ScaleVar_STXS1p2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# STXS 1.2 Scale Variations Uncertainties"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "See [the notebook about STXS 1.2](https://github.com/bonanomi/STXS_Unc/blob/master/ScaleVar_STXS1p2.ipynb) for more details. Here we use a `NanoAOD` HZZ4L sample to calculate $\\mu_R$ and $\\mu_F$ scale variations of the yields in each STXS 1.1 bin. The final uncertainty assigned corresponds to the maxiumum relative variation w.r.t. the nominal yield in each bin.\n",
    "\n",
    "The advantage of using `NanoAOD` samples is their flexibility and the fact that they come directly with Rivet classification implemented. For STXS 1.2 we didn't have at disposal any ZZ4L `NanoAOD` sample, hence a ggH ntuple from the HZZ4L analysis group was used."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.colors\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import uproot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Not needed on lxplus\n",
    "import os\n",
    "os.system('/opt/exp_soft/cms/t3/eos-login -init -username mbonanom')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "url = 'https://gist.githubusercontent.com/bonanomi/d14780f7562cb2a22fdd753a9d4459d4/raw/b3c6d77505b90eb38a886c19a31725df8ceb908f/MyMPLStyle'\n",
    "plt.style.use(url)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "tree = 'Events'\n",
    "fname = '/eos/user/m/mbonanom/nanoAOD_GluGluHToZZTo4L_powheg2NNLOPS_M125.root'\n",
    "branches = ['HTXS_stage1_2_cat_pTjet30GeV', 'LHEScaleWeight']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "evt = uproot.open(fname)[tree]\n",
    "evts = evt.arrays(branches)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "binning = np.linspace(0, 500, 500)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "h_stxs = np.histogram(evts[b'HTXS_stage1_2_cat_pTjet30GeV'], binning)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "d = {}\n",
    "for i in range(9):\n",
    "    d[i] = [j[i] for j in evts[b'LHEScaleWeight']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 684 ms, sys: 50.5 ms, total: 735 ms\n",
      "Wall time: 733 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "d_scales = {}\n",
    "for i in range(9):\n",
    "    d_scales[i] = np.histogram(evts[b'HTXS_stage1_2_cat_pTjet30GeV'], binning, weights=d[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "int_nominal = [i for i in h_stxs[0] if i!=0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 2.46 ms, sys: 0 ns, total: 2.46 ms\n",
      "Wall time: 2.46 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "int_scale = {}\n",
    "for i in range(9):\n",
    "    if (i==2) | (i==6):\n",
    "        continue\n",
    "    int_scale[i] = [j for j, k in zip(d_scales[i][0], h_stxs[0]) if k != 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "bins_c =  0.5*(h_stxs[1][1:]+h_stxs[1][:-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "stxs_bins = [round(j, 0) for i, j in zip(h_stxs[0], bins_c) if i!=0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[100] => 50643 ::: 86460.02,  86199.33,  50279.04,  50643.00,  50244.25,  40326.19,  40156.08, \n",
      "\n",
      "[101] => 6945 ::: 8641.23,  8458.84,  7155.85,  6945.00,  6770.81,  5759.45,  5578.52, \n",
      "\n",
      "[102] => 1706 ::: 2100.80,  2045.54,  1774.50,  1706.00,  1648.59,  1425.75,  1366.12, \n",
      "\n",
      "[103] => 247 ::: 298.84,  291.20,  258.26,  247.00,  237.27,  208.90,  198.41, \n",
      "\n",
