BTA_Example.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import uproot\n",
    "import awkward\n",
    "import uproot_methods\n",
    "import numpy as np\n",
    "\n",
    "from coffea import hist, processor\n",
    "from coffea.hist import plot\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import mplhep as hep"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "63192.0"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pat = 'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2016/9_4_X_v04/QCD_Pt-800to1000_MuEnrichedPt5_TuneCUETP8M1_13TeV_pythia8_9_4_X_v04/'\n",
    "fi = 'JetTree_mc_FatJets_Subjets_BoostedCommissioning_2.root'\n",
    "f = uproot.open(pat+fi)\n",
    "\n",
    "f[\"allEvents/hEventCount\"].values[0]\n",
    "#f['otree'].keys()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def ja(df, counts, prefix, bnames):\n",
    "    objp4 = awkward.JaggedArray.fromcounts(df[counts],\n",
    "        uproot_methods.TLorentzVectorArray.from_ptetaphim(\n",
    "            df[prefix + 'pt'].content,\n",
    "            df[prefix + 'eta'].content,\n",
    "            df[prefix + 'phi'].content,\n",
    "            df[prefix + 'mass'].content if prefix + 'mass' in [br.decode('ASCII') for br in df.available] else df[prefix + 'pt'].zeros_like().content,\n",
    "        )\n",
    "    )\n",
    "    branches = {k: df[prefix.split('.')[0]+'.'+v] for k,v in bnames.items()}\n",
    "    obj = awkward.JaggedArray.zip(p4=objp4, **branches)\n",
    "    obj['mass'] = obj['p4'].mass\n",
    "    obj['pt'] = obj['p4'].pt\n",
    "    obj['phi'] = obj['p4'].phi\n",
    "    obj['eta'] = obj['p4'].eta\n",
    "    return obj\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "class JetProcessor(processor.ProcessorABC):        \n",
    "    def __init__(self):\n",
    "        dataset_axis = hist.Cat(\"dataset\", \"Dataset\")        \n",
    "        pthat_axis = hist.Bin(\"pthat\", r\"$p_T$\", 40,150,1200)\n",
    "        \n",
    "        axes = {\n",
    "            'mass' : hist.Bin(\"mass\", r\"$m_{jet}$ [GeV]\", 50, 0, 300),\n",
    "            'pt' : hist.Bin(\"pt\", r\"$p_T$\", 20,150,1200),\n",
    "            'eta' : hist.Bin(\"eta\", r\"$\\eta$\", 20, -3,3),\n",
    "            'flav' : hist.Bin(\"flav\", r\"flavour\", [0,1,2,3,4,5,6]),\n",
    "        }\n",
    "        \n",
    "        _hist_dict = {\n",
    "            'pthat': hist.Hist(\"Counts\", dataset_axis, pthat_axis),\n",
    "            'mass': hist.Hist(\"Counts\", dataset_axis, axes['mass'], axes['flav']),\n",
    "            'mass_wcuts': hist.Hist(\"Counts\", dataset_axis, axes['mass'], axes['flav']),\n",
    "            \n",
    "            'flav': hist.Hist(\"Counts\", dataset_axis, axes['flav']),\n",
    "            \n",
    "        }   \n",
    "        self._accumulator = processor.dict_accumulator(_hist_dict)\n",
    "        \n",
    "    @property\n",
    "    def accumulator(self):\n",
    "        return self._accumulator\n",
    "    \n",
    "    def process(self, df):\n",
    "        output = self.accumulator.identity()\n",
    "        output['sumw'] = processor.dict_accumulator()\n",
    "        \n",
    "        dataset = df['dataset']    \n",
    "        pthat = df['pthat']        \n",
    "                                                \n",
    "        subjets = ja(df, 'SoftDropPuppiSubJetInfo.nJet', 'SoftDropPuppiSubJetInfo.Jet_', \n",
    "                        {\n",
    "                            'nsv':'Jet_SV_multi',\n",
    "                        }\n",
    "                    )\n",
    "\n",
    "        muons = ja(df, 'SoftDropPuppiSubJetInfo.nPFMuon', 'SoftDropPuppiSubJetInfo.PFMuon_', \n",
    "                        {\n",
    "                            'subjetIdx' : 'PFMuon_IdxJet',\n",
    "                            \n",
    "                            'isGlobal'  : 'PFMuon_isGlobal',                            \n",
    "                        }\n",
    "                    )\n",
    "        \n",
    "        jets = ja(df, 'FatJetInfo.nJet', 'FatJetInfo.Jet_', \n",
    "                        {\n",
    "                            'nsj'   : 'Jet_nSubJets_SoftDropPuppi',\n",
