ATLAS_SigvsBkg_LL-TTvsOther_emu.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c157cd32-2e52-4c7e-9b91-05a41aec33bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.3.5\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CleanJet_pt_1</th>\n",
       "      <th>CleanJet_pt_2</th>\n",
       "      <th>mll</th>\n",
       "      <th>MET_significance</th>\n",
       "      <th>zeta</th>\n",
       "      <th>mjj</th>\n",
       "      <th>detajj</th>\n",
       "      <th>dphijj</th>\n",
       "      <th>m2ljj30</th>\n",
       "      <th>isLL</th>\n",
       "      <th>isMIX</th>\n",
       "      <th>isTT</th>\n",
       "      <th>isBKG</th>\n",
       "      <th>isVBS</th>\n",
       "      <th>notLL</th>\n",
       "      <th>notTT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>53.054813</td>\n",
       "      <td>29.876976</td>\n",
       "      <td>119.088409</td>\n",
       "      <td>1.595703</td>\n",
       "      <td>1.021057</td>\n",
       "      <td>276.330872</td>\n",
       "      <td>3.893066</td>\n",
       "      <td>1.000959</td>\n",
       "      <td>271.210632</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>30.664513</td>\n",
       "      <td>29.299437</td>\n",
       "      <td>128.700623</td>\n",
       "      <td>15.085938</td>\n",
       "      <td>2.046753</td>\n",
       "      <td>495.311188</td>\n",
       "      <td>5.605469</td>\n",
       "      <td>1.686401</td>\n",
       "      <td>239.743408</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>30.099604</td>\n",
       "      <td>28.075436</td>\n",
       "      <td>111.508141</td>\n",
       "      <td>10.007812</td>\n",
       "      <td>0.949219</td>\n",
       "      <td>104.640396</td>\n",
       "      <td>2.666260</td>\n",
       "      <td>0.274885</td>\n",
       "      <td>156.898926</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>91.604919</td>\n",
       "      <td>32.597763</td>\n",
       "      <td>267.004730</td>\n",
       "      <td>8.101562</td>\n",
       "      <td>1.380615</td>\n",
       "      <td>551.808350</td>\n",
       "      <td>4.623047</td>\n",
       "      <td>1.564453</td>\n",
       "      <td>840.582458</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>59.228630</td>\n",
       "      <td>48.039818</td>\n",
       "      <td>81.836113</td>\n",
       "      <td>6.886719</td>\n",
       "      <td>1.799194</td>\n",
       "      <td>628.104187</td>\n",
       "      <td>4.930176</td>\n",
       "      <td>1.513550</td>\n",
       "      <td>791.871887</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32287</th>\n",
       "      <td>158.753632</td>\n",
       "      <td>80.266327</td>\n",
       "      <td>85.201508</td>\n",
       "      <td>18.312500</td>\n",
       "      <td>0.529480</td>\n",
       "      <td>1368.427856</td>\n",
       "      <td>4.981445</td>\n",
       "      <td>2.118530</td>\n",
       "      <td>1592.925781</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32288</th>\n",
       "      <td>190.165131</td>\n",
       "      <td>109.592522</td>\n",
       "      <td>112.974281</td>\n",
       "      <td>44.375000</td>\n",
       "      <td>1.710358</td>\n",
       "      <td>1039.798218</td>\n",
       "      <td>3.931885</td>\n",
       "      <td>1.972290</td>\n",
       "      <td>1177.098145</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32289</th>\n",
       "      <td>68.281357</td>\n",
       "      <td>32.656868</td>\n",
       "      <td>129.935699</td>\n",
       "      <td>3.316406</td>\n",
       "      <td>1.657227</td>\n",
       "      <td>903.292419</td>\n",
       "      <td>5.904785</td>\n",
       "      <td>1.004395</td>\n",
       "      <td>1104.163574</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32290</th>\n",
       "      <td>153.700607</td>\n",
       "      <td>113.567513</td>\n",
       "      <td>316.587463</td>\n",
       "      <td>11.273438</td>\n",
       "      <td>0.893066</td>\n",
       "      <td>2401.952637</td>\n",
       "      <td>5.796875</td>\n",
       "      <td>2.201477</td>\n",
       "      <td>3160.002930</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32291</th>\n",
       "      <td>90.099380</td>\n",
       "      <td>34.271664</td>\n",
       "      <td>97.926270</td>\n",
       "      <td>1.205078</td>\n",
       "      <td>1.165588</td>\n",
       "      <td>419.576477</td>\n",
       "      <td>4.005798</td>\n",
       "      <td>2.561646</td>\n",
       "      <td>521.605835</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>32292 rows × 16 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       CleanJet_pt_1  CleanJet_pt_2         mll  MET_significance      zeta  \\\n",
       "0          53.054813      29.876976  119.088409          1.595703  1.021057   \n",
       "1          30.664513      29.299437  128.700623         15.085938  2.046753   \n",
       "2          30.099604      28.075436  111.508141         10.007812  0.949219   \n",
       "3          91.604919      32.597763  267.004730          8.101562  1.380615   \n",
       "4          59.228630      48.039818   81.836113          6.886719  1.799194   \n",
       "...              ...            ...         ...               ...       ...   \n",
       "32287     158.753632      80.266327   85.201508         18.312500  0.529480   \n",
       "32288     190.165131     109.592522  112.974281         44.375000  1.710358   \n",
       "32289      68.281357      32.656868  129.935699          3.316406  1.657227   \n",
       "32290     153.700607     113.567513  316.587463         11.273438  0.893066   \n",
       "32291      90.099380      34.271664   97.926270          1.205078  1.165588   \n",
       "\n",
       "               mjj    detajj    dphijj      m2ljj30  isLL  isMIX  isTT  isBKG  \\\n",
       "0       276.330872  3.893066  1.000959   271.210632   0.0    0.0   0.0    1.0   \n",
       "1       495.311188  5.605469  1.686401   239.743408   0.0    0.0   0.0    1.0   \n",
       "2       104.640396  2.666260  0.274885   156.898926   0.0    0.0   0.0    1.0   \n",
       "3       551.808350  4.623047  1.564453   840.582458   0.0    0.0   0.0    1.0   \n",
       "4       628.104187  4.930176  1.513550   791.871887   0.0    0.0   0.0    1.0   \n",
       "...            ...       ...       ...          ...   ...    ...   ...    ...   \n",
       "32287  1368.427856  4.981445  2.118530  1592.925781   0.0    0.0   0.0    0.0   \n",
       "32288  1039.798218  3.931885  1.972290  1177.098145   0.0    0.0   0.0    0.0   \n",
       "32289   903.292419  5.904785  1.004395  1104.163574   0.0    0.0   0.0    0.0   \n",
       "32290  2401.952637  5.796875  2.201477  3160.002930   0.0    0.0   0.0    0.0   \n",
       "32291   419.576477  4.005798  2.561646   521.605835   0.0    0.0   0.0    0.0   \n",
       "\n",
       "       isVBS  notLL  notTT  \n",
       "0        0.0    0.0    0.0  \n",
       "1        0.0    0.0    0.0  \n",
       "2        0.0    0.0    0.0  \n",
       "3        0.0    0.0    0.0  \n",
       "4        0.0    0.0    0.0  \n",
       "...      ...    ...    ...  \n",
       "32287    1.0    0.0    0.0  \n",
       "32288    1.0    0.0    0.0  \n",
       "32289    1.0    0.0    0.0  \n",
       "32290    1.0    0.0    0.0  \n",
       "32291    1.0    0.0    0.0  \n",
       "\n",
       "[32292 rows x 16 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "print(pd. __version__)\n",
    "from sklearn.utils import shuffle\n",
    "\n",
    "#df = pd.read_pickle('https://pandora.infn.it/public/24236c/dl/dataset_pol_vbs_full2018v9_VBS_WW_2j_emu_dnn_ATLAS.pkl')\n",
    "df = pd.read_pickle('https://pandora.infn.it/public/3dcf6d/dl/dataset_pol_vbs_full2018v9_VBS_WW_2j_emu_dnn_ATLAS_selection.pkl')\n",
    "pd.set_option('display.max_columns', None)\n",
    "\n",
    "df['notLL'] = df['isTT'] + df['isMIX']\n",
    "df['notTT'] = df['isLL'] + df['isMIX']\n",
    "\n",
    "df_SvB = df[((df['isTT'] + df['isMIX'] + df['isLL']) == 0) & ((df['isVBS'] + df['isBKG']) == 1)]\n",
    "df_SvB = df_SvB[['CleanJet_pt_1', 'CleanJet_pt_2', 'mll', 'MET_significance', 'zeta', 'mjj', 'detajj', 'dphijj', 'm2ljj30', 'isLL', 'isMIX', 'isTT', 'isBKG', 'isVBS', 'notLL', 'notTT']]\n",
    "df_red = df[['CleanJet_pt_1', 'CleanJet_pt_2', 'mll', 'MET_significance', 'zeta', 'mjj', 'detajj', 'dphijj', 'm2ljj30', 'isLL', 'isMIX', 'isTT', 'isBKG', 'isVBS', 'notLL', 'notTT']]\n",
    "df_SvB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b1c30b9d-7706-4fda-8bce-99ffc4a93c08",
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "import numpy as np\n",
    "import tensorflow as tf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "bc26e027-8929-4c05-8995-bc477450a479",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.7.0-rc1\n"
     ]
    }
   ],
   "source": [
    "import tqdm\n",
    "from tensorflow.keras.models import Sequential, Model\n",
    "from tensorflow.keras.layers import Input, Activation, Dense, Dropout, InputLayer\n",
    "from tensorflow.keras.constraints import max_norm\n",
    "from tensorflow.keras import initializers\n",
    "from tensorflow.keras.models import load_model\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "from sklearn.utils import shuffle\n",
    "import copy\n",
    "print(tf.__version__)\n",
    "from tensorflow.python.client import device_lib\n",
    "tf.config.run_functions_eagerly(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "52abbb0d-f65b-408f-9465-36991164434b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "59\n",
      "[[5.30548134e+01 2.98769760e+01 1.19088409e+02 ... 3.89306641e+00\n",
      "  1.00095880e+00 2.71210632e+02]\n",
      " [3.06645126e+01 2.92994366e+01 1.28700623e+02 ... 5.60546875e+00\n",
      "  1.68640137e+00 2.39743408e+02]\n",
      " [3.00996037e+01 2.80754356e+01 1.11508141e+02 ... 2.66625977e+00\n",
      "  2.74884522e-01 1.56898926e+02]\n",
      " ...\n",
      " [6.82813568e+01 3.26568680e+01 1.29935699e+02 ... 5.90478516e+00\n",
      "  1.00439453e+00 1.10416357e+03]\n",
      " [1.53700607e+02 1.13567513e+02 3.16587463e+02 ... 5.79687500e+00\n",
      "  2.20147705e+00 3.16000293e+03]\n",
      " [9.00993805e+01 3.42716637e+01 9.79262695e+01 ... 4.00579834e+00\n",
      "  2.56164551e+00 5.21605835e+02]] [[0. 1.]\n",
      " [0. 1.]\n",
      " [0. 1.]\n",
      " ...\n",
      " [1. 0.]\n",
      " [1. 0.]\n",
      " [1. 0.]]\n"
     ]
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.pyplot import cm\n",
    "import numpy as np\n",
    "import matplotlib as mpl\n",
    "%matplotlib inline\n",
    "mpl.rcParams['figure.dpi'] = 150\n",
    "\n",
    "nclasses = 2\n",
    "features = df.columns[:len(df.columns)-7]\n",
    "\n",
    "NDIM = len(features)\n",
    "print(NDIM)\n",
    "for col in df.columns:\n",
    "    df[col]    = np.hstack(df[col])\n",
    "'''\n",
    "df.replace([np.inf, -np.inf], np.nan, inplace=True)\n",
    "df.dropna(inplace=True)\n",
    "df = shuffle(df)\n",
    "'''\n",
    "# Perform the splitting and define training and validation datasets\n",
    "msk    = np.random.rand(len(df_SvB)) < 0.8\n",
    "\n",
    "df_train    = df_SvB[msk]\n",
    "df_val      = df_SvB[~msk]\n",
    "\n",
    "NDIM = len(df_SvB.columns[:len(df_SvB.columns)-7])\n",
    "X    = df_train.values[:,0:NDIM]\n",
    "Y    = df_train[['isVBS', 'isBKG']].values\n",
    "X_val = df_val.values[:,0:NDIM]\n",
    "Y_val = df_val[['isVBS', 'isBKG']].values\n",
    "\n",
    "print(X,Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2bd8aa6e-8ec9-462c-937d-0430fcdec8da",
   "metadata": {},
   "outputs": [],
   "source": [
    "class SimpleNeuralNetwork ( tf.Module ):\n",
    "    def __init__ (self, nEpochs, learning_rate, N_NODES, n_layers, n_features, n_outputsC=nclasses, ATLAS=False):\n",
    "        self.learning_rate = learning_rate\n",
    "        self.optimizer  = tf.optimizers.Adam (self.learning_rate)\n",
    "        self.nEpochs = nEpochs\n",
    "        self.N_NODES = N_NODES\n",
    "        self.n_layers = n_layers\n",
    "        self.n_features = n_features\n",
    "        self.n_outputsC = n_outputsC\n",
    "        self.ATLAS = ATLAS\n",
    "        self.weights = self.build (self.n_features, self.N_NODES)\n",
    "        \n",
    "                                \n",
    "    # Define the structure of the model\n",
    "    def build (self, n_input, N_NODES):\n",
    "        # initializer = initializers.Ones()\n",
    "\n",
    "        # Classifier model\n",
    "        self.model1 = Sequential()\n",
    "        self.model1.add(Dense (self.N_NODES, activation = 'relu', input_dim  = n_input))\n",
    "        if not self.ATLAS:\n",
    "            for i in range(self.n_layers):\n",
    "                self.model1.add(Dense (self.N_NODES, activation = 'relu', kernel_constraint=max_norm(3)))\n",
    "                self.model1.add(Dropout (0.1))\n",
    "        else:\n",
    "            self.model1.add(Dense (108, activation = 'relu', kernel_constraint=max_norm(3)))\n",
    "            self.model1.add(Dropout (0.1))\n",
    "            self.model1.add(Dense (60, activation = 'relu', kernel_constraint=max_norm(3)))\n",
    "            self.model1.add(Dropout (0.1))\n",
    "        if self.n_outputsC > 1:\n",
    "            self.model1.add(Dense (self.n_outputsC, activation = 'softmax',input_dim = self.N_NODES))\n",
    "        else:\n",
    "            self.model1.add(Dense (self.n_outputsC, activation = 'sigmoid',input_dim = self.N_NODES))\n",
    "        \n",
    "        return self.model1.weights\n",
    "     \n",
    "    # Performs the epochs loop and the actual training.\n",
    "    # Monitors the training and validation loss functions, both for the classifier and the adversary.\n",
    "    # Returns the classifier categorical accuracy.\n",
    "    def fit (self, X, Y, X_val, Y_val, show_loss = False):\n",
    "        losses = []\n",
    "        losses_val = []\n",
    "\n",
    "        self.means = np.mean ( X, axis = 0)\n",
    "        self.sigmas = np.std ( X, axis = 0)\n",
    "\n",
    "        for iEpoch in tqdm.tqdm(range(self.nEpochs)):\n",
    "                l, l_val = self._train (X, Y, X_val, Y_val)\n",
    "                losses.append ( l )\n",
    "                losses_val.append ( l_val )\n",
    "\n",
    "        losses = np.array(losses)               \n",
    "        losses_val = np.array(losses_val)\n",
    "               \n",
    "        plt.plot (losses, color = \"c\", label='Training set')\n",
    "        plt.plot (losses_val, color ='tab:blue', label = \"Validation set\")\n",
    "        plt.xlabel (\"Epoch\"); plt.ylabel (\"Loss\")\n",
    "        plt.legend(frameon=False)\n",
    "        plt.show()\n",
    "         \n",
    "        ca = tf.keras.metrics.CategoricalAccuracy()\n",
    "        if self.n_outputsC <= 1:\n",
    "            ca = tf.keras.metrics.BinaryAccuracy()\n",
    "        ca.update_state(Y, self.predict_proba(X))\n",
    "        \n",
    "        return ca.result().numpy()\n",
    "\n",
    "    def save_weights(self, model_name):\n",
    "        self.model1.save_weights(model_name+'_weights_1')\n",
    "    \n",
    "    def load_weights(self, model_name):\n",
    "        self.model1.load_weights(model_name+'_weights_1')\n",
    "        \n",
    "    def save_model(self, model_name):\n",
    "        self.model1.save(\"saved_models/\"+model_name+\"_1\")\n",
    "\n",
    "    def reset_optimizers(self):\n",
    "        self.optimizer  = tf.optimizers.Adam (self.learning_rate)\n",
    "        \n",
    "    def set_epochs(self, epochs):\n",
    "        self.nEpochs = epochs\n",
    "        \n",
    "    # Applies a pre-processing to the input features and returns the classifier representation.\n",
    "    #@tf.function\n",
    "    def predict_proba (self, X):\n",
    "        ppX = (X - self.means)/self.sigmas\n",
    "        return  tf.clip_by_value ( self.model1 (ppX) , 1e-7, 1. - 1e-7 )\n",
    "\n",
    "\n",
    "    @tf.function\n",
    "    def _train (self, X, Y, X_val, Y_val):\n",
    "        Y_true = tf.cast (Y, tf.float32)\n",
    "        Y_true_val = tf.cast (Y_val, tf.float32)\n",
    "\n",
    "        with tf.GradientTape() as gt:\n",
    "            #gt.watch ( self.weightsC )\n",
    "            Y_hat = self.predict_proba (X)\n",
    "            Y_hat_val = self.predict_proba (X_val) #N3(N1(x)) validation set\n",
    "            \n",
    "            ## Training set\n",
    "            # Use the categorical cross-entropy as loss function for the classifier\n",
    "            cce = 0.\n",
    "            cce_val = 0.\n",
    "            bce = 0.\n",
    "            bce_val = 0.\n",
    "            loss = 0.\n",
    "            loss_val = 0.\n",
    "            \n",
    "            if self.n_outputsC > 1:\n",
    "                cce = tf.keras.losses.CategoricalCrossentropy()\n",
    "                loss = tf.reduce_mean ( cce( Y_true, Y_hat ) )\n",
    "            \n",
    "                ## Validation set\n",
    "                cce_val = tf.keras.losses.CategoricalCrossentropy()\n",
    "                loss_val = tf.reduce_mean (cce_val( Y_true_val, Y_hat_val ) )\n",
    "            else:\n",
    "                bce = tf.keras.losses.BinaryCrossentropy()\n",
    "                loss = tf.reduce_mean ( bce( Y_true, Y_hat ) )\n",
    "            \n",
    "                ## Validation set\n",
    "                bce_val = tf.keras.losses.BinaryCrossentropy()\n",
    "                loss_val = tf.reduce_mean (bce_val( Y_true_val, Y_hat_val ) )\n",
    "            \n",
    "            # Compute the gradient of the overall loss with respect to the classifier weights\n",
    "            gradients = gt.gradient ( loss, self.weights )\n",
    "\n",
    "        # Apply the gradients\n",
    "        self.optimizer.apply_gradients ( zip(gradients, self.weights) )\n",
    "        \n",
    "        return loss, loss_val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "270faf23-c06e-46c2-b801-40b74b8ee17b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-07-27 08:09:32.231311: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.240424: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.241045: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.242475: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 AVX512F FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2023-07-27 08:09:32.242993: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.243573: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.244116: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.955742: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.956358: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.956897: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2023-07-27 08:09:32.957412: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1525] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 13789 MB memory:  -> device: 0, name: Tesla T4, pci bus id: 0000:00:05.0, compute capability: 7.5\n"
     ]
    }
   ],
   "source": [
    "dnn = SimpleNeuralNetwork(120, learning_rate=0.001, N_NODES=50, n_layers=8, n_features=X.shape[1], n_outputsC=2, ATLAS=True)\n",
    "dnn.save_weights(\"ATLAS_selection_SigVsBkg_emu_RDF\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "4878e7f1-b129-4343-a100-f1de1e3fc119",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(25844, 9) (25844, 2)\n",
      "[[5.30548134e+01 2.98769760e+01 1.19088409e+02 ... 3.89306641e+00\n",
      "  1.00095880e+00 2.71210632e+02]\n",
      " [3.06645126e+01 2.92994366e+01 1.28700623e+02 ... 5.60546875e+00\n",
      "  1.68640137e+00 2.39743408e+02]\n",
      " [3.00996037e+01 2.80754356e+01 1.11508141e+02 ... 2.66625977e+00\n",
      "  2.74884522e-01 1.56898926e+02]\n",
      " ...\n",
      " [6.82813568e+01 3.26568680e+01 1.29935699e+02 ... 5.90478516e+00\n",
      "  1.00439453e+00 1.10416357e+03]\n",
      " [1.53700607e+02 1.13567513e+02 3.16587463e+02 ... 5.79687500e+00\n",
      "  2.20147705e+00 3.16000293e+03]\n",
      " [9.00993805e+01 3.42716637e+01 9.79262695e+01 ... 4.00579834e+00\n",
      "  2.56164551e+00 5.21605835e+02]]\n"
     ]
    }
   ],
   "source": [
    "print(X.shape, Y.shape)\n",
    "print(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "03845452-4fd3-4f29-b16f-b0d2ee0ef2c1",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 120/120 [00:02<00:00, 44.11it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "acc = dnn.fit (X.astype(np.float32), Y.astype(np.float32), X_val.astype(np.float32), Y_val.astype(np.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "fe6f8ce9-8957-4250-9622-e570e00551d7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Y_predict_train = dnn.predict_proba(X)\n",
    "Y_predict_val = dnn.predict_proba(X_val)\n",
    "\n",
    "axis = np.linspace(0,1,20)\n",
    "\n",
    "targets = ['isLL', 'isMIX', 'isTT', 'isBKG']\n",
    "predictions = ['isVBS']\n",
    "colors = ['r', 'g', 'y', 'b']\n",
    " \n",
    "Y_targets = {}\n",
    "Y_predictions = {}\n",
    "\n",
    "for target in targets:\n",
    "    #Y_targets[target] = dnn.predict_proba( df_val[ df_val[target]==1 ].values[:,0:NDIM] )\n",
    "    if target != 'isBKG':\n",
    "        Y_targets[target] = dnn.predict_proba( df_red[ df_red[target]==1 ].values[:,0:NDIM] )\n",
    "        #Y_targets[target] = dnn.predict_proba( df[ df[target]==1 ].values[:,0:NDIM] )\n",
    "    else:\n",
    "        Y_targets[target] = dnn.predict_proba( df_val[ df_val[target]==1 ].values[:,0:NDIM] )\n",
    "    Y_predictions[target] = {}\n",
    "    for iprediction, prediction in enumerate(predictions):\n",
    "        Y_predictions[target][prediction] = Y_targets[target][:,iprediction]\n",
    "\n",
    "for prediction in predictions:\n",
    "    for target, mycolor in zip(targets, colors):\n",
    "        plt.hist(Y_predictions[target][prediction].numpy(), bins = axis, label = prediction + ' class - ' + target + ' sample', histtype='step', color=mycolor,  density=True, linewidth=2 )\n",
    "    plt.title('Validation dataset')\n",
    "    plt.xlabel(prediction + ' Probability')\n",
    "    plt.legend(frameon=False, bbox_to_anchor=(1.05, 1.))\n",
    "    #plt.yscale('log')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "db374a53-c7bd-4337-9216-68f5e143c8ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "import itertools\n",
    "#from matplotlib.backends.backend_pdf import PdfPages\n",
    "\n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45, fontsize = 16)\n",
    "    plt.yticks(tick_marks, classes, fontsize = 16)\n",
    "\n",
    "    thresh = cm.max() / 1.2\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, cm[i, j],\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\",fontsize=10)\n",
    "\n",
    "    plt.xlabel(\"Predicted label\", fontsize=16)\n",
    "    plt.ylabel(\"True label\", fontsize=16)\n",
    "\n",
    "    \n",
    "    plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "f3ec098a-fc62-4ab3-87a8-7ee81376b653",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classifier categorical accuracy =  0.8343672456575683\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "from scipy.stats import ks_2samp, anderson_ksamp\n",
    "\n",
    "X = df_val.values[:,0:NDIM]\n",
    "Y_true = df_val[['isVBS', 'isBKG']].values\n",
    "Y_pred = dnn.predict_proba(X)\n",
    "\n",
    "# The classifier categorical accuracy is a good and simple metrics to evaluate the overall classification performance\n",
    "Y_true_max = np.argmax(Y_true, axis=1)\n",
    "Y_pred_max = np.argmax(Y_pred, axis=1)\n",