      "[104] => 31 ::: 37.43,  36.34,  32.69,  31.00,  29.48,  26.36,  24.74, \n",
      "\n",
      "[105] => 67231 ::: 118155.49,  139054.38,  55387.01,  67231.00,  74399.33,  52310.32,  58811.20, \n",
      "\n",
      "[106] => 200757 ::: 292391.88,  299904.19,  188899.20,  200757.00,  208916.97,  166216.28,  174309.62, \n",
      "\n",
      "[107] => 66458 ::: 86479.88,  87627.44,  64561.94,  66458.00,  67897.75,  56622.25,  57937.41, \n",
      "\n",
      "[108] => 52621 ::: 66629.19,  67027.69,  52206.34,  52621.00,  52941.81,  44140.16,  44333.53, \n",
      "\n",
      "[109] => 10493 ::: 13494.50,  13359.56,  10638.84,  10493.00,  10371.06,  8600.94,  8468.03, \n",
      "\n",
      "[110] => 14219 ::: 18559.75,  18789.38,  13893.56,  14219.00,  14463.25,  12024.25,  12239.31, \n",
      "\n",
      "[111] => 23014 ::: 29182.38,  29254.00,  22988.56,  23014.00,  23027.06,  19152.66,  19114.91, \n",
      "\n",
      "[112] => 13705 ::: 17505.50,  17297.69,  13938.22,  13705.00,  13510.75,  11250.81,  11042.91, \n",
      "\n",
      "[113] => 2743 ::: 3496.81,  3505.25,  2733.78,  2743.00,  2748.31,  2295.41,  2295.31, \n",
      "\n",
      "[114] => 5423 ::: 6913.50,  6890.12,  5458.22,  5423.00,  5392.12,  4510.47,  4469.81, \n",
      "\n",
      "[115] => 1126 ::: 1478.25,  1460.31,  1139.44,  1126.00,  1114.44,  936.66,  924.97, \n",
      "\n",
      "[116] => 2318 ::: 2960.81,  2923.12,  2363.22,  2318.00,  2280.75,  1932.72,  1893.44, \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for stxs, nom_int in zip(enumerate(stxs_bins), int_nominal):\n",
    "    k = stxs[0]; stxs_bin = stxs[1]\n",
    "    print('[{}] => {} :::'.format(int(stxs_bin), nom_int), end='')\n",
    "    for i in range(9):\n",
    "        if (i==2) | (i==6):\n",
    "            continue\n",
    "        print(' {0:.2f}, '.format(int_scale[i][k]), end='')\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[100] => 50643 ::: 70.72%,  70.21%,  -0.72%,  0.00%,  -0.79%,  -20.37%,  -20.71%, \n",
      "\n",
      "[101] => 6945 ::: 24.42%,  21.80%,  3.04%,  0.00%,  -2.51%,  -17.07%,  -19.68%, \n",
      "\n",
      "[102] => 1706 ::: 23.14%,  19.90%,  4.02%,  0.00%,  -3.37%,  -16.43%,  -19.92%, \n",
      "\n",
      "[103] => 247 ::: 20.99%,  17.89%,  4.56%,  0.00%,  -3.94%,  -15.43%,  -19.67%, \n",
      "\n",
      "[104] => 31 ::: 20.74%,  17.24%,  5.46%,  0.00%,  -4.90%,  -14.96%,  -20.20%, \n",
      "\n",
      "[105] => 67231 ::: 75.75%,  106.83%,  -17.62%,  0.00%,  10.66%,  -22.19%,  -12.52%, \n",
      "\n",
      "[106] => 200757 ::: 45.64%,  49.39%,  -5.91%,  0.00%,  4.06%,  -17.21%,  -13.17%, \n",
      "\n",
      "[107] => 66458 ::: 30.13%,  31.85%,  -2.85%,  0.00%,  2.17%,  -14.80%,  -12.82%, \n",
      "\n",
      "[108] => 52621 ::: 26.62%,  27.38%,  -0.79%,  0.00%,  0.61%,  -16.12%,  -15.75%, \n",
      "\n",
      "[109] => 10493 ::: 28.60%,  27.32%,  1.39%,  0.00%,  -1.16%,  -18.03%,  -19.30%, \n",
      "\n",
      "[110] => 14219 ::: 30.53%,  32.14%,  -2.29%,  0.00%,  1.72%,  -15.44%,  -13.92%, \n",
      "\n",
      "[111] => 23014 ::: 26.80%,  27.11%,  -0.11%,  0.00%,  0.06%,  -16.78%,  -16.94%, \n",
      "\n",
      "[112] => 13705 ::: 27.73%,  26.21%,  1.70%,  0.00%,  -1.42%,  -17.91%,  -19.42%, \n",
      "\n",
      "[113] => 2743 ::: 27.48%,  27.79%,  -0.34%,  0.00%,  0.19%,  -16.32%,  -16.32%, \n",
      "\n",
      "[114] => 5423 ::: 27.48%,  27.05%,  0.65%,  0.00%,  -0.57%,  -16.83%,  -17.58%, \n",
      "\n",
