v0.0.3

• updates and fixes of jet and hadron multiplicity analyses
• preliminary stable version of LcpK0s and D0 MBvspt_perc_v0m and V0vspt_perc_v0m analyses
  • machine_learning_hep/data/data_prod_20200304/database_ml_parameters_LcpK0spp_0304.yml
  • machine_learning_hep/data/data_prod_20200304/database_ml_parameters_LcpK0spp_0304_HM_V0.yml
  • machine_learning_hep/data/data_prod_20200417/database_ml_parameters_D0pp_0417.yml
  • machine_learning_hep/data/data_prod_20200417/database_ml_parameters_D0pp_0417_HM_V0.yml
• HardProbes2020 results can be reproduced with this tag

v0.0.2

Features and changes

• small fixes and clean-up of jet analysis
• new plotting function utilities.make_plot
• extend/update validation plots for multiplicity analyses
• include both trigger correction methods for ntrkl analyses, first one based on raw histogram and the other based on the fitted function
• add database for HM triggered for LcpK0spp
• move from Travis to GitHub actions
• clean per-period-results if not needed when passing -c or --clean flag

v0.0.1

First release of MachineLearningHEP

This package is used to conduct HEP analyses. Starting from flat root candidate trees it produces final results incorporating ML techniques to optimise the physics analyses. Training, testing and optimisation of the ML algorithms used is included in the package workflow:

• (pre-)processing and preparing data
• ML training, testing and optimisation
• conduct physics analyses