    "print(\"Classifier categorical accuracy = \", accuracy_score(Y_true_max, Y_pred_max))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ca1fc441-c2f5-4a0d-a4b2-7a451354a1ab",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAroAAAJJCAYAAABF+5SuAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAABcSAAAXEgFnn9JSAAB7tUlEQVR4nO3dd3xUVfrH8c9DrwIKKlIEsaOIggWxYF+769p7XwuWtbv2turae1m7rvpTV8XeFREFOygWEKWjAtIh1Of3x7lDbsJMMknuJMPM9+1rXndyy7lnQmKeOfOc55i7IyIiIiJSaOrVdQdERERERHJBga6IiIiIFCQFuiIiIiJSkBToioiIiEhBUqArIiIiIgVJga6IiIiIFCQFuiIiIiJSkBToioiIiEhBUqArIiIiIgVJga6IiIiIFCQFuiIiIiJSkBToioiIiEhBUqArIiIiIgWpQV13QCQTM/sNaAaMr+u+iIhIwegEzHP31eu6I+mY2ctAtxw0Pdrd98lBu3lNga7ks2ZY/ZbWpPWGdd0RkXy3XpfV6roLIiuEX0aPZuHCBXXdjYp0w+ptaI1bJdagL5gJvjSx9lYkCnQln423Jq03bLzJCXXdD5G8N+T9a+u6CyIrhK169eDHH77P608KrXErGq9/WGLtLfjxKbxkemLtrUgU6IqIiIjkFQNLchqVJdjWikWBroiIiEi+seINTpOkqgsiIiIiUpA0oisiIiKSbxJNXSheCnRFRERE8o1SFxKhtwsiIiIiUpA0oisiIiKSV1R1ISka0RURERGRgqQRXREREZF8YiSbo1u8A7oKdEVERETyjqouJELfRREREREpSBrRFREREck3Ki+WCAW6IiIiInlFVReSotQFERERESlIGtEVERERyTdKXUiEAl0RERGRfKOqC4nQd1FERERECpICXREREZF8klowIrFHgl0za2pmV5nZSDMrMbNJZvawmXWoRlu7mNlrZjbFzBaZ2TQze9vM/ppUfxXoioiIiEilzKwJ8D5wKdACGACMB44FvjaztarQ1lnA28DuwEjgf8CPwM7AC2Z2bRJ9VqArIiIiklei8mJJPZIb0r0E2Ar4FFjX3Q929y2Bc4B2wMNZvTqzdsD1wCJgB3fv6+6HuHtfoB+wALioKoFzJgp0RURERPJNooFuAt0xawT0j748zd3npI65+y3AcGB7M+uVRXNbAo2B9919YPyAu38EvEWIznvXtN8KdEVERESkMn2BVsBod/86zfHno+3eWbS1IMt7TsvyvIwU6IqIiIjkm3qW3CMZm0TbrzIcT+3vkUVbnwEzgB3NbPv4ATPbDtgNGAUMqno3y1IdXREREZG8kpdLAHeOthMyHE/tX7Oyhtx9ppkdDzwFfGBmn0TXdwS2BgYDR7n7wpp1WYGuiIiISDHoZmYj0h1w9+5ZXN8i2s7LcHxutG2ZTWfc/QUz2x14lpAWkTKLUI1hYjbtVEapCyIiIiL5JI/r6CbFzM4B3gU+IqQ7tIi27wNXAS8kcR+N6IqIiIgUvtFZjtxmkqqy0CzD8ebRdnZlDZlZP+AmQl7vge6+NDr0rZkdAHwB7Glmu7v7G9XuMRrRFREREck/eVZeDBgXbTtmOJ7aPzaLto6Mti/GglwA3H0JpaO521Wph2loRFdEREQk31je5RsMi7abZTie2j88i7ZSQfHMDMdT+9tk0VaFNKIrIiIiIpUZTAhAu5lZzzTHD4i2r2TR1m/RNtOCEJtH2zHZdi4TBboiIiIieSX/lgCOSn3dFX15t5mlcnIxs7MJE8kGuvuXsf39zexHM7uuXHMvRdvDzWyvMq/cbF/gMGAp8GJN+63UBREREZF8k3+pCwDXADsTat2OMrNBhLq5WwJTgOPKnd8WWA9oX27/S8BzwIHAK2b2BfAr0JXSUd6L3f2nmnZYI7oiIiIiUil3LwF2AK4m1NPdjxDoPgps5u6/ZNmOAwcDxxPKi60N/BXoArwO7O7u/0qizxrRFREREck3ia6Mlhx3nw9cFj0qO/cK4IoMxxx4OHrkTH5+F0VEREREakgjuiIiIiL5JLUyWpLtFSkFuiIiIiJ5xRJOXSjeSFepCyIiIiJSkDSiKyIiIpJv8rO82ApHga6IiIhIvsnTqgsrGn0XRURERKQgaURXREREJN9oRDcR+i6KiIiISEHSiK6IiIhIXrGEJ6MV78Q2BboiIiIi+cRINnWheONcpS6IiIiISGHSiK6IiIhIvlEd3UQo0BURERHJN6q6kAh9F0VERESkIGlEV0RERCSvqOpCUhToioiIiOQZU45uIpS6ICIiIiIFSSO6IiIiInlGI7rJ0IiuiIiIiBQkjeiKiIiI5BMj2fljRTw4rEBXREREJM8odSEZSl0QERERkYKkEV0RERGRPGIkO6JrgCfW2opFga6IiIhInlHqQjKUuiAiIiIiBUkjuiIiIiJ5xRIe0S3e0WGN6IqIiIhIQdKIroiIiEi+Kd5B2EQp0BURERHJJ5bwZLQiDpqVuiAiIiIiBUkjuiIiIiJ5RuXFkqFAV0RERCTPKNBNhlIXRERERKQgaURXREREJI/kYgngYqURXREREREpSBrRFREREck3xTwMmyAFuiIiIiJ5RpPRkqHUBREREREpSBrRFREREcknZgmvjFa8o8MKdEVERETyjFIXkqHUBREREREpSAp0RURERPKNJfhIsltmTc3sKjMbaWYlZjbJzB42sw5VaOMYM/MsHkfVtL9KXRARERHJM/mYumBmTYD3ga2AycAAoAtwLLCXmW3l7r9k0dTPwGMZjrUC9ouef1yT/oICXRERERHJziWEIPdTYFd3nwNgZmcDNwMPA/0qa8TdPyZDEGtmpxAC3cFZBs0VUuqCiIiISJ6xqPJCEo+E+tMI6B99eVoqyAVw91uA4cD2Ztarhrc6Ito+UcN2AAW6IiIiIlK5voS0gtHu/nWa489H272rewMz6wpsDSwEnq1uO3FKXRARERHJI0ayOboJtbRJtP0qw/HU/h41uEdqNPc1d59eg3aWUaArIiIikk8s4cloyTTVOdpOyHA8tX/NGtwj0bQFUKArIiIiUgy6mdmIdAfcvXsW17eItvMyHJ8bbVtWtWMAZrYFsC7wJ/BaddpIR4GuSAFp0rgB5x21Awfu3INOq7Vm+qz5vD10JFc98DaTpsyqUls7br42/Q/eht4bdqR1y6bMmlvC1z9O5D8vDuXlgWX/X9m0cUN23nId9thmA7bu0YXOq7dmyVJn9IRpvPThd9zx9CDmzl9Y5hozY+sea7LHNhvQr3c31unUjkYN6zPxj5m89/kobn5iIGMnJ/LJlchy5s+fzy03Xs8Lzz/LhPHjaNNmZXbaZTcuvuxK1uiQdTlQPh40kI8/GshXX3zOl19+zrSpU+nUeU2+/XF02vPHjh3DJhusXWm7hx95DHff/2CZfTNnzuT2m//N66+9wtgxv7J06VI6dV6TXXb7C2efeyHtVl01637LCiD/qovlWmo091l3X1jhmVWgQFekQDRu1IA37zyRLTdek8lTZvHqoO9Zs30bjt6rN7tvvT7bn3gPYyb9mVVb/Q/uy41n7c3SpUsZ+t04Jvw+k46rtWLHzddm5y3X5YZH3+eK+99edv7Bu/bk3n/+DYAffv2d1z7+gZbNG7PVxmty2Ym7cNAum7DrqfczZfrcZdd07bAy7953MgCTp87iwy9Hs3TpUnpv2IkT/7oVB+/ak7+e/QifDB+b4HdJBEpKSthn9535/LOhrL56e/bYax/GjR3Df594lLfefI13PxxMl65rZdXWheeezXffDsv63i2at+DQIzLXwH/x+WcpKSmhT99tyuyfNnUqu+64LaN/HsVqq61Ovx13BuCrLz7n3rvu4MXnn+PtDz+mc+eafGos+SO5agmp9giTyLIZuc0kVWWhWYbjzaPt7Ko2bGYNgIOjLxNLWwAFuiIF48JjdmTLjddkyLdj2evMh5aNoJ5xyDbccOZe3H/xAex22gOVttO2dXOuPuUvLFy0mD3PfIiPv/512bG+Pbvy6m3Hcd5R/Xj0lS+WBc6LlizhwZeGctczH/PT2CnLzl99lZa8cPMxbLpeB248a2+OufyZZcfcnXeHjuSmJwYy8MvS0a9GDetz5/l/5ai9evPIFYfQ/cAbWbxkaY2/PyIpN11/LZ9/NpQtttyKF155kxYtwieyd91xK5dceB6nnXwCr731flZt7bjzzuy3/9/YrFdv1ujQka16VTwPZ5W2bbn3gYfTHvvpxx94+snHadq0Kfvst3+ZYzffeB2jfx7F7nvuzSNPPE2TJk2AELSfcMwRvPryS/zr6iu47z+PZNVvkWoYF207Zjie2l+d0YldgVWBX9z9k2pcn1HRlxczszGxpea61HV/RKqjYYP6nHxAHwDOumlAmTSBO575mOGjJrPdZmux6XqVfyS7efdONGnckA+/HF0myAUY/M2vvDN0FPXq1aPXBqVt/ff1rzj9hhfLBLkAv02bzT9uGgDAvtt3p2GD+suO/TrxT/Y+6+EyQS7AwkVLOPOml5gxez6d27dhqx4aoZLkLFy4kAfuvweAG2+9c1mQC9D/jH/QfeMeDB70Ed989WVW7V117Q2ce8E/2XHnXWnTZuUa9e3Zp/8LwB577cNKK61U5tgnH4fa+mefd8GyIBegSZMmnHfhxQB8/eUXNbq/5Jd8q6MLpD662CzD8dT+4dVoO5W28GQ1rq1Q0Qe6uWJmDczst1gQfVUN2jot1s4cM2sRO9Yvi7WiF5rZFDMbama3m1nvKt6/sZkdamb/NbMfzGy6mS2O+jLezD42s/vN7AQzy+7zPklUnx5r0rplU0ZPmMqwkZOWO/7iB98CsMc2G1Ta1oKFi7O657SZmeYjlDV81GQAmjRuyCqtMn3iVVbJgsX8PH4qAGu0XamSs0WyN+TTwcyaOZOua3Vjk56bLnd832gk9Y3XX63Vfrk7zz0bPvE4+NDDlzveuHGjSttYeeWaBdoilRgMzCRMauuZ5vgB0faVqjQaxTT7Rl8q0F1RuPti4L+xXUdY9d9WHR17/nx8NZIsNQTaAlsAZwCfm9kzZlbpzEgz2wsYDTwFHAasD7QG6hPycToSikifBPwHGG1md1axf1JDPdZpD8A3Py0f5Ib9EwHYeO3VK23ri+/HM33WfPr16sY2m3Ytc6xvz67ssuU6jBo3hcHfjMmqb107hD++Cxct5s9Z2QXHZkan1VsDYVRYJCnffRsGm9IFufH9I777ttb6BPDpJx8zbuwY2q26KjvuvOtyx3fYaRcAbr3p35SUlCzbX1JSwo3XXwvAEUcfVzudlVqRbyO60QSxu6Iv7zazVE5uagngHsBAd/8ytr+/mf1oZtdV0PT+hLzfIe4+KpHOxihHN7ceA86OnncFtgU+qkoDZrY+sHm5Nityd5p9TQkB6daUlgc5GOhoZv2ioDzdvY8FHqLs3M9RhI8lpgGNCTk1PYH2sXPaVNJHSVin1VoDMPGPmWmPT/wjVFxIBY8VmTV3Aaf863kevfIQ3rrrRIZ8O46Jf8ykw6qt2Grjznw6fCzHX/UsixYvyapvpx3UF4C3h4xk4aLsrjl4l01YbeWW/PHnHIZ8q8lokpwJ40OaYabKCmt0CGmG48eNS3s8V559+ikA9j/gYBo0WP5P8+lnncPgjz/i9VdfZpMN1qbX5lsA8OXnn1GyoISrr/s3hx959HLXyQosP6suXAPsTIgnRpnZIELd3C2BKUD5d1ttgfUoGyOUl3jt3LiiD3TdvUsO2x5uZt8QAkGAI6lioAvEp+eOBT6s5J79Mx2L3n1dT+la1amR2HvSnLs2cC+lv2qfEta2TrfsH2a2LuFd2fEV9U9yo3nT8LHmvJJFaY/PLQk5uy2bNc6qvQEDR7Dv2Y/w5DWHsfUmXZbtnzmnhHc/G5V1qbLd+qzHMXv3ZuGixVz1wDtZXdNx1Vb8+6y9ALj6wXeyDo5FsjF3Tqj80axp+jSa5s3DINWcObX3ScKCBQt46cWweuohadIWUv169oVXOLP/yTz79H95/dWXlx3bdvt+9OnTt1b6KsXN3UvMbAfgIsKnvPsR6t4+Clzq7pkWk0jLzNoDOwKLgP9LtLMRpS7kXnwE9kAza5LxzHKiVIf4//WecHevbkfcfa67nw68Fdud/v+q8A/CiC3ACGDHTEFu1PZId7+eUOz5sur2UfLDmYduy+t3nMDgb8bQ+4hbWWWHS+l9xK0M/HI0l5+0K89cf0Slbay7ZjsevuJg6tWrxz/veoNvf55c6TXNmjTkmeuOoF2bFrw8cAQPvjg0iZcjktfeeuN1Zkyfzrrrrc+mvdJPoRg/fhw7bdeHd99+k/sefJSfx07m57GTufc/j/Ddt8PZ6y878cngQbXcc8kZSzh1IcHRYXef7+6Xufva7t7Y3du7+7Hpglx3v8Ldzd2PydDWZHdv4O6N3H1acr0spUA3954CUqkBrShNuM7GDpQuuQfweIJ9StkwwznxJLE73b0kw3llePBLtXsm1ZKqstCsScO0x5s3CSO+s+ctqLStbTddi+vP2JNhoyZx2MX/ZcTo35lXsogRo3/n0H8+yTcjJ7FH3w3Ydat1M7axRruVePnW41h5pWbc/tQg7n52cKX3bVC/Hv+99nB6bdiJwd/8ytGXP13pNSJV1bxFGLGdNz99vvjcuWHEt0WLai3uVC3PPhOmc6SbhJZyyonH8v2I77jj7gc45LAjaNuuHW3btePQw4/k9rvup6SkhEsvOr+2uiw5ZiQb6OZnFkTtKPpA17IoL2ZmLczsZDN7zczGmdk8M1tkZjOjJOtXzOyfZrZR+Wvd/Q/gjdiuI6vQvXjawicJJmnHh9aaZzgnnsCmJMk8N/73GQB0WLVV2uMdVg2VC8b/NqPStg7bPUzGeXng95T/AGHpUmfAh98BLDdRLaXNSk155bbjWbN9Gx579QsuvLPylRzNjAcvO4i/bL0+34ycxN/Oe4ySBdlVfxCpio6dwtjBpIkT0x6fNDEMSnXq3Dnt8aTNmDGDd956AzPjwEMOS3vOhAnj+fijgTRu3Jjd99xrueN77r0PjRo14qsvvygzUU1ElKNbKTPrAzxH2cAvZaXosR6wF3CtmTVMM7nrMWDv6PluZrZqFABXdN9mwN/KtZGU+NT73zKcE6/Qnz6ikbyRKuHVc7010h7vGdXP/fbnTP/cpTq0C8HyrDnp/2Cm9rdu2XS5Y82bNmLALcey4Vqr8dIH33Hqdf+rvPPALefsw8G79mTk2Cnsc9ZDzMxwb5Ga2mjjsKDDsG/SZ2Kl9nffaONa6c9L/3uOBQsWsPU222Zc1SwVfDdr3pz69esvd7x+/fo0a96cGdOnM3PGDJqsXnl1Fcl/SVVLKHZFP6JbETPrRMhnTQW5i4BPCCkE9xFmCA4CKpuZ8wohWRvCm4v0b9vL2p/SCgklwLNZd7xyB8eeZ0rqilfxPz0KvCVPfTp8LDNmz6dbx7bLSo3F/XWH8Ef79Y9/qLSt3/8Mk3A22yD9rPReG4RZ6eMmTy+zv1HD+jz376PYvHtn3h7yE0dd9jRLl1aeUn75Sbty8t/6MG7ydPY688EyywSLJG2rPn1ZqVUrfv1lNMOHfbPc8QEvvQDA7nssP3KaC/8XpS0ccmjmvPdVVwuB6/Q//2TMmF+XO/7rL6OZMX06zZs3Z5W2bXPTUZEVlALdip0NpBK1BgFd3L2vux/t7qe4+1Huvh2wCtCPUDd3ub/sUe25Z2K7sklfiKctDHD3GdXofxlm1tTMbgT2jHYtBm7OcPpLsecbAEPN7PBsau9K7Vu0eAn3Pf8pALedu2+ZXN0zDtmGHuu056OvfuHrn0o/rj35gD5888zZXHXKbmXaeuWj7wE4ZNee7N53/TLH9tp2Qw7etSdLlixlwMARy/bXq2c8ftWh7NB7bT7++lcOufDJrMqPnX7INlx47I5MnjqLPc54kPG/py+PJpKURo0acdLfTwXgvH+cviwnF8ISwCO+HU7fbbej52a9lu1/4N672bxnd6687J+J9mXcuLEM+WQwTZo0Yd/9D8h4XpcuXekejUT/4/RTmDmz9PdkxowZnHX6KQDsufe+aUuTyYoowfxcM/K1Vllt0G9ExbaNPT/O3dNW449SFQZGj0weA06Nnm9mZt3dfUS6E81sDWCnctdmxczuSrO7MaV1dFPLTM0CDnL3rzI0dRthoYrUZ2kbEVYsWRSVTPsc+IKQO/xTtv2T3Ln+0ffZcfO16dOjC989ex6Dh/1K59XbsMVGnfnjzzn8/drny5y/SqvmrLfmqqy+yvgy+18eOIL/vTecv+3UgxduOoYvvx/PmMnT6dK+Db027ATAZfe9yahxU5ddc8oBW7Nvv5CiPm3mXG4/b7+0fbzozteWrajWY532XH/6HgCMnTydC47ZMe01j778GZ8MV5q4JOfcCy/mww/eY+iQT+m18fr06bsN48eN5YvPP6Ntu3bcfd+DZc6fNm0qo0b+xG+/LZ/68/gjD/H4ow8BsGhRyFr7/bfJ7Lz91svOuem2u+i56fKrpj73zFO4O3/ZYy9atUqfX59y+533st9eu/HBe++y2cbr0bt3qKP7+edD+XPaNDqv2YUrr7m+at8IyW/FG5smSoFuxeJrj06pSUPu/pmZ/UhYWQzCiO0FGU4/gtLR9t+At6twq9OyOOe/wJkVlfJw9+lmtivwAtA9dqghYQGLZYtYmNlvhPp3d1dnwpyZpQ34gW5VbauYLVi4mN36P8B5R+3Awbtuwt7bdefPWfN4/NUvuOqBt5mYZe1bgCMueYq3h4zkiD02Y6Nu7emx7hrMmD2fNwb/yL3Pf8I7Q0aWOT+er5sKeNO55sF3lwW6rVo0pV698GO+1cZrstXG6fMTP/rqFwW6kqgmTZrwypvvccuN1/P8s8/w2isDaNNmZQ474mguvuxKOnTsmHVbEydO4IvPPyuzb+HChWX2zZ6d/nfvuf8LlUUOOiRztYWU3ltsyaAhX3LbzTcy8MP3+fCD96hXrx5rdunK0ccezxlnnUsbLQEsshyrQVnWgmBmYygdtezq7mNix94jFDIGuNDdb6jhvS4EUsvgTQQ6u/vSNOd9SxhBBbjJ3c+roM1+wAfV6M54QrD7YiV9bkRYBOIUoLLZGYuAm4BL0r2uCu6RMdC1pqs0brzJCdk2JVK0fnv/2rrugsgKYatePfjxh++/d/fulZ9d+8xsRMOVO2/Y8eh7E2tzwmOnsOjPcXn7mnNJOboVi08Au97M3jWzY80s+7f7ZT1JaTWDDpQG0cuY2WaUBrlQxWoLUWHmMg/CyP1qwB5AajmdTsALZnZSJe0tdPd73b0HYTGIE4GHga8prQ+c0pCwWsojVexz93QPyk6IExERKQ55vGDEikaBbsUepOykrJ0IQd54MxtrZk+a2fFmltU012jVkPdiu45Kc1p8sfKv3P27KvY53X2XuPsf7v6Gu+8LXBk7fIeF5XuzaWeUuz/o7se7+2ZAG0J1iHfLnXqUme1f036LiIiI1IQC3Qq4+xJCIHcC8H25w50Jy+c+CEwyswfNLJsEqfgI7f5mtmzBBjNrABya4dwkXQOMi543Bs6oTiPuPsfdX3T3XYAzyx0u/7WIiIhkySy5RzFToFuJaEnbh6KP0tcDTiIEoPFlbhsS8lg/M7N2lTT5IjA7et6cEEin7A6krl9E2aV6ExNViYiPLO+U6dwqtHkHZSfN9YkCdxEREZE6oUC3Ctx9pLv/x92PcfduhMD3FiBVMLQbcHklbcwjrLSWclSG56+7+1RyJ74McPrp7lX3Zux5Q0J9YREREakCI9kc3WIe1FWgWwNR4HsOZYPbfbK4NJ6SsKOZdTCz1pQuE1z+nFyIr3SWdYWESpRft3VBQu2KiIgUFaUuJEOBbjJejj1fLYvzBwGpdRzrEXJ9DybkywJMA15LrHfpxauXT8x4VtVsEns+O4nV3ERERESqS4FuBbKtpkAo1ZXyR2Uneyhe/Hhs15GUTVt4Olo2OCfMrDdlV317L805t5rZelVocz3KLm38ZqZzRUREpGLJLgFcvBToVmycmd1vZtubWdrvVRQ03hnb9UaWbT8OpFbr2IiwPG9KztIWzGxnYAClVfUWU7b/KYcCI8xsgJn9LV4dolx7DczsIMLyx6l0iKVAjRbXEBERKWZKXUiGZsVXrCmhysJJwGwz+wYYC8wF2hKW842vMjIFuCKbht39FzP7mLIjqwDfu/sX1e2wmd2VZnd9YGWgF8svq3uuu/+Qobn6hJzjfYAlZjacUG1iGuFnZw3CUsDlJ52d7+5fVu8ViIiIiCRDgW7F5gAtouctCUFp+cA0ZRhwiLtPqkL7j6Vpr6ajuadled6fhCWAn8xw/H+EUd020df1gU2jRyYTgbPc/fks+yAiIiLlGdSrl+BQbBGP6irQrdgqwHbA9oSRy3UIk82aAPOACcCXhKDwZXevavWC54A7KPuRf6bAsyacULv3D+Ab4C3gGXefk/EC99PM7EygDyEY700op9aeEPQvBmYRRriHESbPvZ7L3GIREZFiEMqLJdtesSr6QNfdu1RwbCFhedvyS9wmde9ZhEUjatLGh+ToZzhaWGJQ9BARERFZoRR9oCsiIiKSX5KullC8Y7qquiAiIiIiBUkjuiIiIiJ5ptjLgiVFga6IiIhInin2hR6SotQFERERESlIGtEVERERySeW8IhuEQ8OK9AVERERyTPKXEiGUhdEREREpCBpRFdEREQkj4SV0ZIb0i3mwWGN6IqIiIhIQdKIroiIiEieUY5uMqod6JrZkiQ7koG7u4JxERERKSqqo5uMmgSRBjjFnfohIiIiInmqpqOlCnJFREREEqYB3WRUO9B1d01kExEREUmaFoxIjIJVERERESlImuglIiIikmeUupAMjeiKiIiISEGq9RFdM2sKtAEauPu42r6/iIiISD4zLOGV0Yp3eDjnga6Ff6kDgMOBbQhBLoTSZA3KndsWOCj6cpS7v5Pr/omIiIjkG6UuJCOnga6ZrQf8H7Bxalcll0wDzgK6Ab+ZWSd3X5q7HoqIiIhIocpZjq6ZbQB8SghyLXrMjR5pubsD90bnrg70y1X/RERERPKVmSX2SLhfTc3sKjMbaWYlZjbJzB42sw7VbK+Lmd1nZr+a2QIzm2pmn5rZeUn0NyeBrpnVB14AWhOC1veBPu7eEni0ksufjT3fNRf9ExEREclnZsk9kuuTNSHEdJcCLYABwHjgWOBrM1uriu3tDowATiJ8qv8C8BXQBfh7En3OVerC4cB6hDzcF4GDsk1BcPeJZvYr4UX2zlH/RERERKRqLgG2Inxiv6u7zwEws7OBm4GHyfLTeDNbnxDYzgZ2cfdPYsfqAZsl0eFcpS78NdqWAKdUI8/2O8JI8DqJ9kpEREQk31nCqQsJjOqaWSOgf/TlaakgF8DdbwGGA9ubWa8sm7wFaAIcEw9yo/aWuvsXNe917gLdXoTR3I/dfUo1rp8abVdJrksiIiIiK4Y8TF3oC7QCRrv712mOPx9t9678tVknYDfgF3d/PbEeppGr1IV20XZMNa9fFG21cpuIiIhI3dsk2n6V4Xhqf48s2upHGGz9xMwaAPsTAun6hE/1/8/dp1e/q6VyFUguABoBDat5fSpQTuRFioiIiKwoDBJeMCIRnaPthAzHU/vXzKKtDaPtHGAQIe837lozO8DdP6haF5eXq0D3d6Al1c+x3YKQ+jA+sR6JiIiIFK9uZjYi3QF3757F9S2i7bwMx1PlY1tm0VZq8bATCMHuYcCbhIHOS4EjgBfNrLu7T8yivYxylaP7KeENxOZmtlpVLjSzXYBULbaPku6YiIiISL7L1zq6CUnFnw2Av7v70+4+3d1HuvuRwOeEfOBTa3qjXI3ovgQcRUhduA44LpuLzKwlcEds13OJ90xEREQkz+UgPh2d5chtJqkqC80yHG8ebWdXoa05pI/1HgE2B7bPuncZ5GRE191fAoYRRnWPNrNborIUGZnZxoQR3FT93ffcfWgu+iciIiIiVTIu2nbMcDy1f2wWbaXOGRetilvemGi7anZdyyyXVQ2OJgSuLYEzgYPN7AVKZ+1hZmcSlvrtC2xNab70VOD4HPZNREREJG/lYcrBsGibaSGH1P7hWbSVKk/WJsPxlaPtnAzHs5azQNfdh5vZPoQlfVclBLSpXItU9H5L7JLUv+hkYF9310Q0ERERKT4JL92bUNmFwcBMwqS2nu7+TbnjB0TbV7Jo6xPCkr+rm9l67v5TueOplIV09XqrJFeT0QBw948II7iPEGrjWgWPxcCjQO+kVsMQERERkZpz94XAXdGXd5tZKic3tQRwD2Cgu38Z29/fzH40s+vKtbWYMNhpUVsrxa7ZGTiGMCh6f037nfMFGdz9d+B4M7sA2BnoA6xBmE03l1CKbCjwlrtPynV/RERERPJb0tUSEmvrGkIstzUwyswGEermbglMYfniA20Jc6/ap2nrRmCHqL2RZjYkOn8rwsIRF7v7ZzXtcK2tPObuU4FnooeIiIiIrEDcvcTMdgAuItS+3Q/4k/CJ/KXunmkxiXRtLTKzPYB/ECp17QYsBAYCt7r7q0n0WUvsioiIiOSRsDJasu0lxd3nA5dFj8rOvQK4ooLji4B/R4+cUKArIiIikmfq5V/VhRVSrQa6ZrYusClhibeWhKLCU4Gv3H1kbfZFRERERApbzgNdM2tFyL84njAJLdN5k4EHgdvcfUau+yUiIiKSrzSgm4yclhczs12A74FLgQ5UXF5sjei8781st1z2S0RERCSfmVlij2KWsxFdM9ubsH5xQ0rzoJcAPxGWkZtLWBe5E7A+oZQEhIUlXjazA9395Vz1T0REREQKW04CXTNbjVBqolG0aypwFfBkurSEKL3hCMKIbjtCcPyImW0Y1eEVERERKQpmUC/JqgtFPKibq9SF/oT1ix34DtjY3e/KlHvr7jPd/W7CKmojot2to3ZERERERKosV4HuPtF2KXBQtqOy0XkHR9fF2xEREREpGsrRTUauAt0uhNHcwe7+Y1UudPcfgI8Jeb1dEu+ZiIiISJ4zS+5RzHJadQGobm1c1dQVERERkRrJVaCbWuu4RTWvT12X9ZrJIiIiIoXCEvyvmOWqvNjbwAbANlW90EIyyTaE1Id3E+6XiIiISN5LsupCMcvViO59wAKgg5mdWcVr+wMdgYVROyIiIiIiVZaTQNfdfwLOIEwou8nMzjWzCu9lwdnArdGus6KJaSIiIiJFw0i26kIxDw5XO3XBzDpXcspbwFnAv4EbgJPN7CngU8LKaPOAZkBnYCvgUKAbYST3fOANM+vs7uOq20cRERERKV41ydEdQ8ijzYYBawEXV3IOhNXUboseTg6XKRYRERHJR8VeFiwpSQSRlf1TOKUBcUXnxoNm/fOKiIhIcTKjXpKRbhFHzTUJdMeR/YiuiIiIiEitqnag6+5dEuyHiIiIiESKeBA2Ucp/FREREckzpkg3EbleAlhEREREpE5oRFdEREQkj4Q6usm2V6wU6IqIiIjkmUSrLhSxWg90zawV0JIs0ya0YISIiIiIVEfOA10zWxM4GdgZ2BhoWIXLtWCEiIiIFB2N5yYjp0GkmZ0LXENpcKt/NxERERGpFTkLdM3sPOCG2K45hBHaltF2XPS8DaUBsAMlwB+56peIiIhIvlN5sWTkpLyYmXUijORCCHAPBloDj6fOcfeu7t422r8n8Boh4G0I3B8d75qL/omIiIjkKzOol+CjmGPmXNXR/TshYHWgv7s/5+5L053o7rPd/Q133xs4NLrmWjO7LEd9ExEREZEikKtAd4doOxV4ItuL3P3/gLMJI7uXmtkmOeibiIiISF4zs8QexSxXgW43wsjsUHf3dCeYWab84HuAyVHfjstN90RERETyl1lyj2KWq0C3TbSdXG7/gtjzZukujALjQYRR3R2T75qIiIiIFINcVV1YGLVdfjR3Vux5R+D7DNfPibYdEu6XiIiISN4r9pSDpORqRDdVHqxVuf1jYs83q+D6taJt06Q6JCIiIiLFJVeB7veE1IO1y+3/Ovb80HQXmtm6QF/CaPCknPROREREJI8lWV6smOUq0B0cbbubWePUTnf/FhhJCIL/YmYXm1n91HEz6wI8RelKah/kqH8iIiIieclIruKCmWFFvDBtrgLdt6NtY6BfuWPXxZ5fBfxhZoPN7GtgFLBpdGwxcGuO+iciIiIiBS4nga67fwV8QcjV3bvcsceARwmjukao0LAV0AOoH+1bCpzu7iNy0T8RERGRfGYJPopZrqou4O5bVHDsODMbApwDrEPpv4MDQ4BL3f39XPVNREREJG8Z1Euy6kIRR7s5C3Qr4+4PAA+YWUdgDcIo7q/uPq2u+iQiIiIihaPOAt0Ud58ATKjrfoiIiIjkC5XRTUauJqOJiIiISIExs6ZmdpWZjTSzEjObZGYPm1mVFvkyszFm5hU81k+iv3U+oisiIiIiZeXjymhm1gR4n1BEYDIwAOgCHAvsZWZbufsvVWz2sQz7Z1a3n3EKdEVERETyiJFs6kKCTV1CCHI/BXZ19zkAZnY2cDPwMMuXla2Qux+TXPeWV+1A18weTrIjGbi7H18L9xERERGRDMysEdA/+vK0VJAL4O63mNnRwPZm1svdv6yTTqZRkxHdYwjlwHJNga6IiIgUlUTLiyWjL9AKGO3uX6c5/jxhTYS9gYIIdCH3ldlqI5AWERERySv5F+eySbT9KsPx1P4eVWnUzM4DugELgBHAi+4+pVo9TKMmge6xSXVCRERERPJa52ibqSRsav+aVWz33+W+vtXMTnf3RFJkqx3oRkv5ioiIiEiSLOGqC6GpbmY2It1hd++eRSstou28DMfnRtuWWfbqZeADQprDFGAt4DjgTOBBM5vm7gOybCsjVV2QvLZ+19X4atD1dd0NkbzXZvP+lZ8kIiz4dXJdd6FSRrILHeRfFgS4+xnldo0AzjGzH4EHgBsI5ctqRIGuiIiISOEbneXIbSapKgvNMhxvHm1n1+AeAA8B1wDrmVkXdx9Tk8a0MpqIiIhInjGzxB4JGRdtO2Y4nto/tiY3cfelwOjoy/Y1aQsU6IqIiIhI5YZF280yHE/tH57AvdpE27kVnpUFpS6IiIiI5Jl6+ZdYO5iwLG83M+vp7t+UO35AtH2lJjcxs+7AeoRJbz/WpC3QiK6IiIhI3qlnyT2S4O4LgbuiL+82s1RObmoJ4B7AwPiqaGbW38x+NLPr4m2Z2R5mtmP5e5hZD+A5wvy5B6N71ohGdEVEREQkG9cAOwNbA6PMbBChbu6WhBJhx5U7vy1hdLZ8ru0WwOVmNpaQEjGPUF5sM0Js+iFwYRIdVqArIiIikkeMRCeRYQkVGHP3EjPbAbgIOAzYD/gTeBS41N0zLSZR3ltAJ2BzSpcWngV8DPwXeMTdlyTRZwW6IiIiInkmD3N0AXD3+cBl0aOyc68Arkiz/1Pg06T7lo5ydEVERESkIGlEV0RERCSfGCS5AnBeLo1WSzSiKyIiIiIFSSO6IiIiInmmXqJDusWr1gNdM2tKWPGigbuPq+x8ERERkWJiJPuRezGHzDkPdC3UxzgAOBzYhtJl3bz8/c2sLXBQ9OUod38n1/0TERERkcKU00DXzNYD/g/YOLWrkkumAWcB3YDfzKyTuy/NXQ9FRERE8o8yF5KRs8loZrYBoUbaxoQA14C50SMtd3fg3ujc1YF+ueqfiIiISL6qZ5bYo5jlJNA1s/rAC0BrQtD6PtDH3VsSVs+oyLOx57vmon8iIiIiUvhylbpwOGFtYwdeBA7KNgXB3Sea2a9AF6B3jvonIiIikreKfCA2MblKXfhrtC0BTqlGnu13hJHgdRLtlYiIiIgUjVyN6PYijOZ+7O5TqnH91Gi7SnJdEhEREcl/BtRLcES3mAeHcxXotou2Y6p5/aJoqwUtREREpLhYwgtGFHGkm6vUhQXRtmE1r08FytMT6IuIiIiIFKFcjZj+DrSk+jm2WxBSH8Yn1iMRERGRFYQmoyUjVyO6nxIGyjc3s9WqcqGZ7QJ0iL78KOmOiYiIiOSzVI5uUo9ijplzFei+FG0bAtdle5GZtQTuiO16LsE+iYiIiEgRyUmg6+4vAcMIbyKONrNbzKxRRdeY2caEEdxU/d333H1oLvonIiIiks8swf+KWS6rGhxNCFxbAmcCB5vZC8AmqRPM7EzCUr99ga0pHV2fChyfw76JiIiISIHLWaDr7sPNbB/Ckr6rEgLaU1OHo+0tsUtSQe5kYF9310Q0ERERKUpJ1tEtZrnK0QXA3T8ijOA+QqiNaxU8FgOPAr3d/Ytc9ktEREQknyU5Ga2Y5XxBBnf/HTjezC4Adgb6AGsArYC5hFJkQ4G33H1SrvsjIiIiIsWh1lYec/epwDPRQ0RERETSMAxLsJBuMU9I0xK7IiIiIvkk6ZSD4o1zc5ujKyIiIiJSVzSiKyIiIpJntARwMnIS6JrZUUm15e6PJ9WWiIiIyIqgniLdRORqRPdRSmvl1oQDCnRFREREpMpymbpQ1bciXo1rRERERAqKkexktGIOrnIV6D6W5Xn1CPV0Nwa6RvtKgOeApTnol4iIiIgUiZwEuu5+bFWvMbPewO2EBSVWBw5091lJ901EREQk3ylFNxl5U14sWvZ3O+Btwgpqys0VERGRolQPS+xRzPIm0AVw9yXACcAiYG8z27+OuyQiIiIiK6i8CnQB3H0CMJiQO13lFAgRERGRFZ1Zco9ilneBbmR0tN2kTnshIiIiUsssWgI4qUcxB7v5Gug2ibar1mkvRERERGSFlXdLAJtZPcKkNICZddkXERERkbqgldGSkY8jutcAnQkLSHxVx30RERERkRVUTkZ0zaxzFfuwCtATOJpQRzcl24UnRERERAqGBnSTkavUhTGEEdmaeN3dn0mgLyIiIiIrjLAEcHKRbjHHzLlOXbBqPBy4Bzggx30TERERkQKWqxHdcWQ/orsImEUYBR4K/J+7j8tRv0RERETynlIXkpGTQNfdu+SiXREREZFikI/VAgDMrClwEXAIoXjAn8CbwKXuPrEG7a4DDCeUmH3P3XdOoLt5+30UERERkTxiZk2A94FLgRbAAGA8YSXbr81srRo0/wDQuMadLCcnga6Z9Yg96ufiHiIiIiKFyswSeyToEmAr4FNgXXc/2N23BM4B2gEPV6dRMzse6Af8J6F+LpOrEd1vgK+Bl9x9SY7uISIiIiK1wMwaAf2jL09z9zmpY+5+CyHtYHsz61XFdlcDbgTeAZ5OqLvL5CrQXRRth+SofREREZGCVZ2yVZkeCekLtAJGu/vXaY4/H233rmK7twNNgVNr0LeMclV14TegIzCnshNFREREpJSZJVtHN5m2Nom2mVatTe3vkW2DZrYHcDBwmbv/bGYda9C/tHI1ovsj4U3EmjlqX0RERERqT2rV2wkZjqf2ZxX7mVlzwroJPwE31KxrmeVqRPdZYBdgGzNbxd2n5eg+IiIiIgUnB2V0u5nZiHQH3L17Fte3iLbzMhyfG21bZtmfawhB8Q7uvjDLa6osVyO6/wW+J9RCuztH9xAREREpSGbJPfKNmfUGzgAed/cPc3mvXC0YUWJmBxAKCB9oZisBZ7n7yFzcT0REREQqNDrLkdtMUvOummU43jzazq6oETNrQCgjNgM4twb9yUqNAl0zuyx6+pm7v5lm/8vAycBuwA9mNhz4EpgCzM/mHu5+VU36KCIiIrKiSbj+bRLGRdtME8ZS+8dW0k5HoCehcMFz5V5n62jby8w+BHD3flXrZlk1HdG9AnBCesKbafbHGWEmXtaz8SIKdEVERKRoGMnmliYUMg+LtptlOJ7aPzzL9laPHum0BrbPsp0K5XIJ4HRl3Oqo9JuIiIiI1MBgYCZhUlvPNMcPiLavVNSIu49xd0v3AHaITnsvtq9GclV14coctSsiIiJS8PItdcHdF5rZXcDFwN1mtqu7zwUws7MJn9gPdPcvU9eYWX/CamovuvtFddHvXE1GU6ArIiIiUliuAXYGtgZGmdkgQomwLQnzr44rd35bYD2gfW12Mi6XqQsiIiIiUg15uAQw7l5CSC+4mlBPdz9CoPsosJm7/5Lg7RKRq9QFEREREammfEtdSHH3+cBl0aOyc68gFCjItu0PSXiOlkZ0RURERKQgaURXREREJM9oJDIZSQW6fzWzjRJqK87dfacctCsiIiKSl4xkUxfyMwmidiQV6K4RPZJkLL/ohIiIiIhIVpIKdIv5zYKIiIhIohRYJSOpQPcz4I2E2hIRERERqbHEAl0tEiEiIiKSAINEq4sV8fCwqi6IiIiI5Jl6xRydJkjVK0RERESkIGlEV0RERCTP5OnCaCscBboiIiIiecSi/5Jsr1gpdUFERERECpJGdEVERETyjFIXkpHEiK7+KUREREQk79R0RLdrtJ1V046IiIiISKDyYsmoUaDr7mOT6oiIiIiIBEpdSIYmo4mIiIhIQdJkNBEREZE8oxHdZGhEV6RAzJ8/n6uuuIyNN1yX1i2a0LXzGvz9hOOYOHFildoZ9NFArr36Sv66z550at+Opg2N9dbuUuE1Jx53DE0bWsbHf+6/L6t7n3zi8cuuGfzxx1Xqt0hVNGnckEtP2ZPhL13G9CG38svb13Lf5YezRrtWVW5rxy3X54U7Tmbc+9cx67PbmfDBDbxyz2nss0OPrK5fuVVzxr53HfO/vovvBlxe4bkNGtSj/2H9GPTEufw+6EamDL6Z4S9dVu2+S34ySmvpJvNf8dKIrkgBKCkp4S+77MhnQ4ewevv27LXPvowdM4bHH3uEN15/lYEfD6HrWmtl1da5/ziT4cOHVasfu+y6G6uttvpy+9ddb71Krx344Qc89ujDmBnuXq37i2SjcaMGvPnAGWzZoyuTp8zk1Q+Hs+Yaq3D0fn3YfbuN2P6omxgzcVpWbfU/rB83nncAS5cuZejwX5nw+ww6rtaaHbdcj537bMAND77FFXe/UmEb15/9V9q2bl7pvdqs1IxX7u1Prw07M3nKTD4Y+hMAa3Vqx9H79eHxAZ8yacrMrPotUiwU6IoUgOv/dQ2fDR3Cllv14dU33qZFixYA3H7rLVx4/jn8/cTjePu9D7Nqa6dddmX/Aw6kV+/N6dChI5tt0j3rfpx7/oVst32/Kve/pKSE/qf+nQ27d2ellVox5NNPqtyGSLYuPOEvbNmjK0OG/cJep9zF3PkLATjjiB254Zz9uf+KI9jtxNsrbadtmxZcfca+LFy0mD1PuYuPv/x52bG+m3Xj1Xv6c95xu/DoS59kDJz7bbEuR+6zFQ8+/zEnHLBNhfd76sbj6bVhZ66573Wuf/BNlixZuuxYlw6rMHtuSTYvX1YQ9Yp5GDZBRZW6YGZjzMyjR5e67o9IEhYuXMh999wFwG133L0syAU48x9ns/HGPRj00UC++vLLrNr71/X/5oKLLmbnXXalzcor56TP5V137dWM/vln7rjrPho2bFgr95Ti1LBBfU4+ZDsAzrru2WVBLsAdT77P8JET2K73Omy6QadK29p8oy40adyQDz8bWSbIBRj81Wje+fQH6tWrR68NO6e9vknjhtx18aF8P3oytz3xXoX3+tsum9Jvi/X439tfce39r5cJcgHGTJzGtBlzK+2zSLEpqkC3NphZl1gwnemx2MymmdnXZna/me2QZdv94u1keY2Z2T3l7v+omdXP4totzOxiM3vXzEaa2Z9mtijajjazV83sajPb1sz0s1RHPv1kMDNnzmStbt3ouemmyx3/698OAOD11yr++LSufPftt9x6840cfcxx9N2m4hEtkZrq03MtWrdsxuhxUxj204Tljr/47jcA7LHdxpW2tWDh4qzuOW1m+gD04r/vTteOq3DGtc+waNGSCts4dv++ANz7zMCs7ikrPmXoJkOpC3WjPrBy9OgJnGRm7wFHuvvkpG4SBZ8PAsfGdt8HnOoVJEGa2S7AFcDWGU5pEz3WAvYELgEmm9k9wO3uPrvmvZdsDR8W8ml7brpZ2uOp/d9+OzznfXnpxRd48YX/sXTJEtbs2pU999yb9dZfP+P5S5cu5bRTTqJ169Zce/2/c94/kR7rdgDgmx/Hpz3+zQ9h/8brrFFpW1+MGMP0WfPot8W6bNNr7eVSF3bpswGjxv7B4K9GL3ftRuuswZlH7MTjLw9h8Nej6dw+86cnDRrUY+uea7Fo0RI+/24sG62zBvvvsint2rRk0pQZvPrht3w7smqTTiXPWcJVF4o41lWgm3uPA+UDv0ZAe0Igmfq/207Ah2a2ubvXeKU5M2sQ3fvQ2O7b3f2sCq6pD9wAnFPu0ELgS2Ai8CfQEliVEKSvEp3THrgaOBDYpKb9l+yNHz8OgA4dOqY9nto/fmzu13e59+47y3x9yUUXcNLfT+GmW2+nQYPl/3dz3z1389nQITz48GOsXEtpElLcOkUB5cQ/ZqQ9ntrfqYLAM2XWnBJOufK/PPqvY3jrgTMYMuxXJv4xgw6rtmarTbry6Te/cPylT7BocdnRWjPj3ssOY8aceVx824BK79O1Q1uaNmnEb1NnccYRO3DFaXtTv37ph2iX/H0P7n7qQ86/+YVK2xIpNkUV6Lp7lzq47eXuPibdATNrRAgq/xXtWpcwOnp+TW5oZg2BZ4D9Y7uvc/d/VnCNAU8TAtWUHwnB60vuPi/NNfWATYEjgeOBFkDlU4clUXPnzAGgWbNmaY83bx7+SWbPyd1A+yY9N2XLrfrQb4cd6dCxI7//9htvvfkGV15+Cfffdw8NGzXixptvLXPNhAkTuOKyi9lu+34cfuRROeubSFzzpo0BmBfLzY2bO38BAC2bN8mqvQHvD2Pf/vfw5L+PY+tNuy3bP3P2fN4d8iOT0gTUpx6yPb036sKJlz3BnxnSGuLarBR+t1dp1Zyrz9iX+/7vI25/4j1mzSlhr34bc8sFB3L6ETsyevxU7n/2o6z6Lfmv2FMOkqK8yjrk7gvd/Trg/tjuw2vSppk1Bl6gbJB7eUVBbuQCyga5jwI93P2pdEEugLsvdfcvo1HizoTXobpQRaj/GWdywkl/Z+111qFp06Z06dqVv59yKu9+MIhGjRpx3z13MX582Y+K/3HGaSxYsIA77rq3jnotUnNnHrkjr993OoO/Gk3vA//FKn3OpveB/2Lg5yO5/NS9eObmE8uc32n1Nlx+2l589MUonnxlaFb3qBdNv2/YsD5vfjyCf1z/LGMmTuPPmXN5fMAQ/nnrSwCcd9wuib42qTtGqLqQ1KOYQ2YFuvnhqdjzNcysWlW/zawp8AqwV2z3+e5+VSXXdQOuie16yd2PdfdF2d7b3ae7+8nAwVXps9Rc86jKwrx5ad+PMHduGDFq2aJlrfUpZcPu3dlz731YvHgxH75fOqv8xRf+x6uvvMw5511QYQ6vSNJSI7bNmjZKezw14ptNqa5te63D9Wfvz7CfJnDYeQ8x4udJzCtZyIifJ3HoeQ/yzY/j2WO7jdi174bLrrn1woNo1LA+p1/7TNZ9njNvwbLnTwwYstzxJ14J+zqs1oa1OrXNul2RYlBUqQtmNgZYM/qya7qUAjNrARwB7A1sDLQFGgLzgMnAKOBT4GV3/y6hrpWfgNYcqFLV76jfrwLbR7scOMPd78ri8vMJE+Qg5BOfWMG5FXL3r6p7rVRPp06hdNHEicvPII/v77TmmmmP59raa68DwG+/lf6Yv/5qqADx3rvv8PGgsh+1Dh/2DQDnnHU6K7VqxZFHHcORRx9TK32Vwjd+8p8AdFi1ddrjqf2p8ypy2F6bA/DyB8OWW+Rk6VJnwPvD6Ll+J7bZbG3eHvw9AHtuvzHTZ83jzosPKXN+k0bhz/Eaq7birf+cCcBRFz7M79NmMy7Wl7GTlq/HO79kEb9Pm8Vqq6zEqiu35JfxUyvtu+Q/pS4ko6gC3cqYWR/gOaBDmsMrRY/1CCOm15pZQ3fPrr5MxeJLSS0GsluSJxKNAL9OaZWEpcDf3f3BLK5tCsQTJJ9wd/1fcgXSY5Mw9++br9O/x0jt33jj7JYjTdqM6dMBaNZs+fTtz4YuPzqVMiwKeKuzAIVIJsOj6gQ9109fJ7dnVD/321GTKm2rw6ptgDApLZ1Zc+YD0HqlsvnzbVZqxna910l7TdMmjZYda9yo4bL2f50wla4d2y7XFoTJba1bNgXKjv7Kii3RqgtFTIFuxMw6AW8RKgoALAI+B34mjOY2B7oQKgqslPDt4x/3f+buWf+fysxWJvS7d7RrCXCMuz+ZZRNbAfFZF/+X7b0lP/TZui+tWrXil9GjGfbNN2zSs2eZ4y/+73kA9thz71rv24IFC3jjjdeAsuXP/vPwo/zn4UfTXrPrTv0Y9NFA3v1gkOrqSuI+/eYXZsyeR7fO7eixbodlgW/KX3fuCcDrH31baVu/TwsFcjbLsCBErw3DpyjjYqOwTTftn/bczu1X5qfXr2L0uClstO+Vyx1/beC39D98B7brvQ7vDfmxzLEte3ShcaOGzJu/kJFj/qi03yLFRDm6pc6mNMgdBHRx977ufrS7n+LuR7n7doRyWv2A/1LDiVdm1sDMzgZOje2+oQrXtwM+oDTIXQQcUoUgF2Db2PMlwBdVuFbyQKNGjTj51PDH86wzTluWkwthCeBvvx3Ottttz2a9ei3bf+/dd7HJRutz6cUX1fj+P/34I089+QQLFpR9fzZlyhSOPPwQJowfT48em7B13741vpdITS1avIT7ngnpMrdddBDNmpTm6p5xxI70WLcjH30xiq9/KJ08efLB2/HNC5dw1en7lGnrlQ9CDetDdu/N7ttuVObYXv025uDde7NkyVIGvD+sxv2+878fsGDhIk4+eDu22LjLsv2rtG7Ojef+DYDHXx7CwkVJfMgo+cASfBQzjeiWigd8x7l72s+tolSFgdEjG1eaWfm6Tg0J6QpbE3KAIdSqPcndX86+ywwENoieLwAOquL1EKolpIzLVGFB8tuF/7yE9997lyGffsJGG6xD3222ZdzYsXz+2VDatWvH/f95uMz506ZNZeRPP/Hb5OXXJ3nkoQd55OGQ9bJ4UZiP+NvkyWzXd6tl59x+5z1sulkYof399984/tijOPfsM9msV2/atm3H5MmT+PqrL5k9ezYdOnbkyaefxfQ5nOSJ6x98kx23XI8+Pbvx3cuXM/irn+ncfmW26NGVP/6czd+vKDtWsErrFqzXdXVW/3ZMmf0vfzCc/739FX/bdTNeuONkvhwxljETp9Glwyr06h5Gcy+782VGja35KOu4yX9yxr/+j3svO4x3HjqLocN/ZeacErbq0ZW2bVrw1ffjuOT2l2p8H5FCo0C3VDwdYUqC7WZTIPQtQpA7roptbxB7fnw1glwoXbACYEZlJ0epEhVWcQBed/fXs+2AmY3IcKhbhv1STpMmTXjr3Q+48Ybr+L9nnuKVAS+x8sorc+RRx3DZlVfTsWP6xSTSmThxAp9/Vrbs0cKFC8vsmz27dE2TtddZl/5nnMVnQ4cw4rtvmTZtGo0bN2adddZlj7325rTTz6RNmzY1f5EiCVmwcDG7nXQH5x23Kwf/pTd779CDP2fO4/EBQ7jqnlczLiaRzhEXPMzbn3zPEXtvyUbrdKDHuh2ZMWcebwz6jnufGcg7n/yQWL8fHzCEXydM49xjd2HzjbvQtHFDfp04lXufGchtj7/HvJL0tYFlxWMY9RIcHCjmiW1WwUqwBaeiqgvRErw7Rl9e6O5ZpxCUu0cX4NdqXDoVuNjdH6ig7X6EVIV0hgM7uHvlU4XLtvkOsHP05cfuvm0l53eh8td3pbtfUYU+ZAx0N9hww8ZfDct0WERS2myePvdTRMpa8ONTeMn07929e133JR0zG9F17fU3fOqNTxNr87Dd+/Drzz/m7WvOJeXolno29vx6M3vXzI41s+yHwtLr6u4WfxBKebUlBNaPEXJ92wL3m9m/KmirvHiCZQ/g7WrU4I2nVdTJqmbu3j3dA1h+gXgRERGRLCnQLfUg8FLs652Ah4HxZjbWzJ40s+PNrMbVuKMVxaa5+wfufgxwXOzwRdHIbTbtXA/Ep+f2At40s6qsDBAfAW6dxT3HlA/co+A925xlERERqYxmoyVCgW7E3ZcQls09Afi+3OHOhKV5HwQmmdmDUa5qUvd+FPgktuvcKlx7BWUrNWwFvGZmyxdbTG9s7HlnM6uTUV0REREpZQn+V8wU6MZ48FD0sfl6wEmE1IJfYqc1BI4HPovKeyXlrdjzfmZWP+OZ5bj7hcBtsV3bAq+YWZP0V5QxKPa8PmFUWERERGSFp0A3A3cf6e7/cfdj3L0bIfC9hVBrFkJFgMsTvGW8zlNzQr3erLn7P4B7Y7t2BF4ws/QLupcaAsSX9Tkk04kiIiJSCyysjJbUo5gHdRXoZikKfM+hbHC7T6bzq6F8qsHSarRxGiGvOGV34Fkzy1hGzt1LCKPWKUckkYcsIiIiUtcU6FZdvFbtagm2u1nseQkwLdOJmXioFXciYdW2lH2BpypJhbiR0pHqlsB/qnpvERERSY7moiVDgW6kCqOYnWLPE1lU3Mw6AQfEdn3g1Sxw7O5LgaOB52K7DwQeNbO0/97uPhq4OLZrPzN7xMwaVqcPIiIiUkN5GumaWVMzu8rMRppZiZlNMrOHzaxDFdpoYGZXmNlrZvaLmc2O2hplZveY2ZqVt5IdBbqlxpnZ/Wa2faaA0Mx6A3fGdr1R05uaWS/CRLR46sItNWkzqiBxOGVHn48A/mOZ12H9N2WD42OA4WZ2aEUVHMxsTTO7DtiiJn0WERGR/BZNcn8fuBRoAQwAxgPHAl+b2VpZNtWEkAq6HWGO0puEWKgRcAoh/uidRJ+1BHCppoQqCycBs83sG0LprbmExRzWB+IrikwBrsii3SvNbHa5ffWAVsAm5doEuM3d361q58tz90VmdiChNvDu0e7jCGkRp6U5383sUGAC8I9o9/rAU8BCM/sCmEiou9uQMFluA2Ddck39Dnxc0/6LiIgUszwtC3YJoYzpp8Cu7j4HwMzOBm4mzBPql0U7JcA2wFB3X5zaGaVZXgNcCNwH1DjYVaBbag7h3QmEPNVto0c6w4BD3H1SFu0eleX95wOXuvvNWZ5fKXdfaGb7A68SFsAAONXMFrj72WnOXwKcbWZvEhai2Co61AjYupLbTQAeAm529/KBvYiIiGTJiKolJNhejdsIVZxSa42flgpyAdz9FjM7GtjezHq5+5cVtRUFt4PT7F9iZpcCZwG9zKyVu8+sSb8V6JZahTCEvj2wObAOYbJZE2AeIZD7Evgf8HKUC1sTcwgTzoYTPgZ40t2n1rDN5bh7iZntQ/hYIBW4/yMKdi/KcM3bhOWEtwR2Jbw7W5PwPWpBWDZ4OjAS+JywKtoHCXxPREREJD/1JXwaPdrdv05z/HmgB7A3IV6qLidMkHdgYQ3aAYos0HX3LhUcWwi8Gz1qco8x5GiSo7t/WJ223X0eIYiv6nVDgaHA1VW9VkRERKovDxMXNom2X2U4ntrfo7o3iOYRXUBYT+B9d59f3bZSiirQFREREZFq6RxtJ2Q4ntpfpYoJZnYD4RP0lQhBcjfgB+CEavRxOQp0RURERPJN8kO63cxsRLoD7l5+Ynw6qXlM8zIcnxttW1axX38jBLcpw4Ej3P3XKraTlsqLiYiIiOQZS/C/fObua7u7Ae2AvwCLgC+jyW01phFdERERkcI3OsuR20xSVRYy1dZvHm2rVXkpmpD/lpkNAb4F7jWz9919fHXaS9GIroiIiEieMUvukZBx0bZjhuOp/WNrcpOonNgrhPUNdqlJW6ARXREREZG8k4cJB8Oi7WYZjqf2D0/gXqlyq+1q2pBGdEVERESkMoOBmYRJbT3THD8g2r6SwL22j7aja9qQAl0RERGRfGMJPhIQrTdwV/Tl3WaWyslNLQHcAxgYXxXNzPqb2Y9mdl2Zl2a2p5ktt+KqmTUzs2sJge5vhMWuakSpCyIiIiJ5Jk+rJVwD7AxsDYwys0GEurlbAlOA48qd3xZYD2hfbv/mwOVmNhH4hjBSvDrQE1g5+vqg+DLD1aVAV0REREQq5e4lZrYDcBFwGLAf8CfwKHCpu2daTKK8Fwj1drclBL0rA/OBn4H7gTvdfXISfVagKyIiIpJHEq6WkGhb0bK8l0WPys69Argizf7hwDnJ9Soz5eiKiIiISEHSiK6IiIhInsnLDN0VkAJdERERkXyjSDcRSl0QERERkYKkEV0RERGRPJOn5cVWOAp0RURERPJMkpUSiplSF0RERESkIGlEV0RERCTPaEA3GRrRFREREZGCpBFdERERkXyjId1EKNAVERERySuWcNWF4o2albogIiIiIgVJI7oiIiIiecRItrxY8Y7nKtAVERERyTvFHJwmSakLIiIiIlKQNKIrIiIikm80pJsIjeiKiIiISEHSiK6IiIhInkm2vFjxUqArIiIikk8s2aoLxRwzK3VBRERERAqSRnRFRERE8kwRD8ImSoGuiIiISL5RpJsIpS6IiIiISEHSiK6IiIhIHjGSrbpQzIPDGtEVERERkYKkEV0RERGRPJNoebEipkBXREREJM8ozk2GUhdEREREpCBpRFdEREQk32hINxEKdEVERETyTJJVF4qZUhdEREREpCBpRFdEREQkz6jqQjIU6IqIiIjkGcW5yVDqgoiIiIgUJI3oioiIiOQRI9nUhWIeHdaIroiIiIgUJI3oioiIiOSdYh6HTY4CXREREZF8YglXXSjimFmpCyIiIiJSkDSiKyIiIpJningQNlEKdEVERETyjBaMSIZSF0RERESkICnQFREREckzluB/ifbLrKmZXWVmI82sxMwmmdnDZtahCm20NrPDzOxpM/vVzBaa2WwzG2pmZ5pZw6T6q9QFEREREamUmTUB3ge2AiYDA4AuwLHAXma2lbv/kkVT5wIXAw58AwwF2gF9gS2AA8xsN3efV9M+a0RXREREJN9Ygo/kXEIIcj8F1nX3g919S+AcQqD6cJbtzAX+DXRx983c/RB33wnYGBgHbBPdq8YU6IqIiIjkmXyLc82sEdA/+vI0d5+TOubutwDDge3NrFdlbbn7de5+gbuPK7d/FHBh9OWhSfRbga6IiIiIVKYv0AoY7e5fpzn+fLTdu4b3GRZt16hhO4BydEVERETyipFsebGEmtok2n6V4Xhqf48a3metaPtbDdsBFOiKiIiI5JmkqyUk0lbnaDshw/HU/jVreJ8zo+2AGrYDKNAVERERKQbdzGxEugPu3j2L61tE20yVEOZG25ZV7ViKmZ0M7AzMAK6vbjtxCnRFRERE8k2RrYxmZtsCtxNKjh3n7pOSaFeBroiIiEjhG53lyG0mqSoLzTIcbx5tZ1e1YTPbiJCq0Ag4w91frHr30lOgKyIiIpJn8nBAN1UKrGOG46n9Y6vSqJl1Bd4G2gBXuPud1eteegp0RURERPJMklUXEpIq+7VZhuOp/cOzbdDM2gPvAO2B2939yup3Lz3V0RURERGRygwGZhImtfVMc/yAaPtKNo2ZWRvgLaAb8AjwjwT6uBwFuiIiIiJ5xhL8LwnuvhC4K/rybjNL5eRiZmcT6ucOdPcvY/v7m9mPZnZdmddm1gx4jbDk77PAie7uiXS0HKUuiIiIiOQTSzh1Ibm2riGU/9oaGGVmgwh1c7cEpgDHlTu/LbAeITUh7lqgD7AEWAw8ZGlesLsfU9MOK9AVERERkUq5e4mZ7QBcBBwG7Af8CTwKXOrumRaTKK9NtK0ftZPJMdXqaIxSF0REREQkK+4+390vc/e13b2xu7d392PTBbnufoW7W/mRWXc/Jtpf4SOJ/mpEV0RERCSPGMmmLuRfAYfaoxFdERERESlIGtEVERERyTNJVUsodhrRFREREZGCpBFdERERkTyThyujrZAU6IqIiIjkGcW5yVDqgoiIiIgUJI3oioiIiOQbDekmQoGuiIiISJ5R1YVkKHVBRERERAqSRnRFRERE8oyqLiRDI7oiIiIiUpA0oisiIiKSR4xk56IV8+CwAl0RERGRfFPM0WmClLogIiIiIgVJI7oiIiIieUblxZKhQFdEREQkn1jCVReKOGY2d6/rPoikZWazGjVu3LJbt2513RWRvPfjL5PrugsiKwRfMBN86Wx3X6mu+5KOmY1o1KjRht3WXjuxNkf//DMLFy783t27J9boCkKBruQtM/sNaAaMr+u+SBmpdx6j67QXIvlPvyv5qRMwz91Xr+uOpGNmL1P6s5Ok0e6+Tw7azWsKdEWkSsxsBEAxjgyIVIV+V0TqnqouiIiIiEhBUqArIiIiIgVJga6IiIiIFCQFuiIiIiJSkBToioiIiEhBUtUFERERESlIGtEVERERkYKkQFdERERECpICXREREREpSAp0RURERKQgKdAVERERkYKkQFdERERECpICXREREREpSAp0RURERKQgKdAVERERkYKkQFdERERECpICXREREREpSAp0RURERKQgKdAVkTpjZlbXfRARkcKlQFdE6oy7u5nVM7PWoMBXpCJmttzfbDOrXxd9EVlRKNAVkTphZm3NbDvgdeBGM+vg7l7X/RLJV+6+1IKVzWyHaN8SBbsimTWo6w6ISPExs02AE4C9gc7ADKCBmf3D3WfUYddE8pKZNQNWB64G1gF6m9lD7n6iuy+p296J5C8FuiJSK8zMolSFHYFbgLWAFsCfwGTgI2BRHXZRJC+ZWXfgMOBAYO3YoePN7A93v7hueiaS/xToikitiILcPYBngWbR7h+Ac4Ef3H1MXfVNJF9FKQo3EEZxWwHTgd+Ap4BRwJdmVs/dl9ZdL0XylwJdEakVZrY78Gr05XjgLeACd58eO8eUpysSRG8MnwZaRru+By4kvDEcXWcdE1mBKNAVkZwzs22BJ6IvRwH/BW5199lmVj+VY6ggVyQws79Q+sZwEmHS5vnxHPbyI7l6oyiyPFVdEJGcMrNOhFGolYFZwPPAzVGQWy+biTQqOybFJHpj+Gj05c/Af4Bz3H1GvMJCpiDXzDY0sza12GWRvKVAV0RyIlbzszuwafT8K+B6d5+bKa8wXVBbfpQqXT1RkUJgZmsD1wCrEvJxXwRuquiNYbQ/FeQeDlwHnG5mq9Vi10XyklIXRCQnYkHsGYSySHMIObll0hXi4vvNrCGwBdAxuv5bYLy7j4rqiWoCjhSM2M9zP2D9aPenwL8qeWO4bL+ZHQycCvQBNgeWmtl/3P33WnkRInlIga6I5Eys9udSoAT4vaJ0hViQ+09CkLtP/DDwo5k96e7XpYrnKydRCkH089wM6A+0I5Tc6+/uMyv69KNckNufEORC+L07KTrtAQW7UqwU6IpILrUA2lKaJrWg/B/sWH3dVkBP4CJgV0JwDLAQMML/rzYArjWz9u5+RnRd2tFhkRVF7A3bXoSfcQfeByZXYST3NKBvdPgPYBXCpyEnRuco2JWipDw3EcmlhcDi6Hkz4C9m1ih1MJVbaGZdgUuA+whBLpT+/2kpUJ8Q7Kb0N7MboHQUWGRFFftUogfQkPCzPtDdl3tjGLsmFeQeApwCbBMdugQ4klDdZAmlwe5JytmVYqRAV0RyJiqF9BxhxbNmhFGnbc0sVRd05WiltNeBk4H1Ype/Q5hUsyuwO3A38FPs+HlmdlhOX4BILYhVUkj9/M8HpkbHMv6dNrPewD+B7aJd17r7v9z9HeAKlg92T1SwK8VGqQsikmtDgAWEkarehFHbsWb2I2EEawsgNcq7hPAH/h7gRkKqQ2o2+VDC5JzLCatELSFMuHmq1l6JSG6kRnRXjbZNgMZQtoRYGsOAz4CNCEHupbAsFeJZM1sCXE34fUkFuw3N7A53n5b8yxDJPxrRFZFqs0j0vF6G0mAvAv+K7eoG7EiYHd6X0iB3IfA/4BR3v9rdS6K0BovamQW8BnwYnV8f2MPMWif+wkRqUSyYTQWfTqg7XWENaXdfBJwO7BcLcuN1dv9HeGM4kvDGsBNwPKXl/kQKngJdEak2j0TPl8YrIEQxcL3o2PWEj1jnEP7gEm2N8DHtBOA4wspPL0XXp65NjejWc/eZhBHc1H3qRw+RvBZ7Q7idmW1Q7ljqb3Fqslg94FgzW7WyqiLuPt/dX47aqe/uS8q9QXwOuACYEV2yRuy5SMFT6oKIVIuZtQfWBE4AViKURHoV+MzdB6X+2Mb++F5vZj8Qyh/9lVAMfzzwBvCOu4+PtW1e7iPb2NerEQLkpdH1c3P6QkVqKFZZ5K/A48DW8eOxn+1XiUqCAe2BjYH3sq0sEj+nXEWSHwijxSsTPhH5vuavSmTFoEBXRKrMzPoC5wNbUppXCCEVYZqZXeruD6YWdkiVQnL3AcAAM7vS3eeXa3NZuaR0o1ix46sRRnTrAT+4e0luXqVIMqKgcz9Cas5CwlLY6fwKTCTk03YgjMS+5+5LqllGL/V71IuQMmTAJ+4+r4rtiKywlLogIlmJffS6OzAA2IMQ5C6ITllCePO8GvCAmf0dlqU0lJ9Qsyw4jaUoZJx0kxrhNbP1COWTjDAqdW+8byL5JnqjtypwfbTrd0oD0DLc/XvgydSXwM5mdk90bEk8/zaL+8Z/Z24ipPh8AtyaOl6d1yOyolGgKyKVin30ujdhQtjKhGD1K+BcQmmwJwgjUinXmdmuyzVG2RHbypbxjd27I3ApoRD+fOAjYEz59kTySfTzPYPS4LYhYVS3jFie7l3Ae5TWjT7azK6O2soq2I39znQilBnrQFhE4jlgdtSWfmekKCh1QUQqVC7IHRDtHgW8SChpNDs67xlC7uHNhNWdmhJSGd5O4N6dCAH1ToQ36D8C16fuLZLnGlP693YJaQaZYm/4ficEpF0I6QZNgb9HqTsXV5bGEPud6UD4ndmREDR/BTwTVWoQKRoa0RWRjDIEuT8ADwNXu/tsM2sAy8p/DQRuJ+QgNgZ2SB2vyj3L3Xt9Qk3dgwhpEZOBQ919XM1foUhuRSOwTShbRi/tpxjRz/wS4DHgGWBKdKgtcLaZ3R87J/670iB2vZvZ2sC1wAGESaKjgNO1BLAUIwW6IpJWhiD3R8Ks8TvdfW40ypRa4pdogtlbRB+PAnPjx8u3n+7r2Eeqq5vZXoQR4f0pDXJ3dveRibxIkRyLgtL4hMl6hJUC053r0e/UAsKqgM8TUg4gvHE8EXjXzPYxszVipf1Sv2PNzWw74P+AvxEqN/wB7OvuoxN+aSIrBKUuiMhyMgS53wFPE4LceelKgEUjS9MJOYkdgP8zs87AhoTc2hmElZymlit/lPoj3wxoDRxMSFPYhZDTOA/4GjhSQa6soLKa/JWqVBL9jp1LWCnwEMLqZgA7EFZC+9PMngb+BBYTFoPYFNiZ0r/t3wEHunt86WyRomLKRxeRdMxsH+Cl6Mv5wF3ufkF0rF6aIDcVHPckzO5eTEhl2LNc018Ag4Fzo3zDBqkRKTPbmTAZZ11CLmN9QqrEe8CN8Vq7Ivkk9fOfZn99wqcR3xDe7P0AbOPuMyppr14U9DYGDiWkIexR7rSlhBHihYTUCCcE1L8S3hie5+6/IlLElLogIsuJRmY3iu1qChxlZqunTil3fnzS2HWEnMQWlAa58Zq5vYEzgbdSQW5qJrm7vws8GJ1XH3g5OvcSBbmSj2IpOE2jr8t8Uhp9YpFaBTB1bjaLP6RGdhcQ0oWOIFRQiC/2kPobnsr/LQFeB84B/q4gV0QjuiKSgZmtDBxPqP+Z+gP9G2E06pdU2kEsyF2NUOD+UMII1nzgFcJI1jTCaFQfyi4w8RqwX5qR3YOANu5+f85fqEg1xX72DwDuA7Zy95/LV0Uws1bAT4Sf/ZHA5tWtGGJmaxIqMuxIWFiiJSF94WtgpLt/UJPXJFJoFOiKSPwPdlt3nxrb34awJOm/SB/sNnL3hVGQezYhl7ATYTLawYRVmGbF2juOEDz3IXzMupgwWntjdDxdSsRy+0TyRbkUn/HADuXfCBJ+J4YTlsr+hvD7U6XVyarye5ApjUKkGCl1QaTIxYLcvwEjLaykBIC7TwceAP5JacH71YGPzWztKMjtSPioNB7kbu/ub0bPMbNGUXsPA3cTRreMMNGsT+x+y/0hV5Ar+SpNik8nwu/GWrHFHYwwOSyV0rCQ8AavStK9AYy2ZuUWkVCQK1JKga5IkYuC3L0IRepbA9tCaa5hBcHuQDPbkjBCezClQe627v5NNKKVKn+0MFY+7Cng3VgXdjGzdRBZwUSpNvcRUnbKvxFcy92XRAHqPEpLijUh5J/X9N5Lo61nWjxCRBToihQ9M2sC7B7bdTiEP+KpUaMMwW57QsB6EqVB7nbuPrx8jmLUhsdGnm4jLCrhhP8PaQRKVkju/idhAmW6Tz3Wir5uQfj0Asot/xufvGZm9cuN1MaPNYgtEFEvtj+rsmUixUqBrkiRc/cS4P3Yru3NrH90bGnsvHTBblNCwDsX2NDdh6ULcmNtpPavHF27hDBRrSTd+SIrgkpSfNYFZgLNomO/AxuaWRszWxVob2atoxrSqwArRQFuc8ICEPWjrxsTjQRHFRnaRc/1JlGkAlowQkRw9/+Z2X+JRnOBLc3s7vJ/RN19upk9EH35L8Kb5dTKTzOj/RlHmGITalYl/P/HgC/dfUJyr0ak9qX53TBCsPseYUWzCYQqCXsSFkOZS/gdqE+oULKYUCZsAeH3qT6hTm4JpcHzAjObT3hzOcLMjnT31O+diKShqgsiRS42Ge1QQr5hy+jQ3u7+WoZr0lVj+J0wm3x0ulHd2H26AC8CmxBmqZ8dBdqaKS55K/7zWdHPaobfjYWEwDWVupNa6KH8c6eCN4qx46lc+OHVezUixUOpCyJFLvYH+0VCNYSUo6OPV5f7w5vho9rVgEHlZpwDZYLcNYBzCXVAIZRaGliuHyJ5J/7zGQt4l5tUluF3ox4hyF1EqHc7g7BU9rTo+azoMZcwujuXMJK7mBAkp/J6jVAzt4+CXJHsaERXRIjV/NwfeAhoBUwEdnT3UZlqeGZRZ7cBsCQW5J4DHAR0AH4BdnP30Tl/gSLVZGatCak2hxAmlc0DRgADomoi2f5uLCYEr/sCQwi/A07IvV0aPW9G6ZK+TaNHSbSvVXTeKK14JpI9Bboiskw0ceZFYINo1/8BR6ZWLMtwTYXBbnROKsg9kJCnOB3o6+4/5uJ1iCTBzLYCTibk1HaIHZpHGJk92N0nZUplyPC7MYlQneSXnHZeRAClLohIjLuPBO6I7eoObAhlSxqVu6aiGeerRqO651Ea5P5JCIIV5EreiZXw2g14nlAjugNlF3loBvQF3jazNcuVzlsmw+/GGsBgM+sa3UeTwkVySCO6IgKUyaPtCvwX2Co6dLW7X57F9elGr6YArwA7A50JI7nbuPsPSfdfpKZivwN7AwNih36Kvm5D+DneLXbsc6Cfu8+voN3KUnwyluQTkZpRoCtS4KpTzcDMbgHOir6cCOzr7l9lcV26P+glhNWgFORK3soQ5P4MvApc5u5zovNWI6wGeE10ziLgZHd/pJL2FeyK1AGlLogUmNhHr61h2YpkWa2eFEtPuAb4PnreFuhR7nhaGT6qbUKYWb6tglzJRxmC3B+Bh4FL3X2OmTUGcPffgZsoTfFpCKxf2T0qSfFZrlKJiCRDga5IAYn9wd4XeM7MLjGzpkT1O7MIVJdGQfF84KNod2PgEjNbPd3s8jRtpP6gXxDtWkgYtfo+81UidaOCIPdR4A53nxtVVliQusbdFwLvED6lgFANoVJZBrvK2RVJkFIXRAqMmR0APBvb9Snhj/Kj7j4mdl7askix45sCgwkjsnOBc939/squi12/MnAk8I6CXMlHsSB3H+ClaPcPwOPAXVGQWyb1J3ZNd+BjQtmvXYDhhFrS7YBRwEx3nx2/JtZGujSG8cBO7v5z7l6xSPFRoCtSIKKR2MbAIGA9Qs3PlKWE9IFbgE/d/YPYdcsFrrE/5jcC/yB8+vOmu+8RP55Fn7IKikXqipkdBDwTfTkVuA24xd1LKvnd2IWQv/sbYcLlwYQ3hU0JC0EMBe5z99cz3DddsDsS2BhYrAVURJKhQFekwES5uZsQRlNTf3xTuX+LCDVAHwNeIAS9i2LXlh95Kj86fKK7P5TTFyBSS6K827eBbYEFhDeK97v7KRnOTwW5nQk1prdMc1r5ZXxPdvcH4tfH2mtDmNj2b8KnJlu7+7c1f2UikqJAV6SAlB+BMrPdgS0IdWwbAfH8v3GECWeXA+OiSTbLtWNmTxFWhYKQt3gCYY6b/uchK7xoMZOHgV1ju+8GzolycVPnpYLc1YELgcMIEzVLCOlBPxGC5T0ItXKbx9o7xt0fz3B/pfiI5JACXZEClCbg3YDwh3k3oHe506cCwwiTZN6IlVFq7O4LzOxvhD/8q0bn7+DuA3P9GkRqi5m1B54AdoztvouQl75smd+otNg5hE9KOgFzgGMIn4xMjtrakLDM9SmEfF2AJYTltAdluL9SfERyRIGuSIGLjUQ1JOTaXkj4yPUvaU4fAAwhfJSaKk22CmEyW8/onGcIH8fOynXfRWpLBcHu+VG+7urA2ZQGubMJC0V8HfsdSwXEqxMC4POB1oRV1a5398uqU9daRKpPga5IkSiXjrAy4aPa/sA6lI48pQwl5PC+5u7fR8uhPkVYGWoMsIu7j9ZIlBSSDMHurcDtwGmEFJ5UkLuduw/L9DtgZmsD9wI7RbuGRtcsKn+uiOSOAl2RApJutMjMGrj74nTHzawD4Q/3lYSi951il86PHlcTPqI9HOhLKJD/mLsfm8vXIlIXotHYJygNUAG+JuTjlg9yK1zNzMyOIuS1Q8jh7ePuM3LRbxFJT4GuSIGI/9E1sy5AH+CZ6CPVMn+Q0wS8jYDtgL0Ik80aRg8I+YU/EQrbr0yYVf4dcJC7/5TzFyZSy8oFu/EqCrMJi598W1GQG0tl+BvwXLR7MLCbu8/LcfdFJEYro4kUgGjUNhXkngP8B/gvMDD6o1vmD3K5ILeeuy9093fd/Sxgb0Ilhj8If+TrAxsQ0haWEvINNwb2zPkLE6kD7v4boRLCe4QgdwnhZ//1KMg1Kl4NLRUYrxTb95OCXJHapxFdkRVcudSEh4ADgJbR4dHAv9z9kSzaKV+poTNwHCFfcZtod2p0ax7hY1jV/JQVTmUpPrF96dIY7gXO9tiSwOnaNrNuwJtAN8Lv4THuPliT0URql0Z0RVZg0cenqSD3FeBYoBkwEziXkL6Qtn5neeWC3HruPo6Qn7sTcAHwGiHIXQhsriBXVkTR74xHz7uY2aFR8LnYzOrHzy03sptyCnBTlO5Tvu1UkNuBUHGhfXRoCKFmNQpyRWqXRnRFVlDlqii8BOxDmDw2GLjT3V9Jd24V72GxoKAVsDUwxt1/SOAliNSqcp9+nEMosbcT8DGwfaYgNMPIbplFJWJB7hqEWrsHAR2AUcBf3P3XHL0sEamAAl2RFZyZ3UEoE7aA8FHpTe4+ODpWPh0hHhxn9RGqPmqVQlDTFJ8Kgt1zU2kMsSD3QKAj8CfQV5M2ReqOUhdE8pCZta7kuEXbnSlduvQn4I5YkGuxoDb1u74sYI2N1KYmzqSlIFdWdEmk+GRIYzgNuDlqd3WWD3K3U5ArUrcaVH6KiNQmM3sV6GBmu0d/XJcTCz63A9aNnt/u7h9EbRghnzZ1XkMzWw841czWAloAXwIPRys7aeEHKUjRz3aqIslLhGoh1UrxcfffzOxIyo7snmpmzQlB8/6EIHc6sK1SfETqnlIXRPKImT0KHBV9+SZwortPzHBuQ2AgsBUwHujl7lPT1MzdkZC/ezzQPNaEA7OAfdx9kFIUpJAlmeKTIY1hJtCKEORuoyBXJD8odUEkvzxIKN1VQpgoc10FqQX1gdbR83pAIwB3X2Jm9cysvpmdTfhI9u+EIHdudP4CwohvK+BZM9tYQa6saGo7xSfVXpo0hgWUBrkayRXJIwp0RfKIu38M7AA0IUyQuSBTAOruJYSSX0sIs7uvNLPVzKwtIQfxCeAmYA2gMfAJcBFhNvj5hBxCCMHyX6DyfF2RfBGl+HwQja6mVYUUn5SGZtYDuNfM3jazT8zsTjPbNGorXbD7FuH3609CkPt9gi9TRGpIqQsiecjMNgEmu/sfsX0ru/uf5c47CLgdWJXwB/sHQv5ta0pnlH8NvAv8s1xKw2HAQ4Q/0q+7+145e0EiCcqnFJ+o0sLdwOXuPjyp1ygiydCIrkgecvdh5YLcz4D3UqNXsVGlZ4E7gTnRqRsAnSgNcu8Eznf3C6KUhvgE1KmE9IelhI9cRVYUeZPi4+6TgAMU5IrkJ1VdEMlz0Ue0vaMvHzezo919cqouqLv/y8z+AHYD+hJGn74E/ufuL8TaqRet/pSaaNOY0v8H/F57r0ikZtz9YzPbARhKFik+ZvYasA6lKT6XEFJ+9iVMKDskdsknwDPAb4SVzS4HVqY0xefb8qO68ZFhEckvSl0QyXNmtg3wBiFvtz5hAsxR8WA3Oq8J0IYQ6C5OFbEv11Zq9aa1gP8BmwAjgUOjMmOqvCArDKX4iEhllLogkueiCWo7AYsI+YI7EUZ226dGaKNTF7j7ZHefG51bRizI7UCYlLZ2dCg1KqbFIWSFohQfEamMUhdEVgDu/pmZ9QM+JIwspYLdo9x9cnRO/KPUMkXvU+kKZtYNOA/YnZCLOJIwiWZWbbwOkVxRio+IpKPUBZEViJltQWmwa8TSGNKcW352eC/Cx7fdCZNrpqAlSqVAKMVHRNJR6oLICsTdPwP6EWaDl0ljSHNuqtD9nmZ2BfAxsDUhyB2JglwpIErxEZF0NKIrsgLKdmTXzI4glE1aSnhjOx74HDjP3X+tzT6L1IaqfOqR5tryKT57ERZcGQn8xd3H5KrfIpIbCnRFVlBVCHbvBk4BPgCeBF5x96m121uR2qMUHxFJUaArsgKr6A+6mTV090XReQcTcg2Xlp+oJlKIqjqya2Z7ApsDF0TXQBjJ3VdBrsiKS4GuyAou22A3OleTaKRoKMVHRBToihSADH/Qj3T33+qyXyJ1TSk+IsVNga5IgUjzB/0LYG93V+1PKWpK8REpXgp0RQpI9Af9bWClaFcXdx9Xh10SyQtK8REpTqqjK1JAojq7ewKTge4KckWCCmpQrx4PcqNzFeSKFAiN6IoUIDNr6u7z67ofIvlGKT4ixUUjuiIFSEGuSHqxkd3Z0a7elJYTE5ECo0BXRESKilJ8RIqHUhdERKQoKcVHpPAp0BURERGRgqTUBREREREpSAp0RURERKQgKdAVERERkYKkQFdERERECpICXREREREpSAp0RURERKQgKdAVERERkYKkQFdERERECpICXREREREpSAp0RURERKQgKdAVERERkYKkQFdEioaZfWhmHj36ZTjn0dg5x9RqB1cQZnZM7Hv0aALt1fr33My6xO45pjbuWR1m1i/Wzw/ruj8iKxoFuiJFolyQl+4x28zGmNkAMzvdzFrVdZ9FRERqQoGuiKS0ANYE9gHuAMaZ2VF126XCF3+zUdd9EREpNA3qugMiUic+Bz6LfW1Aa2BzYJ1o30rAY2bWxN0fqN3uiYiI1JwCXZHi9Lq7X5HugJn9FXgESKUu3GFmr7v7hNrqXF1y92OAY+q4GyIikgClLohIGe7+InB4bFdj4NQ66o6IiEi1KdAVkeW4+2vAsNiuneuqLyIiItWlQFdEMvkk9nyt+AEzuyI2ieqKaF9TMzvezN42s3FmtjA63jNd42a2k5ndZ2YjzOxPM1tgZpPM7C0z629mTbPtqJnVM7OjzewdM/vNzEpiFST2q8qLrk6pKzPbxsxuN7OvzewPM1tkZrPM7Fsze8zMDo2/nnjJqHLtZKqI0aWCe3cys0vNbFD0/VsQfT+/NrObzGzdKr7+zczsP2b2i5nNN7MpZvaZmZ1vZitXpa2kmVlDM9vNzP5tZh9Er7ck6ucEM3vDzM4ysxY1uMdOZva0mY2Ovf5B0c9k4yq21dzMTjGzV8xsrJnNi6qbjDKzh81sx+r2U0SyoxxdEclkeuz5ShWdaGYbAM8B3Str1Mw6AY8D/dIcbh89dgUuMrND3H1QJe2tDrwEbFnu0JrRYx8zexE4urK+VZWZdQQeBnZJc7glsFH0OAoYCmyV4L3rAVcA5wFNyh1uBLQBegJnmtm/gUvcvcLKDmZ2DXAhUD+2uwnQljBR8XQzOzCJ/ldV9HPzNbBKhlM6RI+/AJeY2aHu/k4V2m8I3AWcVO5QE2Cb6HGqme3n7iOzaO9AQvWS1dMcXjt6HGtmrwJHuPvMbPsqItlToCsimbSJPa/oj/AqwJtAZ6AE+BgYSyhXViawiwLi9wjBLIADXwHfA/MJgcp2hCBxDeAdM9vd3T9Id2Mzaw28D2wQ2/0r8CmwgBB4bwH8FVha0YutKjPrDrwTey0AfxBGwqcQAqRuwKZAU8oGoxOBu6Pnp8X23016s8rduz7wf8DfyrX5WXTvFoTAvxvh//P/BNqxfBAXb/NfwEWxXfMI39vJhGBtR6Aj8DpwW6Z2cqg5pUHudGAE4edsDiGw70r4eWsSnfe6mW3v7p+kaSudGyj9/gwHviFUI+kFbBjt3wB438z6uPv4TA2Z2T+Am6PrIfz7fQpMILyJ6A70jo7vBXxoZn3dfV6WfRWRbLm7HnroUQQP4ENCYOnAFVmcPyx2/mfljl0RO7Yo2j4HtCt3Xj2gYfS8OSGgTV33OtAtzX1XAu6JnTcJaJWhjw/FzlsAHJfmnC2AMbFzUuf3y9Dmo7FzjslwzkrAyNh5U4BDAUtzbnPgMODhDG2l2vAq/FteFbtuMrB/hnsfCMyInXtQhva2I7wRSJ33HNCm3DmtgKfTfB8fTeBnM5vv+ZqEEdItgHoV/LvcFGvrpwrO7RI7b2G0nQrsmubcvQlv9lLnv1nBa9kJWBL7Pl0ANEtzXk9CsJ5q854M7fWLnfNhTb/XeuhRbA/l6IrIcsxsT6BHbNd7FZzeAHgbONjdp8QPuPtSd18UfXk2pSOvLwJ7ufvo8o25+yx3PxV4LNrVHjg5TR/XBY6N7TrJ3R9O095nhFSIeYSRvyScT2m94ZnANu7+tLsvlxrg7nPd/Sl3Py6JG0f5uv+MvvwzuvcLGe79HGE0O+UKM7Py5wHXUTr6+B5wiLvHU1fw8NH64YR/66S+j1lz97Hufoa7f+buaUfno5+dc4H7ol3rArtl0XxDQqC/j7u/nabdVyg7er5buvzaKJ3kXkrnvxzi7jd4mpFad/+GEBT/Hu06IUqFEZEEKdAVkTKiyVtPxnYtIIywVuSsTMFH1GZDoH+svZMrOj/yT8IoFpQtd5ZyPKXB2Wfu/liacwDwkFN5WyX3y0o0ISmebnChu/+URNtZOpPSHNqr0r1ZiPOQ9vFW9OUGhFSKZaJ0kq1ju85w9yUZ2loKnE7pv0u+eiT2PNuKIf/1CtIc3P1d4IXYrhPTnLY3pW+AXvJQqi8jd/+N0p/LhsBBWfZVRLKkHF2R4rSHmbUtt6814WPhdcrtP9sryEcEhrv7D5XcrzewavT8PXf/o7IOuvskM/uREJxtZGatvOyEnR1iz5+orD3CCPE/Kz2rclsRvlcAsykdea4te8SeP5XlNe9TOrK5DSEvOiX+ffzS3b+vqCF3H2lmQ4A+Wd47cdEbpy2BTQj5wy0p+/esZex5zyybfTyLcx4jpIlA2e9bSnX/bVK2AW7J8joRyYICXZHitHn0qMhs4Ex3f6SS877M4n7xoKijmd2VxTVQGlAaYSLUTIDo4/dNYud9WllDUYD2J1DTElnxCXZD3H1+DdvLmpmtQvg4HkJe6eXpMxGWs2Hseadyx+IjvJV+H2Pn1XqgG5Vo+ychlaX8G7VMsjnPCVUxKhP//qxmZu3dfXJsX/x78jcz2z6LNlvFnpf/txGRGlKgKyIpc4BphBnn7wKPu/uMLK6bUvkprBF73oOy+b/ZileBaEXZPNFxWbYxjpoHuqvFnv9Sw7aqKl7hoRFlUyiy1abc1+1iz6vyfaxVZtaGMPrZs4qXtqz8FKa7++zKTnL3KWZWQmkFjXaEyYAp8Z/zg7Pv4jLl/21EpIYU6IoUpyvd/YqE2spmRLNV5adUKv7/q/ILAmRblmluAv2IB05zEmivKpL+PkLZ72Vtfh+r6m5Kg9yFhFSDV4AfCMHm/FRucTRh79fo3GzmolSlrNdcSgPd8kF0Tf999DdZJGH6pRKR2hAPjO5w9zNr2F75ALMZ2QVfzWt4XwgpHSnVXoGrmuKvcZa7JxH4xr+XzbK8JonvY9bMrANwSPTlUuAvnqG2ciSbUdy4bF83lH3t5UeB51Ia7G7m7l9XsR8ikjBVXRCR2vB77Hm6laKqaiZhVC+lc5bXJZEDGX8tXRNor7r3XsnMqhKgZRJPPanN72NV7EhphY03KglyIdTcrYo22SwbHE3gjC/8MbXcKUn/nItIDSnQFZHaEJ/os3WGWq5Zi2rGDovtqnRpXTNbh8zLx1bFkNjzPtEEqVoRTXyKV8DYOtO5VRAfdcx2ieLanogWz339Novzt6ti+8byS0inE3/dv7v7pHLH4z/nfavYBxHJAQW6IlIbBhNW6IJQPWHvBNqMj+odkcX5RyVwTwiBbmoxhZYJtFuSehKVzarMq7Hnp9bw3lD2+9jbzNav6GQzW5vaD3TjNZcrHMWORrmr829yZBbnxNtNN6oc/7c5zsyapDlHRGqRAl0RyTl3X0DZBRvuifIus2Jmq6XZ/VDs+VZmljHYjYKzf2R7v4pEryW+gMYNZrZeDZqcFnuezffkZsISswB/NbNjsr2RmS33cXpUAzm+UMLt0Qpf6a6vR1iGt0Yj8tUQr26xh5nVz3hm+P6k+3mpzBFmlnFU18x2oOzqaA+mOe1/wM/R8/aEn/Osvldm1sLMajX3WaQYKNAVkdpyMzAiet4B+MLMDqwgqGprZieZ2VfAeeWPR6udPRrb9aCZHZ2mnd7AO4RJRAvLH6+mfwOpFclaAR+b2SHpghoza2Zmh5rZcssTR76LPT+wshtHK6FdE9v1sJndlGYBkNT9G5jZrmb2BGXTFOIupnS1s12Bp8ysdbl2ViIszLE7yX0fs/U+pZUR1gYeS9c/M3uAUGO3qlUhFhFWm3vVzJZbSS1aEvtFSgP8d9x9uWWxo6oPp1D6RuRY4LVo9bm0zKynmd1ASEmp7ZxvkYKnqgsiUivcfY6Z7UOo0duVMFnnWWBqtNLWb4RAYmXCAgfrUPpm/P3lWwTgbMLH6OsBjYFHzewyQmH/BUB3wmpvRli+dRUgmyL+lb2WWWa2PyGAXpWwKMHTwG1m9glhglcToBuwGdCUsjnFcf+jdNWyG8xsd8IbggWxc6519+mxr68EugBHR6/tHOB0M/uCEIDPA1aKzulBaaWA+Ohx/PV8aGY3UfqG4mBgLzN7n/DvshphQlgLQtrG7cAVGV5P4tx9etS/y6JdhwO7m9lQYCJh9LQf4XUuJqR0VGXFukmEQPYs4B0zGwZ8Q/je9iL8HKVMJv3yv6m+vmtmpwD3EoLn3YG/mNn3hBrVswjpF+0Ji560y9SWiNScAl0RqTXu/ks0wnofcAAhkGgL7FXBZTPIMAEpCoB2BAYQlhkGWCt6xL0MHEOou5oIdx9uZlsQ6rmmJj+tBvw1wyWZau4+Ssgx3o7w/diB5ZeXvYvSvODUZLxjzOxLQtDbhrCAxNZknqDmhFzpTK/nfDNbApxPeIPRnOVzqScR/t1qkqpRXVcRAvdUnuzKhCAybgZhFPWbarR/PiHn+nhCALpJmnN+AvZz97EVNeTu/zGzn4H7CW/YjBAsd6/gshHAn9Xot4hUQIGuiNQqd/8TOMjMNgIOJYzEdSWMti4lBCs/A18RRn/fcfeStI2F9iaZ2VaEAOhwwghmK0Kpp2GEkb3/ubvXsNhDunuPBbY3s50IaQfbEkbqViJ8fD6WsETya4RgO10bi6KPy48n5IBuRAjiGqU7v9y1d5rZo4SJVLtQOkLYhFDjdQIhgPoQeN3dx6dvaVl7F5nZ84QR0R2j1zIHGEMYEX/A3afWMCe5WqK0gKPN7DngJEKVhDaENwDjCG92Ho5+HrpUo/1FwAlR+8cTlshuT/h3/AH4P8LrX5C5lTLtfRClLOwH7EmoaLE64WdjHuHn80dCfvQb7v5NVfssIpWzMDAgIiIiIlJYNBlNRERERAqSAl0RERERKUgKdEVERESkICnQFREREZGCpEBXRERERAqSAl0RERERKUgKdEVERESkICnQFREREZGCpEBXRERERAqSAl0RERERKUgKdEVERESkICnQFREREZGCpEBXRERERAqSAl0RERERKUgKdEVERESkICnQFREREZGCpEBXRERERAqSAl0RERERKUgKdEVERESkICnQFREREZGCpEBXRERERArS/wNQJfnW8jf02wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "C = confusion_matrix(Y_true_max, Y_pred_max, normalize=\"true\")\n",
    "C = np.around(C, decimals=3)\n",
    "\n",
    "plt.figure()\n",
    "plot_confusion_matrix(C, classes=['isVBS', 'isBKG'], title='')\n",
    "plt.show(1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "d364ee44-a77c-4a0b-9a4c-845fdf4aa08e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "48438\n"
     ]
    }
   ],
   "source": [
    "X    = df_red.values[:,0:NDIM]\n",
    "print(len(dnn.predict_proba(X)[...,0]))\n",
    "df['DNN_SigVsBkg'] = dnn.predict_proba(X)[...,0].numpy().tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "e412b290-c218-47e4-8ec9-2905c2ea8cf9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['isLL', 'isMIX', 'isTT', 'isBKG', 'isVBS', 'notLL', 'notTT', 'DNN_SigVsBkg']\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CleanJet_eta_1</th>\n",
       "      <th>CleanJet_eta_2</th>\n",
       "      <th>CleanJet_phi_1</th>\n",
       "      <th>CleanJet_phi_2</th>\n",
       "      <th>CleanJet_pt_1</th>\n",
       "      <th>CleanJet_pt_2</th>\n",
       "      <th>Lepton_eta_1</th>\n",
       "      <th>Lepton_eta_2</th>\n",
       "      <th>Lepton_phi_1</th>\n",
       "      <th>Lepton_phi_2</th>\n",
       "      <th>Lepton_pt_1</th>\n",
       "      <th>Lepton_pt_2</th>\n",
       "      <th>MET_significance</th>\n",
       "      <th>Rpt</th>\n",
       "      <th>Zepp_l1</th>\n",
       "      <th>Zepp_l2</th>\n",
       "      <th>Zepp_ll</th>\n",
       "      <th>detajj</th>\n",
       "      <th>detall</th>\n",
       "      <th>dphijj</th>\n",
       "      <th>dphilep1jet1</th>\n",
       "      <th>dphilep1jet2</th>\n",
       "      <th>dphilep1jj</th>\n",
       "      <th>dphilep2jet1</th>\n",
       "      <th>dphilep2jet2</th>\n",
       "      <th>dphilep2jj</th>\n",
       "      <th>dphill</th>\n",
       "      <th>dphilljet</th>\n",
       "      <th>dphilljetjet</th>\n",
       "      <th>dphillmet</th>\n",
       "      <th>dphilmet1</th>\n",
       "      <th>dphilmet2</th>\n",
       "      <th>dr_l1j1</th>\n",
       "      <th>dr_l1j2</th>\n",
       "      <th>dr_l2j1</th>\n",
       "      <th>dr_l2j2</th>\n",
       "      <th>drll</th>\n",
       "      <th>ht</th>\n",
       "      <th>m2ljj30</th>\n",
       "      <th>mT2</th>\n",
       "      <th>mTi</th>\n",
       "      <th>m_l1j1</th>\n",
       "      <th>m_l1j2</th>\n",
       "      <th>m_l2j1</th>\n",
       "      <th>m_l2j2</th>\n",
       "      <th>mcoll</th>\n",
       "      <th>mcollWW</th>\n",
       "      <th>mjj</th>\n",
       "      <th>mll</th>\n",
       "      <th>mtw1</th>\n",
       "      <th>mtw2</th>\n",
       "      <th>phi_miss</th>\n",
       "      <th>proxyW_l1</th>\n",
       "      <th>proxyW_l2</th>\n",
       "      <th>pt_miss</th>\n",
       "      <th>ptll</th>\n",
       "      <th>recoil</th>\n",
       "      <th>yll</th>\n",
       "      <th>zeta</th>\n",
       "      <th>DNN_SigVsBkg</th>\n",
       "      <th>isLL</th>\n",
       "      <th>isMIX</th>\n",
       "      <th>isTT</th>\n",
       "      <th>isBKG</th>\n",
       "      <th>isVBS</th>\n",
       "      <th>notLL</th>\n",
       "      <th>notTT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.369629</td>\n",
       "      <td>2.523438</td>\n",
       "      <td>2.955566</td>\n",
       "      <td>-2.326660</td>\n",
       "      <td>53.054813</td>\n",
       "      <td>29.876976</td>\n",
       "      <td>0.719849</td>\n",
       "      <td>-0.348572</td>\n",
       "      <td>0.324402</td>\n",
       "      <td>-0.898926</td>\n",
       "      <td>79.900345</td>\n",
       "      <td>68.976547</td>\n",
       "      <td>1.595703</td>\n",
       "      <td>3.476872</td>\n",
       "      <td>0.142944</td>\n",
       "      <td>-0.925476</td>\n",
       "      <td>0.391266</td>\n",
       "      <td>3.893066</td>\n",
       "      <td>1.068420</td>\n",
       "      <td>1.000959</td>\n",
       "      <td>2.631165</td>\n",
       "      <td>2.651062</td>\n",
       "      <td>2.979963</td>\n",
       "      <td>2.428693</td>\n",
       "      <td>1.427734</td>\n",
       "      <td>2.079895</td>\n",
       "      <td>1.223328</td>\n",
       "      <td>3.091776</td>\n",
       "      <td>2.742978</td>\n",
       "      <td>2.321738</td>\n",
       "      <td>1.761494</td>\n",
       "      <td>2.984821</td>\n",
       "      <td>3.359902</td>\n",
       "      <td>3.206410</td>\n",
       "      <td>3.987439</td>\n",
       "      <td>3.207314</td>\n",
       "      <td>1.624208</td>\n",
       "      <td>268.945618</td>\n",
       "      <td>271.210632</td>\n",
       "      <td>67.454170</td>\n",
       "      <td>220.818268</td>\n",
       "      <td>205.699677</td>\n",
       "      <td>138.472168</td>\n",
       "      <td>130.887726</td>\n",
       "      <td>189.783020</td>\n",
       "      <td>21.942909</td>\n",
       "      <td>178.228348</td>\n",
       "      <td>276.330872</td>\n",
       "      <td>119.088409</td>\n",
       "      <td>112.865700</td>\n",
       "      <td>135.558762</td>\n",
       "      <td>2.085896</td>\n",
       "      <td>125.040550</td>\n",
       "      <td>137.635452</td>\n",
       "      <td>67.013908</td>\n",
       "      <td>122.045883</td>\n",
       "      <td>90.691017</td>\n",
       "      <td>0.221504</td>\n",
       "      <td>1.021057</td>\n",
       "      <td>0.038441</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.438477</td>\n",
       "      <td>-4.166992</td>\n",
       "      <td>0.698364</td>\n",
       "      <td>2.384766</td>\n",
       "      <td>30.664513</td>\n",
       "      <td>29.299437</td>\n",
       "      <td>-0.608276</td>\n",
       "      <td>-1.422363</td>\n",
       "      <td>0.585815</td>\n",
       "      <td>-1.950195</td>\n",
       "      <td>97.959366</td>\n",
       "      <td>38.920242</td>\n",
       "      <td>15.085938</td>\n",
       "      <td>4.243519</td>\n",
       "      <td>0.755981</td>\n",
       "      <td>-0.058105</td>\n",
       "      <td>0.348938</td>\n",
       "      <td>5.605469</td>\n",
       "      <td>0.814087</td>\n",
       "      <td>1.686401</td>\n",
       "      <td>0.112549</td>\n",
       "      <td>1.798950</td>\n",
       "      <td>0.930194</td>\n",
       "      <td>2.648560</td>\n",
       "      <td>1.948224</td>\n",
       "      <td>2.816981</td>\n",
       "      <td>2.536011</td>\n",
       "      <td>0.436593</td>\n",
       "      <td>1.254238</td>\n",
       "      <td>2.441383</td>\n",
       "      <td>2.765428</td>\n",
       "      <td>0.229417</td>\n",
       "      <td>2.049845</td>\n",
       "      <td>3.987566</td>\n",
       "      <td>3.898624</td>\n",
       "      <td>5.130777</td>\n",
       "      <td>2.663473</td>\n",
       "      <td>271.538696</td>\n",
       "      <td>239.743408</td>\n",
       "      <td>138.635910</td>\n",
       "      <td>263.918365</td>\n",
       "      <td>133.487793</td>\n",
       "      <td>319.718811</td>\n",
       "      <td>152.140701</td>\n",
       "      <td>136.750137</td>\n",
       "      <td>27.437832</td>\n",
       "      <td>252.764832</td>\n",
       "      <td>495.311188</td>\n",
       "      <td>128.700623</td>\n",
       "      <td>198.303772</td>\n",
       "      <td>14.563496</td>\n",
       "      <td>-2.179612</td>\n",
       "      <td>207.873352</td>\n",
       "      <td>99.386734</td>\n",
       "      <td>103.994568</td>\n",
       "      <td>69.581612</td>\n",
       "      <td>67.745537</td>\n",
       "      <td>-0.847283</td>\n",
       "      <td>2.046753</td>\n",
       "      <td>0.145060</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-1.018311</td>\n",
       "      <td>-3.684570</td>\n",
       "      <td>3.027344</td>\n",
       "      <td>-2.980957</td>\n",
       "      <td>30.099604</td>\n",
       "      <td>28.075436</td>\n",
       "      <td>-1.967529</td>\n",
       "      <td>-2.218262</td>\n",
       "      <td>-0.597900</td>\n",
       "      <td>2.771973</td>\n",
