The way this works it (should) turn off all but the nominal for the main overlap removal algorithm, and then schedules extra algorithms to copy over the decision from the nominal to all shallow copies, so that it can be used for preselections in other algorithms (if desired).

I did test that it runs and defined some test cases, but I didn't test at all that it does anything meaningful or correct.  I leave that to physics validation.

Baptiste Ravina authored 1 year ago and Edward Moyse committed 1 year ago

CPAlgorithms: common bootstrapping algorithm

Baptiste Ravina authored 1 year ago and Vakhtang Tsulaia committed 1 year ago

CP Algorithms: event cut flow

The main thing this does is to introduce an AsgUnionPreselectionAlg
that allows to combine systematics dependent selection flags into a
uniform preselection per systematic copy.  The logic here is that in
preselections I (typically) just want to filter out objects for
computational efficiency, but if the object fails for one systematic
on the shallow copy I anyways have to process it.  Having such a joint
selection decoration prevents the systematics of the selection to
propagate into outputs that would otherwise be unaffected by it.

This then in turn allows me to add such a decoration (on request) as
part of the overlap removal config, as that is the only config block
that has systematics dependent selection decorations.

I also added a bunch of options to the OverlapRemovalConfig to allow
changing the selection name to whatever the user wants/needs.

Teng Jian Khoo authored 1 year ago and Vakhtang Tsulaia committed 1 year ago

Define algorithms that cross-link nominal and systematic variations objects, and use these to build union selections

R D Schaffer authored 1 year ago and Tadej Novak committed 1 year ago

Improvements and fixes to PhysLite, primarily as seen from the H4l analysis perspective. Running isolationCloseBy tool in PHYS and PHYSLITE to provide isolation variables with overlap from muons, electrons and photons already removed. Added in config to PhysLite for H4l vertex calculation. Added in STXS information to PhysLite for Higgs analyses, fixing units. Reducing the electron id requirement from LH Loose to SiHits for PhysLite.

This is essentially a heavily modified version of
AsgxAODNTupleMakerAlg for the specific purpose at hand, i.e. writing
out MET variables as `float` instead of `std::vector<float>`.

There is probably some overlap of code with the original algorithm,
but I'd rather clean that up at a later point, as I anyways have to
revisit how I handle the output n-tuples in a better way.

This goes back to an old discussion that in principle AsgSelectionAlg
should not have a SysCopyHandle, as it doesn't introduce new momentum
systematics.  The reason it has one currently is that if there is a
selection before the correction algorithm it does need to make a
shallow copy that can then subsequently be modified at will.

The solution implemented here is what we discussed back then: To
implement an algorithm that does nothing more than make a shallow copy
(based on the view container algorithm and the copy mechanism from
SysCopyHandle) and schedule that before AsgSelectionAlg as needed.

For now this just replaces the algorithm with a service, and
communicates the systematics list via the service interface instead of
via the event store.

The plan is to add more functionality in future merge requests that
will move a lot of the systematics handling functionality from the
configuration layer into the service.

Nils Erik Krumnack authored 4 years ago and Atlas Nightlybuild committed 4 years ago

Add CP::IClassificationTool + CP::AsgClassificationDecorationAlg

See merge request !34893

(cherry picked from commit 31f39fba)

060f439e Add CP::IClassificationTool interface
bd02da17 Add CP::AsgClassificationDecorationAlg

Originally I had tried sweeping/cherry-picking this from 21.2, but
with all the three-way merges and commits that involve other
packages/directories that was rather painful, so this is now just a
direct copy from the current state in 21.2.