Giordon Stark
Giordon Stark
The only thing I might need some help with is to figure out the re-ordering of some of the algorithms. I know there are subtle changes that break existing logic...
> I just spent a set of grid jobs making some empty trees. It's not the end of the world, but I've seen others make this mistake in the past...
> A better solution would be for the info switch classes to throw an error if it detects a string it cannot parse. We can't do this during config time...
> There are a lot of reports in valgrind from the ElectronEfficiencyCorrector and MuonEfficiencyCorrector algorithms, because of the `new` CP tools that aren't deleted anywhere. It was recommended that we...
Again, in cases when there's a new (and no delete) -- it's a memory leak because it's not going to get deleted. In cases when there's a new and a...
Ok, so I consider myself an argparse expert (having played with the internals a lot) and can tell you that this situation occurs when you're trying to check if two...
(where is Namespace equality happening though?!)
And it looks like it might be because `pyAMI` is catching an error and ignoring it -- because that might be the only way `parse_args` returns `None`: ```python # =====================================...
Pinging @kkrizka . I suspect that you can just re-run the overlap removal twice.
submit a PR :)