       "      <td>67.096413</td>\n",
       "      <td>46.198463</td>\n",
       "      <td>10.007812</td>\n",
       "      <td>3.668086</td>\n",
       "      <td>0.383911</td>\n",
       "      <td>0.133179</td>\n",
       "      <td>0.258545</td>\n",
       "      <td>2.666260</td>\n",
       "      <td>0.250732</td>\n",
       "      <td>0.274885</td>\n",
       "      <td>2.657941</td>\n",
       "      <td>2.383057</td>\n",
       "      <td>2.525311</td>\n",
       "      <td>0.255371</td>\n",
       "      <td>0.530256</td>\n",
       "      <td>0.388001</td>\n",
       "      <td>2.913312</td>\n",
       "      <td>2.216007</td>\n",
       "      <td>2.083377</td>\n",
       "      <td>1.396015</td>\n",
       "      <td>0.954081</td>\n",
       "      <td>2.415792</td>\n",
       "      <td>3.747454</td>\n",
       "      <td>2.937208</td>\n",
       "      <td>1.226824</td>\n",
       "      <td>5.936856</td>\n",
       "      <td>2.924082</td>\n",
       "      <td>203.235275</td>\n",
       "      <td>156.898926</td>\n",
       "      <td>45.883656</td>\n",
       "      <td>258.683807</td>\n",
       "      <td>98.626106</td>\n",
       "      <td>116.712708</td>\n",
       "      <td>49.805935</td>\n",
       "      <td>61.152802</td>\n",
       "      <td>24.233461</td>\n",
       "      <td>264.130157</td>\n",
       "      <td>104.640396</td>\n",
       "      <td>111.508141</td>\n",
       "      <td>58.188202</td>\n",
       "      <td>98.309204</td>\n",
       "      <td>0.356181</td>\n",
       "      <td>156.961578</td>\n",
       "      <td>172.760864</td>\n",
       "      <td>59.840794</td>\n",
       "      <td>24.445000</td>\n",
       "      <td>68.463287</td>\n",
       "      <td>-2.069887</td>\n",
       "      <td>0.949219</td>\n",
       "      <td>0.005695</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-2.482910</td>\n",
       "      <td>2.140137</td>\n",
       "      <td>-2.079590</td>\n",
       "      <td>-0.515137</td>\n",
       "      <td>91.604919</td>\n",
       "      <td>32.597763</td>\n",
       "      <td>-1.102295</td>\n",
       "      <td>0.483093</td>\n",
       "      <td>1.666992</td>\n",
       "      <td>-0.923462</td>\n",
       "      <td>128.451630</td>\n",
       "      <td>81.739906</td>\n",
       "      <td>8.101562</td>\n",
       "      <td>3.516148</td>\n",
       "      <td>-0.930908</td>\n",
       "      <td>0.654480</td>\n",
       "      <td>0.138214</td>\n",
       "      <td>4.623047</td>\n",
       "      <td>1.585388</td>\n",
       "      <td>1.564453</td>\n",
       "      <td>2.536603</td>\n",
       "      <td>2.182129</td>\n",
       "      <td>2.877764</td>\n",
       "      <td>1.156128</td>\n",
       "      <td>0.408325</td>\n",
       "      <td>0.814967</td>\n",
       "      <td>2.590454</td>\n",
       "      <td>3.117462</td>\n",
       "      <td>2.776301</td>\n",
       "      <td>0.852475</td>\n",
       "      <td>0.223355</td>\n",
       "      <td>2.813809</td>\n",
       "      <td>3.992866</td>\n",
       "      <td>3.908331</td>\n",
       "      <td>3.183364</td>\n",
       "      <td>1.706611</td>\n",
       "      <td>3.037089</td>\n",
       "      <td>389.537445</td>\n",
       "      <td>840.582458</td>\n",
       "      <td>59.463200</td>\n",
       "      <td>315.879150</td>\n",
       "      <td>263.529846</td>\n",
       "      <td>335.037811</td>\n",
       "      <td>374.165955</td>\n",
       "      <td>98.272255</td>\n",
       "      <td>191.155640</td>\n",
       "      <td>289.987244</td>\n",
       "      <td>551.808350</td>\n",
       "      <td>267.004730</td>\n",
       "      <td>18.758913</td>\n",
       "      <td>132.475067</td>\n",
       "      <td>1.890347</td>\n",
       "      <td>99.327400</td>\n",
       "      <td>136.463150</td>\n",
       "      <td>55.143219</td>\n",
       "      <td>72.741791</td>\n",
       "      <td>116.669960</td>\n",
       "      <td>-0.457326</td>\n",
       "      <td>1.380615</td>\n",
       "      <td>0.405764</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-3.537109</td>\n",
       "      <td>1.393066</td>\n",
       "      <td>-0.335815</td>\n",
       "      <td>1.177734</td>\n",
       "      <td>59.228630</td>\n",
       "      <td>48.039818</td>\n",
       "      <td>-0.406128</td>\n",
       "      <td>-1.342041</td>\n",
       "      <td>-3.069336</td>\n",
       "      <td>2.493652</td>\n",
       "      <td>81.435837</td>\n",
       "      <td>57.172844</td>\n",
       "      <td>6.886719</td>\n",
       "      <td>1.636335</td>\n",
       "      <td>0.665894</td>\n",
       "      <td>-0.270020</td>\n",
       "      <td>0.197937</td>\n",
       "      <td>4.930176</td>\n",
       "      <td>0.935913</td>\n",
       "      <td>1.513550</td>\n",
       "      <td>2.733521</td>\n",
       "      <td>2.036115</td>\n",
       "      <td>2.891073</td>\n",
       "      <td>2.829468</td>\n",
       "      <td>1.315918</td>\n",
       "      <td>2.170876</td>\n",
       "      <td>0.720197</td>\n",
       "      <td>3.027807</td>\n",
       "      <td>2.596787</td>\n",
       "      <td>2.431972</td>\n",
       "      <td>2.137686</td>\n",
       "      <td>2.857883</td>\n",
       "      <td>4.156342</td>\n",
       "      <td>4.612451</td>\n",
       "      <td>3.581091</td>\n",
       "      <td>3.035202</td>\n",
       "      <td>1.180939</td>\n",
       "      <td>333.340424</td>\n",
       "      <td>791.871887</td>\n",
       "      <td>91.928360</td>\n",
       "      <td>244.596680</td>\n",
       "      <td>345.862061</td>\n",
       "      <td>167.208679</td>\n",
       "      <td>193.272903</td>\n",
       "      <td>203.217087</td>\n",
       "      <td>36.447132</td>\n",
       "      <td>165.542557</td>\n",
       "      <td>628.104187</td>\n",
       "      <td>81.836113</td>\n",
       "      <td>147.970154</td>\n",
       "      <td>140.008194</td>\n",
       "      <td>-0.931650</td>\n",
       "      <td>151.941467</td>\n",
       "      <td>173.332565</td>\n",
       "      <td>87.463287</td>\n",
       "      <td>130.000046</td>\n",
       "      <td>85.432556</td>\n",
       "      <td>-0.797519</td>\n",
       "      <td>1.799194</td>\n",
       "      <td>0.316078</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48433</th>\n",
       "      <td>-3.129395</td>\n",
       "      <td>2.741699</td>\n",
       "      <td>-2.973145</td>\n",
       "      <td>0.735474</td>\n",
       "      <td>89.830719</td>\n",
       "      <td>58.886036</td>\n",
       "      <td>1.100830</td>\n",
       "      <td>1.797363</td>\n",
       "      <td>0.155731</td>\n",
       "      <td>-2.492676</td>\n",
       "      <td>146.724854</td>\n",
       "      <td>47.941044</td>\n",
       "      <td>5.027344</td>\n",
       "      <td>1.329762</td>\n",
       "      <td>1.294678</td>\n",
       "      <td>1.991211</td>\n",
       "      <td>1.642944</td>\n",
       "      <td>5.871094</td>\n",
       "      <td>0.696533</td>\n",
       "      <td>2.574567</td>\n",
       "      <td>3.128876</td>\n",
       "      <td>0.579742</td>\n",
       "      <td>2.487191</td>\n",
       "      <td>0.480469</td>\n",
       "      <td>3.055036</td>\n",
       "      <td>1.147588</td>\n",
       "      <td>2.648407</td>\n",
       "      <td>2.914996</td>\n",
       "      <td>2.701071</td>\n",
       "      <td>2.446765</td>\n",
       "      <td>2.660645</td>\n",
       "      <td>0.012238</td>\n",
       "      <td>5.261622</td>\n",
       "      <td>1.740274</td>\n",
       "      <td>4.950130</td>\n",
       "      <td>3.363439</td>\n",
       "      <td>2.738470</td>\n",
       "      <td>408.894348</td>\n",
       "      <td>1865.708130</td>\n",
       "      <td>59.543850</td>\n",
       "      <td>303.881073</td>\n",
       "      <td>965.849487</td>\n",
       "      <td>179.082977</td>\n",
       "      <td>766.007629</td>\n",
       "      <td>119.235420</td>\n",
       "      <td>207.144745</td>\n",
       "      <td>270.647614</td>\n",
       "      <td>1373.288208</td>\n",
       "      <td>173.242432</td>\n",
       "      <td>190.441330</td>\n",
       "      <td>0.685746</td>\n",
       "      <td>-2.504914</td>\n",
       "      <td>221.677887</td>\n",
       "      <td>114.846367</td>\n",
       "      <td>65.511696</td>\n",
       "      <td>106.933487</td>\n",
       "      <td>70.455421</td>\n",
       "      <td>1.277518</td>\n",
       "      <td>0.944336</td>\n",
       "      <td>0.866934</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48434</th>\n",
       "      <td>1.018066</td>\n",
       "      <td>-3.458984</td>\n",
       "      <td>-1.925781</td>\n",
       "      <td>1.144287</td>\n",
       "      <td>321.460907</td>\n",
       "      <td>133.999588</td>\n",
       "      <td>0.184326</td>\n",
       "      <td>-1.049561</td>\n",
       "      <td>2.806641</td>\n",
       "      <td>0.098877</td>\n",
       "      <td>288.248871</td>\n",
       "      <td>127.276970</td>\n",
       "      <td>85.250000</td>\n",
       "      <td>0.851698</td>\n",
       "      <td>1.404785</td>\n",
       "      <td>0.170898</td>\n",
       "      <td>0.787842</td>\n",
       "      <td>4.477051</td>\n",
       "      <td>1.233887</td>\n",
       "      <td>3.070068</td>\n",
       "      <td>1.550763</td>\n",
       "      <td>1.662354</td>\n",
       "      <td>1.601709</td>\n",
       "      <td>2.024658</td>\n",
       "      <td>1.045410</td>\n",
       "      <td>1.973713</td>\n",
       "      <td>2.707764</td>\n",
       "      <td>1.851075</td>\n",
       "      <td>1.902020</td>\n",
       "      <td>2.104769</td>\n",
       "      <td>2.405080</td>\n",
       "      <td>0.302684</td>\n",
       "      <td>4.805303</td>\n",
       "      <td>4.004639</td>\n",
       "      <td>2.893842</td>\n",
       "      <td>2.626444</td>\n",
       "      <td>2.975645</td>\n",
       "      <td>1098.038452</td>\n",
       "      <td>2521.232910</td>\n",
       "      <td>43.703491</td>\n",
       "      <td>686.186340</td>\n",
       "      <td>500.830536</td>\n",
       "      <td>1218.500610</td>\n",
       "      <td>604.488342</td>\n",
       "      <td>417.939453</td>\n",
       "      <td>478.330597</td>\n",
       "      <td>519.611877</td>\n",
       "      <td>1969.721802</td>\n",
       "      <td>450.848633</td>\n",
       "      <td>477.351166</td>\n",
       "      <td>51.258739</td>\n",
       "      <td>0.401561</td>\n",
       "      <td>479.681244</td>\n",
       "      <td>193.631454</td>\n",
       "      <td>227.052155</td>\n",
       "      <td>180.856827</td>\n",
       "      <td>206.063980</td>\n",
       "      <td>-0.216643</td>\n",
       "      <td>0.833740</td>\n",
       "      <td>0.995779</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48435</th>\n",
       "      <td>-1.457275</td>\n",
       "      <td>3.064453</td>\n",
       "      <td>-3.119141</td>\n",
       "      <td>0.177948</td>\n",
       "      <td>69.824425</td>\n",
       "      <td>54.498955</td>\n",
       "      <td>0.562988</td>\n",
       "      <td>-0.216248</td>\n",
       "      <td>-2.235352</td>\n",
       "      <td>0.264465</td>\n",
       "      <td>67.126358</td>\n",
       "      <td>48.581814</td>\n",
       "      <td>4.535156</td>\n",
       "      <td>0.856981</td>\n",
       "      <td>-0.240601</td>\n",
       "      <td>-1.019836</td>\n",
       "      <td>0.630219</td>\n",
       "      <td>4.521729</td>\n",
       "      <td>0.779236</td>\n",
       "      <td>2.986097</td>\n",
       "      <td>0.883789</td>\n",
       "      <td>2.413300</td>\n",
       "      <td>1.369645</td>\n",
       "      <td>2.899579</td>\n",
       "      <td>0.086517</td>\n",
       "      <td>2.413723</td>\n",
       "      <td>2.499817</td>\n",
       "      <td>1.684394</td>\n",
       "      <td>2.170250</td>\n",
       "      <td>1.867713</td>\n",
       "      <td>2.668318</td>\n",
       "      <td>0.168501</td>\n",
       "      <td>2.205119</td>\n",
       "      <td>3.475822</td>\n",
       "      <td>3.604017</td>\n",
       "      <td>3.281841</td>\n",
       "      <td>2.618452</td>\n",
       "      <td>280.925171</td>\n",
       "      <td>732.914551</td>\n",
       "      <td>50.398869</td>\n",
       "      <td>158.898468</td>\n",
       "      <td>173.699463</td>\n",
       "      <td>224.801559</td>\n",
       "      <td>139.414627</td>\n",
       "      <td>255.682388</td>\n",
       "      <td>107.627251</td>\n",
       "      <td>198.407562</td>\n",
       "      <td>598.560303</td>\n",
       "      <td>117.597641</td>\n",
       "      <td>101.866058</td>\n",
       "      <td>7.501616</td>\n",
       "      <td>0.432967</td>\n",
       "      <td>106.160141</td>\n",
       "      <td>12.228662</td>\n",
       "      <td>40.893623</td>\n",
       "      <td>40.516712</td>\n",
       "      <td>48.418770</td>\n",
       "      <td>0.232906</td>\n",
       "      <td>1.241028</td>\n",
       "      <td>0.680826</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48436</th>\n",
       "      <td>2.373535</td>\n",
       "      <td>-3.776855</td>\n",
       "      <td>2.814453</td>\n",
       "      <td>-2.516602</td>\n",
       "      <td>209.293030</td>\n",
       "      <td>71.008499</td>\n",
       "      <td>-1.235352</td>\n",
       "      <td>-2.081055</td>\n",
       "      <td>0.601318</td>\n",
       "      <td>-2.180176</td>\n",
       "      <td>169.232925</td>\n",
       "      <td>115.416367</td>\n",
       "      <td>36.187500</td>\n",
       "      <td>1.314278</td>\n",
       "      <td>-0.533691</td>\n",
       "      <td>-1.379395</td>\n",
       "      <td>0.956543</td>\n",
       "      <td>6.150391</td>\n",
       "      <td>0.845703</td>\n",
       "      <td>0.952131</td>\n",
       "      <td>2.213135</td>\n",
       "      <td>3.117920</td>\n",
       "      <td>2.440106</td>\n",
       "      <td>1.288556</td>\n",
       "      <td>0.336426</td>\n",
       "      <td>1.061585</td>\n",
       "      <td>2.781494</td>\n",
       "      <td>2.799508</td>\n",
       "      <td>3.026480</td>\n",
       "      <td>0.249041</td>\n",
       "      <td>0.835414</td>\n",
       "      <td>1.946080</td>\n",
       "      <td>4.233442</td>\n",
       "      <td>4.022520</td>\n",
       "      <td>6.692510</td>\n",
       "      <td>1.728850</td>\n",
       "      <td>2.907219</td>\n",
       "      <td>705.139282</td>\n",
       "      <td>3266.109863</td>\n",
       "      <td>111.594643</td>\n",
       "      <td>539.113647</td>\n",
       "      <td>1162.617920</td>\n",
       "      <td>421.610809</td>\n",
       "      <td>1437.083130</td>\n",
       "      <td>175.788910</td>\n",
       "      <td>279.400787</td>\n",
       "      <td>326.273926</td>\n",
       "      <td>2636.750977</td>\n",
       "      <td>300.742310</td>\n",
       "      <td>124.967491</td>\n",
       "      <td>210.288467</td>\n",
       "      <td>-0.234096</td>\n",
       "      <td>238.050140</td>\n",
       "      <td>378.853180</td>\n",
       "      <td>140.188492</td>\n",
       "      <td>73.496506</td>\n",
       "      <td>212.192230</td>\n",
       "      <td>-1.582561</td>\n",
       "      <td>1.695801</td>\n",
       "      <td>0.949972</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48437</th>\n",
       "      <td>2.749023</td>\n",
       "      <td>-2.595703</td>\n",
       "      <td>1.461426</td>\n",
       "      <td>-0.806152</td>\n",
       "      <td>134.450607</td>\n",
       "      <td>125.994003</td>\n",
       "      <td>0.742432</td>\n",
       "      <td>2.025879</td>\n",
       "      <td>2.222168</td>\n",
       "      <td>-1.684814</td>\n",
       "      <td>145.817352</td>\n",
       "      <td>125.114052</td>\n",
       "      <td>0.041870</td>\n",
       "      <td>1.076968</td>\n",
       "      <td>0.665771</td>\n",
       "      <td>1.949219</td>\n",
       "      <td>1.307495</td>\n",
       "      <td>5.344727</td>\n",
       "      <td>1.283447</td>\n",
       "      <td>2.267578</td>\n",
       "      <td>0.760742</td>\n",
       "      <td>3.028320</td>\n",
       "      <td>1.825134</td>\n",
       "      <td>3.136945</td>\n",
       "      <td>0.878662</td>\n",
       "      <td>2.081849</td>\n",
       "      <td>2.376203</td>\n",
       "      <td>1.761243</td>\n",
       "      <td>2.825635</td>\n",
       "      <td>1.480161</td>\n",
       "      <td>2.480663</td>\n",
       "      <td>0.104460</td>\n",
       "      <td>2.145959</td>\n",
       "      <td>4.507091</td>\n",
       "      <td>3.228276</td>\n",
       "      <td>4.704367</td>\n",
       "      <td>2.700662</td>\n",
       "      <td>577.253052</td>\n",
       "      <td>2450.030762</td>\n",
       "      <td>145.873016</td>\n",
       "      <td>397.334442</td>\n",
       "      <td>346.891998</td>\n",
       "      <td>744.865601</td>\n",
       "      <td>277.156708</td>\n",
       "      <td>1258.061523</td>\n",
       "      <td>315.771698</td>\n",
       "      <td>377.157806</td>\n",
       "      <td>1889.846924</td>\n",
       "      <td>311.784576</td>\n",
       "      <td>154.729660</td>\n",
       "      <td>7.910470</td>\n",
       "      <td>-1.580355</td>\n",
       "      <td>166.736801</td>\n",
       "      <td>181.283951</td>\n",
       "      <td>45.877026</td>\n",
       "      <td>102.978477</td>\n",
       "      <td>116.466637</td>\n",
       "      <td>1.340872</td>\n",
       "      <td>0.723145</td>\n",
       "      <td>0.985080</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>48438 rows × 67 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       CleanJet_eta_1  CleanJet_eta_2  CleanJet_phi_1  CleanJet_phi_2  \\\n",
       "0           -1.369629        2.523438        2.955566       -2.326660   \n",
       "1            1.438477       -4.166992        0.698364        2.384766   \n",
       "2           -1.018311       -3.684570        3.027344       -2.980957   \n",
       "3           -2.482910        2.140137       -2.079590       -0.515137   \n",
       "4           -3.537109        1.393066       -0.335815        1.177734   \n",
       "...               ...             ...             ...             ...   \n",
       "48433       -3.129395        2.741699       -2.973145        0.735474   \n",
       "48434        1.018066       -3.458984       -1.925781        1.144287   \n",
       "48435       -1.457275        3.064453       -3.119141        0.177948   \n",
       "48436        2.373535       -3.776855        2.814453       -2.516602   \n",
       "48437        2.749023       -2.595703        1.461426       -0.806152   \n",
       "\n",
       "       CleanJet_pt_1  CleanJet_pt_2  Lepton_eta_1  Lepton_eta_2  Lepton_phi_1  \\\n",
       "0          53.054813      29.876976      0.719849     -0.348572      0.324402   \n",
       "1          30.664513      29.299437     -0.608276     -1.422363      0.585815   \n",
       "2          30.099604      28.075436     -1.967529     -2.218262     -0.597900   \n",
       "3          91.604919      32.597763     -1.102295      0.483093      1.666992   \n",
       "4          59.228630      48.039818     -0.406128     -1.342041     -3.069336   \n",
       "...              ...            ...           ...           ...           ...   \n",
       "48433      89.830719      58.886036      1.100830      1.797363      0.155731   \n",
       "48434     321.460907     133.999588      0.184326     -1.049561      2.806641   \n",
       "48435      69.824425      54.498955      0.562988     -0.216248     -2.235352   \n",
       "48436     209.293030      71.008499     -1.235352     -2.081055      0.601318   \n",
       "48437     134.450607     125.994003      0.742432      2.025879      2.222168   \n",
       "\n",
       "       Lepton_phi_2  Lepton_pt_1  Lepton_pt_2  MET_significance       Rpt  \\\n",
       "0         -0.898926    79.900345    68.976547          1.595703  3.476872   \n",
       "1         -1.950195    97.959366    38.920242         15.085938  4.243519   \n",
       "2          2.771973    67.096413    46.198463         10.007812  3.668086   \n",
       "3         -0.923462   128.451630    81.739906          8.101562  3.516148   \n",
       "4          2.493652    81.435837    57.172844          6.886719  1.636335   \n",
       "...             ...          ...          ...               ...       ...   \n",
       "48433     -2.492676   146.724854    47.941044          5.027344  1.329762   \n",
       "48434      0.098877   288.248871   127.276970         85.250000  0.851698   \n",
       "48435      0.264465    67.126358    48.581814          4.535156  0.856981   \n",
       "48436     -2.180176   169.232925   115.416367         36.187500  1.314278   \n",
       "48437     -1.684814   145.817352   125.114052          0.041870  1.076968   \n",
       "\n",
       "        Zepp_l1   Zepp_l2   Zepp_ll    detajj    detall    dphijj  \\\n",
       "0      0.142944 -0.925476  0.391266  3.893066  1.068420  1.000959   \n",
       "1      0.755981 -0.058105  0.348938  5.605469  0.814087  1.686401   \n",
       "2      0.383911  0.133179  0.258545  2.666260  0.250732  0.274885   \n",
       "3     -0.930908  0.654480  0.138214  4.623047  1.585388  1.564453   \n",
       "4      0.665894 -0.270020  0.197937  4.930176  0.935913  1.513550   \n",
       "...         ...       ...       ...       ...       ...       ...   \n",
       "48433  1.294678  1.991211  1.642944  5.871094  0.696533  2.574567   \n",
       "48434  1.404785  0.170898  0.787842  4.477051  1.233887  3.070068   \n",
       "48435 -0.240601 -1.019836  0.630219  4.521729  0.779236  2.986097   \n",
       "48436 -0.533691 -1.379395  0.956543  6.150391  0.845703  0.952131   \n",
       "48437  0.665771  1.949219  1.307495  5.344727  1.283447  2.267578   \n",
       "\n",
       "       dphilep1jet1  dphilep1jet2  dphilep1jj  dphilep2jet1  dphilep2jet2  \\\n",
       "0          2.631165      2.651062    2.979963      2.428693      1.427734   \n",
       "1          0.112549      1.798950    0.930194      2.648560      1.948224   \n",
       "2          2.657941      2.383057    2.525311      0.255371      0.530256   \n",
       "3          2.536603      2.182129    2.877764      1.156128      0.408325   \n",
       "4          2.733521      2.036115    2.891073      2.829468      1.315918   \n",
       "...             ...           ...         ...           ...           ...   \n",
       "48433      3.128876      0.579742    2.487191      0.480469      3.055036   \n",
       "48434      1.550763      1.662354    1.601709      2.024658      1.045410   \n",
       "48435      0.883789      2.413300    1.369645      2.899579      0.086517   \n",
       "48436      2.213135      3.117920    2.440106      1.288556      0.336426   \n",
       "48437      0.760742      3.028320    1.825134      3.136945      0.878662   \n",
       "\n",
       "       dphilep2jj    dphill  dphilljet  dphilljetjet  dphillmet  dphilmet1  \\\n",
       "0        2.079895  1.223328   3.091776      2.742978   2.321738   1.761494   \n",
       "1        2.816981  2.536011   0.436593      1.254238   2.441383   2.765428   \n",
       "2        0.388001  2.913312   2.216007      2.083377   1.396015   0.954081   \n",
       "3        0.814967  2.590454   3.117462      2.776301   0.852475   0.223355   \n",
       "4        2.170876  0.720197   3.027807      2.596787   2.431972   2.137686   \n",
       "...           ...       ...        ...           ...        ...        ...   \n",
       "48433    1.147588  2.648407   2.914996      2.701071   2.446765   2.660645   \n",
       "48434    1.973713  2.707764   1.851075      1.902020   2.104769   2.405080   \n",
       "48435    2.413723  2.499817   1.684394      2.170250   1.867713   2.668318   \n",
       "48436    1.061585  2.781494   2.799508      3.026480   0.249041   0.835414   \n",
       "48437    2.081849  2.376203   1.761243      2.825635   1.480161   2.480663   \n",
       "\n",
       "       dphilmet2   dr_l1j1   dr_l1j2   dr_l2j1   dr_l2j2      drll  \\\n",
       "0       2.984821  3.359902  3.206410  3.987439  3.207314  1.624208   \n",
       "1       0.229417  2.049845  3.987566  3.898624  5.130777  2.663473   \n",
       "2       2.415792  3.747454  2.937208  1.226824  5.936856  2.924082   \n",
       "3       2.813809  3.992866  3.908331  3.183364  1.706611  3.037089   \n",
       "4       2.857883  4.156342  4.612451  3.581091  3.035202  1.180939   \n",
       "...          ...       ...       ...       ...       ...       ...   \n",
       "48433   0.012238  5.261622  1.740274  4.950130  3.363439  2.738470   \n",
