_0 pypi libffi 3.4.4 hd77b12b_0 defaults libuv 1.44.2 h2bbff1b_0 defaults lightgbm 4.0.0 pypi_0 pypi littleutils 0.2.2 pypi_0 pypi llvmlite 0.40.1 pypi_0 pypi markupsafe 2.1.3 pypi_0 pypi matplotlib 3.7.2 pypi_0 pypi mkl 2023.1.0 h6b88ed4_46357 defaults mkl-service 2.4.0 py38h2bbff1b_1 defaults mkl_fft 1.3.6 py38hf11a4ad_1 defaults mkl_random 1.2.2 py38hf11a4ad_1 defaults mpmath 1.3.0 py38haa95532_0 defaults netlsd 1.0.2 pypi_0 pypi networkx 3.1 py38haa95532_0 defaults ninja 1.10.2 haa95532_5 defaults ninja-base 1.10.2 h6d14046_5 defaults nni 2.8 pypi_0 pypi numba 0.57.1 pypi_0 pypi numpy 1.24.3 py38h79a8e48_1 defaults numpy-base 1.24.3 py38h8a87ada_1 defaults ogb 1.3.6 pypi_0 pypi openssl 3.0.10 h2bbff1b_0 defaults outdated 0.2.2 pypi_0 pypi packaging 23.0 py38haa95532_0 defaults pandas 2.0.3 pypi_0 pypi pillow 10.0.0 pypi_0 pypi pip 23.2.1 py38haa95532_0 defaults pooch 1.4.0 pyhd3eb1b0_0 defaults prettytable 3.8.0 pypi_0 pypi protobuf 4.24.0 pypi_0 pypi psutil 5.9.0 py38h2bbff1b_0 defaults pycparser 2.21 pyhd3eb1b0_0 defaults pymongo 4.4.1 pypi_0 pypi pyopenssl 23.2.0 py38haa95532_0 defaults pyparsing 3.0.9 pypi_0 pypi pysocks 1.7.1 py38haa95532_0 defaults python 3.8.17 h1aa4202_0 defaults python-dateutil 2.8.2 pypi_0 pypi pythonwebhdfs 0.2.3 pypi_0 pypi pytorch 2.0.1 cpu_py38hb0bdfb8_0 defaults pytz 2023.3 pypi_0 pypi pywavelets 1.4.1 pypi_0 pypi pyyaml 6.0.1 pypi_0 pypi requests 2.31.0 py38haa95532_0 defaults responses 0.23.3 pypi_0 pypi schema 0.7.5 pypi_0 pypi scikit-image 0.21.0 pypi_0 pypi scikit-learn 1.3.0 pypi_0 pypi scipy 1.10.1 py38hdcfc7df_1 defaults setuptools 68.0.0 py38haa95532_0 defaults simplejson 3.19.1 pypi_0 pypi smart-open 6.3.0 pypi_0 pypi sqlite 3.41.2 h2bbff1b_0 defaults sympy 1.12 pypi_0 pypi tabulate 0.9.0 pypi_0 pypi tbb 2021.8.0 h59b6b97_0 defaults tensorboardx 2.6.2 pypi_0 pypi texttable 1.6.7 pypi_0 pypi threadpoolctl 3.2.0 pypi_0 pypi tifffile 2023.7.10 pypi_0 pypi torch-sparse 0.6.17+pt20cpu pypi_0 pypi torchvision 0.15.2 pypi_0 pypi tqdm 4.65.0 py38hd4e2768_0 defaults typeguard 4.1.0 pypi_0 pypi types-pyyaml 6.0.12.11 pypi_0 pypi typing-extensions 4.7.1 py38haa95532_0 defaults typing_extensions 4.7.1 py38haa95532_0 defaults tzdata 2023.3 pypi_0 pypi urllib3 1.26.16 py38haa95532_0 defaults vc 14.2 h21ff451_1 defaults vs2015_runtime 14.27.29016 h5e58377_2 defaults wcwidth 0.2.6 pypi_0 pypi websockets 11.0.3 pypi_0 pypi wheel 0.38.4 py38haa95532_0 defaults win_inet_pton 1.1.0 py38haa95532_0 defaults zipp 3.16.2 pypi_0 pypi ]
Additional Info (Optional)
提示的定位不到的动态链接库_version_cpu，本地可以找到几个，替换报错位置的文档，也没有解决。