-
Notifications
You must be signed in to change notification settings - Fork 67
Add ability to read matrix in "XYZ" format; new 'flatten' command #877
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
This pull request introduces 2 alerts when merging 12e0236 into 702b1e5 - view on LGTM.com new alerts:
|
I'm getting messages like |
That's a transient Jenkins error. It was stashed in the first step and should be available in every future step. Re-running should fix it. |
I'm not sure why, but Jenkins seems unstable lately. One run (number 5) took ~14 hours to fail; problem was a node unexpectedly closing down:
The following restarted run failed on the Intel step because it couldn't extract the source:
I guess I will try restarting, but I eventually may need to remove the required status from Jenkins just so I can get this merged. |
The PGI build in Jenkins failed. |
I think I found the issue with premature shutdowns and I've fixed it. I think switching the required builds to several of the GHA builds is a good move - I'd recommend switching regardless (I don't think Jenkins does any CMake builds, for instance). Requiring the CMake builds that take 11 and 15 minutes will test a wider variety of options (OpenMP, MPI, sander API, etc.) should increase coverage. |
Non-square matrices can now be read in via the
read2d
keyword, i.e. matrix data in the format:The
nosquare2d
keyword needs to be specified. I should probably try to do a better job autodetecting this in the future.Also adds the
flatten
command, used to convert a matrix to 1D array by dividing matrix elements up among the array somehow.Useful for e.g. distributing elements of a pairwise matrix across its elements. For example, given a matrix with values like this:
The “flattened” 1D array with mode SUM would be determined as follows:
And the final 1D array would look like so: