training-dataset-dumper merge requestshttps://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests2024-03-28T07:20:11+01:00https://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/651Make ca config accumulator more module for taucomp2024-03-28T07:20:11+01:00Nikita Ivvan PondMake ca config accumulator more module for taucompImplements the changes recommended in !649Implements the changes recommended in !649https://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/648Draft: Error running rich configuration (EMPFlowRetagAthena), TRK_RES_D0_MEAS...2024-03-27T19:50:57+01:00Thomas Benjamin DingleyDraft: Error running rich configuration (EMPFlowRetagAthena), TRK_RES_D0_MEAS within EMPFlowRetag causes error.Hi @dguest, here's a draft MR of the bug we discuss on MM.Hi @dguest, here's a draft MR of the bug we discuss on MM.https://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/641Draft: Caloclusters22024-03-27T20:13:27+01:00Ivan OleksiyukDraft: Caloclusters2### Description
Make it possible to dump calo clusters as constituents of the jet
Related to this issue:
atlas-flavor-tagging-tools/training-dataset-dumper#128
Review checklist:
- [ ] CI Passing
- [ ] Comments addressed
- [ ] Source b...### Description
Make it possible to dump calo clusters as constituents of the jet
Related to this issue:
atlas-flavor-tagging-tools/training-dataset-dumper#128
Review checklist:
- [ ] CI Passing
- [ ] Comments addressed
- [ ] Source branch is up to date with targethttps://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/634Draft: Decorating Jes variables directly to jets without btagging link2024-03-25T09:35:01+01:00Wei Sheng LaiDraft: Decorating Jes variables directly to jets without btagging link<!--
If this is your first time making an MR please read
ftag.docs.cern.ch/getting_involved/how_to_ftag/#code-gitlab
-->
### Description
<!--Please provide a detailed description of your changes here-->
Allow decoration of Jes variable...<!--
If this is your first time making an MR please read
ftag.docs.cern.ch/getting_involved/how_to_ftag/#code-gitlab
-->
### Description
<!--Please provide a detailed description of your changes here-->
Allow decoration of Jes variables for jets without a btagging link.
<br>
We write the Jes variables directly to the jet collection with a `decorateJes` function.
<br>
Note that the requirement of a btagging link is removed.
<!--Please link any relevant issues:-->
<!--Closes #issue will automatically close the issue when the MR is merged-->
<!--Relates to #issue will link the issue to the MR-->
<!--The following items need to be addressed before your request can be merged-->
<!--Please proactively take care of them!-->
Review checklist:
- [ ] CI Passing
- [ ] Comments addressed
- [ ] Source branch is up to date with targethttps://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/630Draft: Initial commit for adding vector of MC event weights from event info -...2024-03-12T21:28:55+01:00Thomas Benjamin DingleyDraft: Initial commit for adding vector of MC event weights from event info - bugged at the moment.Hi @dguest & @svanstro ,
I'm looking into implementing MC event weights within the TDD. There's a function directly from event info to return the array of weights https://ucatlas.github.io/RootCoreDocumentation/2.4.28/df/df8/classxAOD_...Hi @dguest & @svanstro ,
I'm looking into implementing MC event weights within the TDD. There's a function directly from event info to return the array of weights https://ucatlas.github.io/RootCoreDocumentation/2.4.28/df/df8/classxAOD_1_1EventInfo__v1.html#aaa66353a56b59ede862025d543c2f97f, once it's working I will add some reference within the TDD for exactly which MC weight each index corresponds to. We already have mcEventWeight implemented, so I looked to do the same as a vector of floats for the mcEventWeights.
I've had a first go here but keep getting errors with https://gitlab.cern.ch/atlas/athena/-/tree/main/PhysicsAnalysis/AnalysisCommon/HDF5Utils/HDF5Utils. I've attached the error message and wondered if you had any pointers to solve the problem?
[error_error.txt](/uploads/f075f0889f6f0f4ac56bbb089d81de28/error_error.txt)
If this doesn't work, then there should be other options via PMG (https://twiki.cern.ch/twiki/bin/view/AtlasProtected/PmgEventWeights) and accessing TruthMetaData within the DAOD. Let me know what you think :smile:
Thanks,
Tomhttps://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/625Draft: Update Xbb Dumping2024-03-05T17:19:11+01:00Jackson BarrDraft: Update Xbb Dumping<!--
If this is your first time making an MR please read
ftag.docs.cern.ch/getting_involved/how_to_ftag/#code-gitlab
-->
### Description
Update fat jet dumping script that included some VR track jet hacks that are no longer needed
<!--...<!--
If this is your first time making an MR please read
ftag.docs.cern.ch/getting_involved/how_to_ftag/#code-gitlab
-->
### Description
Update fat jet dumping script that included some VR track jet hacks that are no longer needed
<!--Please provide a detailed description of your changes here-->
<!--Please link any relevant issues:-->
<!--Closes #issue will automatically close the issue when the MR is merged-->
<!--Relates to #issue will link the issue to the MR-->
<!--The following items need to be addressed before your request can be merged-->
<!--Please proactively take care of them!-->
Review checklist:
- [ ] CI Passing
- [ ] Comments addressed
- [ ] Source branch is up to date with targethttps://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/612Draft: Automatic check on test-dumper outputs2024-03-12T10:23:34+01:00Nikita Ivvan PondDraft: Automatic check on test-dumper outputsThis MR introduces a new bash script, `compare-output`, which acts as a wrapper around the existing `test-dumper` script. When running compare-output, a mode is provided. The test-dumper is then run for this mode, and the outputs compare...This MR introduces a new bash script, `compare-output`, which acts as a wrapper around the existing `test-dumper` script. When running compare-output, a mode is provided. The test-dumper is then run for this mode, and the outputs compared to the 'expected' outputs.