      "[115] => 1126 ::: 31.28%,  29.69%,  1.19%,  0.00%,  -1.03%,  -16.82%,  -17.85%, \n",
      "\n",
      "[116] => 2318 ::: 27.73%,  26.11%,  1.95%,  0.00%,  -1.61%,  -16.62%,  -18.32%, \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for stxs, nom_int in zip(enumerate(stxs_bins), int_nominal):\n",
    "    k = stxs[0]; stxs_bin = stxs[1]\n",
    "    print('[{}] => {} :::'.format(int(stxs_bin), nom_int), end='')\n",
    "    for i in range(9):\n",
    "        if (i==2) | (i==6):\n",
    "            continue\n",
    "        norm_stxs = (int_scale[i][k] - nom_int)/nom_int * 1e2\n",
    "        print(' {0:.2f}%, '.format(norm_stxs), end='')\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[100] => ::: 0.5857,  0.5875,  1.0072,  1.0000,  1.0079,  1.2558,  1.2612, \n",
      "\n",
      "[101] => ::: 0.8037,  0.8210,  0.9705,  1.0000,  1.0257,  1.2058,  1.2450, \n",
      "\n",
      "[102] => ::: 0.8121,  0.8340,  0.9614,  1.0000,  1.0348,  1.1966,  1.2488, \n",
      "\n",
      "[103] => ::: 0.8265,  0.8482,  0.9564,  1.0000,  1.0410,  1.1824,  1.2449, \n",
      "\n",
      "[104] => ::: 0.8282,  0.8530,  0.9483,  1.0000,  1.0515,  1.1759,  1.2531, \n",
      "\n",
      "[105] => ::: 0.5690,  0.4835,  1.2138,  1.0000,  0.9037,  1.2852,  1.1432, \n",
      "\n",
      "[106] => ::: 0.6866,  0.6694,  1.0628,  1.0000,  0.9609,  1.2078,  1.1517, \n",
      "\n",
      "[107] => ::: 0.7685,  0.7584,  1.0294,  1.0000,  0.9788,  1.1737,  1.1471, \n",
      "\n",
      "[108] => ::: 0.7898,  0.7851,  1.0079,  1.0000,  0.9939,  1.1921,  1.1869, \n",
      "\n",
      "[109] => ::: 0.7776,  0.7854,  0.9863,  1.0000,  1.0118,  1.2200,  1.2391, \n",
      "\n",
      "[110] => ::: 0.7661,  0.7568,  1.0234,  1.0000,  0.9831,  1.1825,  1.1617, \n",
      "\n",
      "[111] => ::: 0.7886,  0.7867,  1.0011,  1.0000,  0.9994,  1.2016,  1.2040, \n",
      "\n",
      "[112] => ::: 0.7829,  0.7923,  0.9833,  1.0000,  1.0144,  1.2181,  1.2411, \n",
      "\n",
      "[113] => ::: 0.7844,  0.7825,  1.0034,  1.0000,  0.9981,  1.1950,  1.1950, \n",
      "\n",
      "[114] => ::: 0.7844,  0.7871,  0.9935,  1.0000,  1.0057,  1.2023,  1.2132, \n",
      "\n",
      "[115] => ::: 0.7617,  0.7711,  0.9882,  1.0000,  1.0104,  1.2021,  1.2173, \n",
      "\n",
      "[116] => ::: 0.7829,  0.7930,  0.9809,  1.0000,  1.0163,  1.1993,  1.2242, \n",
      "\n"
     ]
    }
   ],
   "source": [
    "for stxs, nom_int in zip(enumerate(stxs_bins), int_nominal):\n",
    "    k = stxs[0]; stxs_bin = stxs[1]\n",
    "    print('[{}] => :::'.format(int(stxs_bin)), end='')\n",
    "    for i in range(9):\n",
    "        if (i==2) | (i==6):\n",
    "            continue\n",
    "        norm_stxs = nom_int/(int_scale[i][k])\n",
    "        print(' {0:.4f}, '.format(norm_stxs), end='')\n",
    "    print('\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "integrals = {i: {int(k): [] for k in stxs_bins} for i in range(9) if ((i!= 2) & (i!= 6))}\n",
    "for stxs, nom_int in zip(enumerate(stxs_bins), int_nominal):\n",
    "    k = stxs[0]; stxs_bin = int(stxs[1])\n",
    "    for i in range(9):\n",
    "        if (i==2) | (i==6):\n",
    "            continue\n",
    "        int_varied = int_scale[i][k]\n",
    "        integrals[i][stxs_bin] = (int_varied)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# for i in range(9):\n",
    "#     if (i==2) | (i==6):\n",
    "#         continue\n",
    "\n",
    "#     integrals[i][104] += integrals[i][105]\n",
