{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "8109cc2b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-09-12T22:42:48.072732Z",
     "start_time": "2023-09-12T22:42:47.716626Z"
    }
   },
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "\n",
    "import awkward as ak\n",
    "import numpy as np\n",
    "\n",
    "import h5py as h5\n",
    "\n",
    "import numba as nb\n",
    "\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import mplhep as hep\n",
    "plt.style.use(hep.style.CMS)\n",
    "hep.style.use(\"CMS\")\n",
    "plt.rcParams.update({\"font.size\": 16})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f5491f26",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-09-12T22:52:43.225350Z",
     "start_time": "2023-09-12T22:52:43.222740Z"
    }
   },
   "outputs": [],
   "source": [
    "proj_dir = Path.cwd().parent.parent\n",
    "\n",
    "test_file = proj_dir.joinpath('reports/bv2/hhh_test.h5')\n",
    "pred_file = proj_dir.joinpath('reports/bv2/pred_v22.h5')\n",
    "\n",
    "plot_dir = proj_dir.joinpath('reports/bv2/figures')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "8f51df5f",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-09-12T22:52:58.120492Z",
     "start_time": "2023-09-12T22:52:58.004623Z"
    }
   },
   "outputs": [],
   "source": [
    "testfile = h5.File(test_file)\n",
    "predfile = h5.File(pred_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e073d10c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-09-12T22:53:15.289153Z",
     "start_time": "2023-09-12T22:53:15.285287Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KeysViewHDF5 ['b1', 'b2', 'mask', 'pt']>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testfile['TARGETS']['h1'].keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6d6876c5",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2023-09-12T22:54:56.866346Z",
     "start_time": "2023-09-12T22:54:56.861623Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KeysViewHDF5 ['assignment_probability', 'bb', 'detection_probability', 'marginal_probability']>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predfile['TARGETS']['bh2'].keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "799e22ea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<KeysViewHDF5 ['MASK', 'fj_charge', 'fj_chargedenergyfrac', 'fj_cosphi', 'fj_ehadovereem', 'fj_eta', 'fj_mass', 'fj_ncharged', 'fj_neutralenergyfrac', 'fj_nneutral', 'fj_phi', 'fj_pt', 'fj_sdmass', 'fj_sinphi', 'fj_tau21', 'fj_tau32']>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testfile['INPUTS']['BoostedJets'].keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2c80bac3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0,  6,  6, 10, 10,  4,  2,  6, 11,  7,  5, 10,  1,  6,  7, 10, 10,\n",
       "        6, 10,  0, 10,  5,  1,  1,  6,  4, 12,  3, 11, 10, 11, 10, 10,  4,\n",
       "       10, 10, 10, 10, 10,  4,  8,  0,  1,  1,  6, 10,  6, 10,  2,  1,  0,\n",
       "       11,  3,  7,  3,  9, 10,  5, 10,  4,  5,  0,  6,  0, 11, 12,  0, 11,\n",
       "        6,  0, 11,  2,  2, 10, 11,  8,  0,  3,  7, 10,  1, 10,  4, 11, 10,\n",
       "        3, 10,  6,  0,  5, 10, 10,  0,  1, 11, 10,  0,  0,  1,  5])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predfile[\"TARGETS\"][\"bh1\"]['bb'][0:100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "cd50151f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([False, False,  True, False, False, False, False, False,  True,\n",
       "        True, False, False, False, False, False, False,  True, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False,  True,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "        True, False, False, False, False,  True, False, False, False,\n",
       "       False, False, False, False, False, False, False, False, False,\n",
       "       False])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testfile[\"TARGETS\"][\"bh1\"]['mask'][0:100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e96fcc84",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-1, -1,  0, -1, -1, -1, -1, -1,  0,  1, -1, -1, -1, -1, -1, -1,  0,\n",
       "       -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,\n",
       "       -1,  0, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,\n",