       "48434   0.302684  4.805303  4.004639  2.893842  2.626444  2.975645   \n",
       "48435   0.168501  2.205119  3.475822  3.604017  3.281841  2.618452   \n",
       "48436   1.946080  4.233442  4.022520  6.692510  1.728850  2.907219   \n",
       "48437   0.104460  2.145959  4.507091  3.228276  4.704367  2.700662   \n",
       "\n",
       "                ht      m2ljj30         mT2         mTi       m_l1j1  \\\n",
       "0       268.945618   271.210632   67.454170  220.818268   205.699677   \n",
       "1       271.538696   239.743408  138.635910  263.918365   133.487793   \n",
       "2       203.235275   156.898926   45.883656  258.683807    98.626106   \n",
       "3       389.537445   840.582458   59.463200  315.879150   263.529846   \n",
       "4       333.340424   791.871887   91.928360  244.596680   345.862061   \n",
       "...            ...          ...         ...         ...          ...   \n",
       "48433   408.894348  1865.708130   59.543850  303.881073   965.849487   \n",
       "48434  1098.038452  2521.232910   43.703491  686.186340   500.830536   \n",
       "48435   280.925171   732.914551   50.398869  158.898468   173.699463   \n",
       "48436   705.139282  3266.109863  111.594643  539.113647  1162.617920   \n",
       "48437   577.253052  2450.030762  145.873016  397.334442   346.891998   \n",
       "\n",
       "            m_l1j2       m_l2j1       m_l2j2       mcoll     mcollWW  \\\n",
       "0       138.472168   130.887726   189.783020   21.942909  178.228348   \n",
       "1       319.718811   152.140701   136.750137   27.437832  252.764832   \n",
       "2       116.712708    49.805935    61.152802   24.233461  264.130157   \n",
       "3       335.037811   374.165955    98.272255  191.155640  289.987244   \n",
       "4       167.208679   193.272903   203.217087   36.447132  165.542557   \n",
       "...            ...          ...          ...         ...         ...   \n",
       "48433   179.082977   766.007629   119.235420  207.144745  270.647614   \n",
       "48434  1218.500610   604.488342   417.939453  478.330597  519.611877   \n",
       "48435   224.801559   139.414627   255.682388  107.627251  198.407562   \n",
       "48436   421.610809  1437.083130   175.788910  279.400787  326.273926   \n",
       "48437   744.865601   277.156708  1258.061523  315.771698  377.157806   \n",
       "\n",
       "               mjj         mll        mtw1        mtw2  phi_miss   proxyW_l1  \\\n",
       "0       276.330872  119.088409  112.865700  135.558762  2.085896  125.040550   \n",
       "1       495.311188  128.700623  198.303772   14.563496 -2.179612  207.873352   \n",
       "2       104.640396  111.508141   58.188202   98.309204  0.356181  156.961578   \n",
       "3       551.808350  267.004730   18.758913  132.475067  1.890347   99.327400   \n",
       "4       628.104187   81.836113  147.970154  140.008194 -0.931650  151.941467   \n",
       "...            ...         ...         ...         ...       ...         ...   \n",
       "48433  1373.288208  173.242432  190.441330    0.685746 -2.504914  221.677887   \n",
       "48434  1969.721802  450.848633  477.351166   51.258739  0.401561  479.681244   \n",
       "48435   598.560303  117.597641  101.866058    7.501616  0.432967  106.160141   \n",
       "48436  2636.750977  300.742310  124.967491  210.288467 -0.234096  238.050140   \n",
       "48437  1889.846924  311.784576  154.729660    7.910470 -1.580355  166.736801   \n",
       "\n",
       "        proxyW_l2     pt_miss        ptll      recoil       yll      zeta  \\\n",
       "0      137.635452   67.013908  122.045883   90.691017  0.221504  1.021057   \n",
       "1       99.386734  103.994568   69.581612   67.745537 -0.847283  2.046753   \n",
       "2      172.760864   59.840794   24.445000   68.463287 -2.069887  0.949219   \n",
       "3      136.463150   55.143219   72.741791  116.669960 -0.457326  1.380615   \n",
       "4      173.332565   87.463287  130.000046   85.432556 -0.797519  1.799194   \n",
       "...           ...         ...         ...         ...       ...       ...   \n",
       "48433  114.846367   65.511696  106.933487   70.455421  1.277518  0.944336   \n",
       "48434  193.631454  227.052155  180.856827  206.063980 -0.216643  0.833740   \n",
       "48435   12.228662   40.893623   40.516712   48.418770  0.232906  1.241028   \n",
       "48436  378.853180  140.188492   73.496506  212.192230 -1.582561  1.695801   \n",
       "48437  181.283951   45.877026  102.978477  116.466637  1.340872  0.723145   \n",
       "\n",
       "       DNN_SigVsBkg  isLL  isMIX  isTT  isBKG  isVBS  notLL  notTT  \n",
       "0          0.038441   0.0    0.0   0.0    1.0    0.0    0.0    0.0  \n",
       "1          0.145060   0.0    0.0   0.0    1.0    0.0    0.0    0.0  \n",
       "2          0.005695   0.0    0.0   0.0    1.0    0.0    0.0    0.0  \n",
       "3          0.405764   0.0    0.0   0.0    1.0    0.0    0.0    0.0  \n",
       "4          0.316078   0.0    0.0   0.0    1.0    0.0    0.0    0.0  \n",
       "...             ...   ...    ...   ...    ...    ...    ...    ...  \n",
       "48433      0.866934   0.0    0.0   1.0    0.0    1.0    1.0    0.0  \n",
       "48434      0.995779   0.0    0.0   1.0    0.0    1.0    1.0    0.0  \n",
       "48435      0.680826   0.0    0.0   1.0    0.0    1.0    1.0    0.0  \n",
       "48436      0.949972   0.0    0.0   1.0    0.0    1.0    1.0    0.0  \n",
       "48437      0.985080   0.0    0.0   1.0    0.0    1.0    1.0    0.0  \n",
       "\n",
       "[48438 rows x 67 columns]"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df.columns.values[NDIM:].tolist())\n",
    "df = df[[col for col in df if col not in df.columns.values[NDIM:].tolist()] + ['DNN_SigVsBkg'] + df.columns.values[NDIM:-1].tolist()]\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "af3362d5-95ce-44db-bc16-b545f6e919c5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "48438\n",
      "2560\n",
      "3732\n",
      "4356\n",
      "4212\n",
      "5382\n",
      "2686\n",
      "2696\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CleanJet_eta_1</th>\n",
       "      <th>CleanJet_eta_2</th>\n",
       "      <th>CleanJet_phi_1</th>\n",
       "      <th>CleanJet_phi_2</th>\n",
       "      <th>CleanJet_pt_1</th>\n",
       "      <th>CleanJet_pt_2</th>\n",
       "      <th>Lepton_eta_1</th>\n",
       "      <th>Lepton_eta_2</th>\n",
       "      <th>Lepton_phi_1</th>\n",
       "      <th>Lepton_phi_2</th>\n",
       "      <th>Lepton_pt_1</th>\n",
       "      <th>Lepton_pt_2</th>\n",
       "      <th>MET_significance</th>\n",
       "      <th>Rpt</th>\n",
       "      <th>Zepp_l1</th>\n",
       "      <th>Zepp_l2</th>\n",
       "      <th>Zepp_ll</th>\n",
       "      <th>detajj</th>\n",
       "      <th>detall</th>\n",
       "      <th>dphijj</th>\n",
       "      <th>dphilep1jet1</th>\n",
       "      <th>dphilep1jet2</th>\n",
       "      <th>dphilep1jj</th>\n",
       "      <th>dphilep2jet1</th>\n",
       "      <th>dphilep2jet2</th>\n",
       "      <th>dphilep2jj</th>\n",
       "      <th>dphill</th>\n",
       "      <th>dphilljet</th>\n",
       "      <th>dphilljetjet</th>\n",
       "      <th>dphillmet</th>\n",
       "      <th>dphilmet1</th>\n",
       "      <th>dphilmet2</th>\n",
       "      <th>dr_l1j1</th>\n",
       "      <th>dr_l1j2</th>\n",
       "      <th>dr_l2j1</th>\n",
       "      <th>dr_l2j2</th>\n",
       "      <th>drll</th>\n",
       "      <th>ht</th>\n",
       "      <th>m2ljj30</th>\n",
       "      <th>mT2</th>\n",
       "      <th>mTi</th>\n",
       "      <th>m_l1j1</th>\n",
       "      <th>m_l1j2</th>\n",
       "      <th>m_l2j1</th>\n",
       "      <th>m_l2j2</th>\n",
       "      <th>mcoll</th>\n",
       "      <th>mcollWW</th>\n",
       "      <th>mjj</th>\n",
       "      <th>mll</th>\n",
       "      <th>mtw1</th>\n",
       "      <th>mtw2</th>\n",
       "      <th>phi_miss</th>\n",
       "      <th>proxyW_l1</th>\n",
       "      <th>proxyW_l2</th>\n",
       "      <th>pt_miss</th>\n",
       "      <th>ptll</th>\n",
       "      <th>recoil</th>\n",
       "      <th>yll</th>\n",
       "      <th>zeta</th>\n",
       "      <th>DNN_SigVsBkg</th>\n",
       "      <th>isLL</th>\n",
       "      <th>isMIX</th>\n",
       "      <th>isTT</th>\n",
       "      <th>isBKG</th>\n",
       "      <th>isVBS</th>\n",
       "      <th>notLL</th>\n",
       "      <th>notTT</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>40518</th>\n",
       "      <td>-0.477417</td>\n",
       "      <td>2.986816</td>\n",
       "      <td>1.836426</td>\n",
       "      <td>-1.672363</td>\n",
       "      <td>124.805344</td>\n",
       "      <td>80.120354</td>\n",
       "      <td>0.226135</td>\n",
       "      <td>2.252441</td>\n",
       "      <td>0.036774</td>\n",
       "      <td>2.776367</td>\n",
       "      <td>84.481834</td>\n",
       "      <td>64.644058</td>\n",
       "      <td>13.671875</td>\n",
       "      <td>0.546155</td>\n",
       "      <td>-1.028564</td>\n",
       "      <td>0.997742</td>\n",
       "      <td>0.015411</td>\n",
       "      <td>3.464233</td>\n",
       "      <td>2.026306</td>\n",
       "      <td>2.774396</td>\n",
       "      <td>1.799652</td>\n",
       "      <td>1.709137</td>\n",
       "      <td>2.321460</td>\n",
       "      <td>0.939941</td>\n",
       "      <td>1.834455</td>\n",
       "      <td>0.418133</td>\n",
       "      <td>2.739594</td>\n",
       "      <td>1.008260</td>\n",
       "      <td>1.530068</td>\n",
       "      <td>2.142135</td>\n",
       "      <td>1.350743</td>\n",
       "      <td>2.192849</td>\n",
       "      <td>1.932287</td>\n",
       "      <td>3.246923</td>\n",
       "      <td>2.887147</td>\n",
       "      <td>4.508936</td>\n",
       "      <td>3.407534</td>\n",
       "      <td>437.157532</td>\n",
       "      <td>815.015503</td>\n",
       "      <td>3.160135e+01</td>\n",
       "      <td>348.510468</td>\n",
       "      <td>177.942291</td>\n",
       "      <td>330.811035</td>\n",
       "      <td>339.706390</td>\n",
       "      <td>127.020271</td>\n",
       "      <td>126.127602</td>\n",
       "      <td>205.137039</td>\n",
       "      <td>583.159729</td>\n",
       "      <td>228.474884</td>\n",
       "      <td>104.770233</td>\n",
       "      <td>130.405304</td>\n",
       "      <td>-1.313969</td>\n",
       "      <td>106.477066</td>\n",
       "      <td>240.668655</td>\n",
       "      <td>83.105957</td>\n",
       "      <td>35.556599</td>\n",
       "      <td>70.533859</td>\n",
       "      <td>1.136892</td>\n",
       "      <td>0.703552</td>\n",
       "      <td>0.698907</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44449</th>\n",
       "      <td>1.980225</td>\n",
       "      <td>-3.985840</td>\n",
       "      <td>2.588867</td>\n",
       "      <td>1.987793</td>\n",
       "      <td>193.126358</td>\n",
       "      <td>54.814369</td>\n",
       "      <td>-0.160583</td>\n",
       "      <td>1.335205</td>\n",
       "      <td>-1.342529</td>\n",
       "      <td>1.149414</td>\n",
       "      <td>266.018677</td>\n",
       "      <td>31.155691</td>\n",
       "      <td>116.187500</td>\n",
       "      <td>0.782913</td>\n",
       "      <td>0.842224</td>\n",
       "      <td>2.338013</td>\n",
       "      <td>1.590118</td>\n",
       "      <td>5.966064</td>\n",
       "      <td>1.495789</td>\n",
       "      <td>0.601074</td>\n",
       "      <td>2.351789</td>\n",
       "      <td>2.952863</td>\n",
       "      <td>2.481129</td>\n",
       "      <td>1.439453</td>\n",
       "      <td>0.838379</td>\n",
       "      <td>1.310113</td>\n",
       "      <td>2.491943</td>\n",
       "      <td>2.429763</td>\n",
       "      <td>2.559103</td>\n",
       "      <td>1.907331</td>\n",
       "      <td>1.985305</td>\n",
       "      <td>0.506638</td>\n",
       "      <td>4.476487</td>\n",
       "      <td>5.071847</td>\n",
       "      <td>1.577363</td>\n",
       "      <td>5.386687</td>\n",
       "      <td>2.906401</td>\n",
       "      <td>692.228027</td>\n",
       "      <td>2395.236816</td>\n",
       "      <td>2.388153e+02</td>\n",
       "      <td>416.484070</td>\n",
       "      <td>718.230225</td>\n",
       "      <td>835.173340</td>\n",
       "      <td>115.836853</td>\n",
       "      <td>589.225464</td>\n",
       "      <td>455.491150</td>\n",
       "      <td>484.218079</td>\n",
       "      <td>2027.787598</td>\n",
       "      <td>228.132904</td>\n",
       "      <td>331.348602</td>\n",
       "      <td>33.934185</td>\n",
       "      <td>0.642776</td>\n",
       "      <td>332.871216</td>\n",
       "      <td>103.010315</td>\n",
       "      <td>147.112946</td>\n",
       "      <td>241.944656</td>\n",
       "      <td>238.059662</td>\n",
       "      <td>0.036701</td>\n",
       "      <td>0.645020</td>\n",
       "      <td>0.878863</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39659</th>\n",
       "      <td>-3.074219</td>\n",
       "      <td>2.674805</td>\n",
       "      <td>1.145020</td>\n",
       "      <td>0.401611</td>\n",
       "      <td>95.718979</td>\n",
       "      <td>39.599072</td>\n",
       "      <td>-1.658447</td>\n",
       "      <td>-0.592651</td>\n",
       "      <td>-2.795410</td>\n",
       "      <td>-1.236084</td>\n",
       "      <td>66.873520</td>\n",
       "      <td>52.486717</td>\n",
       "      <td>3.103516</td>\n",
       "      <td>0.926020</td>\n",
       "      <td>-1.458740</td>\n",
       "      <td>-0.392944</td>\n",
       "      <td>0.925842</td>\n",
       "      <td>5.749023</td>\n",
       "      <td>1.065796</td>\n",
       "      <td>0.743408</td>\n",
       "      <td>2.342756</td>\n",
       "      <td>3.086164</td>\n",
       "      <td>2.554176</td>\n",
       "      <td>2.381104</td>\n",
       "      <td>1.637695</td>\n",
       "      <td>2.169683</td>\n",
       "      <td>1.559326</td>\n",
       "      <td>3.003820</td>\n",
       "      <td>3.067945</td>\n",
       "      <td>0.003075</td>\n",
       "      <td>0.664139</td>\n",
       "      <td>0.895187</td>\n",
       "      <td>4.187051</td>\n",
       "      <td>5.384981</td>\n",
       "      <td>3.439161</td>\n",
       "      <td>3.654903</td>\n",
       "      <td>1.888761</td>\n",
       "      <td>316.240601</td>\n",
       "      <td>1250.063110</td>\n",
       "      <td>4.807276e+01</td>\n",
       "      <td>175.575394</td>\n",
       "      <td>192.404648</td>\n",
       "      <td>455.175323</td>\n",
       "      <td>260.644897</td>\n",
       "      <td>234.469772</td>\n",
       "      <td>183.966873</td>\n",
       "      <td>257.434845</td>\n",
       "      <td>1088.404053</td>\n",
       "      <td>106.388313</td>\n",
       "      <td>41.834286</td>\n",
       "      <td>49.203587</td>\n",
       "      <td>-2.131271</td>\n",
       "      <td>126.169441</td>\n",
       "      <td>59.912575</td>\n",
       "      <td>61.562298</td>\n",
       "      <td>85.483574</td>\n",
       "      <td>147.045700</td>\n",
       "      <td>-1.184388</td>\n",
       "      <td>1.415771</td>\n",
       "      <td>0.428050</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39627</th>\n",
       "      <td>3.229492</td>\n",
       "      <td>-1.982422</td>\n",
       "      <td>2.911621</td>\n",
       "      <td>0.068192</td>\n",
       "      <td>171.464676</td>\n",
       "      <td>26.969557</td>\n",
       "      <td>-1.343262</td>\n",
       "      <td>2.045410</td>\n",
       "      <td>-0.316284</td>\n",
       "      <td>-1.466553</td>\n",
       "      <td>155.979950</td>\n",
       "      <td>38.758762</td>\n",
       "      <td>9.328125</td>\n",
       "      <td>1.307345</td>\n",
       "      <td>-1.966797</td>\n",
       "      <td>1.421875</td>\n",
       "      <td>0.272461</td>\n",
       "      <td>5.211914</td>\n",
       "      <td>3.388672</td>\n",
       "      <td>2.843430</td>\n",
       "      <td>3.055280</td>\n",
       "      <td>0.384476</td>\n",
       "      <td>3.109609</td>\n",
       "      <td>1.905012</td>\n",
       "      <td>1.534744</td>\n",
       "      <td>1.959340</td>\n",
       "      <td>1.150269</td>\n",
       "      <td>2.852175</td>\n",
       "      <td>2.906504</td>\n",
       "      <td>1.929827</td>\n",
       "      <td>1.726722</td>\n",
       "      <td>2.876991</td>\n",
       "      <td>5.597272</td>\n",
       "      <td>0.745887</td>\n",
       "      <td>4.535467</td>\n",
       "      <td>4.310321</td>\n",
       "      <td>3.578577</td>\n",
       "      <td>432.700623</td>\n",
       "      <td>1686.925415</td>\n",
       "      <td>8.678886e+01</td>\n",
       "      <td>495.263245</td>\n",
       "      <td>1625.792969</td>\n",
       "      <td>50.484715</td>\n",
       "      <td>168.873550</td>\n",
       "      <td>242.322266</td>\n",
       "      <td>326.821198</td>\n",
       "      <td>373.291901</td>\n",
       "      <td>926.510620</td>\n",
       "      <td>417.585999</td>\n",
       "      <td>154.809525</td>\n",
       "      <td>100.647934</td>\n",
       "      <td>1.410438</td>\n",
       "      <td>213.705505</td>\n",
       "      <td>158.863327</td>\n",
       "      <td>66.497269</td>\n",
       "      <td>175.408432</td>\n",
       "      <td>164.296051</td>\n",
       "      <td>-0.285602</td>\n",
       "      <td>0.639160</td>\n",
       "      <td>0.869667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41551</th>\n",
       "      <td>1.042725</td>\n",
       "      <td>-4.070312</td>\n",
       "      <td>-2.477051</td>\n",
       "      <td>-0.161530</td>\n",
       "      <td>181.452820</td>\n",
       "      <td>71.989296</td>\n",
       "      <td>-1.866943</td>\n",
       "      <td>0.174805</td>\n",
       "      <td>1.780273</td>\n",
       "      <td>0.376526</td>\n",
       "      <td>95.604736</td>\n",
       "      <td>57.316761</td>\n",
       "      <td>7.296875</td>\n",
       "      <td>0.419498</td>\n",
       "      <td>-0.353149</td>\n",
       "      <td>1.688599</td>\n",
       "      <td>0.667725</td>\n",
       "      <td>5.113037</td>\n",
       "      <td>2.041748</td>\n",
       "      <td>2.315521</td>\n",
       "      <td>2.025861</td>\n",
       "      <td>1.941803</td>\n",
       "      <td>2.405507</td>\n",
       "      <td>2.853577</td>\n",
       "      <td>0.538055</td>\n",
       "      <td>2.473931</td>\n",
       "      <td>1.403748</td>\n",
       "      <td>2.519122</td>\n",
       "      <td>2.898768</td>\n",
       "      <td>0.656762</td>\n",
       "      <td>1.150023</td>\n",
       "      <td>0.253725</td>\n",
       "      <td>5.156644</td>\n",
       "      <td>2.936909</td>\n",
       "      <td>2.982647</td>\n",
       "      <td>4.279080</td>\n",
       "      <td>2.477749</td>\n",
       "      <td>508.310608</td>\n",
       "      <td>1724.230713</td>\n",
       "      <td>3.882069e+01</td>\n",
       "      <td>313.446869</td>\n",
       "      <td>580.387268</td>\n",
       "      <td>261.430328</td>\n",
       "      <td>224.522217</td>\n",
       "      <td>530.164307</td>\n",
       "      <td>400.757690</td>\n",
       "      <td>439.905487</td>\n",
       "      <td>1482.520996</td>\n",
       "      <td>202.744644</td>\n",
       "      <td>107.382027</td>\n",
       "      <td>19.343044</td>\n",
       "      <td>0.630250</td>\n",
       "      <td>237.887344</td>\n",
       "      <td>23.519346</td>\n",
       "      <td>101.947006</td>\n",
       "      <td>119.363922</td>\n",
       "      <td>209.560638</td>\n",
       "      <td>-1.041360</td>\n",
       "      <td>0.867920</td>\n",
       "      <td>0.935274</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37669</th>\n",
       "      <td>2.833984</td>\n",
       "      <td>-3.181641</td>\n",
       "      <td>-1.418213</td>\n",
       "      <td>3.117676</td>\n",
       "      <td>63.195469</td>\n",
       "      <td>36.514484</td>\n",
       "      <td>0.735229</td>\n",
       "      <td>0.741577</td>\n",
       "      <td>0.540527</td>\n",
       "      <td>-2.371582</td>\n",
       "      <td>189.691284</td>\n",
       "      <td>117.041931</td>\n",
       "      <td>13.195312</td>\n",
       "      <td>9.621387</td>\n",
       "      <td>0.909058</td>\n",
       "      <td>0.915405</td>\n",
       "      <td>0.912231</td>\n",
       "      <td>6.015625</td>\n",
       "      <td>0.006348</td>\n",
       "      <td>1.747297</td>\n",
       "      <td>1.958740</td>\n",
       "      <td>2.577148</td>\n",
       "      <td>2.523108</td>\n",
       "      <td>0.953369</td>\n",
       "      <td>0.793927</td>\n",
       "      <td>0.389001</td>\n",
       "      <td>2.912109</td>\n",
       "      <td>1.620621</td>\n",
       "      <td>2.184989</td>\n",
       "      <td>2.068194</td>\n",
       "      <td>1.730075</td>\n",
       "      <td>1.641001</td>\n",
       "      <td>2.870790</td>\n",
       "      <td>4.688663</td>\n",
       "      <td>2.299365</td>\n",
       "      <td>6.747117</td>\n",
       "      <td>2.912116</td>\n",
       "      <td>517.966431</td>\n",
       "      <td>1328.529663</td>\n",
       "      <td>2.281642e-07</td>\n",
       "      <td>429.039886</td>\n",
       "      <td>329.290680</td>\n",
       "      <td>599.910767</td>\n",
       "      <td>228.910873</td>\n",
       "      <td>458.500214</td>\n",
       "      <td>272.322754</td>\n",
       "      <td>317.963745</td>\n",
       "      <td>972.946045</td>\n",
       "      <td>296.047546</td>\n",
       "      <td>221.405945</td>\n",
       "      <td>167.143814</td>\n",
       "      <td>2.270602</td>\n",
       "      <td>246.942703</td>\n",
       "      <td>188.282700</td>\n",
       "      <td>111.523277</td>\n",
       "      <td>80.262146</td>\n",
       "      <td>101.673973</td>\n",
       "      <td>0.737652</td>\n",
       "      <td>2.092407</td>\n",
       "      <td>0.577388</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37670</th>\n",
       "      <td>-4.213867</td>\n",
       "      <td>1.827148</td>\n",
       "      <td>-2.466797</td>\n",
       "      <td>2.848145</td>\n",
       "      <td>39.767376</td>\n",
       "      <td>26.642946</td>\n",
       "      <td>0.659668</td>\n",
       "      <td>-0.826172</td>\n",
       "      <td>0.046516</td>\n",
       "      <td>-3.013184</td>\n",
       "      <td>94.072266</td>\n",
       "      <td>82.209488</td>\n",
       "      <td>5.750000</td>\n",
       "      <td>7.299185</td>\n",
       "      <td>1.853027</td>\n",
       "      <td>0.367188</td>\n",
       "      <td>1.110107</td>\n",
       "      <td>6.041016</td>\n",
       "      <td>1.485840</td>\n",
       "      <td>0.968244</td>\n",
       "      <td>2.513313</td>\n",
       "      <td>2.801628</td>\n",
       "      <td>2.893883</td>\n",
       "      <td>0.546387</td>\n",
       "      <td>0.421857</td>\n",
       "      <td>0.165817</td>\n",
       "      <td>3.059700</td>\n",
       "      <td>2.007390</td>\n",
       "      <td>2.387960</td>\n",
       "      <td>0.502933</td>\n",
       "      <td>0.002990</td>\n",
       "      <td>3.056710</td>\n",
       "      <td>5.483438</td>\n",
       "      <td>3.035149</td>\n",
       "      <td>3.431475</td>\n",
       "      <td>6.433916</td>\n",
       "      <td>3.401394</td>\n",
       "      <td>290.504486</td>\n",
       "      <td>798.779358</td>\n",
       "      <td>7.162418e-01</td>\n",
       "      <td>288.719421</td>\n",
       "      <td>703.859497</td>\n",
       "      <td>116.662880</td>\n",
       "      <td>302.265350</td>\n",
       "      <td>165.260284</td>\n",
       "      <td>94.734863</td>\n",
       "      <td>230.970078</td>\n",
       "      <td>666.759277</td>\n",
       "      <td>226.579025</td>\n",
       "      <td>0.250244</td>\n",
       "      <td>156.331924</td>\n",
       "      <td>0.043527</td>\n",
       "      <td>56.215313</td>\n",
       "      <td>169.884338</td>\n",
       "      <td>74.455353</td>\n",
       "      <td>13.876649</td>\n",
       "      <td>86.871559</td>\n",
       "      <td>-0.040770</td>\n",
       "      <td>1.167480</td>\n",
       "      <td>0.183776</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37671</th>\n",
       "      <td>1.232666</td>\n",
       "      <td>-3.455078</td>\n",
       "      <td>-1.707764</td>\n",
       "      <td>1.163818</td>\n",
       "      <td>132.806534</td>\n",
       "      <td>108.308319</td>\n",
       "      <td>0.077759</td>\n",
       "      <td>-1.061035</td>\n",
       "      <td>2.754395</td>\n",
       "      <td>-2.368164</td>\n",