    "#     integrals[i][104] += integrals[i][106]\n",
    "#     integrals[i][104] += integrals[i][107]\n",
    "    \n",
    "#     integrals[i][108] += integrals[i][109]\n",
    "#     integrals[i][108] += integrals[i][110]\n",
    "#     integrals[i][108] += integrals[i][111]\n",
    "#     integrals[i][108] += integrals[i][101]\n",
    "#     integrals[i][108] += integrals[i][102]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "yields = open(\"Yields_1p2.txt\",\"w+\")\n",
    "yields.write('STXS\\t\\t DD \\t\\t DN \\t\\t ND \\t\\t NN \\t\\t NU \\t\\t UN \\t\\t UU')\n",
    "yields.write('\\n')\n",
    "\n",
    "for stxs, nom_int in zip(enumerate(stxs_bins), int_nominal):\n",
    "    k = stxs[0]; stxs_bin = int(stxs[1])\n",
    "    yields.write('{} \\t'.format(int(stxs_bin)))\n",
    "    for i in range(9):\n",
    "        if (i==2) | (i==6):\n",
    "            continue\n",
    "        yields.write('\\t{0:.2f},\\t ' .format(integrals[i][stxs_bin]))\n",
    "    yields.write('\\n')\n",
    "yields.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "normalization = open(\"Normalization_1p2.txt\",\"w+\")\n",
    "normalization.write('STXS\\t\\t DD \\t\\t DN \\t\\t ND \\t\\t NN \\t\\t NU \\t\\t UN \\t\\t UU')\n",
    "normalization.write('\\n')\n",
    "for stxs, nom_int in zip(enumerate(stxs_bins), int_nominal):\n",
    "    k = stxs[0]; stxs_bin = int(stxs[1])\n",
    "    normalization.write('{} \\t'.format(int(stxs_bin)))\n",
    "    for i in range(9):\n",
    "        if (i==2) | (i==6):\n",
    "            continue\n",
    "        norm_int = nom_int/integrals[i][stxs_bin]\n",
    "        normalization.write('\\t{0:.2f},\\t ' .format(norm_int))\n",
    "    normalization.write('\\n')\n",
    "normalization.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "histos = {}\n",
    "for i in range(9):\n",
    "    if (i==2) | (i==6):\n",
    "        continue    \n",
    "    values = []\n",
    "    for stxs, nom_int in zip(enumerate(stxs_bins), int_nominal):\n",
    "        k = stxs[0]; stxs_bin = int(stxs[1])\n",
    "        norm_int = nom_int/integrals[i][stxs_bin]\n",
    "        values.append(norm_int)\n",
    "    histos[i] = values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "scale_var = {\n",
    "    0: r'$\\mu_r$ = 0.5, $\\mu_f$ = 0.5',\n",
    "    1: r'$\\mu_r$ = 0.5, $\\mu_f$ = 1.0',\n",
    "    3: r'$\\mu_r$ = 1.0, $\\mu_f$ = 0.5',\n",
    "    4: r'$\\mu_r$ = 1.0, $\\mu_f$ = 1.0',\n",
    "    5: r'$\\mu_r$ = 1.0, $\\mu_f$ = 2.0',\n",
    "    7: r'$\\mu_r$ = 2.0, $\\mu_f$ = 1.0',\n",
    "    8: r'$\\mu_r$ = 2.0, $\\mu_f$ = 2.0'\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x750 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (6,5))\n",
    "cmap = plt.cm.coolwarm(np.linspace(0, 1., 9.))\n",
    "for i in range(9):\n",
    "    if (i==2) | (i==6):\n",
    "        continue\n",
    "    plt.step(stxs_bins, histos[i], where = 'mid', label = scale_var[i], color = cmap[i], linewidth = 1.5)\n",
    "    plt.grid(b = None)\n",
    "plt.legend(bbox_to_anchor=(1., 0.9))\n",
    "plt.xticks(stxs_bins, rotation = 45)\n",
    "plt.tick_params(which='minor', length=0)\n",
    "plt.ylim(0.5, 1.5)\n",
    "plt.title('STXS 1.2 bins uncertainties')\n",
    "plt.xlabel('STXS 1.2 Bin', ha='right', x=1.0, fontsize = 12)\n",
    "plt.ylabel(r'$\\sigma_{nom}/\\sigma_{var}$', ha='right', y=1.0, fontsize = 12)\n",
    "# plt.savefig('STXS1p2_Unc.pdf', bbox_inches='tight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}