I downloaded and installed CUDA10.0 pyhton3.6 pytorch1.2. , also loaded cuda/cuda-10.0-x86_6 envirnment.
Commands like below
module load cuda/cuda-10.0-x86_64
# CUDA 10.0
conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0 -c pytorch
Because it needs to be installed on a NO-NETWOEK centOs server! Therefore, all pakages need be downloaded in advance and run command like below.
conda install --offline pytorch-1.2.0-cuda100py36h938c94c_0.tar.bz2
......
The error occured below.
Python 3.6.0 |Anaconda 4.3.0 (64-bit)| (default, Dec 23 2016, 12:22:00)
[GCC 4.4.7 20120313 (Red Hat 4.4.7-1)] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/data/users/CHDHPC/2020224009/anaconda3/lib/python3.6/site-packages/torch/__init__.py", line 81, in <module>
from torch._C import *
ImportError: libmkl_gnu_thread.so: cannot open shared object file: No such file or directory
I have no idea to solve it! Anyone can help me! Thank you!!
Finally I fixed this issue by changing Python version to 3.6.2 and re-installing the cuda and cudnn.
Pecfect!
Related
After following the Ubuntu/Debian installation instructions for the Nest simulator I can only import the nest module in python2.x, not python3.x
$ python3
Python 3.6.8 (default, Aug 20 2019, 17:12:48)
[GCC 8.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import nest
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/nest/lib/python2.7/site-packages/nest/__init__.py", line 26, in <module>
from . import ll_api # noqa
File "/nest/lib/python2.7/site-packages/nest/ll_api.py", line 72, in <module>
from . import pynestkernel as kernel # noqa
ImportError: dynamic module does not define module export function (PyInit_pynestkernel)
The default install compiles with the default Python version which is still 2 in Ubuntu.
To use Python 3, run:
cmake -Dwith-python=3 -DCMAKE_INSTALL_PREFIX:PATH=</install/path> </path/to/NEST/src>
Mentioned here in the doc.
NB: don't forget to clear the build folder to avoid issues
I am using python3 and have installed tensorflow-gpu using:
pip3 install tensorflow-gpu==1.13.1
And have checked that it is installed by:
pip3 show tensorflow-gpu
Name: tensorflow-gpu
Version: 1.13.1
Summary: TensorFlow is an open source machine learning framework for
everyone.
Home-page: https://www.tensorflow.org/
Author: Google Inc.
Author-email: opensource#google.com
License: Apache 2.0
Location: /usr/lib64/python3.6/site-packages
Requires: tensorflow-estimator, keras-applications, termcolor, absl-py, six,
astor, protobuf, wheel, keras-preprocessing, gast, grpcio, numpy,
tensorboard
Required-by:
but when I do the following it gives me an error:
user:/home/mydirectory # python3
Python 3.6.5 (default, Mar 31 2018, 19:45:04) [GCC] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/usr/lib/python3.6/site-packages/tensorflow/__init__.py", line 24,
in <module>
from tensorflow.python import pywrap_tensorflow # pylint:
disable=unused-import
File "/usr/lib/python3.6/site-packages/tensorflow/python/pywrap_tensorflow.py", line 25, in <module>
from tensorflow.python.platform import self_check
ModuleNotFoundError: No module named 'tensorflow.python.platform'
I see from the error that its looking for tensorflow in directory
/usr/lib/python3.6/site-packages/tensorflow
But pip indicates that its installed at
/usr/lib64/python3.6/site-packages
How can I fix this issue?
Please follow below steps..
0) please login with root permission or sudo and write this command in terminal
1)pip3 install tensorflow-gpu==1.13.1
2)pip3 --version
pip 8.1.1 from /usr/lib/python3/dist-packages (python 3.5)
3)shekh#shekh:~$ python3
Python 3.5.2 (default, Nov 12 2018, 13:43:14)
[GCC 5.4.0 20160609] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>>
https://www.tensorflow.org/install/pip
See https://github.com/tensorflow/tensorflow/issues/374
Are you opening the terminal from within the tensorflow directory?
"Actually I had this issue because I was trying to import tensorflow from a python session inside the tensorflow repo folder, going to some other folder and starting python and importing tensorflow worked for me."
I also had this error and it was solved with this command
python2:
PYTHONPATH="${PYTHONPATH}:/usr/local/lib/python2.7/dist-packages/"
export PYTHONPATH
python3:
PYTHONPATH="${PYTHONPATH}:/usr/local/lib/python3.6/dist-packages/"
export PYTHONPATH
I encountered a similar error on a Windows 10 system.
There, I had a second tensorflow install under AppData\Roaming\Python\.... This install was likely a product of some previous work that I did not clean up properly, so for me, removing the content of the site-packages directory solved the issue. After that I could import tensorflow without any problems.
When trying to install pgadmin4 in desktop mode on my Ubuntu system, I received a ModuleNotFoundError for _ctypes.
I did some research and found that _ctypes requires the libffi-dev package to be installed. However it seems that libffi-dev and thus _ctypes was installed for Python 2.7, when I run import ctypes it seems to work:
$ python2
Python 2.7.15+ (default, Nov 27 2018, 23:36:35)
[GCC 7.3.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import ctypes
>>>
When I try to do the same for Python 3.7, it doesn't work:
$ python
Python 3.7.3 (default, Jun 21 2019, 12:46:58)
[GCC 7.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import ctypes
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/usr/local/lib/python3.7/ctypes/__init__.py", line 7, in <module>
from _ctypes import Union, Structure, Array
ModuleNotFoundError: No module named '_ctypes'
How could I add the module _ctypes to my Python 3.7 configuration?
If you install python from source file, you have to install some required packages manually as mentioned in https://superuser.com/questions/1412975/how-to-build-and-install-python-3-7-x-from-source-on-debian-9-8.
Actually you are supposed to see some errors after make due to libffinot found as shown in the screenshot below. However, you can still run make install despite the error. When you open python after the installation and import the module, it then gives you such error.
To solve this problem, you can install the dependent package i.e libffi or libffi-devel(redhat) prior to ./configure, make and make install as mentioned in:
Package libffi was not found in the pkg-config search path REDHAT6.5 and
https://bugs.python.org/issue31652.
I have installed graph-tool:
brew --prefix graph-tool
'/usr/local/opt/graph-tool
but when trying to import it:
Python 3.6.4 |Anaconda, Inc.| (default, Jan 16 2018, 12:04:33)
[GCC 4.2.1 Compatible Clang 4.0.1 (tags/RELEASE_401/final)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> from graph_tool.all import *
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ModuleNotFoundError: No module named 'graph_tool'
:(
Does anyone have a solution for this?
Thanks
Homebrew installs the library using its own Python interpreter, which has its module path somewhere in /usr/local/opt/. You are using the anaconda Python interpreter, which knows nothing about homebrew-installed modules. The solution here is simply to use homebrew's Python interpreter, not anaconda's.
I'm using Python 3.6.4 with pillow 5.0.0, through Anaconda.
Python 3.6.4 |Anaconda, Inc.| (default, Jan 16 2018, 10:22:32) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> from PIL import Image
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "F:\Anaconda2\envs\py3\lib\site-packages\PIL\Image.py", line 58, in <module>
from . import _imaging as core
ImportError: DLL load failed: The specified module could not be found.
Probably the same as this question. But that was 10 months ago and the answer says it was fixed with Python 3.6.1.
If you're using the Anaconda distribution, try removing it with Conda:
conda uninstall pillow
and installing it using pip:
pip install pillow
You can test that it's working:
python -c "from PIL import Image"