Closed max-radin closed 1 week ago
Note: Combining these changes with those I made still in progress
Just want to make sure we're not stepping on each other here...
(A) I believe Josh is updating problem_instance
files and this PR also updates problem_instance
files.
(B) Josh also has an open PR for fixes to the solution.schema
in PR#19. When I check git diff max-radin/schema-fixes jtcantin-patch-1 -- schemas/solution.schema.0.0.1.json
, I do see there are some differences between the solution.schema
fixes... most appear to be formatting.
Talk to each other to ensure that (A) and (B) are good. Let me know the resolution.
I've merged Max's schema changes into mine Problem instances still pending
@max-radin It looks like most of the updates are just adding task ids, while some problem instances have additional reformatting. Are you updating from Nicole's latest problem instances (PR#20: https://github.com/isi-usc-edu/qb-gsee-benchmark/pull/20 ), or from a version previous to that?
No, it doesn't include Nicole's updated versions of ru_macho and fe_red. It only updates the problem instances that got marked as invalid after #14 got merged. Regarding the changes to the problem instance files, the only materials changes are renaming instance_data to tasks and adds task_uuid fields. The formatting changes are not important. Also I see now that git diff is showing no differences in the schema between #22 and #19.
@jtcantin is it safe to say that jtcantin-problem_instance_updates is more or less ready to merge in? If so, I'd suggest we close this PR and then merge #19 and then merge that branch.
On the other hand, if it's going to take some time to get jtcantin-problem_instance_updates merged in, I'd suggest we merge this PR in first. Mainly I want to make sure we have at least some problem instances available soon so that #13 isn't blocked.
As an aside: I think whenever we make breaking changes to the schema, we should try to update the JSON files in the same PR.
Ok, since there are no more changes that I need to incorporate, I will go ahead with PR19 and then merge in the problem instances and the DMRG solution files.
Addressed in #19
This PR fixes a few mistakes in the solution schema and also updates the problem instances that were removed by CI in the last commit to comply with the new schemas. Note that each task has been assigned a new UUID that differs from the fcidump UUIDs, as discussed in https://github.com/isi-usc-edu/qb-gsee-benchmark/pull/14.