Skip to content

Analysis

This section tries to provide helpful information to our analysis toolchain.

Analysis conventions

We run analyzes as a part of simulation campaigns. In order for farm runners to run an analysis it has to comply to CLI flags convention:

Convention TLDR:

bash
script.py input1.csv input2.csv .... --beam=10x100 -o output_folder
  1. Scripts accept multiple input files. Scripts should be written so that they won't go out of memory if all files for particular energy are used (e.g. 200 files total is 1m events).

  2. Scripts provide -o, --output argument that sets the output Directory for plots and artifacts (e.g. statistics text, csvs)

  3. Scripts provide -b, --beam argument where beam in form (ElectionE)x(IonE) e.g. 10x100 is provided. this flag might be used e.g. for histogram axis limits. Or it might do nothing (but farm scripts will provide it)

  4. Optional: -e, --events, Max number of events to process, not required, but is nice for debugging. None => process all

If multiple csvs are processed at the same time The script is responsible to match input_001.type1.cvs input_001.type2.csv together

bash
   script.py input_001_type1.cvs input_001_type2.csv input_002_type1.csv ...  -o output_folder

Suggestions

A good example of python CSV analysis is considered:

  • analysis/csv_reco_dis_analysis/scattered_electron.py

Released under the MIT License.