A program to apply multivariate classification techniques to top analyses.

Requires python>=2.7.13 and root>=6.06.08 with python 2 bindings enabled. Python 3 is not supported.

To install run

pip install -e .

from the project's root directory.

For multi-layer perception support, specify:

pip install -e .[MLP]

This will install the Keras package, but not a (required!) Keras backend. Consult the Keras documentation for availible backends, but note only the Theano backend has been tested.

For gradient boosted decision trees using the XGBoost library, specify:

pip install -e .[xgboost]

For gradient boosted decision trees using the LightGBM library, specify:

pip install -e .[lightgbm]

This tool is used via the tact command-line utility. It takes only one argument - a YAML configuration file, well-documented examples of which can be found in the "configs/" directory. Alternatively, the --stdin argument can be specified to read a configuration file from stdin.

tact performs multivariate classifier training and application, producing files which can then be used in THETA or the Higgs Analysis Combined Limit tools. It should be the only program needed for most use-cases.

• Corin Hoad

This project is licensed under the Revised BSD License. See the LICENSE.txt for details.