key4hep / dmx Goto Github PK

View Code? Open in Web Editor NEW

3.0 4.0 4.0 7.34 MB

data model eXplorer [Experimental]

Home Page: https://key4hep.github.io/dmx/

License: BSD 2-Clause "Simplified" License

JavaScript 93.40% CSS 3.74% HTML 2.86%

dmx's Introduction

dmX

data model eXplorer

Warning: Experimental

Explore your events at dmX.

To generate your .json file use edm4hep2json available in the Key4hep stack. Example usage for the events from FCC winter2023 campaign:

source /cvmfs/sw.hsf.org/key4hep/setup.sh
edm4hep2json -l Particle \
             -n 10 \
             -o p8_ee_WW_ecm240.json \
             /eos/experiment/fcc/ee/generation/DelphesEvents/winter2023/IDEA/p8_ee_WW_ecm240/events_059793334.root

Development

The tool is written in pure JS and draws on HTML Cavas. To run a local version, clone the repo and create simple web server:

python -m http.server

after that visit localhost (http://0.0.0.0:8000/) in your browser.

ToDo

Filters:
- generator status
- generation
- ancestor
Event switcher
Status box
Details box

dmx's People

Contributors

Stargazers

Watchers

Forkers

kjvbrt suraj-sharma27 zehvogel brauliorivas

dmx's Issues

Make it possible to switch events without having to reload the page

To switch events it is necessary to re-load the page and then re-load the file. It would be nice to be able to switch to another event without doing that.

I have a doubt. Every particle ID is associated with a name. So for example, $d$ quark has an ID 1. This is based on the montecarlo scheme. However, as explained in the last meeting, every particle has its antiparticle. So $d$ quark with ID 1, has the antiquark $\bar{d}$ with ID = -1.
I understand this logic but don't know where it comes from. The cited paper shows the particles but not the antiparticles. In dmX, many particles have their equivalent antiparticle definition. I would like to know more about this because many newly incorporated particles lack their antiparticle.

Setup basic CI and some basic (unit) tests

It would be nice to have at least some basic checks that a PR doesn't break anything and be able to run that in CI.

PDG -5122 is not mapped to a particle

This is a minor issue I just stumbled over when looking at the prieview. The PDG code -5122 is not mapped to a particle name

This can be reproduced with event 1 in the example file where it appears in row 11.

JSON file with MCParticle tree cannot be opened

This is not the highest priority, but we have encountered an issue with opening some json files, e.g. the following one (which contains a rather large MCParticle tree from a calorimeter shower). I don't think the size of the file is a problem, but I suspect that potentially a slightly newer version of podio or EDM4hep could be one?

test_mc_shower.edm4hep.json

What we get is the following error:

If you know where the error is coming from, I can also try to have a look myself to see if I can fix this.

Preview from PRs

It would be very nice to have a preview of the site from a pull request. I am not entirely sure whether it's easily possible to actually do this, but it would make reviewing changes that are mainly visual or that at least have a visual component to their functionality much easier. E.g. new buttons / toggles or interactive filtering, etc...

Display quark charges as fractions

to avoid this

Filtering on simulation status is not yet working as intended.

In order to not blow up #27 too much, I have pulled this into a separate issue that can then also be fixed by a separate PR.

The sim status filtering is not currently working as intended. If I look here (which I hope is the right place):

dmx/js/menu/filter/parameters.js

Lines 114 to 118 in 23a1eb2

 buildCondition() { 

 if (!this.checked) return null; 

 return (particle) => particle[this.property] === this.value; 

 }

This checks whether the sim status is exactly the value of the check boxes that appear on the page:

However, these are the bits in the sim status that are set if a condition is met:
https://github.com/key4hep/EDM4hep/blob/0784e02caf09ee89a728b4dd17ed74ef082e4d66/edm4hep.yaml#L297-L304

So there are two things that would need to be improved / fixed

The current display should not display the bit numbers, but rather their meaning. E.g. 23 -> Overlay
The actual filtering should be checking the bits of the simulator status, not the actual value

Add possibility to filter MCParticles via generatorStatus

Similar to the simulator status it might be useful to filter on the generator status. This would allow it to filter out all the intermediate states that are present, e.g. during hadronization.

In this case the inputs are not as clear cut as for the simulation status, so we would probably need a list of input values to filter(?). Or potentially even the possibility for a user defined filter condition, based on the status value(?).

	buildCondition() {
	if (!this.checked) return null;

	return (particle) => particle[this.property] === this.value;
	}