Umami

The Umami documentation is avaliable here:

Below is included a brief summary on how to get started fast.

Installation

Docker image

singularity exec docker://gitlab-registry.cern.ch/atlas-flavor-tagging-tools/algorithms/umami:latest bash

besides the CPU image, there is also a GPU image available which is especially useful for the training step

singularity exec --nv docker://gitlab-registry.cern.ch/atlas-flavor-tagging-tools/algorithms/umami:latest-gpu bash

Manual setup

Alternatively you can also check out this repository via git clone and then run

python setup.py install

this will install the umami package

If you want to modify the code you should run instead

python setup.py develop

which creates a symlink to the repository.

If you want to commit changes it is recommended to install the pre-commit hooks by doing the following:

pre-commit install

This will run isort, black and flake8 on staged python files when commiting

Testing & Linter

The test suite can be run via

pytest ./umami/tests/ -v

If you want to only run unit tests, this can be done via

pytest ./umami/tests/unit/ -v

and the integration test similarly via

pytest ./umami/tests/integration/ -v

In order to run the code style checker flake8 use the following command

flake8 ./umami

DL1r instructions

If you want to train or evaluate DL1r please follow the DL1r-instructions.

DIPS instructions

If you want to train or evaluate DIPS please follow the DIPS-instructions

Preprocessing

For the training of umami the ntuples are used as specified in the section MC Samples.

The ntuples need to be preprocessed following the preprocessing instructions.