PyOpenCL January 2019

pyopencl@tiker.net

5 participants
2 discussions

by William Smith

Dear all, My colleagues and I use PyOpenCL as part of an cell-based modelling program (Tim Rudge's CellModeller, here <http://haselofflab.github.io/CellModeller/> ). We would like to use CellModeller alongside a PDE Solver (FEniCS <https://fenicsproject.org/>), whose NumPy requirements (>=1.13) conflict with those of PyOpenCL (1.9). Is it possible to make PyOpenCL compatible with later NumPy versions? Here's the conflict in my conda environment: $ conda install numpy=1.13 Solving environment: failed UnsatisfiableError: The following specifications were found to be in conflict: - numpy=1.13 - pyopencl -> numpy=1.9.2 Use "conda info <package>" to see the dependencies for each package. It looks like even the latest version of PyOpenCL (2018.2.2) still only uses NumPy 1.9.3. What scope is there for overcoming this compatibility issue? Many thanks and best wishes, W Dr. William P J Smith _____________________________________________ Postdoctoral researcher, Foster lab, University of Oxford

2 years, 5 months

Issue during installation with pip install pyopencl: missing numpy

by Goncalo Morgado

Hey hey :) I am fairly new with pyopencl. Still getting the grip of it... Although the docs suggest installing pyopencl with conda I have been using pip, which seems to work quite well, apart from an issue with dependencies. pip install pyopencl on a freshly created virtual environment will complain about missing numpy and warn about missing pybind11 and mako. Regarding numpy, and if I understand correctly, some effort has been put ( https://github.com/inducer/pyopencl/blob/master/aksetup_helper.py#L69) to fix the issue with the class "ExtensionUsingNumpy" as opposed to "NumpyExtension". I gave it a try with the latest version in the github repo, and `pip install .`, but it still complains about missing numpy. Should I install dependencies manually beforehand? Shouldn't this be handled in setup.py with setup_requires=[] and install_requires=[]? From a user perspective it makes a lot of sense to just run `pip install pyopencl` and have all dependencies installed, similar to what happens when we install scipy or matplotlib. Thanks in advance, Goncalo

2 years, 5 months

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

PyOpenCL January 2019