If changes are made to the framework that should result in changes to the outputs, then we can simply run the script with the additional flag `-o/--overwrite`, which will overwrite the relevant output file instead of comparing.
This MR closes #127https://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/597Draft: Adding lines for running in retag-fatjet mode on the grid2024-02-26T12:49:56+01:00Lucrezia RambelliDraft: Adding lines for running in retag-fatjet mode on the gridWhen running on the grid if the -T option is called then it will replace the default case otherwise the Nominal case will runWhen running on the grid if the -T option is called then it will replace the default case otherwise the Nominal case will runhttps://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/554Draft: More output dataseets2024-01-25T15:14:00+01:00Dan GuestDraft: More output dataseetsThis is attempting to get at #123. It adds another dataset alongside the `jets` one, which is configurable the same way but uncompressed.
I'm also rewriting the configuration structure a bit to remove (a bit of) the configuration mungin...This is attempting to get at #123. It adds another dataset alongside the `jets` one, which is configurable the same way but uncompressed.
I'm also rewriting the configuration structure a bit to remove (a bit of) the configuration munging.
To do:
- [x] Clean up configuration structures
- [ ] Add another uncompressed jet output datasethttps://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/526Draft: Adding truth_barcode for signal2024-03-16T11:03:37+01:00Spyros ArgyropoulosDraft: Adding truth_barcode for signalSaving the `truth_barcode` for all hits.
![Screenshot_2023-07-18_at_15.00.10](/uploads/a9aea1811b9db15100fbb7b21241d4a9/Screenshot_2023-07-18_at_15.00.10.png)
## Things to discuss
- [x] Currently saving first barcode - in all of the c...Saving the `truth_barcode` for all hits.
![Screenshot_2023-07-18_at_15.00.10](/uploads/a9aea1811b9db15100fbb7b21241d4a9/Screenshot_2023-07-18_at_15.00.10.png)
## Things to discuss
- [x] Currently saving first barcode - in all of the cases I saw there is only 1 barcode but there could be cases with more. Ideally we would save a list but I'm unable to make it compile with a list maybe @dguest knows how to do it
- [x] Do we also want to add some logic to already save the pdgId or a label (like isFromB) for the hits which have a barcode @svanstro suggested that we do it later (but maybe having it here would be an opportunity to harmonise)
- [ ] CI fails because `truth_barcode` is not in the test file we have - we probably need to make a new one from the ttbar sample - however that will also fail before is !508 merged
tag @mtanasin
see also https://gitlab.cern.ch/atlas/athena/-/merge_requests/69866https://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/445Draft: Jet-associating and dumping neutral V0 xAOD::Vertex objects2023-10-16T13:28:52+02:00Waltteri LeinonenDraft: Jet-associating and dumping neutral V0 xAOD::Vertex objects# Contents of MR
This MR contains code to dump neutral vertices (v0's) produced in the InDetV0Finder. This was tested on small samples produced by Ligang in the new CA-style FTAG1 derivation format.
I introduce five new files, as well ...# Contents of MR
This MR contains code to dump neutral vertices (v0's) produced in the InDetV0Finder. This was tested on small samples produced by Ligang in the new CA-style FTAG1 derivation format.
I introduce five new files, as well as multiple modifications to the configuration, outputs, and tools code, as well as the necessary modifications to the event loop code.
The `JetNeutralVertexAssociator` currently does a simple check on the compatibility of the reconstructed neutral vertex and a jet, linking the two if all compatibility selections are satisfied. In the future, more complex associations will be attempted (leveraging larger samples) in order to reconstruct semi-inclusive decay chains of charmed mesons.
**Note:** it seems that the `beamSpotWeight` config option doesn't work with this sample. Still not sure the cause, but could be due to the sample.https://gitlab.cern.ch/atlas-flavor-tagging-tools/training-dataset-dumper/-/merge_requests/385Draft: First import of vertex blocks2023-07-26T16:07:58+02:00Carlo SchiaviDraft: First import of vertex blocksThis first import adds the functionality to dump vertex (xAOD::Vertex and xAOD::BTagVertex) blocks.
To test it, the additional EMPFlowVtx.json configuration can be used.
Additional work on available variables will follow, along with some...This first import adds the functionality to dump vertex (xAOD::Vertex and xAOD::BTagVertex) blocks.
To test it, the additional EMPFlowVtx.json configuration can be used.
Additional work on available variables will follow, along with some more polishing (along the lines of the track block).