Skip to content
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

Testing: Test of restart code needed #232

Closed
prckent opened this issue May 26, 2017 · 7 comments
Closed

Testing: Test of restart code needed #232

prckent opened this issue May 26, 2017 · 7 comments
Assignees

Comments

@prckent
Copy link
Contributor

prckent commented May 26, 2017

Clear tests (and therefore examples) of working restarts with VMC and DMC are needed.

Based on these tests, the restart code can be updated/cleaned/rewritten as needed. No need for backward compatibility.

@prckent
Copy link
Contributor Author

prckent commented May 26, 2017

@markdewing Delegate/ask for helpers as needed

@prckent
Copy link
Contributor Author

prckent commented May 26, 2017

This is also needed to interface with the ECP VeloC project.

@ye-luo
Copy link
Contributor

ye-luo commented May 26, 2017

This part of my project for the summer student. Deal with the restart and paralell IO.

@prckent
Copy link
Contributor Author

prckent commented May 26, 2017

Tests are due by end of June. Fixes later.

@prckent prckent assigned markdewing and ye-luo and unassigned markdewing May 26, 2017
@prckent prckent added this to the v3.1.0 Release milestone May 29, 2017
@prckent
Copy link
Contributor Author

prckent commented May 29, 2017

Provisionally added as a requirement to the next release (3.1) particularly in light of bug referenced in #2 . Would prefer to have fixed unless it looks to be overly complicated.

@ye-luo
Copy link
Contributor

ye-luo commented Jun 1, 2017

I have prepared the test case for checking restart. Not integrated into ctest yet.
The electron coordinates and random numbers seem loaded correctly.

  1. The extra VMC block (no electron move) and the restart run with the same VMC block agree on KE, ElecElec, LocalECP but NonLocalECP is slightly different. I can't trace all the places where random numbers are used.
  2. Run the restart run many times and I got exactly the same results even the initial random seed is not fixed.

So I will set up the checking based on the energy components except NonLocalECP.

@prckent
Copy link
Contributor Author

prckent commented Jun 16, 2017

I'll close this since the VMC restart tests are incorporated ( #249 ). DMC restart currently also looks clean ( #267 ) so we'll need a specific reproducer/bug report with inputs to make progress, if still needed.

@prckent prckent closed this as completed Jun 16, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

4 participants