       "       -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,\n",
       "       -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1,  1, -1, -1, -1,\n",
       "       -1,  0, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "testfile[\"TARGETS\"][\"bh1\"]['bb'][0:100]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed6afa9d",
   "metadata": {},
   "source": [
    "### Test if indices in pred bb are out of boundary for target mask =True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5b380738",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0  1  2  3  4  5  6  7  8  9 10 11 12]\n"
     ]
    }
   ],
   "source": [
    "print(np.unique(predfile[\"TARGETS\"][\"bh3\"]['bb'][:]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "25bd40f5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[-1  0  1  2]\n"
     ]
    }
   ],
   "source": [
    "print(np.unique(testfile[\"TARGETS\"][\"bh3\"]['bb'][:]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e63c807",
   "metadata": {},
   "source": [
    "### OptionL: examaine pt order"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96c44c80",
   "metadata": {},
   "source": [
    "### Obtain python arrays for calculating purity and efficiency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "dc664b0b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Collect H pt, mask, target and predicted jet and fjets for 3 Hs in each event\n",
    "# h pt\n",
    "h1_pt = np.array(testfile['TARGETS']['h1']['pt'])\n",
    "h2_pt = np.array(testfile['TARGETS']['h2']['pt'])\n",
    "h3_pt = np.array(testfile['TARGETS']['h3']['pt'])\n",
    "\n",
    "bh1_pt = np.array(testfile['TARGETS']['bh1']['pt'])\n",
    "bh2_pt = np.array(testfile['TARGETS']['bh2']['pt'])\n",
    "bh3_pt = np.array(testfile['TARGETS']['bh3']['pt'])\n",
    "\n",
    "# mask\n",
    "h1_mask = np.array(testfile['TARGETS']['h1']['mask'])\n",
    "h2_mask = np.array(testfile['TARGETS']['h2']['mask'])\n",
    "h3_mask = np.array(testfile['TARGETS']['h3']['mask'])\n",
    "\n",
    "bh1_mask = np.array(testfile['TARGETS']['bh1']['mask'])\n",
    "bh2_mask = np.array(testfile['TARGETS']['bh2']['mask'])\n",
    "bh3_mask = np.array(testfile['TARGETS']['bh3']['mask'])\n",
    "\n",
    "# target jet/fjets\n",
    "b1_h1_t = np.array(testfile[\"TARGETS\"][\"h1\"]['b1'])\n",
    "b1_h2_t = np.array(testfile[\"TARGETS\"][\"h2\"]['b1'])\n",
    "b1_h3_t = np.array(testfile[\"TARGETS\"][\"h3\"]['b1'])\n",
    "\n",
    "b2_h1_t = np.array(testfile[\"TARGETS\"][\"h1\"]['b2'])\n",
    "b2_h2_t = np.array(testfile[\"TARGETS\"][\"h2\"]['b2'])\n",
    "b2_h3_t = np.array(testfile[\"TARGETS\"][\"h3\"]['b2'])\n",
    "\n",
    "bb_bh1_t = np.array(testfile[\"TARGETS\"][\"bh1\"]['bb'])\n",
    "bb_bh2_t = np.array(testfile[\"TARGETS\"][\"bh2\"]['bb'])\n",
    "bb_bh3_t = np.array(testfile[\"TARGETS\"][\"bh3\"]['bb'])\n",
    "\n",
    "# pred jet/fjets\n",
    "b1_h1_p = np.array(predfile[\"TARGETS\"][\"h1\"]['b1'])\n",
    "b1_h2_p = np.array(predfile[\"TARGETS\"][\"h2\"]['b1'])\n",
    "b1_h3_p = np.array(predfile[\"TARGETS\"][\"h3\"]['b1'])\n",
    "\n",
    "b2_h1_p = np.array(predfile[\"TARGETS\"][\"h1\"]['b2'])\n",
    "b2_h2_p = np.array(predfile[\"TARGETS\"][\"h2\"]['b2'])\n",
    "b2_h3_p = np.array(predfile[\"TARGETS\"][\"h3\"]['b2'])\n",
    "\n",
    "bb_bh1_p = np.array(predfile[\"TARGETS\"][\"bh1\"]['bb'])\n",
    "bb_bh2_p = np.array(predfile[\"TARGETS\"][\"bh2\"]['bb'])\n",
    "bb_bh3_p = np.array(predfile[\"TARGETS\"][\"bh3\"]['bb'])\n",
    "\n",
    "\n",
    "# fatjet assignment probability\n",
    "ap_bh1 = np.array(predfile[\"TARGETS\"][\"bh1\"]['assignment_probability'])\n",
    "ap_bh2 = np.array(predfile[\"TARGETS\"][\"bh2\"]['assignment_probability'])\n",
    "ap_bh3 = np.array(predfile[\"TARGETS\"][\"bh3\"]['assignment_probability'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "a7385080",
   "metadata": {},
   "outputs": [],
   "source": [
    "fj_pts = np.array(predfile[\"INPUTS\"][\"BoostedJets\"][\"fj_pt\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "188d9d1c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(ap_bh1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "bcaf5554",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(ap_bh2)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea3fc91f",
