{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fitting with polarimetric weights\n",
    "\n",
    "The Neural Net (NN) designed by Lawrence Peirson generates polarimetric weights for each event. More precisely, these weights are estimates for each event's individual modulation factor. LeakageLib will automatically use these polarization weights when NN-reconstructed data is loaded."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The traditional Moments weights were designed for PCUBE-like analyses. While they do improve PCUBE analyses, maximum likelihood (MLE) analyses such as LeakageLib outperform moments-weighted PCUBE analyses. There is not yet a way to use the Moments weights in MLE analyses. Therefore, LeakageLib is only compatible with NN polarimetric weights."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the user wishes to use their own per-event modulation factors in a LeakageLib fit, they can replace the `IXPEData.evt_mus` field with their own modulation factors after loading the data."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}