    "                            'firstsj' : 'Jet_nFirstSJ_SoftDropPuppi',\n",
    "\n",
    "                            'doubleb':'Jet_DoubleSV',\n",
    "                            'hflav' : 'Jet_flavour',\n",
    "                            'bhad'  : 'Jet_nbHadrons',\n",
    "                            'chad'  : 'Jet_ncHadrons',\n",
    "                            'mass_prun' : 'Jet_massPruned',\n",
    "                        }\n",
    "                    )\n",
    "        \n",
    "        # Build subjets\n",
    "        jnsj = df['FatJetInfo.Jet_nSubJets_SoftDropPuppi']\n",
    "        sjidx = df['FatJetInfo.SubJetIdx_SoftDropPuppi']\n",
    "        # build double-jagged index array\n",
    "        subjetidx = awkward.JaggedArray.fromoffsets(jnsj.offsets, awkward.JaggedArray.fromcounts(jnsj.content, sjidx.content))\n",
    "        # take advantage of broadcasting to turn into absolute index\n",
    "        subjetidx = (subjetidx + subjets.starts).flatten().flatten()\n",
    "        # use indexed array into actual subjet content\n",
    "        jet_subjet_content = awkward.IndexedArray(subjetidx, subjets.content)\n",
    "        # re-build double-jagged array now with content\n",
    "        jet_subjet = awkward.JaggedArray.fromoffsets(jnsj.offsets, awkward.JaggedArray.fromcounts(jnsj.content, jet_subjet_content))\n",
    "        jets['subjets'] = jet_subjet\n",
    "\n",
    "        # Build muons\n",
    "        # convert to absolute index\n",
    "        muon_absjetidx = (muons['subjetIdx'] + subjets.starts).flatten()\n",
    "        # use np.unique to get the number of muons associated with each subjet\n",
    "        subjet_nmuons = np.zeros(subjets.content.size, dtype=subjets.offsets.dtype)\n",
    "        subjets_wmuons, jet_muoncounts = np.unique(muon_absjetidx, return_counts=True)\n",
    "        subjet_nmuons[subjets_wmuons] = jet_muoncounts\n",
    "        # build the indexed table\n",
    "        subjet_muons = awkward.JaggedArray.fromoffsets(subjets.offsets, awkward.JaggedArray.fromcounts(subjet_nmuons, muons.content))\n",
    "        subjet_muons['dR'] =  subjet_muons['p4'].delta_r(subjets['p4'])\n",
    "        subjets['muons'] = subjet_muons      \n",
    "      \n",
    "        # Cuts\n",
    "        per_sj = ((jets['subjets']['nsv'] >=1).sum() >=2)\n",
    "        per_mu = (((jets['subjets']['muons']['pt'] >=2 ).sum() >= 1).sum() >=2)\n",
    "          \n",
    "        # Define flavors\n",
    "        _f_gbb = (abs(jets['hflav']) == 5) & (jets['bhad'] >= 2)\n",
    "        _f_gcc = (abs(jets['hflav']) == 4) & (jets['bhad'] == 0) & (jets['chad'] >= 2)\n",
    "        _f_b = (abs(jets['hflav']) == 5) & np.invert(_f_gbb)\n",
    "        _f_c = (abs(jets['hflav']) == 4) & np.invert(_f_gcc)\n",
    "        _f_l = (abs(jets['hflav']) < 4) | (abs(jets['hflav']) == 21)  #& (jets.bhad == 0) & (jets.chad == 0)\n",
    "        _f = _f_l*1 + _f_c*2 + _f_b*3 + _f_gcc*4 + _f_gbb*5\n",
    "        jets['flav'] = _f        \n",
    "                \n",
    "        output['pthat'].fill(dataset=dataset, pthat=pthat)\n",
    "        output['flav'].fill(dataset=dataset, flav=jets['flav'].flatten())\n",
    "        output['mass'].fill(dataset=dataset, mass=jets['mass'].flatten(), flav=jets['flav'].flatten())\n",
    "        \n",
    "        return output\n",
    "\n",
    "    def postprocess(self, accumulator):\n",
    "        return accumulator\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228': [...],\n",
      " 'QCD_Pt-170to300_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228': [...]}\n"
     ]
    }
   ],
   "source": [
    "json_dict = {'QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228': ['root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_1.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_10.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_100.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_101.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_102.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_103.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_104.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-1000toInf_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_105.root',\n",
    "                                                                           ],\n",
    "    'QCD_Pt-170to300_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228': ['root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-170to300_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_1.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-170to300_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_104.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-170to300_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_105.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-170to300_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_107.root',\n",
    "  'root://xrootd-cms.infn.it///eos/cms/store/group/phys_btag/BoostedBTag/BTagNTuples/2017/9_4_X_v04/QCD_Pt-170to300_MuEnrichedPt5_TuneCP5_13TeV_pythia8_v04_20190228/JetTree_mc_FatJets_Subjets_BoostedCommissioning_108.root',\n",
    "                                                                        ]\n",
    "}\n",
    "\n",
    "from pprint import pprint\n",
    "pprint(json_dict, depth=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Preprocessing: 100%|██████████| 2/2 [00:01<00:00,  1.73it/s]\n",
      "Processing: 100%|██████████| 13/13 [00:25<00:00,  1.93s/items]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "({'pthat': <Hist (dataset,pthat) instance at 0x2aaca3e0c668>, 'mass': <Hist (dataset,mass,flav) instance at 0x2aaca3de0470>, 'flav': <Hist (dataset,flav) instance at 0x2aaca3de0320>, 'sumw': {}}, {'bytesread': value_accumulator(int, 172556288), 'dataservers': set_accumulator({'p06636710y50388.cern.ch:1095', 'p06253947u35885.cern.ch:1095', 'p06636710d12165.cern.ch:1095', 'p06636710k42552.cern.ch:1095', 'p06636710j45805.cern.ch:1095', 'p06636710y63608.cern.ch:1095', 'p06636710t83205.cern.ch:1095', 'p06636710y67710.cern.ch:1095', 'p05614920c84527.cern.ch:1095', 'p06253937e38526.cern.ch:1095', 'p06253937k46044.cern.ch:1095', 'p06636710p77867.cern.ch:1095'}), 'columns': set_accumulator({'FatJetInfo.Jet_massPruned', 'FatJetInfo.Jet_flavour', 'FatJetInfo.Jet_nbHadrons', 'FatJetInfo.Jet_phi', 'SoftDropPuppiSubJetInfo.nPFMuon', 'FatJetInfo.Jet_ncHadrons', 'SoftDropPuppiSubJetInfo.Jet_eta', 'SoftDropPuppiSubJetInfo.Jet_SV_multi', 'FatJetInfo.Jet_eta', 'FatJetInfo.nJet', 'FatJetInfo.Jet_mass', 'SoftDropPuppiSubJetInfo.nJet', 'FatJetInfo.Jet_pt', 'FatJetInfo.Jet_nSubJets_SoftDropPuppi', 'SoftDropPuppiSubJetInfo.Jet_pt', 'SoftDropPuppiSubJetInfo.PFMuon_phi', 'FatJetInfo.Jet_nFirstSJ_SoftDropPuppi', 'SoftDropPuppiSubJetInfo.PFMuon_eta', 'SoftDropPuppiSubJetInfo.PFMuon_IdxJet', 'SoftDropPuppiSubJetInfo.PFMuon_isGlobal', 'FatJetInfo.SubJetIdx_SoftDropPuppi', 'SoftDropPuppiSubJetInfo.PFMuon_pt', 'SoftDropPuppiSubJetInfo.Jet_mass', 'SoftDropPuppiSubJetInfo.Jet_phi', 'FatJetInfo.Jet_DoubleSV', 'pthat'}), 'entries': value_accumulator(int, 580710), 'processtime': value_accumulator(float, 175.8769428730011)})\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "tstart = time.time()    \n",
    "\n",
    "output = processor.run_uproot_job(json_dict,\n",
    "                                  treename='btaganaFatJets/ttree',\n",
    "                                  processor_instance=JetProcessor(),\n",
    "                                  executor=processor.futures_executor,\n",
    "                                  #executor=processor.iterative_executor,\n",
    "                                  executor_args={'workers': 20, 'savemetrics': True},\n",
    "                                  chunksize=500000,\n",
    "                                 )\n",
    "\n",
    "elapsed = time.time() - tstart\n",
    "print(output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "histograms = output[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2aacdc178128>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.style.use(hep.style.ROOT)\n",
    "fig, ax, _ = plot.plot1d(histograms['pthat'], overlay='dataset', density=True)\n",
    "ax.legend([\"QCD1\",\"QCD2\"])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}