       "      <td>54.110516</td>\n",
       "      <td>50.435898</td>\n",
       "      <td>9.203125</td>\n",
       "      <td>0.189732</td>\n",
       "      <td>1.188965</td>\n",
       "      <td>0.050171</td>\n",
       "      <td>0.619568</td>\n",
       "      <td>4.687744</td>\n",
       "      <td>1.138794</td>\n",
       "      <td>2.871582</td>\n",
       "      <td>1.821027</td>\n",
       "      <td>1.590576</td>\n",
       "      <td>2.614589</td>\n",
       "      <td>0.660400</td>\n",
       "      <td>2.751203</td>\n",
       "      <td>1.453962</td>\n",
       "      <td>1.160627</td>\n",
       "      <td>1.263753</td>\n",
       "      <td>2.057316</td>\n",
       "      <td>2.725532</td>\n",
       "      <td>2.168258</td>\n",
       "      <td>2.954300</td>\n",
       "      <td>4.609194</td>\n",
       "      <td>3.874386</td>\n",
       "      <td>2.386879</td>\n",
       "      <td>4.266889</td>\n",
       "      <td>1.626009</td>\n",
       "      <td>431.401123</td>\n",
       "      <td>1398.959961</td>\n",
       "      <td>7.795755e+01</td>\n",
       "      <td>210.323166</td>\n",
       "      <td>170.419586</td>\n",
       "      <td>448.646027</td>\n",
       "      <td>238.611542</td>\n",
       "      <td>266.062195</td>\n",
       "      <td>22.928062</td>\n",
       "      <td>168.751099</td>\n",
       "      <td>1262.189941</td>\n",
       "      <td>84.972610</td>\n",
       "      <td>120.410286</td>\n",
       "      <td>130.943558</td>\n",
       "      <td>0.586136</td>\n",
       "      <td>120.526772</td>\n",
       "      <td>149.962631</td>\n",
       "      <td>85.739853</td>\n",
       "      <td>87.454422</td>\n",
       "      <td>35.809673</td>\n",
       "      <td>-0.473538</td>\n",
       "      <td>1.154907</td>\n",
       "      <td>0.963875</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37672</th>\n",
       "      <td>3.669434</td>\n",
       "      <td>-1.520264</td>\n",
       "      <td>-2.648926</td>\n",
       "      <td>1.991699</td>\n",
       "      <td>59.859196</td>\n",
       "      <td>42.727608</td>\n",
       "      <td>0.366821</td>\n",
       "      <td>2.274902</td>\n",
       "      <td>0.063507</td>\n",
       "      <td>-0.135101</td>\n",
       "      <td>56.704430</td>\n",
       "      <td>42.705894</td>\n",
       "      <td>2.458984</td>\n",
       "      <td>0.946815</td>\n",
       "      <td>-0.707764</td>\n",
       "      <td>1.200317</td>\n",
       "      <td>0.246277</td>\n",
       "      <td>5.189697</td>\n",
       "      <td>1.908081</td>\n",
       "      <td>1.642560</td>\n",
       "      <td>2.712433</td>\n",
       "      <td>1.928192</td>\n",
       "      <td>2.927015</td>\n",
       "      <td>2.513824</td>\n",
       "      <td>2.126801</td>\n",
       "      <td>3.125623</td>\n",
       "      <td>0.198608</td>\n",
       "      <td>2.627157</td>\n",
       "      <td>3.012290</td>\n",
       "      <td>2.212120</td>\n",
       "      <td>2.297395</td>\n",
       "      <td>2.098787</td>\n",
       "      <td>4.273703</td>\n",
       "      <td>2.697965</td>\n",
       "      <td>2.874723</td>\n",
       "      <td>4.350467</td>\n",
       "      <td>1.918390</td>\n",
       "      <td>231.997330</td>\n",
       "      <td>828.698059</td>\n",
       "      <td>6.574512e+01</td>\n",
       "      <td>178.078583</td>\n",
       "      <td>314.126190</td>\n",
       "      <td>134.585953</td>\n",
       "      <td>123.302994</td>\n",
       "      <td>288.541351</td>\n",
       "      <td>70.711540</td>\n",
       "      <td>188.359192</td>\n",
       "      <td>678.115967</td>\n",
       "      <td>109.239563</td>\n",
       "      <td>75.249825</td>\n",
       "      <td>62.074875</td>\n",
       "      <td>-2.233888</td>\n",
       "      <td>76.772499</td>\n",
       "      <td>117.833290</td>\n",
       "      <td>30.000200</td>\n",
       "      <td>98.930305</td>\n",
       "      <td>84.475143</td>\n",
       "      <td>1.216050</td>\n",
       "      <td>1.394531</td>\n",
       "      <td>0.269940</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37673</th>\n",
       "      <td>-3.261230</td>\n",
       "      <td>1.885986</td>\n",
       "      <td>2.270996</td>\n",
       "      <td>-1.406006</td>\n",
       "      <td>183.750626</td>\n",
       "      <td>88.951378</td>\n",
       "      <td>-1.693115</td>\n",
       "      <td>0.587524</td>\n",
       "      <td>-0.489441</td>\n",
       "      <td>0.068649</td>\n",
       "      <td>56.152721</td>\n",
       "      <td>44.826107</td>\n",
       "      <td>0.509277</td>\n",
       "      <td>0.154000</td>\n",
       "      <td>-1.005493</td>\n",
       "      <td>1.275146</td>\n",
       "      <td>0.134827</td>\n",
       "      <td>5.147217</td>\n",
       "      <td>2.280640</td>\n",
       "      <td>2.606183</td>\n",
       "      <td>2.760437</td>\n",
       "      <td>0.916565</td>\n",
       "      <td>3.122440</td>\n",
       "      <td>2.202347</td>\n",
       "      <td>1.474655</td>\n",
       "      <td>2.602655</td>\n",
       "      <td>0.558090</td>\n",
       "      <td>2.513519</td>\n",
       "      <td>2.913828</td>\n",
       "      <td>1.154842</td>\n",
       "      <td>0.907925</td>\n",
       "      <td>1.466015</td>\n",
       "      <td>3.174744</td>\n",
       "      <td>3.694599</td>\n",
       "      <td>4.434326</td>\n",
       "      <td>1.964844</td>\n",
       "      <td>2.347931</td>\n",
       "      <td>393.857178</td>\n",
       "      <td>1875.090332</td>\n",
       "      <td>2.737496e+01</td>\n",
       "      <td>165.908920</td>\n",
       "      <td>267.109619</td>\n",
       "      <td>416.159607</td>\n",
       "      <td>630.088013</td>\n",
       "      <td>122.527405</td>\n",
       "      <td>162.339218</td>\n",
       "      <td>242.306015</td>\n",
       "      <td>1685.213257</td>\n",
       "      <td>143.567535</td>\n",
       "      <td>29.521320</td>\n",
       "      <td>40.245060</td>\n",
       "      <td>-1.397366</td>\n",
       "      <td>70.520828</td>\n",
       "      <td>44.056137</td>\n",
       "      <td>20.176321</td>\n",
       "      <td>97.122978</td>\n",
       "      <td>106.880989</td>\n",
       "      <td>-0.644414</td>\n",
       "      <td>1.298462</td>\n",
       "      <td>0.962174</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10764 rows × 67 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       CleanJet_eta_1  CleanJet_eta_2  CleanJet_phi_1  CleanJet_phi_2  \\\n",
       "40518       -0.477417        2.986816        1.836426       -1.672363   \n",
       "44449        1.980225       -3.985840        2.588867        1.987793   \n",
       "39659       -3.074219        2.674805        1.145020        0.401611   \n",
       "39627        3.229492       -1.982422        2.911621        0.068192   \n",
       "41551        1.042725       -4.070312       -2.477051       -0.161530   \n",
       "...               ...             ...             ...             ...   \n",
       "37669        2.833984       -3.181641       -1.418213        3.117676   \n",
       "37670       -4.213867        1.827148       -2.466797        2.848145   \n",
       "37671        1.232666       -3.455078       -1.707764        1.163818   \n",
       "37672        3.669434       -1.520264       -2.648926        1.991699   \n",
       "37673       -3.261230        1.885986        2.270996       -1.406006   \n",
       "\n",
       "       CleanJet_pt_1  CleanJet_pt_2  Lepton_eta_1  Lepton_eta_2  Lepton_phi_1  \\\n",
       "40518     124.805344      80.120354      0.226135      2.252441      0.036774   \n",
       "44449     193.126358      54.814369     -0.160583      1.335205     -1.342529   \n",
       "39659      95.718979      39.599072     -1.658447     -0.592651     -2.795410   \n",
       "39627     171.464676      26.969557     -1.343262      2.045410     -0.316284   \n",
       "41551     181.452820      71.989296     -1.866943      0.174805      1.780273   \n",
       "...              ...            ...           ...           ...           ...   \n",
       "37669      63.195469      36.514484      0.735229      0.741577      0.540527   \n",
       "37670      39.767376      26.642946      0.659668     -0.826172      0.046516   \n",
       "37671     132.806534     108.308319      0.077759     -1.061035      2.754395   \n",
       "37672      59.859196      42.727608      0.366821      2.274902      0.063507   \n",
       "37673     183.750626      88.951378     -1.693115      0.587524     -0.489441   \n",
       "\n",
       "       Lepton_phi_2  Lepton_pt_1  Lepton_pt_2  MET_significance       Rpt  \\\n",
       "40518      2.776367    84.481834    64.644058         13.671875  0.546155   \n",
       "44449      1.149414   266.018677    31.155691        116.187500  0.782913   \n",
       "39659     -1.236084    66.873520    52.486717          3.103516  0.926020   \n",
       "39627     -1.466553   155.979950    38.758762          9.328125  1.307345   \n",
       "41551      0.376526    95.604736    57.316761          7.296875  0.419498   \n",
       "...             ...          ...          ...               ...       ...   \n",
       "37669     -2.371582   189.691284   117.041931         13.195312  9.621387   \n",
       "37670     -3.013184    94.072266    82.209488          5.750000  7.299185   \n",
       "37671     -2.368164    54.110516    50.435898          9.203125  0.189732   \n",
       "37672     -0.135101    56.704430    42.705894          2.458984  0.946815   \n",
       "37673      0.068649    56.152721    44.826107          0.509277  0.154000   \n",
       "\n",
       "        Zepp_l1   Zepp_l2   Zepp_ll    detajj    detall    dphijj  \\\n",
       "40518 -1.028564  0.997742  0.015411  3.464233  2.026306  2.774396   \n",
       "44449  0.842224  2.338013  1.590118  5.966064  1.495789  0.601074   \n",
       "39659 -1.458740 -0.392944  0.925842  5.749023  1.065796  0.743408   \n",
       "39627 -1.966797  1.421875  0.272461  5.211914  3.388672  2.843430   \n",
       "41551 -0.353149  1.688599  0.667725  5.113037  2.041748  2.315521   \n",
       "...         ...       ...       ...       ...       ...       ...   \n",
       "37669  0.909058  0.915405  0.912231  6.015625  0.006348  1.747297   \n",
       "37670  1.853027  0.367188  1.110107  6.041016  1.485840  0.968244   \n",
       "37671  1.188965  0.050171  0.619568  4.687744  1.138794  2.871582   \n",
       "37672 -0.707764  1.200317  0.246277  5.189697  1.908081  1.642560   \n",
       "37673 -1.005493  1.275146  0.134827  5.147217  2.280640  2.606183   \n",
       "\n",
       "       dphilep1jet1  dphilep1jet2  dphilep1jj  dphilep2jet1  dphilep2jet2  \\\n",
       "40518      1.799652      1.709137    2.321460      0.939941      1.834455   \n",
       "44449      2.351789      2.952863    2.481129      1.439453      0.838379   \n",
       "39659      2.342756      3.086164    2.554176      2.381104      1.637695   \n",
       "39627      3.055280      0.384476    3.109609      1.905012      1.534744   \n",
       "41551      2.025861      1.941803    2.405507      2.853577      0.538055   \n",
       "...             ...           ...         ...           ...           ...   \n",
       "37669      1.958740      2.577148    2.523108      0.953369      0.793927   \n",
       "37670      2.513313      2.801628    2.893883      0.546387      0.421857   \n",
       "37671      1.821027      1.590576    2.614589      0.660400      2.751203   \n",
       "37672      2.712433      1.928192    2.927015      2.513824      2.126801   \n",
       "37673      2.760437      0.916565    3.122440      2.202347      1.474655   \n",
       "\n",
       "       dphilep2jj    dphill  dphilljet  dphilljetjet  dphillmet  dphilmet1  \\\n",
       "40518    0.418133  2.739594   1.008260      1.530068   2.142135   1.350743   \n",
       "44449    1.310113  2.491943   2.429763      2.559103   1.907331   1.985305   \n",
       "39659    2.169683  1.559326   3.003820      3.067945   0.003075   0.664139   \n",
       "39627    1.959340  1.150269   2.852175      2.906504   1.929827   1.726722   \n",
       "41551    2.473931  1.403748   2.519122      2.898768   0.656762   1.150023   \n",
       "...           ...       ...        ...           ...        ...        ...   \n",
       "37669    0.389001  2.912109   1.620621      2.184989   2.068194   1.730075   \n",
       "37670    0.165817  3.059700   2.007390      2.387960   0.502933   0.002990   \n",
       "37671    1.453962  1.160627   1.263753      2.057316   2.725532   2.168258   \n",
       "37672    3.125623  0.198608   2.627157      3.012290   2.212120   2.297395   \n",
       "37673    2.602655  0.558090   2.513519      2.913828   1.154842   0.907925   \n",
       "\n",
       "       dphilmet2   dr_l1j1   dr_l1j2   dr_l2j1   dr_l2j2      drll  \\\n",
       "40518   2.192849  1.932287  3.246923  2.887147  4.508936  3.407534   \n",
       "44449   0.506638  4.476487  5.071847  1.577363  5.386687  2.906401   \n",
       "39659   0.895187  4.187051  5.384981  3.439161  3.654903  1.888761   \n",
       "39627   2.876991  5.597272  0.745887  4.535467  4.310321  3.578577   \n",
       "41551   0.253725  5.156644  2.936909  2.982647  4.279080  2.477749   \n",
       "...          ...       ...       ...       ...       ...       ...   \n",
       "37669   1.641001  2.870790  4.688663  2.299365  6.747117  2.912116   \n",
       "37670   3.056710  5.483438  3.035149  3.431475  6.433916  3.401394   \n",
       "37671   2.954300  4.609194  3.874386  2.386879  4.266889  1.626009   \n",
       "37672   2.098787  4.273703  2.697965  2.874723  4.350467  1.918390   \n",
       "37673   1.466015  3.174744  3.694599  4.434326  1.964844  2.347931   \n",
       "\n",
       "               ht      m2ljj30           mT2         mTi       m_l1j1  \\\n",
       "40518  437.157532   815.015503  3.160135e+01  348.510468   177.942291   \n",
       "44449  692.228027  2395.236816  2.388153e+02  416.484070   718.230225   \n",
       "39659  316.240601  1250.063110  4.807276e+01  175.575394   192.404648   \n",
       "39627  432.700623  1686.925415  8.678886e+01  495.263245  1625.792969   \n",
       "41551  508.310608  1724.230713  3.882069e+01  313.446869   580.387268   \n",
       "...           ...          ...           ...         ...          ...   \n",
       "37669  517.966431  1328.529663  2.281642e-07  429.039886   329.290680   \n",
       "37670  290.504486   798.779358  7.162418e-01  288.719421   703.859497   \n",
       "37671  431.401123  1398.959961  7.795755e+01  210.323166   170.419586   \n",
       "37672  231.997330   828.698059  6.574512e+01  178.078583   314.126190   \n",
       "37673  393.857178  1875.090332  2.737496e+01  165.908920   267.109619   \n",
       "\n",
       "           m_l1j2      m_l2j1      m_l2j2       mcoll     mcollWW  \\\n",
       "40518  330.811035  339.706390  127.020271  126.127602  205.137039   \n",
       "44449  835.173340  115.836853  589.225464  455.491150  484.218079   \n",
       "39659  455.175323  260.644897  234.469772  183.966873  257.434845   \n",
       "39627   50.484715  168.873550  242.322266  326.821198  373.291901   \n",
       "41551  261.430328  224.522217  530.164307  400.757690  439.905487   \n",
       "...           ...         ...         ...         ...         ...   \n",
       "37669  599.910767  228.910873  458.500214  272.322754  317.963745   \n",
       "37670  116.662880  302.265350  165.260284   94.734863  230.970078   \n",
       "37671  448.646027  238.611542  266.062195   22.928062  168.751099   \n",
       "37672  134.585953  123.302994  288.541351   70.711540  188.359192   \n",
       "37673  416.159607  630.088013  122.527405  162.339218  242.306015   \n",
       "\n",
       "               mjj         mll        mtw1        mtw2  phi_miss   proxyW_l1  \\\n",
       "40518   583.159729  228.474884  104.770233  130.405304 -1.313969  106.477066   \n",
       "44449  2027.787598  228.132904  331.348602   33.934185  0.642776  332.871216   \n",
       "39659  1088.404053  106.388313   41.834286   49.203587 -2.131271  126.169441   \n",
       "39627   926.510620  417.585999  154.809525  100.647934  1.410438  213.705505   \n",
       "41551  1482.520996  202.744644  107.382027   19.343044  0.630250  237.887344   \n",
       "...            ...         ...         ...         ...       ...         ...   \n",
       "37669   972.946045  296.047546  221.405945  167.143814  2.270602  246.942703   \n",
       "37670   666.759277  226.579025    0.250244  156.331924  0.043527   56.215313   \n",
       "37671  1262.189941   84.972610  120.410286  130.943558  0.586136  120.526772   \n",
       "37672   678.115967  109.239563   75.249825   62.074875 -2.233888   76.772499   \n",
       "37673  1685.213257  143.567535   29.521320   40.245060 -1.397366   70.520828   \n",
       "\n",
       "        proxyW_l2     pt_miss        ptll      recoil       yll      zeta  \\\n",
       "40518  240.668655   83.105957   35.556599   70.533859  1.136892  0.703552   \n",
       "44449  103.010315  147.112946  241.944656  238.059662  0.036701  0.645020   \n",
       "39659   59.912575   61.562298   85.483574  147.045700 -1.184388  1.415771   \n",
       "39627  158.863327   66.497269  175.408432  164.296051 -0.285602  0.639160   \n",
       "41551   23.519346  101.947006  119.363922  209.560638 -1.041360  0.867920   \n",
       "...           ...         ...         ...         ...       ...       ...   \n",
       "37669  188.282700  111.523277   80.262146  101.673973  0.737652  2.092407   \n",
       "37670  169.884338   74.455353   13.876649   86.871559 -0.040770  1.167480   \n",
       "37671  149.962631   85.739853   87.454422   35.809673 -0.473538  1.154907   \n",
       "37672  117.833290   30.000200   98.930305   84.475143  1.216050  1.394531   \n",
       "37673   44.056137   20.176321   97.122978  106.880989 -0.644414  1.298462   \n",
       "\n",
       "       DNN_SigVsBkg  isLL  isMIX  isTT  isBKG  isVBS  notLL  notTT  \n",
       "40518      0.698907   0.0    1.0   0.0    0.0    1.0    1.0    1.0  \n",
       "44449      0.878863   0.0    0.0   1.0    0.0    1.0    1.0    0.0  \n",
       "39659      0.428050   0.0    1.0   0.0    0.0    1.0    1.0    1.0  \n",
       "39627      0.869667   0.0    1.0   0.0    0.0    1.0    1.0    1.0  \n",
       "41551      0.935274   0.0    1.0   0.0    0.0    1.0    1.0    1.0  \n",
       "...             ...   ...    ...   ...    ...    ...    ...    ...  \n",
       "37669      0.577388   1.0    0.0   0.0    0.0    1.0    0.0    1.0  \n",
       "37670      0.183776   1.0    0.0   0.0    0.0    1.0    0.0    1.0  \n",
       "37671      0.963875   1.0    0.0   0.0    0.0    1.0    0.0    1.0  \n",
       "37672      0.269940   1.0    0.0   0.0    0.0    1.0    0.0    1.0  \n",
       "37673      0.962174   1.0    0.0   0.0    0.0    1.0    0.0    1.0  \n",
       "\n",
       "[10764 rows x 67 columns]"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(len(df))\n",
    "print(len(df[(df['DNN_SigVsBkg'] > 0.5) & (df['isBKG'] == 1)]))\n",
    "print(len(df[(df['DNN_SigVsBkg'] > 0.5) & (df['isLL'] == 1)]))\n",
    "print(len(df[(df['DNN_SigVsBkg'] > 0.5) & (df['isMIX'] == 1)]))\n",
    "print(len(df[(df['DNN_SigVsBkg'] > 0.5) & (df['isTT'] == 1)]))\n",
    "\n",
    "#df_other = df[((df['isTT'] + df['isMIX'] + df['isLL']) == 1) & (df['DNN_SigVsBkg'] > 0.5)]\n",
    "df_other = df[((df['isTT'] + df['isMIX'] + df['isLL']) == 1) & (df['DNN_SigVsBkg'] > 0.)]\n",
    "\n",
    "#df_notLL = df_other[df_other['notLL'] == 1].sample(n=int(len(df_other[df_other['notLL'] == 1])/2))\n",
    "#df_notTT= df_other[df_other['notTT'] == 1].sample(n=int(len(df_other[df_other['notTT'] == 1])/2))\n",
    "\n",
    "df_notLL = df_other[df_other['notLL'] == 1].sample(n=int(len(df_other[df_other['isLL'] == 1])))\n",
    "df_notTT= df_other[df_other['notTT'] == 1].sample(n=int(len(df_other[df_other['isTT'] == 1])))\n",
    "\n",
    "\n",
    "df_LL = pd.concat([df_notLL, df_other[df_other['isLL'] == 1]])\n",
    "df_TT = pd.concat([df_notTT, df_other[df_other['isTT'] == 1]])\n",
    "\n",
    "print(len(df_LL[(df_LL['isLL'] == 1)]))\n",
    "print(len(df_LL[(df_LL['isMIX'] == 1)]))\n",
    "print(len(df_LL[(df_LL['isTT'] == 1)]))\n",
    "df_LL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "be847e8c-9b7e-4377-88d6-573fc207d768",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-0.47741699  2.98681641  1.83642578 ...  1.13689184  0.70355225\n",
      "   0.69890678]\n",
      " [ 1.98022461 -3.98583984  2.58886719 ...  0.03670105  0.64501953\n",
      "   0.87886256]\n",
      " [-3.07421875  2.67480469  1.14501953 ... -1.18438804  1.41577148\n",
      "   0.42805037]\n",
      " ...\n",
      " [ 1.23266602 -3.45507812 -1.70776367 ... -0.47353783  1.15490723\n",
      "   0.963875  ]\n",
      " [ 3.66943359 -1.52026367 -2.64892578 ...  1.21604979  1.39453125\n",
      "   0.26993981]\n",
      " [-3.26123047  1.88598633  2.27099609 ... -0.64441437  1.29846191\n",
      "   0.962174  ]] [[0. 1.]\n",
      " [0. 1.]\n",
      " [0. 1.]\n",
      " ...\n",
      " [1. 0.]\n",
      " [1. 0.]\n",
      " [1. 0.]]\n"
     ]
    }
   ],
   "source": [
    "for col in df.columns:\n",
    "    df_LL[col] = np.hstack(df_LL[col])\n",
    "    df_TT[col] = np.hstack(df_TT[col])\n",
    "'''\n",
    "df.replace([np.inf, -np.inf], np.nan, inplace=True)\n",
    "df.dropna(inplace=True)\n",
    "df = shuffle(df)\n",
    "'''\n",
    "# Perform the splitting and define training and validation datasets\n",
    "msk_LL = np.random.rand(len(df_LL)) < 0.8\n",
    "msk_TT = np.random.rand(len(df_TT)) < 0.8\n",
    "\n",
    "df_train_LL = df_LL[msk_LL]\n",
    "df_train_TT = df_TT[msk_TT]\n",
    "df_val_LL   = df_LL[~msk_LL]\n",
    "df_val_TT   = df_TT[~msk_TT]\n",
    "\n",
    "features = df_LL.columns[:len(df.columns)-7]\n",
    "NDIM = len(features)\n",
    "\n",
    "X_LL = df_train_LL.values[:,0:NDIM]\n",
    "Y_LL = df_train_LL[['isLL', 'notLL']].values\n",
    "#Y_LL = df_train_LL[['isLL']].values\n",
    "X_val_LL = df_val_LL.values[:,0:NDIM]\n",
    "Y_val_LL = df_val_LL[['isLL', 'notLL']].values\n",
    "#Y_val_LL = df_val_LL[['isLL']].values\n",
    "\n",
    "X_TT = df_train_TT.values[:,0:NDIM]\n",
    "Y_TT = df_train_TT[['isTT', 'notTT']].values\n",
    "#Y_TT = df_train_TT[['isTT']].values\n",
    "X_val_TT = df_val_TT.values[:,0:NDIM]\n",
    "Y_val_TT = df_val_TT[['isTT', 'notTT']].values\n",
    "#Y_val_TT = df_val_TT[['isTT']].values\n",
    "\n",
    "#X_val = df_val.values[:,0:NDIM+1]\n",
    "#Y_val = df_val.values[:,NDIM+1:NDIM+1+nclasses] # isLL, isMIX, isTT, isBKG\n",
    "print(X_LL,Y_LL)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "3c21764e-e053-40e7-a288-9e7fe2763a48",
   "metadata": {},
   "outputs": [],
   "source": [
    "dnnLL = SimpleNeuralNetwork(250, learning_rate=0.0001, N_NODES=50, n_layers=8, n_features=X_LL.shape[1], n_outputsC=2, ATLAS=False)\n",
    "dnnLL.save_weights(\"ATLAS_selection_LLvsOther_emu_RDF\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "99f11e98-eafd-4202-8ec6-6044171a86cc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(8637, 60) (8637, 2)\n"
     ]
    }
   ],
   "source": [
    "print(X_LL.shape, Y_LL.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "id": "63f73e91-f4f2-44ef-8baa-b762ce41b801",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 250/250 [00:08<00:00, 30.95it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "acc = dnnLL.fit (X_LL.astype(np.float32), Y_LL.astype(np.float32), X_val_LL.astype(np.float32), Y_val_LL.astype(np.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "id": "66ab7ce2-9739-480f-88e6-2730bda10ed2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Y_predict_train_LL = dnnLL.predict_proba(X_LL)\n",
    "Y_predict_val_LL = dnnLL.predict_proba(X_val_LL)\n",
    "\n",
    "axis = np.linspace(0,1,20)\n",
    "\n",
    "plt.hist(dnnLL.predict_proba( df_val_LL[ df_val_LL['isLL']==1 ].values[:,0:NDIM] ).numpy()[...,0], bins = axis, label = 'isLL class - isLL sample', histtype='step', color='r',  density=True, linewidth=2 )\n",
    "plt.hist(dnnLL.predict_proba( df_val_LL[ df_val_LL['isLL']==0 ].values[:,0:NDIM] ).numpy()[...,0], bins = axis, label = 'isLL class - isnotLL sample', histtype='step', color='b',  density=True, linewidth=2 )\n",
    "plt.title('Validation dataset')\n",
    "plt.xlabel('isLL Probability')\n",
    "plt.legend(frameon=False, bbox_to_anchor=(1.05, 1.))\n",