   "metadata": {},
   "source": [
    "# Calculating efficiency (denominator should be pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "93177b60",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(ap_bh3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72173996",
   "metadata": {},
   "source": [
    "### As prediction h1 can be matched to target h1, 2, 3,  I will do things in the event level first"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "a8871464",
   "metadata": {},
   "outputs": [],
   "source": [
    "# convert some arrays to ak array\n",
    "aps = np.concatenate((ap_bh1.reshape(-1,1), ap_bh2.reshape(-1,1), ap_bh3.reshape(-1,1)), axis=1)\n",
    "aps = ak.Array(aps)\n",
    "bb_ps = np.concatenate((bb_bh1_p.reshape(-1,1), bb_bh2_p.reshape(-1,1), bb_bh3_p.reshape(-1,1)), axis=1)\n",
    "bb_ps = ak.Array(bb_ps)\n",
    "bb_ts = np.concatenate((bb_bh1_t.reshape(-1,1), bb_bh2_t.reshape(-1,1), bb_bh3_t.reshape(-1,1)), axis=1)\n",
    "bb_ts = ak.Array(bb_ts)\n",
    "fj_pts = ak.Array(fj_pts)\n",
    "\n",
    "# p: prediction\n",
    "AP_threshold = 0.5\n",
    "ap_filter = aps > AP_threshold\n",
    "bb_ps_passed = bb_ps.mask[ap_filter]\n",
    "bb_ps_passed = ak.drop_none(bb_ps_passed)\n",
    "\n",
    "aps_passed = aps.mask[ap_filter]\n",
    "aps_passed = ak.drop_none(aps_passed)\n",
    "\n",
    "sort_by_ap = ak.argsort(aps_passed, axis=-1, ascending=False)\n",
    "bb_ps_passed = bb_ps_passed[sort_by_ap]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "147f9f15",
   "metadata": {},
   "outputs": [],
   "source": [
    "bh_effs = []\n",
    "# for each event\n",
    "for bb_p_event, bb_t_event, fj_pt_event in zip(bb_ps_passed, bb_ts, fj_pts):\n",
    "    # for each predicted fatjet, check if the targets have a t fatject same with the p fatjet\n",
    "    for bb_p in bb_p_event:\n",
    "        match = 0\n",
    "        for bb_t in bb_t_event:\n",
    "            if bb_p == bb_t+10:\n",
    "                match = 1\n",
    "        bh_effs.append([fj_pt_event[bb_t], match])\n",
    "bh_effs = np.array(bh_effs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "0767ca2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# set x axis (pT) of the scattered points\n",
    "bins = np.arange(200, 1000, 100)\n",
    "bin_centers = [(bins[i]+bins[i+1])/2 for i in range(bins.size-1)]\n",
    "\n",
    "# group points into bins by fatjet pT\n",
    "eff_inds = np.digitize(bh_effs[:,0], bins)\n",
    "\n",
    "#np.concatenate((ap_bh1.reshape(-1,1), ap_bh2.reshape(-1,1), ap_bh3.reshape(-1,1)), axis=1)\n",
    "effs_per_bin = []\n",
    "for bin_i in range(1, len(bins)):\n",
    "    effs_per_bin.append(bh_effs[:,1][eff_inds==bin_i])\n",
    "effs_per_bin = ak.Array(effs_per_bin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "26627862",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1660.0 out of 5273 assignment are same with target in bin centered at 250.0 GeV\n",
      "1601.0 out of 2767 assignment are same with target in bin centered at 350.0 GeV\n",
      "850.0 out of 1345 assignment are same with target in bin centered at 450.0 GeV\n",
      "381.0 out of 584 assignment are same with target in bin centered at 550.0 GeV\n",
      "197.0 out of 309 assignment are same with target in bin centered at 650.0 GeV\n",
      "92.0 out of 157 assignment are same with target in bin centered at 750.0 GeV\n",
      "54.0 out of 91 assignment are same with target in bin centered at 850.0 GeV\n"
     ]
    }
   ],
   "source": [
    "for effs, bin_c in zip(effs_per_bin, bin_centers):\n",
    "    print(f\"{ak.sum(effs)} out of {ak.count(effs)} assignment are same with target in bin centered at {bin_c} GeV\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "63c11585",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "ax.scatter(x=bin_centers, y=ak.mean(effs_per_bin, axis=-1))\n",
    "ax.set(xlabel=r\"reco H pT\", ylabel=r\"Matching efficiency\", title=\"SPANet Boosted H Matching Efficiency vs. Reco H pT\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(f\"{str(plot_dir)}/test_efficiency.jpg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "39798763",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre>[0.464,\n",
