Conda installers available!

More
4 years 3 months ago #2885 by lbittrich
I did get the same thing on windows with a different CUDA version but only when I forgot to add the coda-forge channel. Ngsolve has a lot of dependencies that can only be matched in this channel so the short version of supplying only the ngsolve channel in the install command does not work.

So if you setup a new environment just do (strict priority might not be needed under windows):

conda config --env --add channels conda-forge
conda config --env --add channels ngsolve
conda config --env --set channel_priority strict
conda install ngsolve

Best,
Lars
More
4 years 3 months ago #2886 by dengcy
Replied by dengcy on topic In compatible with CUDA 11.0
Thanks! I used you way, and I could install the package. However, when I run
Code:
from ngsolve import *
, I got
Code:
DLL load failed: The specified procedure could not be found.

I gave up, and used .msi to install on windows. It worked now.
More
4 years 4 weeks ago #3140 by bakerk
Replied by bakerk on topic In compatible with CUDA 11.0
Im getting this same error, did you ever work it out?
More
3 years 3 months ago #3796 by sjhamilton
Hi there, I am trying to install Netgen on my MacBook. I tried the DMG file but ran into an error with it not finding python etc. I now tried the anaconda approach and the install produces pages and pages of errors with "conflicts"... Here is some of the terminal code/output. I have installed the latest anaconda distribution.

~ % conda install ngsolve
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: /
Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
Examining conflict for pyls-black pylint astroid black typed-ast anaconda: : 349it [1:08:54, 2Examining conflict for pyls-black pylint astroid black typed-ast anaconda: : 350it [1:08:54, 3Examining conflict for pyls-black prometheus_client black scikit-image anaconda appdirs: : 350iExamining conflict for pyls-black prometheus_client black scikit-image anaconda appdirs: : 351iExamining conflict for scikit-image anaconda networkx: : 351it [1:09:05, 3.57s/it] failed

UnsatisfiableError: The following specifications were found
to be incompatible with the existing python installation in your environment:

Specifications: (pages of errors available if you'd like to see them)

____
Any advice you have would be greatly appreciated.

Thank you,
Sarah
More
3 years 3 months ago - 3 years 3 months ago #3823 by nssharma
**Issue Resolved by upgrading my Ubuntu version.**
Hello Matthias,

I have been using Netgen/NGSolve on 64-bit Ubuntu 16.04 LTS for the past 6 months. I installed it using Conda.
Only recently have I encountered the following error
NameError: name 'dx' is not defined

with codes involving dx.
I followed the advice on: Can not run navierstokes.py example
( ngsolve.org/forum/ngspy-forum/903-can-no...okes-py-example#2046 )
but in vain.
More info on my install:
# packages in environment at /home/nssharma/anaconda2/envs/ngsolve_test:
netgen 6.2.2006 py37he358ceb_3 conda-forge
ngsolve 6.2.2006.post48 py37_0 ngsolve

Could you please tell me what I can do to fix this NameError?

Many thanks!
Natasha
Last edit: 3 years 3 months ago by nssharma. Reason: Issue Resolved.
More
2 years 11 months ago #4033 by Chin
Replied by Chin on topic Conda installers available!
I have tried installing using the conda packages on multiple systems and when it attempt to run:

jupyter nbextension install --user --py ngsolve

i get an error similar to

FileNotFoundError: [Errno 2] No such file or directory: '/home/username/anaconda3/envs/ngs/lib/python3.8/site-packages/ngsolve/nbextension/static'

Can anyone provide a fix to this?
Time to create page: 0.111 seconds