    "#plt.yscale('log')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "id": "ac511b3c-889a-41a0-8484-76841bf345c2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classifier categorical accuracy =  0.7005171603196991\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "from scipy.stats import ks_2samp, anderson_ksamp\n",
    "\n",
    "X_LL = df_val_LL.values[:,0:NDIM]\n",
    "Y_true_LL = df_val_LL[['isLL', 'notLL']].values\n",
    "Y_pred_LL = dnnLL.predict_proba(X_LL)\n",
    "\n",
    "# The classifier categorical accuracy is a good and simple metrics to evaluate the overall classification performance\n",
    "Y_true_max_LL = np.argmax(Y_true_LL, axis=1)\n",
    "Y_pred_max_LL = np.argmax(Y_pred_LL, axis=1)\n",
    "print(\"Classifier categorical accuracy = \", accuracy_score(Y_true_max_LL, Y_pred_max_LL))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "id": "21632233-f418-4238-af28-48a0d7c934ba",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "C = confusion_matrix(Y_true_max_LL, Y_pred_max_LL, normalize=\"true\")\n",
    "C = np.around(C, decimals=3)\n",
    "\n",
    "plt.figure()\n",
    "plot_confusion_matrix(C, classes=['isLL', 'notLL'], title='')\n",
    "plt.show(1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "id": "86c893f8-a7b7-441f-b144-e66a18bb4906",
   "metadata": {},
   "outputs": [],
   "source": [
    "dnnTT = SimpleNeuralNetwork(200, learning_rate=0.0001, N_NODES=50, n_layers=8, n_features=X_TT.shape[1], n_outputsC=2, ATLAS=False)\n",
    "dnnTT.save_weights(\"ATLAS_selection_TTvsOther_emu_RDF\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "id": "4c4e300d-5fa1-4e71-a5ef-267d5beb01e9",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 200/200 [00:06<00:00, 29.83it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "acc = dnnTT.fit (X_TT.astype(np.float32), Y_TT.astype(np.float32), X_val_TT.astype(np.float32), Y_val_TT.astype(np.float32))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "c961815e-a9f8-42ae-996e-6679723ba022",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Y_predict_train_TT = dnnTT.predict_proba(X_TT)\n",
    "Y_predict_val_TT = dnnTT.predict_proba(X_val_TT)\n",
    "\n",
    "axis = np.linspace(0,1,20)\n",
    "\n",
    "plt.hist(dnnTT.predict_proba( df_val_TT[ df_val_TT['isTT']==1 ].values[:,0:NDIM] ).numpy()[...,0], bins = axis, label = 'isTT class - isTT sample', histtype='step', color='r',  density=True, linewidth=2 )\n",
    "plt.hist(dnnTT.predict_proba( df_val_TT[ df_val_TT['isTT']==0 ].values[:,0:NDIM] ).numpy()[...,0], bins = axis, label = 'isTT class - isnotTT sample', histtype='step', color='b',  density=True, linewidth=2 )\n",
    "plt.title('Validation dataset')\n",
    "plt.xlabel('isTT Probability')\n",
    "plt.legend(frameon=False, bbox_to_anchor=(1.05, 1.))\n",
    "#plt.yscale('log')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "id": "5948ad04-3997-4577-890b-0cd619f9351c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classifier categorical accuracy =  0.6825876662636033\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "from scipy.stats import ks_2samp, anderson_ksamp\n",
    "\n",
    "X_TT = df_val_TT.values[:,0:NDIM]\n",
    "Y_true_TT = df_val_TT[['isTT', 'notTT']].values\n",
    "Y_pred_TT = dnnTT.predict_proba(X_TT)\n",
    "\n",
    "# The classifier categorical accuracy is a good and simple metrics to evaluate the overall classification performance\n",
    "Y_true_max_TT = np.argmax(Y_true_TT, axis=1)\n",
    "Y_pred_max_TT = np.argmax(Y_pred_TT, axis=1)\n",
    "print(\"Classifier categorical accuracy = \", accuracy_score(Y_true_max_TT, Y_pred_max_TT))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 230,
   "id": "68a8bade-814c-414d-8c1d-fd64024cac1d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "C = confusion_matrix(Y_true_max_TT, Y_pred_max_TT, normalize=\"true\")\n",
    "C = np.around(C, decimals=3)\n",
    "\n",
    "plt.figure()\n",
    "plot_confusion_matrix(C, classes=['isTT', 'notTT'], title='')\n",
    "plt.show(1) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 231,
   "id": "a1234dc6-62fb-4667-bdc4-46e3d3a06b91",
   "metadata": {
    "collapsed": true,
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1608\n",
      "1608\n",
      "       CleanJet_eta_1  CleanJet_eta_2  CleanJet_phi_1  CleanJet_phi_2  \\\n",
      "44665        0.937134       -4.084961        2.667969       -1.143555   \n",
      "41650       -2.507324        1.450439       -0.121002        3.026367   \n",
      "41627        1.963623       -2.478516       -1.513428        2.435059   \n",
      "42323        0.155762        4.254883        2.935547       -2.021973   \n",
      "47697       -2.166016        3.692871        1.979248       -2.877930   \n",
      "...               ...             ...             ...             ...   \n",
      "37641        4.244141       -4.335938        0.567749       -0.154175   \n",
      "37642        3.259277       -1.629883        1.904541       -0.386963   \n",
      "37646        2.483887       -2.088867       -1.333008        1.924072   \n",
      "37648       -4.634766        4.043945       -2.001953       -2.539062   \n",
      "37673       -3.261230        1.885986        2.270996       -1.406006   \n",
      "\n",
      "       CleanJet_pt_1  CleanJet_pt_2  Lepton_eta_1  Lepton_eta_2  Lepton_phi_1  \\\n",
      "44665     122.242439      71.023552     -1.425781      0.062294      0.121185   \n",
      "41650     193.874359      71.332596      0.243408     -1.608643      2.738281   \n",
      "41627     307.649567     184.718430     -0.960449      0.791504      0.534546   \n",
      "42323     286.376648      80.061569      1.792969      0.739014     -0.262146   \n",
      "47697     103.329491      92.026344     -0.973999      0.926147     -0.079361   \n",
      "...              ...            ...           ...           ...           ...   \n",
      "37641      30.856323      30.751810     -1.277588      0.356384      2.858398   \n",
      "37642     149.026794     141.265549      2.214844      0.646362     -1.668213   \n",
      "37646     182.465561     162.498306      0.696533     -0.557129      1.652344   \n",
      "37648      31.566549      29.221609      0.337891     -1.759277      1.644043   \n",
      "37673     183.750626      88.951378     -1.693115      0.587524     -0.489441   \n",
      "\n",
      "       Lepton_phi_2  Lepton_pt_1  Lepton_pt_2  MET_significance        Rpt  \\\n",
      "44665      2.684082    64.729980    41.997463         11.054688   0.313115   \n",
      "41650     -2.049805   114.392410    56.447361          8.687500   0.466909   \n",
      "41627      2.783691   200.978256    52.963169          8.492188   0.187308   \n",
      "42323      2.739746   167.469696    49.984039         77.937500   0.365095   \n",
      "47697     -1.239746    87.339195    72.267685          3.505859   0.663769   \n",
      "...             ...          ...          ...               ...        ...   \n",
      "37641      0.288147   109.018684    44.385818         11.937500   5.099532   \n",
      "37642     -2.779297    51.215000    40.670746          0.191040   0.098942   \n",
      "37646     -1.591797    39.395264    37.514561          8.125000   0.049844   \n",
      "37648     -1.223145   277.956451   156.945694          5.960938  47.292789   \n",
      "37673      0.068649    56.152721    44.826107          0.509277   0.154000   \n",
      "\n",
      "        Zepp_l1   Zepp_l2   Zepp_ll    detajj    detall    dphijj  \\\n",
      "44665  0.148132  1.636208  0.892170  5.022095  1.488075  2.471662   \n",
      "41650  0.771851 -1.080200  0.154175  3.957764  1.852051  3.135816   \n",
      "41627 -0.703003  1.048950  0.172974  4.442139  1.751953  2.334699   \n",
      "42323 -0.412354 -1.466309  0.939331  4.099121  1.053955  1.325666   \n",
      "47697 -1.737427  0.162720  0.787354  5.858887  1.900146  1.426008   \n",
      "...         ...       ...       ...       ...       ...       ...   \n",
      "37641 -1.231689  0.402283  0.414703  8.580078  1.633972  0.721924   \n",
      "37642  1.400146 -0.168335  0.615906  4.889160  1.568481  2.291504   \n",
      "37646  0.499023 -0.754639  0.127808  4.572754  1.253662  3.026105   \n",
      "37648  0.633301 -1.463867  0.415283  8.678711  2.097168  0.537109   \n",
      "37673 -1.005493  1.275146  0.134827  5.147217  2.280640  2.606183   \n",
      "\n",
      "       dphilep1jet1  dphilep1jet2  dphilep1jj  dphilep2jet1  dphilep2jet2  \\\n",
      "44665      2.546783      1.264740    3.131879      0.016113      2.455549   \n",
      "41650      2.859283      0.288086    2.862646      1.928802      1.207013   \n",
      "41627      2.047974      1.900513    2.686078      1.986066      0.348633   \n",
      "42323      3.085492      1.759827    2.836773      0.195801      1.521467   \n",
      "47697      2.058609      2.798569    2.721620      3.064191      1.638184   \n",
      "...             ...           ...         ...           ...           ...   \n",
      "37641      2.290649      3.012573    2.650971      0.279602      0.442322   \n",
      "37642      2.710431      1.281250    2.486001      1.599347      2.392334   \n",
      "37646      2.985352      0.271729    2.570821      0.258789      2.767316   \n",
      "37648      2.637189      2.100080    2.379250      0.778809      1.315918   \n",
      "37673      2.760437      0.916565    3.122440      2.202347      1.474655   \n",
      "\n",
      "       dphilep2jj    dphill  dphilljet  dphilljetjet  dphillmet  dphilmet1  \\\n",
      "44665    0.568983  2.562897   1.886365      2.471461   1.197006   0.536588   \n",
      "41650    1.925440  1.495099   2.980993      2.977630   1.957742   1.514833   \n",
      "41627    1.347962  2.249146   2.289031      2.927135   0.798687   1.039744   \n",
      "42323    0.444520  3.001892   3.138763      2.895702   0.383277   0.442206   \n",
      "47697    2.401180  1.160385   2.576988      3.043187   1.713162   1.194784   \n",
      "...           ...       ...        ...           ...        ...        ...   \n",
      "37641    0.080719  2.570251   1.967544      2.327865   2.421875   2.098769   \n",
      "37642    2.686101  1.111084   2.226004      2.970428   0.140733   0.343695   \n",
      "37646    0.468224  3.039045   2.222981      1.495968   2.567991   1.493138   \n",
      "37648    1.036748  2.867188   2.960599      2.702660   2.758610   3.082020   \n",
      "37673    2.602655  0.558090   2.513519      2.913828   1.154842   0.907925   \n",
      "\n",
      "       dphilmet2   dr_l1j1   dr_l1j2   dr_l2j1   dr_l2j2      drll  \\\n",
      "44665   3.099484  3.474115  2.944623  0.874988  5.643627  2.963580   \n",
      "41650   3.009933  3.967623  1.240934  2.127888  5.926677  2.380213   \n",
      "41627   1.209401  3.569929  2.432380  4.454110  3.288552  2.850964   \n",
      "42323   2.559686  3.592448  3.026221  0.615240  5.919063  3.181537   \n",
      "47697   2.355169  2.378818  5.441660  4.463563  3.215339  2.226444   \n",
      "...          ...       ...       ...       ...       ...       ...   \n",
      "37641   1.614165  5.978006  4.292913  3.897798  4.713123  3.045662   \n",
      "37642   0.767389  3.722286  4.052595  5.363363  3.302204  1.922145   \n",
      "37646   1.751003  3.479505  2.798623  3.052007  3.835043  3.287470   \n",
      "37648   0.214833  6.166084  5.588668  2.979090  5.950549  3.552306   \n",
      "37673   1.466015  3.174744  3.694599  4.434326  1.964844  2.347931   \n",
      "\n",
      "               ht      m2ljj30         mT2         mTi       m_l1j1  \\\n",
      "44665  369.953674  1307.741577   29.727242  202.759033   313.547333   \n",
      "41650  457.887909  1149.574341   25.915377  234.316193   625.763855   \n",
      "41627  822.017700  2590.272705    0.008358  338.493134  1101.025269   \n",
      "42323  781.586609  1440.434570   61.897785  479.567291   593.845276   \n",
      "47697  376.009430  2136.815918   24.983013  221.172806   202.947678   \n",
      "...           ...          ...         ...         ...          ...   \n",
      "37641  264.896576  2493.904053   84.682228  255.523758   919.616089   \n",
      "37642  420.359924  1831.880615    9.034244  156.741898   196.927002   \n",
      "37646  454.736755  1805.719604   34.116341  121.992035   241.771469   \n",
      "37648  535.824463  1344.701904  276.255096  753.437317  1132.496826   \n",
      "37673  393.857178  1875.090332   27.374958  165.908920   267.109619   \n",
      "\n",
      "           m_l1j2      m_l2j1       m_l2j2       mcoll     mcollWW  \\\n",
      "44665  251.747849   66.864952   439.890167  148.482056  261.417145   \n",
      "41650  119.144234  198.510269   288.535004  160.186356  227.625427   \n",
      "41627  449.457184  267.721161   489.645447  370.510345  414.662628   \n",
      "42323  404.527924   77.445404   366.822083  455.225891  514.578857   \n",
      "47697  932.915100  424.201721   327.581604  181.525650  245.019455   \n",
      "...           ...         ...          ...         ...         ...   \n",
      "37641  279.727295  253.511871   382.779144  157.480423  237.671814   \n",
      "37642  578.345276  289.734436   255.228790  156.448380  256.023926   \n",
      "37646  303.556549  361.171722   202.983887   87.547310  183.588165   \n",
      "37648  582.211853  284.859344  1231.855103  691.359314  715.687439   \n",
      "37673  416.159607  630.088013   122.527405  162.339218  242.306015   \n",
      "\n",
      "               mjj         mll        mtw1        mtw2  phi_miss   proxyW_l1  \\\n",
      "44665  1155.018921  131.166321   35.677654  108.385422 -0.415402  110.217148   \n",
      "41650   867.611267  202.946289   68.683624   70.072647  1.223448   69.758156   \n",
      "41627  2215.748291  276.729797  122.554886   71.999779  1.574290  173.690292   \n",
      "42323  1172.855103  208.595016   79.807953  190.456329  0.180060  380.200531   \n",
      "47697  1824.801025  195.226746   48.232533   72.047226  1.115423   64.904007   \n",
      "...            ...         ...         ...         ...       ...         ...   \n",
      "37641  2247.468262  184.066254  127.890533   67.972542 -1.326018  162.811356   \n",
      "37642  1681.108154   92.641579   15.123770   29.503675 -2.011908  120.180771   \n",
      "37646  1711.920898   92.409035   48.871471   53.922359  0.159206   55.157406   \n",
      "37648  2327.915283  666.764709  277.566437   22.370754 -1.437977  281.540588   \n",
      "37673  1685.213257  143.567535   29.521320   40.245060 -1.397366   70.520828   \n",
      "\n",
      "        proxyW_l2     pt_miss        ptll      recoil       yll      zeta  \\\n",
      "44665  108.438026   69.960228   37.443779   90.606445 -0.817088  0.874840   \n",
      "41650   94.078072   21.841217  131.333801  124.742485 -0.430220  0.898682   \n",
      "41627   88.486732   75.708229  172.737946  231.985718 -0.520773  1.172119   \n",
      "42323  204.744919  197.694656  118.177750  310.459534  1.533163  0.583252   \n",
      "47697   81.187431   21.046717  133.743790  132.406891 -0.093900  1.192017   \n",
      "...           ...         ...         ...         ...       ...       ...   \n",
      "37641   70.031944   49.883934   75.594078   50.313580 -0.752336  3.058350   \n",
      "37642   39.270103   38.181824   78.265305  116.193138  1.505924  1.044434   \n",
      "37646   57.448132   32.863087    4.366361   29.291546  0.083296  1.531738   \n",
      "37648  209.350266   69.355751  133.820236   74.164932 -0.489786  2.875488   \n",
      "37673   44.056137   20.176321   97.122978  106.880989 -0.644414  1.298462   \n",
      "\n",
      "       isLL  isMIX  isTT  isBKG  isVBS  notLL  notTT  DNN_SigVsBkg    DNN_LL  \\\n",
      "44665   0.0    0.0   1.0    0.0    1.0    1.0    0.0      0.901813  0.322567   \n",
      "41650   0.0    1.0   0.0    0.0    1.0    1.0    1.0      0.902762  0.268759   \n",
      "41627   0.0    1.0   0.0    0.0    1.0    1.0    1.0      0.994951  0.130585   \n",
      "42323   0.0    1.0   0.0    0.0    1.0    1.0    1.0      0.877414  0.096353   \n",
      "47697   0.0    0.0   1.0    0.0    1.0    1.0    0.0      0.928095  0.460226   \n",
      "...     ...    ...   ...    ...    ...    ...    ...           ...       ...   \n",
      "37641   1.0    0.0   0.0    0.0    1.0    0.0    1.0      0.509756  0.807250   \n",
      "37642   1.0    0.0   0.0    0.0    1.0    0.0    1.0      0.982488  0.766773   \n",
      "37646   1.0    0.0   0.0    0.0    1.0    0.0    1.0      0.992281  0.670426   \n",
      "37648   1.0    0.0   0.0    0.0    1.0    0.0    1.0      0.917321  0.992378   \n",
      "37673   1.0    0.0   0.0    0.0    1.0    0.0    1.0      0.972608  0.810934   \n",
      "\n",
      "         DNN_TT  \n",
      "44665  0.475454  \n",
      "41650  0.298403  \n",
      "41627  0.398932  \n",
      "42323  0.819893  \n",
      "47697  0.683685  \n",
      "...         ...  \n",
      "37641  0.529457  \n",
      "37642  0.213645  \n",
      "37646  0.194817  \n",
      "37648  0.067959  \n",
      "37673  0.102894  \n",
      "\n",
      "[1608 rows x 69 columns]\n",
      "       CleanJet_eta_1  CleanJet_eta_2  CleanJet_phi_1  CleanJet_phi_2  \\\n",
      "7            4.487305       -4.133789       -0.330261       -1.360596   \n",
      "14           2.625488       -2.706055       -0.496765        1.684570   \n",
      "15           2.583984       -2.562988        0.823975       -1.359619   \n",
      "34          -2.309082        3.049316       -2.972656       -0.699829   \n",
      "35          -2.525391        4.080078        3.033691        1.202881   \n",
      "...               ...             ...             ...             ...   \n",
      "16117        2.172852       -1.452393       -1.618164        2.150391   \n",
      "16122        1.209229       -3.539551       -2.887695        1.491699   \n",
      "16125       -2.789551        3.814941       -1.835693        1.411621   \n",
      "16126       -2.604980        1.816650       -0.072327       -2.488770   \n",
      "16131       -2.239746        2.733398       -1.084229       -1.275879   \n",
      "\n",
      "       CleanJet_pt_1  CleanJet_pt_2  Lepton_eta_1  Lepton_eta_2  Lepton_phi_1  \\\n",
      "7          70.209976      28.927614      0.213257      0.876099     -1.307129   \n",
      "14         94.295082      30.142309      1.009033     -0.573120      2.307129   \n",
      "15        100.840218      95.133331      0.580688     -0.225739     -0.531006   \n",
      "34        120.588684      73.183754      0.099243     -0.372986     -1.984619   \n",
      "35         53.810699      52.710476      0.480591     -1.650391     -2.701172   \n",
      "...              ...            ...           ...           ...           ...   \n",
      "16117     305.131165     191.855652      0.457825     -0.248199      2.986816   \n",
      "16122     138.214874      30.915562      0.628784     -1.780762      0.802368   \n",
      "16125      53.157330      34.465637     -1.347168     -0.169373     -1.483643   \n",
      "16126     199.568909      37.331173     -1.425293     -1.649902      2.728027   \n",
      "16131     456.022339      26.130005     -0.998291     -1.638916      1.668457   \n",
      "\n",
      "       Lepton_phi_2  Lepton_pt_1  Lepton_pt_2  MET_significance       Rpt  \\\n",
      "7          2.556641   145.381210    55.317551         61.625000  3.959677   \n",
      "14        -1.626709   153.037750   116.356346          1.350586  6.265029   \n",
      "15         2.911133   185.689270    84.430817         55.312500  1.634261   \n",
      "34         0.841675    49.153587    38.012066         36.062500  0.211717   \n",
      "35        -0.183502    69.344780    57.073471          4.070312  1.395348   \n",
      "...             ...          ...          ...               ...       ...   \n",
      "16117     -0.230438   121.105896    33.076752         57.968750  0.068427   \n",
      "16122     -1.204590   191.111710   111.049286          0.705566  4.966737   \n",
      "16125      0.869629   183.430588    43.796333         20.796875  4.384904   \n",
      "16126     -1.082764   115.511284    27.576643          2.505859  0.427564   \n",
      "16131      2.965332   129.026367    71.955101         88.812500  0.779138   \n",
      "\n",
      "        Zepp_l1   Zepp_l2   Zepp_ll    detajj    detall    dphijj  \\\n",
      "7      0.036499  0.699341  0.367920  8.621094  0.662842  1.030334   \n",
      "14     1.049316 -0.532837  0.258240  5.331543  1.582153  2.181335   \n",
      "15     0.570190 -0.236237  0.166977  5.146973  0.806427  2.183594   \n",
      "34    -0.270874 -0.743103  0.506989  5.358398  0.472229  2.272827   \n",
      "35    -0.296753 -2.427734  1.362244  6.605469  2.130981  1.830811   \n",
      "...         ...       ...       ...       ...       ...       ...   \n",
      "16117  0.097595 -0.608429  0.255417  3.625244  0.706024  2.514631   \n",
      "16122  1.793945 -0.615601  0.589172  4.748779  2.409546  1.903791   \n",
      "16125 -1.859863 -0.682068  1.270966  6.604492  1.177795  3.035871   \n",
      "16126 -1.031128 -1.255737  1.143433  4.421631  0.224609  2.416443   \n",
      "16131 -1.245117 -1.885742  1.565430  4.973145  0.640625  0.191650   \n",
      "\n",
      "       dphilep1jet1  dphilep1jet2  dphilep1jj  dphilep2jet1  dphilep2jet2  \\\n",
      "7          0.976868      0.053467    0.693231      2.886902      2.365949   \n",
      "14         2.803894      0.622559    2.493576      1.129944      2.971906   \n",
      "15         1.354980      0.828613    0.319197      2.087158      2.012433   \n",
      "34         0.988037      1.284790    0.336886      2.468854      1.541504   \n",
      "35         0.548322      2.379133    1.450292      3.065992      1.386383   \n",
      "...             ...           ...         ...           ...           ...   \n",
      "16117      1.678205      0.836426    1.033694      1.387726      2.380829   \n",
      "16122      2.593122      0.689331    2.368892      1.683105      2.696289   \n",
      "16125      0.352051      2.895264    0.542316      2.705322      0.541992   \n",
      "16126      2.800354      1.066388    2.943628      1.010437      1.406006   \n",
      "16131      2.752686      2.944336    2.763018      2.233625      2.041974   \n",
      "\n",
      "       dphilep2jj    dphill  dphilljet  dphilljetjet  dphillmet  dphilmet1  \\\n",
      "7        3.112647  2.419416   1.315347      1.031711   2.656202   2.994681   \n",
      "14       1.440262  2.349347   2.619364      2.929682   1.941311   1.081384   \n",
      "15       3.122941  2.841047   1.588587      0.552804   2.875407   3.109013   \n",
      "34       3.120005  2.826294   1.723973      1.072822   2.594475   2.952775   \n",
      "35       2.315223  2.517670   1.514785      2.416756   1.101123   2.067586   \n",
      "...           ...       ...        ...           ...        ...        ...   \n",
      "16117    2.032236  3.065931   1.649840      1.005330   2.511175   2.482810   \n",
      "16122    1.907336  2.006958   3.080667      2.978288   2.497152   3.106548   \n",
      "16125    2.895587  2.353271   0.552902      0.743167   3.012692   3.069643   \n",
      "16126    0.867163  2.472394   2.980598      3.123872   0.603261   0.783505   \n",
      "16131    2.223292  1.296875   3.094001      3.083668   0.144159   0.292339   \n",
      "\n",
      "       dphilmet2   dr_l1j1   dr_l1j2   dr_l2j1   dr_l2j2      drll  \\\n",
      "7       0.575266  4.384262  4.347375  4.623312  6.359537  2.508572   \n",
      "14      2.852454  3.236472  3.766890  3.392325  3.938779  2.832427   \n",
      "15      0.267967  2.418505  3.251046  3.500110  4.868476  2.953281   \n",
      "34      0.504116  2.603123  3.217704  4.277568  3.753450  2.865473   \n",
      "35      0.450083  6.474919  5.310173  3.334060  5.895789  3.298445   \n",
      "...          ...       ...       ...       ...       ...       ...   \n",
      "16117   0.734444  4.913976  2.085315  2.790568  2.668038  3.146173   \n",
      "16122   1.099590  3.735436  4.224949  3.431164  3.219210  3.135888   \n",
      "16125   0.860271  1.484725  5.918608  3.766179  4.021009  2.631556   \n",
      "16126   1.688890  3.038691  6.142082  1.390380  3.740834  2.482576   \n",
      "16131   1.004536  3.019684  4.753380  4.093890  6.091388  1.446473   \n",
      "\n",
      "               ht      m2ljj30         mT2         mTi      m_l1j1  \\\n",
      "7      450.775726   951.331787  221.455414  385.209747  849.614807   \n",
      "14     463.844635  1135.173584   32.716843  416.160034  321.177582   \n",
      "15     671.367554  1579.380249  161.776794  489.731506  365.556427   \n",
      "34     379.987671  1464.765625   93.999939  168.908554  244.929138   \n",