       " 0.494,\n",
       " 0.482,\n",
       " 0.476,\n",
       " 0.481,\n",
       " 0.493,\n",
       " 0.491]\n",
       "-----------------\n",
       "type: 7 * float64</pre>"
      ],
      "text/plain": [
       "<Array [0.464, 0.494, 0.482, ..., 0.481, 0.493, 0.491] type='7 * float64'>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ak.std(effs_per_bin, axis=-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e904e58c",
   "metadata": {},
   "source": [
    "# Calculating purity (denominator should be target)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "cc951b4b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# get the truth level boosted higgs mask\n",
    "bh_masks = np.concatenate((bh1_mask.reshape(-1,1), bh2_mask.reshape(-1,1), bh3_mask.reshape(-1,1)), axis=1)\n",
    "bh_masks = ak.Array(bh_masks)\n",
    "\n",
    "# applying masks to the target bh's bb indices:\n",
    "bb_ts_selected = bb_ts.mask[bh_masks]\n",
    "bb_ts_selected = ak.drop_none(bb_ts_selected)\n",
    "\n",
    "bh_pts = np.concatenate((bh1_pt.reshape(-1,1), bh2_pt.reshape(-1,1), bh3_pt.reshape(-1,1)), axis=1)\n",
    "bh_pts = ak.Array(bh_pts)\n",
    "bh_selected_pts = bh_pts.mask[bh_masks]\n",
    "bh_selected_pts = ak.drop_none(bh_selected_pts)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "d81a1e8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "bh_purs = []\n",
    "# for each event\n",
    "for bb_t_event, bb_p_event, bh_pt_event in zip(bb_ts_selected, bb_ps_passed, bh_selected_pts):\n",
    "    # for each target fatjet, check if the predictions have a p fatject same with the t fatjet\n",
    "    for i, bb_t in enumerate(bb_t_event):\n",
    "        match = 0\n",
    "        for bb_p in bb_p_event:\n",
    "            if bb_p == bb_t+10:\n",
    "                match = 1\n",
    "        bh_purs.append([bh_pt_event[i], match])\n",
    "bh_purs = np.array(bh_purs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0eeece8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# group points into bins by fatjet pT\n",
    "pur_inds = np.digitize(bh_purs[:,0], bins)\n",
    "\n",
    "# diepense (gen_H_pT, purity) points into bins\n",
    "purs_per_bin = []\n",
    "for bin_i in range(1, len(bins)):\n",
    "    purs_per_bin.append(bh_purs[:,1][pur_inds==bin_i])\n",
    "purs_per_bin = ak.Array(purs_per_bin)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a869b509",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1527.0 out of 1530 target assignments are matched to >=1 predicted assignment in bin centered at 250.0 GeV\n",
      "3869.0 out of 3885 target assignments are matched to >=1 predicted assignment in bin centered at 350.0 GeV\n",
      "1847.0 out of 1849 target assignments are matched to >=1 predicted assignment in bin centered at 450.0 GeV\n",
      "716.0 out of 721 target assignments are matched to >=1 predicted assignment in bin centered at 550.0 GeV\n",
      "309.0 out of 310 target assignments are matched to >=1 predicted assignment in bin centered at 650.0 GeV\n",
      "140.0 out of 140 target assignments are matched to >=1 predicted assignment in bin centered at 750.0 GeV\n",
      "56.0 out of 56 target assignments are matched to >=1 predicted assignment in bin centered at 850.0 GeV\n"
     ]
    }
   ],
   "source": [
    "for purs, bin_c in zip(purs_per_bin, bin_centers):\n",
    "    print(f\"{ak.sum(purs)} out of {ak.count(purs)} target assignments are matched to >=1 predicted assignment in bin centered at {bin_c} GeV\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "73f8134f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(8,6))\n",
    "ax.errorbar(x=bin_centers, y=ak.mean(purs_per_bin, axis=-1),\\\n",
    "            xerr = (bins[1]-bins[0])/2*np.ones(bins.shape[0]-1),\\\n",
    "            yerr=1/np.sqrt(ak.count(purs_per_bin, axis=-1).to_numpy()),\\\n",
    "           fmt='o', capsize=5)\n",
    "ax.set(xlabel=r\"gen H pT\", ylabel=r\"Matching purity\", title=\"SPANet Boosted H Matching purity vs. gen H pT\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(f\"{str(plot_dir)}/test_purity.jpg\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "001510c6",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}