search

pele-python / pele

Python energy landscape explorer
Other
95 stars 41 forks source link

Copy hs wca cell test #99

Closed smcantab closed 9 years ago

smcantab commented 9 years ago

implement iprint and verbosity options for modified_fire, add tests to constructor of hs_wca, implement high level test for hs_wca potential to compare the scaling with and without cell lists

js850 commented 9 years ago

implement high level test for hs_wca potential to compare the scaling with and without cell lists

That sounds more like a benchmark than a test. If it runs slowly we should not have it as a regular test. Maybe you can make a benchmarks folder in the test folder.

kjs73 commented 9 years ago

It is strange that this PR includes some commits which are supposed to belong to PR #97. Maybe there is some confusion here.

smcantab commented 9 years ago

maybe you merged it or created this branch from the one for PR 97 and brought along the changes?

On Mon, Nov 24, 2014 at 7:02 PM, Julian Schrenk notifications@github.com wrote:

It is strange that this PR includes some commits which are supposed to belong to PR #97 https://github.com/pele-python/pele/pull/97. Maybe there is some confusion here.

— Reply to this email directly or view it on GitHub https://github.com/pele-python/pele/pull/99#issuecomment-64244972.

kjs73 commented 9 years ago

Hum, that can be. This version will then fail though, because it only comes with half of meta_dist and does not work correctly. Either we exclude the first three commit of PR #97, listed below, from this PR or we merge first PR #97's branch into this one: 91c1e2f 1a0c205 93b7d76

kjs73 commented 9 years ago

OK, for me PR #97 is now ready to merge. I propose that we merge #97 first, that should sort out this PR here. Then the unrelated commits should be gone from this PR.

js850 commented 9 years ago

merging the current master branch might help also.

kjs73 commented 9 years ago

OK, thanks, now it works fine! At the moment, the cells benchmark is in the playground dir. Is that OK, or should we move it to a new benchmarks folder?

smcantab commented 9 years ago

it's not a unit test, so unless we rewrite is a unit test it should be kept in the playground.

js850 commented 9 years ago

yes, it's fine to leave it in the playground for now. It might be nice to set up a proper benchmarking framework sometime though. scipy has something like that.

Go ahead and pull when you're ready