Put some legacy energy density variables back into the correct hdf5 group. #5258
Conversation
| { | ||
| h5desc.emplace_back(hdf_path{"reference_points"}); | ||
| h5desc.emplace_back(enclosing_path / "reference_points"); |
There was a problem hiding this comment.
Just feel the for-loop at 137 can be replaced with a more elegant range-based loop.
There was a problem hiding this comment.
I agree but since I don't want to write unit tests for dead code I'm going to leave it.
|
Re: python fix. Thanks for the catch Peter. Another example of why we need more of these tests in CI. I recall that these have been broken for a long time in the nightlies, but my assumption was that the simulation was broken... |
|
Q. Does short-diamondC_1x1x1_pp-vmc-estimator-energydensity-cell-batched-4-4-check work now? (link to recent nightly failure) i.e. The batched version? The labels are different between the legacy and batched cases. Why does this difference exist? |
|
This has no effect on the batched test. At least one more PR incoming for batched energydensity. |
|
Test this please |
|
After rerunning the failed sulfur CUDA driver tests passed. Will merge based on Ye's prior approval + easy to iterate further updates. |
Proposed changes
At least on my system check_stat.py cannot successfully read the output of the legacy energy density estimator anymore. This fixes that.
Consequently the
short-diamondC_1x1x1_pp-vmc-estimator-energydensity-cell-4-4-check
short-diamondC_1x1x1_pp-dmc-estimator-energydensity-cell-4-4-check
tests pass
Does this introduce a breaking change?
What systems has this change been tested on?
dgx workstation.
Checklist
Update the following with a yes where the items apply. If you're unsure about any of them, don't hesitate to ask. This is
simply a reminder of what we are going to look for before merging your code.