https://github.com/lukasheinrich/lhoodbinder2

dynamic version of contour binder
https://github.com/lukasheinrich/lhoodbinder2

Last synced: about 1 month ago
JSON representation

dynamic version of contour binder

• Host: GitHub
• URL: https://github.com/lukasheinrich/lhoodbinder2
• Owner: lukasheinrich
• Created: 2019-10-21T19:59:29.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2021-03-29T14:35:28.000Z (almost 4 years ago)
• Last Synced: 2024-12-27T03:42:20.728Z (about 1 month ago)
• Language: Python
• Size: 699 KB
• Stars: 1
• Watchers: 3
• Forks: 1
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# JSON Likelihoods for ATLAS SUSY sbottom multi-b analysis

[![Binder](https://mybinder.org/badge_logo.svg)](https://mybinder.org/v2/gh/lukasheinrich/lhoodbinder2/master?filepath=PlotContour.ipynb)

The JSON likelihoods are serialized for each signal region: [RegionA](RegionA), [RegionB](RegionB), and [RegionC](RegionC). This is done by providing a background-only workspace containing the signal/control/validation channels for each region at `$region/BkgOnly.json` as well as patch files for each mass point on the signal phase-space explored in the analysis.

Each [jsonpatch](http://jsonpatch.com/) file follows the format `$region/patch.sbottom_msb_mn2_mn1.json` where `msb` is the mass of the sbottom squark, `mn2` is the mass of the second-lightest neutralino, and `mn1` is the mass of the lightest supersymmetric particle (LSP).

This particular analysis has two mass scenarios. In the [associated PUB note](https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PUBNOTES/ATL-PHYS-PUB-2019-029/), the reproduction of `mn1 = 60 GeV` is shown. The other mass scenario is where `mn2 - mn1 = 130 GeV`. To get files for each mass scenario, you can use bash like so:

```

find . -name "patch.sbottom*.json" | awk -F_ '$3-$4==130 {print $0}'

find . -name "patch.sbottom*.json" | awk -F_ '$4==60 {print $0}'

```

## Producing signal workspaces

As an example, we use [python jsonpatch](https://python-json-patch.readthedocs.io/en/latest/) here:

```

jsonpatch RegionA/BkgOnly.json RegionA/patch.sbottom_1300_850_60.json > RegionA/sbottom_1300_850_60.json

```

## Computing signal workspaces

For example, with [pyhf](https://diana-hep.org/pyhf/), you can do any of the following:

```

pyhf cls RegionA/BkgOnly.json -p RegionA/patch.sbottom_1300_850_60.json

jsonpatch RegionA/BkgOnly.json RegionA/patch.sbottom_1300_850_60.json | pyhf cls

pyhf cls RegionA/sbottom_1300_850_60.json

```