ML FW analysis and plotting tools - jniedzie/SVJanalysis_wiki GitHub Wiki

In the ML framework directory, in the analysis subdirectory, you will find a few scripts that can visualize training results using the ROOT file created by prepareEvaluationObjects.py. They either take the same config path as an argument as before, or you need to specify paths to the ROOT file with evaluation objects at the top of the script. Here some examples of commands that may be useful:

# Draws metrics evolution
python analysis/drawMetricsEvolution.py -c configs/ae_default.py <extra_flags>

# Produces a table with AUCs for different signals
python drawAUCtable.py -c configs/ae_default.py <extra_flags>

# Draws ROC curves for - point to files with evaluation objects and specify signals to plot
python compareRocs.py

Important note:
A script tailored for the NAE is set up to perform training, evaluation and plotting! See run_nae_training_evaluation_and_plots_example.sh. You can find example usage in this script and run_metrics_evolution_online.sh.