      "35     298.286957  1813.168945   13.138183  260.973785  263.234589   \n",
      "...           ...          ...         ...         ...         ...   \n",
      "16117  916.813965  1675.256470    2.622233  381.100555  469.512360   \n",
      "16122  524.293213  1253.035767  170.148392  575.539062  328.254852   \n",
      "16125  424.045044  1628.740234  210.948715  406.890961  159.547409   \n",
      "16126  448.569519   958.594177   50.440971  214.039185  354.459351   \n",
      "16131  997.323914   684.885803    1.428777  416.309814  577.127197   \n",
      "\n",
      "            m_l1j2      m_l2j1      m_l2j2       mcoll     mcollWW  \\\n",
      "7       562.643555  389.220886  492.117889  162.965485  269.438080   \n",
      "14      426.751221  510.125183  192.337921  243.520447  313.472626   \n",
      "15      621.965698  388.022125  301.773987  158.995590  274.228577   \n",
      "34      258.951508  199.248657  292.105560  157.447784  232.462051   \n",
      "35      373.090027  121.921265  962.255615  212.153885  300.057770   \n",
      "...            ...         ...         ...         ...         ...   \n",
      "16117   360.227448  333.667450  180.789856  304.040955  349.760468   \n",
      "16122   610.541382  557.021729  163.682785  474.512299  506.091431   \n",
      "16125  1056.306763  191.031982  280.415710  206.291504  262.207489   \n",
      "16126   326.331726  103.291588  180.978439  107.607071  239.943649   \n",
      "16131   384.062195  348.953217  388.258698  467.010712  493.914856   \n",
      "\n",
      "               mjj         mll        mtw1        mtw2  phi_miss   proxyW_l1  \\\n",
      "7      3356.354736  178.376160  322.542633   56.594036  1.981375  324.387878   \n",
      "14      769.077515  339.570007  106.560913  178.632523  1.225745  152.387390   \n",
      "15     1288.881348  268.451935  390.420532   35.172081  2.643166  406.996948   \n",
      "34     1373.403931   87.829185  138.929871   30.606205  1.345791  139.102478   \n",
      "35     1448.581421  200.575775  115.682808   27.255415 -0.633586  120.205841   \n",
      "...            ...         ...         ...         ...       ...         ...   \n",
      "16117  1515.234375  134.466873  339.440857   67.307823  0.504006  349.402161   \n",
      "16122   705.200195  505.989624  201.257584   80.173325 -2.304180  211.289185   \n",
      "16125  1164.848145  199.721497  282.869720   57.674030  1.729900  349.582275   \n",
      "16126   795.055969  107.373032   67.965950   65.024200 -2.771653  153.714600   \n",
      "16131  1304.677246  132.240891   61.041130  150.668976  1.960796  226.370209   \n",
      "\n",
      "        proxyW_l2     pt_miss        ptll      recoil       yll      zeta  \\\n",
      "7      106.494087  179.866974  110.120850   97.160965  0.400171  3.611206   \n",
      "14     186.170563   70.013161  109.314186  106.343262  0.307932  1.616455   \n",
      "15      46.086937  205.273895  107.975914  105.014832  0.321930  2.003296   \n",
      "34      38.296635   99.049591   17.559797   84.548004 -0.107231  1.936096   \n",
      "35     115.875893   65.347534   40.520626   91.154236 -0.508284  0.875000   \n",
      "...           ...         ...         ...         ...       ...       ...   \n",
      "16117   71.234901  265.644501   88.159241  201.256454  0.300860  1.204193   \n",
      "16122  174.861572   53.001797  175.851898  137.225555 -0.350993  0.580444   \n",
      "16125   58.854176  109.195160  155.682205   49.427856 -1.095656  1.442383   \n",
      "16126  103.226105   68.581520   95.428329  156.808319 -1.468761  0.955078   \n",
      "16131  323.392517  340.320129  163.854202  503.025757 -1.230407  0.600830   \n",
      "\n",
      "       isLL  isMIX  isTT  isBKG  isVBS  notLL  notTT  DNN_SigVsBkg    DNN_LL  \\\n",
      "7       0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.840906  0.946401   \n",
      "14      0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.608875  0.650176   \n",
      "15      0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.881009  0.254525   \n",
      "34      0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.903084  0.275881   \n",
      "35      0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.553053  0.786904   \n",
      "...     ...    ...   ...    ...    ...    ...    ...           ...       ...   \n",
      "16117   0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.989525  0.091947   \n",
      "16122   0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.701908  0.233796   \n",
      "16125   0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.552594  0.524979   \n",
      "16126   0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.758194  0.246647   \n",
      "16131   0.0    0.0   0.0    1.0    0.0    0.0    0.0      0.514906  0.424866   \n",
      "\n",
      "         DNN_TT  \n",
      "7      0.175750  \n",
      "14     0.414830  \n",
      "15     0.764352  \n",
      "34     0.544369  \n",
      "35     0.309193  \n",
      "...         ...  \n",
      "16117  0.917150  \n",
      "16122  0.664231  \n",
      "16125  0.834772  \n",
      "16126  0.521129  \n",
      "16131  0.493744  \n",
      "\n",
      "[2548 rows x 69 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_257/3038950175.py:3: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_val_LL['DNN_LL'] = dnnLL.predict_proba(X_val_LL)[...,0].numpy().tolist()\n",
      "/tmp/ipykernel_257/3038950175.py:4: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_val_LL['DNN_TT'] = dnnTT.predict_proba(X_val_LL)[...,0].numpy().tolist()\n",
      "/tmp/ipykernel_257/3038950175.py:7: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_val_bkg['DNN_LL'] = dnnLL.predict_proba(X_val_bkg)[...,0].numpy().tolist()\n",
      "/tmp/ipykernel_257/3038950175.py:8: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_val_bkg['DNN_TT'] = dnnTT.predict_proba(X_val_bkg)[...,0].numpy().tolist()\n"
     ]
    }
   ],
   "source": [
    "print(len(dnnLL.predict_proba(X_val_LL)[...,0]))\n",
    "print(len(dnnTT.predict_proba(X_val_LL)[...,0]))\n",
    "df_val_LL['DNN_LL'] = dnnLL.predict_proba(X_val_LL)[...,0].numpy().tolist()\n",
    "df_val_LL['DNN_TT'] = dnnTT.predict_proba(X_val_LL)[...,0].numpy().tolist()\n",
    "df_val_bkg = df[(df['DNN_SigVsBkg'] > 0.5) & (df['isBKG'] == 1)]\n",
    "X_val_bkg = df_val_bkg.values[:,0:NDIM]\n",
    "df_val_bkg['DNN_LL'] = dnnLL.predict_proba(X_val_bkg)[...,0].numpy().tolist()\n",
    "df_val_bkg['DNN_TT'] = dnnTT.predict_proba(X_val_bkg)[...,0].numpy().tolist()\n",
    "print(df_val_LL)\n",
    "print(df_val_bkg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "id": "a93cc097-ea4a-4f84-a256-9703427ff3a8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "LL = df_val_LL[df_val_LL['isLL'] == 1]\n",
    "mix = df_val_LL[df_val_LL['isMIX'] == 1]\n",
    "TT = df_val_LL[df_val_LL['isTT'] == 1]\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "h_isLL = ax.hist2d(LL['DNN_LL'].values, LL['DNN_TT'].values, density=True)\n",
    "fig.colorbar(h_isLL[3], ax=ax)\n",
    "plt.ylabel('DNN TT probability')\n",
    "plt.xlabel('DNN LL probability')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 233,
   "id": "1f0cd41d-696b-4645-8343-fff9ea287e41",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "h_isTT = ax.hist2d(TT['DNN_LL'].values, TT['DNN_TT'].values, density=True)\n",
    "fig.colorbar(h_isTT[3], ax=ax)\n",
    "plt.ylabel('DNN TT probability')\n",
    "plt.xlabel('DNN LL probability')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 234,
   "id": "6afe52fe-8b78-4b8b-bb16-5ad97e7bdf04",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "h_isMIX = ax.hist2d(mix['DNN_LL'].values, mix['DNN_TT'].values, density=True)\n",
    "fig.colorbar(h_isMIX[3], ax=ax)\n",
    "plt.ylabel('DNN TT probability')\n",
    "plt.xlabel('DNN LL probability')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 235,
   "id": "72de65a9-e950-497a-9b1b-f190ff3c6bf8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "h_isBKG = ax.hist2d(df_val_bkg['DNN_LL'].values, df_val_bkg['DNN_TT'].values, density=True)\n",
    "fig.colorbar(h_isBKG[3], ax=ax)\n",
    "plt.ylabel('DNN TT probability')\n",
    "plt.xlabel('DNN LL probability')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 236,
   "id": "ff3b2c7c-47e9-4261-878a-e5b2f6dc9404",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_257/3595125263.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_val_TT['DNN_LL'] = dnnLL.predict_proba(X_val_TT)[...,0].numpy().tolist()\n",
      "/tmp/ipykernel_257/3595125263.py:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df_val_TT['DNN_TT'] = dnnTT.predict_proba(X_val_TT)[...,0].numpy().tolist()\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_val_TT['DNN_LL'] = dnnLL.predict_proba(X_val_TT)[...,0].numpy().tolist()\n",
    "df_val_TT['DNN_TT'] = dnnTT.predict_proba(X_val_TT)[...,0].numpy().tolist()\n",
    "\n",
    "LL = df_val_TT[df_val_TT['isLL'] == 1]\n",
    "mix = df_val_TT[df_val_TT['isMIX'] == 1]\n",
    "TT = df_val_TT[df_val_TT['isTT'] == 1]\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "h_isLL = ax.hist2d(LL['DNN_LL'].values, LL['DNN_TT'].values, density=True)\n",
    "fig.colorbar(h_isLL[3], ax=ax)\n",
    "plt.ylabel('DNN TT probability')\n",
    "plt.xlabel('DNN LL probability')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "id": "c09e08d4-55cb-4265-ba47-5d5ef43c2d47",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "h_isTT = ax.hist2d(TT['DNN_LL'].values, TT['DNN_TT'].values, density=True)\n",
    "fig.colorbar(h_isTT[3], ax=ax)\n",
    "plt.ylabel('DNN TT probability')\n",
    "plt.xlabel('DNN LL probability')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 238,
   "id": "bac8cc9a-7ec8-47be-97e2-7b4262e3433d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAu0AAAIhCAYAAAD6lISoAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/MnkTPAAAACXBIWXMAABcSAAAXEgFnn9JSAAA+0klEQVR4nO3deZgkVZWw8fewrzYoqOytraIwICrYKCowgyiKoogLzKggOjqCy4DLIOIgKi4o6igzroDOgjsyDC6orAoqi4ofKiDIooCC0kADDdh1vj8iUooyszsrI6IzovL9PU880RnLvTcrafrkqXNvRGYiSZIkqb1WGvcAJEmSJC2bQbskSZLUcgbtkiRJUssZtEuSJEktZ9AuSZIktZxBuyRJktRyBu2SJElSyxm0S5IkSS1n0C5JkiS1nEG7JEmS1HIG7ZIkSVLLGbRLkiRJLbfKuAcwKSLiRmAt4Lpxj0WSJGk5NgPuzMyHjnsgEfG/wIIGu7gyM5/bYPu1MGhfcdaKlVZed821Nthq3ANpSiy+a9xDUBUrzeFfvE1NjXsEktQpd7KYpDX/71yw2mqx1SPmr1p7w7+++l7uuSdrb7cJBu0rznVrrrXBVtvvcui4x9GY1U778biHoApWnrfeuIfQmKWLFo17CJLUKefn6dzBba2pDnjE/FX5+dlb1N7uNjtfwy8uv6f2dptg0C5JkqSWS6Yayfx3I8sOTkSVJEmSWs9MuyRJklotgaVZf6a9O3l2g3ZJkiR1wFSnQuz6WR4jSZIktZyZdkmSJLVaQiMTUbuUuzfTLkmSJLWcmXZJkiS1XLI0m8iLdyfXbqZdkiRJajkz7ZIkSWq9SV89xqBdkiRJrZbA0gaC9i59DbA8RpIkSWo5M+2SJElqvUkvjzHTLkmSJLWcmXZJkiS1WkIjSz52KXdvpl2SJElqOTPtkiRJar2pcQ9gzMy0S5IkSS1npl2SJEmt5jrtBu2SJEnqgKVdirAbYHmMJEmS1HJm2iVJktR6TkSVJEmS1Gpm2iVJktRqxUTUaKTdrjDTLkmSJLWcmXZJkiS1W8JUE2nxDqXaDdolSZLUapbHWB4jSZIktZ6ZdkmSJLVeE5n2LjHTLkmSJLWcmXZJkiS1WgJTaU27JEmSpBYz0y5JkqTWm/SadoN2SZIktVoSLG2gQCQ79EXA8hhJkiSp5cy0S5IkqfWamIjaJWbaJUmSpJYz0y5JkqRWS5qZiOqSj5IkSZJqY6ZdkiRJrbc0JzvXPNnvXpIkSeoAM+2SJElquWCqkVxzd1akMWiXJElSqzkR1aB9hZpaBRY/dO7+yNd/6uPGPYRGrXLrknEPoVFLL/nluIfQmFXmbzHuITQqF9067iE0aumiReMegtTXyuutN+4hNOfWlWBq3IPQdHM3gpQkSdKc4URUSZIkSa1m0C5JkqRWS2CKqH2rq6Y9Is6KiFzG9syqfVgeI0mSJNXjq8DiPsd/V7Vhg3ZJkiS1XLC0G0s+vikzr667UTBolyRJUsslzUxE7dKSj9a0S5IkSS1npl2SJEmt18wTUWt3YEQ8iGKV+8uBr2fmtXU0bNAuSZKkSbYgIi7tdyIzt55lW2+f8fqDEfGuzHzXaEO7Tye+skiSJGlyZQZLG9gya5uIeg7wUmABsBawJXA48GfgqIh4Q9UOzLRLkiRpkl05Qkb9fjLzHTMOXQ4cHREXAt8GjoyIT2XmXaP2YaZdkiRJrbeUlWrfmpaZpwMXAusBC6u0ZaZdkiRJrTfVwJKPK8gVwPbARlUa6ey7lyRJkjpg/XJ/R5VGzLRLkiSp1bKhJ6Jm/U9EvZ+I2BB4avny4iptmWmXJEmSRhQRT46I50XEyjOOzwdOBtYG/jczf1ulHzPtkiRJar2l9S3PWLdHAScAN0bExcAiYAvgCcAawKXAq6p2YtAuSZIkje5HwH9QrA6zA0UN+x3AT4EvA/9RZanHHoN2SZIktVoCU43UtNfQRuYvgdfW0NQyWdMuSZIktZyZdkmSJLXe0u6u014Lg3ZJkiS1WhJMNbA8Y9NLPtZpsr+ySJIkSR1gpl2SJEmtN+nlMZP97iVJkqQOMNMuSZKkVktgaUuXfFxROp1pj4g1I+KoiLg8IpZExPURcXxEbDJCW0+PiNMi4qaIuDci/hgRp0fE85sYuyRJkjSszmbaI2IN4AxgR+AG4BRgPnAAsGdE7JiZVw3Z1huBD1N84TofuA7YDNgNeHpEHJ2Zh9f9HiRJkjSEhKlsYKWXDqXaOxu0A2+nCNjPB3bPzMUAEXEI8CHgeGCX5TUSERsC7wPuBZ6emWdPO/c04HTgsIj47LBfAiRJklSnaKQ8Bpd8bFZErAYcXL48qBewA2TmscAlwM4R8YQhmlsIrA6cMT1gL9s6B/g2xSe6fR1jlyRJkmarq5n2nYB5wJWZ+ZM+578CbAs8B7hoOW3dPWSffxx+eJIkSapLAlMNLPnYoeqYbmbagceW+4sHnO8d33aItn4MLAL+NiJ2nn6iLI95BnAFcO7shylJkiRV19VM++bl/rcDzveOb7G8hjLz1og4EPgf4MyIOK+8f1PgycAPgJdl5j3DDCwiLh1wasEw90uSJOmvLe1Q/XkTuhq0r1Pu7xxw/o5yv+4wjWXm1yJiD+BLFKU3PbdRTET93SiDlCRJkurQ1aC9VhFxKPAB4OvAkcBVwMOBo8ptIbDnMG1l5tYD+rgU2Kr6aCVJkiZLEg3VtHcne9/VoL23WsxaA86vXe5vX15DEbEL8EGKOvgXZuZUeernEbEPcCHw7IjYIzO/OfKIJUmSNLJJL4/p6kTUa8v9pgPO945fM0RbLy33J08L2AHIzKXA18qXT5vVCCVJkqSadDXT/rNy//gB53vHLxmirV6Af+uA873j6w/RliRJkhrQRHlMl3T13f+AIpheEBHb9Tm/T7k/dYi2biz3gx6etEO5v3rYwUmSJEl16mTQXi6/+PHy5XER0athJyIOoVif/ezMvGja8YMj4lcR8d4ZzX293P99RNxvsmlE7AXsB0wBJ9f7LiRJkjSMBJbmSrVvXXq4UlfLYwDeDexGsZb6FRFxLsW67AuBm4BXzLh+A2BLYKMZx78OfBl4IXBqRFwI/AZ4GPdl3w/PzMsaeA+SJEnScnUy0w6QmUuAXYF3UazX/jyKoP1E4PGZedWQ7STwYuBA4BzgEcDzgfnAN4A9MvPoekcvSZKk4QVTDWx0aEWaLmfaycy7gHeU2/KuPZJiDfZ+5xI4vtwkSZKkVul00C5JkqS5r1fT3kS7XWHQLkmSpHZLmMoGSlk6FLUbtK9AK9+1lA0vuGXcw2jM9bvO7aXsNz5zybiH0Kh86uPGPYTGLNp49XEPoVHrfHGY58hJqtvSRYvGPYTm3P95k2oBg3ZJkiS1WgJLG1g/pUOJ9u6uHiNJkiRNCjPtkiRJarlopqa9Q0s+mmmXJEmSWs5MuyRJklpvasJzzQbtkiRJarVinfb6S1mciCpJkiSpNmbaJUmS1HrNTETtDjPtkiRJUsuZaZckSVLLBVPZRK65O9l7M+2SJElSy5lplyRJUqslsLSBrHiXVo8xaJckSVLrORFVkiRJUquZaZckSVKrZdLIRNTsUH2MmXZJkiSp5cy0S5IkqfWmOrQ8YxPMtEuSJEktZ6ZdkiRJLRcsbWT1mO5k7820S5IkSS1npl2SJEmtljS0ekztLTbHoF2SJEmt58OVJEmSJLWamXZJkiS1nks+SpIkSWo1M+2SJElqtWIiav2Z9i5NRDXTLkmSJLWcmXZJkiS1XDSy5GOXHq5k0C5JkqR2y4aWfOxQfYzlMZIkSVLLmWmXJElSqyXNLPnYoUS7mXZJkiSp7cy0S5IkqfUaqWnvEDPtkiRJUsuZaZckSVLrTXqm3aBdkiRJrZZEQ09E7c4XActjJEmSpJpExIMi4g8RkRHx67raNdMuSZKk1utQecyHgA3qbrS2THtE7FxXW5IkSVLXRMTfAS8HPl1323WWx5wZEb+MiEMjovZvF5IkSZpcU0TtW50iYk3gk8AvgA/W2jj1Bu3fBR4FfAD4bUR8ofy2IUmSJM11/wo8HHgNcG/djdcWtGfm7sAC4L3ATcCLgNMj4sqI+JeIeGhdfUmSJGlyJEVNe91b1jS+iNgWOBQ4ITPPranZ+6l19ZjMvDoz3w5sAewFnAZsDhwNXBsRX4uIZ0ZEZ2YSSJIkaU5bEBGX9tuGuTkiVgI+AywC3tLUIBtZ8jEzpzLz1Mx8LkXQfgRwHfcF8ldHxBFm3yVJkrRc2UymvaZU++uAHYA3Z+Yfa2mxj8aXfMzMGyLiQmB74GHl4c2AdwJvi4jjgMMys/ban7ZZuubK3LTD+uMeRmMWP/GucQ+hUdftMrcfa3DvpWuOewiNWeWOcY+gWevN32LcQ2jUooUbjXsIjZr37V+NewiNWrpo0biHoDmiwSUfr8zMrUe5MSI2B94NnJ2ZJ9Y6qhkaC9ojYhPgFeW2ORDAxcAngDOAF1N8M/lnYFXgDU2NRZIkSWrAccBqFJNPG1Vr0F7W9OwJvAp4JrAycCdwAvCJzLxw2uXvLbPsP6EI4A3aJUmS9FeSaCTTntWXfdyTopb9EzOmbK5R7jeJiLPKP78kM28ctaPagvaIeDdwAPBQiqz6pRRrVX4+M2/rd09m3hYR5wIvrWsckiRJ0gq0HjDoIaNrTDu3xoBrhlJnpv1twN3ASRRZ9e8Ped9Z1DUNQJIkSXNSNlfTPrIcMKiImA/8hqJe/hF19FVn0P5m4MTZzpoti/ZPrHEckiRJ0pxSZ9B+AfAgYJlBe0Q8EtgoM8+psW9JkiTNYVPV6887rc417M4E3jrEdW8pr5UkSZKWq+1PRF0R6sy0R7kNc50kSZI0J2Xm1dQc8zb+cKU+NgYWj6FfSZIkdVQbJ6KuSJWC9oh42YxDj+hzbHpfWwK7AT+s0q8kSZI0Sapm2k/kvuUaE9ip3AYJYAlwVMV+JUmSNCmSRh6u1KWi9qpB+1EUbzeAdwA/BU4ZcO09wPXA6Zl5Q8V+JUmSpIlRKWjPzCN7f46I/YHvZuY7K45JkiRJmiYaqmnvTp18bRNRM3N+XW1JkiRJPb0lH5totyvqXKddkiRJUgNGzrRHxPEUX1Delpm/L18PKzPzwFH7liRJ0mTJLqXFG1ClPGZ/iqD9/cDvy9fDSsCgXZIkSRpClaB913J/7YzXkiRJUq2mOjRptAkjB+2ZefayXkuSJEmqR22rx0iSJElNaWbJx+5w9RhJkiSp5aqsHnNGhX4zM/+uwv2SJEmaEK7TXq08ZpcK93bpZyRJkqRxyoaWfOxQRFolaH9YbaOQJEmSNFCV1WOuqXMgkiRJ0iBORJUkSZLUai75KEmSpJaLhjLt3cneV1k95iqK8v3dMvM35ethZWYuGLVvSZIkaZJUybTPL/erzngtSZIk1aqJJR+7pMpE1JWW9VqSJEmqQ9LMko8dWvHRiaiSJElS23U6aI+INSPiqIi4PCKWRMT1EXF8RGwyYnvzI+ITEfGbiLg7Im6OiPMj4s11j12SJEnDy4zaty6pffWYiNgIOAB4KrBxefh64FzgxMy8vqZ+1gDOAHYEbgBOoairPwDYMyJ2zMyhJ8dGxB7AV4A1gYuBHwIPArYBXg0cU8e4JUmSpNmqNWiPiJcDHwfW4v5r6GwD7A68LSJel5kn1NDd2ykC9vOB3TNzcTmGQ4APAccDuww57kcDXwNuB56emedNO7cS8PgaxitJkqRRZEMPV+pQUXtt5TFlpvp4YA3gS8DzgO3KbS/gi8DqwGfKa6v0tRpwcPnyoF7ADpCZxwKXADtHxBOGbPLYctz7Tw/Yy/amMvPCKuOVJEmSqqizpv1wiu8re2Xmvpn5v5l5Sbmdmpn7Ac8tr31bxb52AuYBV2bmT/qc/0q5f87yGoqIzYBnAFdl5jcqjkuSJEkNyAa2LqmzPGY74NxlBb6Z+c2IOAfYoWJfjy33Fw843zu+7RBt7ULx5eW8iFgF2JviS8HKwP8DvpiZt4w+VEmSJFXVtYmjdaszaF9CMeF0eW4or61i83L/2wHne8e3GKKtrcr9YorJsjvOOP+eiNgnM88cZmARcemAUz4BVpIkSSOpszzmbGBhOXGzr4hYGVgInFOxr3XK/Z0Dzt9R7tcdoq31y/0rgUcD+wEPBLYE/qv888mjLiMpSZKkGkx4fUydQfthFEskfiYi5s08GREPAD5FESQfVmO/VfV+BqsAr87MkzLzlsy8PDNfClxAUT//2mEay8yt+23Alc0MX5IkSXPdyOUxEfGOPodPBfYHXhARpwPXlMe3oFjycR3gv4EXAe8atW+KUhYolpbsZ+1yf/ss2loMfLnP+RMoavB3Hnp0kiRJqpU17aM7kuIXC/1+gusCLxhw3z+U91UJ2q8t95sOON87fs2A89P1rrk2M/v9ouTqcv/g4YYmSZIk1atK0H5AbaOYvZ+V+0EPPeodv2SItnpLRq4/4PwDy/3iAeclSZLUoAT6plZraLcrRg7aM/NzdQ5kln4A3AosiIjtMvOnM87vU+5PHaKt84A/Ag+NiC0z87IZ53tlMf3Wg5ckSZIaV+eSjytMZt4TER+neKDTcRGxe2beARARh1Csz352Zl7UuyciDqZ4iurJmXnYtLb+HBHHAu8p29o7M28r79mNokY/gU9WHffKdy1lwwvm8pLvg35ZMTdc8K4Txz2ERr1ww93GPYTG/OqmuV3dduXac3txq82/fde4h9CoWO+v1m6YU1Ye9wA0mltXgqlxD2K6aKimvTt18p0M2kvvBnYDngxcERHnUkx4XQjcBLxixvUbUCzjuFGfto4Bdi3buzwiflhevyPF/28Oz8wfN/EmJEmStBwJNBG0d6g+ptagPSIC+HtgL+CRFBNS+/2EMzMrPWwoM5dExK4Uy0fuBzwP+BNwInBEZg568FK/tu6NiGcB/wy8DHgGcA/F2vMfzsz/qzJWSZIkqYragvaIWA04DfhbBv+uYdBqMyPJzLuAd5Tb8q49kmLFm0Hn7wU+UG6SJElqkSYmonZJnQ9XOhT4O+D/KLLs/0kRpK8OPIYiYL4DOCYz6+xXkiRJmtPqLI95MUV5yn6ZeUdETMFfMtiXAUdFxJnAmRFxWWYeX2PfkiRJmsvMtNfmEcCPe6u4UM45joi/TBzPzHMplmt8bY39SpIkSXNanUH7Uoq103t6wfuGM677HcUqLpIkSdJQMqP2rUvqDNp/B2w67fWvy/2OM67bFp8uKkmSpNnIBrYOqTNo/yHwNxGxevn6G+X+IxHxzIjYJiI+RjEp9Uc19itJkiTNaXUG7V8FlgC7A2Tmr4GPAJtTLAX5U+Ag4E7gLTX2K0mSpDlu0stjals9JjNPY8bTRjPz0Ii4gOLBR+sDlwP/lplX1NWvJEmSNNfV+kTUfjLzC8AXmu5HkiRJc1RTNegdqmv3IUeSJElSy9WeaY+IjYADgKcCG5eHrwe+D5yQmdfX3ackSZLmum7VoNet1qA9Il4OfBxYi/v/ZLehmKB6WES8LjNPqLNfSZIkaS6rLWiPiD2A4ymehPol4CTg6vL0FsC+wAuBz0TEjZn5zbr6liRJ0hzXofrzJtSZaT+c4se5V2Z+Y8a5S4BTI+I/gf8D3gYYtEuSJGk4Ex601zkRdTvg3D4B+1+U2fVzgMfV2K8kSZI0p9WZaV9CMeF0eW4or5UkSZKG07GHIdWtzkz72cDCiBjYZkSsDCykyLZLkiRJGkKdQfthwIMoJprOm3kyIh4AfIriyaiH1divJEmS5rjM+rcuGbk8JiLe0efwqcD+wAsi4nTgmvL4FhRLPq4D/DfwIuBdo/YtSZIkTZIqNe1HUszj7VdgtC7wggH3/UN5n0G7JEmShtOxzHjdqgTtB9Q2CkmSJGmQjGYmonZocuvIQXtmfq7OgUiSJEnqr86JqJIkSVIjIuvfahtbxCER8bWIuCIibo2IuyPimoj4fERsU0cfda7T/hcR8STgqcAm5aHfUTx46fwm+pMkSZLG6G3A2sAlwM/LY1sDLwVeEhF7Z+b/Vemg1qA9Ih4F/Cewfe9Quc/y/IXAP2TmFXX2K0mSpDmu3RNR9wIuysz7PUA0Il4LHEexJPqmmfnnUTuoLWiPiI0oHrD0EIono34ZuJriRzwfeCGwA3BWRGyfmTfU1bckSZI0Lpn5gwHH/z0iDgEWAFtRZOJHUmem/e0UAfuHgcMy857pJyPircB7gUMofoXwuhr7liRJ0lzWoZVeZri33N+zzKuWo86JqM8CLsvMQ2cG7ACZeS/wZuAyYM8a+5UkSdJclw1sDYuIlwJbAleU28jqzLRvBHx1WRdkZkbExQx+8JIkSZK0Ii2IiEv7ncjMrWfTUES8mWIC6trAY8o/Xw/sm5lLqwyyzqD9NmCzIa7brLxWkiRJGk67J6L2PAP4u2mvrwFelpkXVW24zvKY84GdIuLZgy6IiGcBOwHn1divJEmSNKorM3PrfttsG8rM3TIzgPWBp1GUxJwdEYdXHWSdQfv7gCng5Ij4z4jYIyIeU27PjIjPASeX17yvxn4lSZI0lzVRz95gXXtmLsrMcynmfF4EvCsidqjSZm3lMZl5fkQcAHwS+HtgvxmXBHAX8OrM/GFd/UqSJEltlJn3RsQXgScAzwEuGLWtWh+ulJn/FRFnAa8CngJsXJ66HjgX+GxmXldnn5IkSZoA3V3y8eZyv2GVRup8uNLrgTsz8zPAv9bVriRJktRhO5f7K6s0UmdN+4co0v6SJElSrSLr32oZV8RO5fzNlWYcXzUiXge8lKJE/ItV+qmzPOZGYEmN7UmSJEmF9i75+EjgBODmiLgI+COwAbANxXOMlgD7Vy0RrzNo/zawR0Ss1u+JqJIkSdIcdDZwNEUZzLYUAfs9wNXAV4B/y8xfV+2kzqD9cGB34L8j4vWZeUONbUuSJEmtk5m/oYiDG1Vn0P5e4GfA3sCzI+Ji4Fr6l8xkZh5YY9+SJEnSnFVn0L7/tD+vATy53PpJwKBdkiRJQ6lr4mhX1Rm071pjW5IkSZJKdT4R9ey62pqrYiqJ2+4c9zAatP64B9CoLyye2+/vjRufPu4hNOYj7D7uITTqV1uPewTNuuU388Y9hEY9eI4/cnDpokXjHoJGkVPjHsFf6+7DlWpR6xNRJUmSpNolzSz52KGSm9qD9ohYHXgB8FRg4/Lw9cD3ga9mpmu5S5IkSbNQa9AeEbsBJ1IsJD/zdxj/CHwgIvbPzO/U2a8kSZLmuA5lxZtQW9AeEQuB/wNWA34EnESxqDzAFsC+wI7AqRGxc2b+qK6+JUmSpLmszkz7u4BVgX/KzE/2Of+xiPhH4BPAUcAzauxbkiRJc9ikL/m4Uo1tLQQuHBCwA5CZnwIuoMi4S5IkSRpCnUH7FPDrIa77NRNflSRJkqRZyQa2DqmzPObHwLZDXLdtea0kSZI0nI4F2XWrM9N+BPDIiHhnRPxVu1F4J/DI8lpJkiRJQ6gz0/5o4HPA24GXRsRXgWvKc1sAewPzgU8DW0bEltNvzszP1zgWSZIkzSGTPhG1zqD9RIpfXARFcH4o9/0iY/qa7a8qN6adS8CgXZIkSeqjzqD9KCa+2kiSJEm1SyBnPrezpnY7oragPTOPrKstSZIkSfepM9MuSZIkNaNDWfEm1Ll6jCRJkqQGmGmXJElSqwXNrB7TQJV8YwzaJUmS1H6Wx0iSJElqMzPtkiRJar1Jf7iSmXZJkiSp5UYO2iNiaUR8ts7BSJIkSX1lA1uHVMm0B92adCtJkiR1kjXtkiRJaremMuMdyrYbtEuSJKn1nIgqSZIkqdWqZtqfEhHHj3BfZuaBFfuWJEmSJkLVoH0B8IgR7kvAoF2SJEkaQtWg/WfAKXUMRJIkSRpowmvaqwbtP83Md9YyEkmSJEl9uXqMJEmSWm/SV48xaJckSVL7TXjQ7pKPkiRJUstVybRfAzylroFIkiRJA5lpH9kWwPfrGogkSZKk/qxplyRJUrtlQxNRO5S9t6ZdkiRJajkz7ZIkSWq/DmXFm1A1aF8nIjYf5cbMvLZi30TEmsBhwEuAzYE/Ad8CjsjM31Vo95HAJcAawPcyc7eqY5UkSZJGVTVof0G5zVZW7Tsi1gDOAHYEbgBOAeYDBwB7RsSOmXnViM1/Cli9yvgkSZJUn0l/uFLVmvYYcaujlv7tFAH7+cCjMvPFmbkQOBTYEDh+lEYj4kBgF+DTNYxRkiRJdcgGtg6pGjyfmJkrjbJV6TQiVgMOLl8elJmLe+cy81iK0padI+IJs2z3IcAxwHeAk6qMUZIkSapLV1eP2QmYB1yZmT/pc/4r5f45s2z3o8CawGsrjE2SJEl1M9PeSY8t9xcPON87vu2wDUbEs4AXA0dn5q8rjE2SJEmqVVeXfOytWPPbAed7x7cYprGIWBv4d+Ay4P1VBhYRlw44taBKu5IkSZMqaGYiatTfZGO6GrSvU+7vHHD+jnK/7pDtvZsiwN81M++pMrBlyaVLyUW3NtX82K1/+QPHPYRGvf0bLxr3EBq18Va/H/cQGnP9TeuNewjNumluL3aVDx73CFTFyuutN+4hNGbpokXjHoImSJWg/QCg82UkEbE98Hrg85l5VtX2MnPrAf1cCmxVtX1JkqSJ01QNeofq2kcO2jPzc3UOZJZ6q8WsNeD82uX+9mU1EhGrUCztuAh4Uy0jkyRJUv06FGA3oavlMb2nqW464Hzv+DXLaWdTYDvgRuDLEferbFqv3D8hIs4CyMxdZjdMSZIkqbquBu0/K/ePH3C+d/ySIdt7aLn1sx6w85DtSJIkqQE+EbWbfgDcCiyIiO36nN+n3J+6rEYy8+rMjH4bsGt52femHZMkSZJWuE4G7eUKLx8vXx5XLtkIQEQcQrE++9mZedG04wdHxK8i4r0rdrSSJEmqbMIfrtTV8hgolmncDXgycEVEnEuxbONC4CbgFTOu3wDYEthoRQ5SkiRJqqqTmXaAzFxCUcLyLor12p9HEbSfCDw+M68a2+AkSZJUq8j6ty7pcqadzLwLeEe5Le/aI4EjZ9H2WXTrQVmSJElzV8eC7LqNHLRHxNIK/WZmdvoLgyRJkrSiVAmc72X233lWpcMlOZIkSRoDn4ha6Ymoawx7bURsBLwNeCWwOsVyjZIkSZKG0GjWOyIeHBEfBn4NvBa4m2Li6MOa7FeSJElzSzSwdUkjdeURsQHwVuCfgLWA24H3Ah/KzFua6FOSJEmaq2oN2iPigcBbgIOAtYE7gPcDx2Tmn+rsS5IkSROkpfXnEbEWsDvwHOApFEuQL6WoNPkqcGxmLq7aTy1Be0SsB7wZOBhYl2Ld9GMogvWb6+hDkiRJaqH9gE+Xf/4l8L/AAygeAPpOYN+I2Dkz/1Clk0pBe0TMAw4B3lAO7i7gWOD9mXlTlbYlSZKknhY/DOle4FPARzLzl72D5UIspwGPAz5CEdyPrMo67e8A3gjMo5hg+hGKYP33VQYkSZIk/ZWWBu2Z+Tngc32O3xARBwHnAXtHxGqZec+o/VTJtB9J8eO7CzgRuBk4MGK4ubiZeXSFviVJkqS2+1m5Xx14EHDDqA1VrWkPitVh/mnaa1j2d6Eozxu0S5IkaTgtzbQvx8PL/b1ApUVZqgTt76zSsSRJktQCCyLi0n4nMnPrim2/odx/KzPvrtJQlSeiGrRLkiRphWjxRNS+IuJZwIEUWfYjqrbXyMOVJEmSpI64soaM+v1ExKOB/6IoC39zZv5sObcsl0G7JEmS2i1ppqa9gTYjYhPgW8D6FA9W+mgd7VZd8nFkmXlUlfslSZI0GYJmymOGW/NwFu1FPBA4neKpqCcAb6qr7TqWfJzN+53+4zZolyRJ0pwQEesA3wS2Ar4GvCoza/uqUSVoP2yW129CUYy/Jl1dtEeSJEnj0eLoMSJWB04Bngh8G9g3M5fW2UeV1WPeP8x1EfEQ4G3AK4E1gNsonp4qSZIkdVpErAycBPwtcC6wd5Unnw7S2ETUiNgQ+BfgNRTB+mLgQ8CHMnNRU/1KkiRp7mnxko8HA88v/3wz8O8RfavH35SZN4/aSe1Be0Q8CHgrxVNS16YI1t9HEaxXehKUJEmS1DLrT/vz8wdeVcwHHX/QHhHrA2+m+LaxDnAH8AHgmMz8Y139SJIkaQK1NNOemUdSBOSNqhy0R8Q8iuVsXgc8ALiTogzm/VV+BSBJkiT9RUuD9hWlyjrtDwAOAd4AzAPuAj5MEaz/oZ7hSZIkSaqSab+GIrO+BPgo8L7M/H0to5IkSZJ6sqGJqB3K3lcJ2udRvNXVgYOAgwbMlO0nM3P1Cn1LkiRJE6NqTXuU20o1jEWSJEnqr0NZ8SZUebiSgbokSZK0AjT2cCVJkiSpLpGTnWo3Wy5JkiS1XJ0PV3oysCvwGIonQyXwJ+AXwJmZ+aO6+pIkSdKEmexEey0PV9oWOB54XO/QjEuyvO7HwIGZ+YuqfUqSJGmyNLLkY4dUCtojYgfgDGBt4A7gm8BPgZspgvcNKIL5ZwALgfMjYpfM/EmVfiVJkqRJUuWJqCsD/00RsH8WODQzbxtw7QOAY4FXAP8TEVtlTvhsAkmSJA1vwiPHKhNR9wIeAXwxM181KGAHyMzbMvOVwJeBRwHPqdCvJEmSNFGqlMc8B5gC3jaLew4D9gGeB/xvhb47KVZemVhv3riHoRFt9p2pcQ+hUbf8ZqNxD6Ex64x7AA1bsuG4R9CsLU763biHoCo2n7v/b2HRonGPYHJkQzXtHcreV8m0PwG4LDN/M+wNmXkV8KvyXkmSJElDqBK0bwRcPsJ9lwMbV+hXkiRJkyYb2DqkSnnMPODWEe67DXhAhX4lSZI0QYJmymNmrlPeZlUy7atQ1LTP1hQ1PtRJkiRJmusMniVJktR+HStnqVvVoP3lEfHyWkYiSZIkqa+qQfuopUAT/l1JkiRJs9HIko8dMnLQnplV6uElSZIkDcmadkmSJLVfTnaq3aBdkiRJ7eYTUSst+ShJkiRpBTDTLkmSpPbrUFa8CWbaJUmSpJYz0y5JkqTWi6lxj2C8zLRLkiRJLWemXZIkSe1nTbskSZKkNjPTLkmSpNZrZJ32DjFolyRJUrslzTwRtUNfBCyPkSRJklrOTLskSZJab9LLY8y0S5IkSS1npl2SJEntZ6ZdkiRJUpuZaZckSVKrBc3UtEf9TTbGoF2SJEnt18SSjx1ieYwkSZLUcmbaJUmS1Hou+ShJkiSp1cy0S5Ikqd2SZpZ87FD23ky7JEmS1HJm2iVJktR6k17TbtAuSZKklkuYmuz6GMtjJEmSpJYz0y5JkqT2605SvBFm2iVJkqSWM9MuSZKk1pv0iahm2iVJkqSWM9MuSZKkdksgG0i1dyh7b6ZdkiRJajkz7ZIkSWq9Sa9pN2iXJElS+0140G55jCRJktRynQ7aI2LNiDgqIi6PiCURcX1EHB8Rm8yijfUiYr+IOCkifhMR90TE7RHxo4h4Q0Ss2uR7kCRJ0vJFZu1bl3S2PCYi1gDOAHYEbgBOAeYDBwB7RsSOmXnVEE29CTic4pcuPwV+BGwI7AQ8EdgnIp6RmXdWHXMuXUouurVqM621yhx+bwCrrjdv3ENo1Kq3PXDcQ9CI7th49XEPoVGXvXGjcQ+hUQ//6t3jHkKjVr3uT+MegjQndDnT/naKgP184FGZ+eLMXAgcShF0Hz9kO3cAHwDmZ+bjM/Mlmfl3wDbAtcBTyr4kSZI0LlMNbB3SyaA9IlYDDi5fHpSZi3vnMvNY4BJg54h4wvLaysz3ZuZbM/PaGcevAP6lfLlvPSOXJEmSZq+TQTtF6co84MrM/Emf818p98+p2M/Pyv3GFduRJEnSiCKbqWnv0jKSXa1pf2y5v3jA+d7xbSv28/Byf2PFdiRJklRFiwPssrrj6RTzIZ8IbAKQmVFXH10N2jcv978dcL53fIuK/byh3J9SsR1JkiTNXUcAezXZQVeD9nXK/aAVXe4o9+uO2kFEvAbYDVgEvG8W91064NSCUcciSZI08dq9ROP5FHMqLyi3q4Fal/bqatDeqIh4KvBRil/EvCIzrx/zkCRJktRSmfn+6a8jaquK+YuuBu291WLWGnB+7XJ/+2wbjoi/oSiHWQ14fWaePJv7M3PrAe1eCmw12/FIkiSJTk0abUJXV4/pLc+46YDzvePXzKbRiHgYcDqwPnBkZn5stOFJkiRJ9elqpr23FOPjB5zvHb9k2AYjYiPgO8BGwEcz852jD0+SJEm1andNe+O6GrT/ALgVWBAR22XmT2ec36fcnzpMYxGxPvBtismiJwD/XNM4JUmSVFVCNPEE0+J7wIJBC4kMKnseh06Wx2TmPcDHy5fHRUSvhp2IOIRiffazM/OiaccPjohfRcR7p7cVEWsBpwHbAF8CXpU54V/lJEmS1CpdzbQDvJtiScYnA1dExLkU67IvBG4CXjHj+g2ALSnKX6Z7D/AkYCnwZ+Cz/Wb8Zub+NY5dkiRJQ8uGymMS4Mo2ZdQH6WzQnplLImJX4DBgP+B5wJ+AE4EjMnPQg5dmWr/cr1y2M8j+Iw1UkiRJqqizQTtAZt4FvKPclnftkcCRfY7vjwG5JElSu0148XIna9olSZKkSdLpTLskSZImQ0z4OiEG7ZIkSVIFEfFs4Ihph1Yrj/9w2rF3ZeZpo/Zh0C5JkqT2a3emfUOKFQxnWjjjmpEZtEuSJKndEmju4UrVm8k8kWIFw8Y4EVWSJElqOTPtkiRJar1Jn4hqpl2SJElqOTPtkiRJaj8z7ZIkSZLazEy7JEmS2m/CM+0G7ZIkSWq/JpZ87BDLYyRJkqSWM9MuSZKkdstsZsnHDpXcmGmXJEmSWs5MuyRJktqvQ1nxJphplyRJklrOTLskSZLab8Iz7QbtkiRJar8JD9otj5EkSZJazky7JEmS2s+HK0mSJElqMzPtkiRJarVIGnm4UnSoTN5MuyRJktRyZtolSZLUfq4eI0mSJKnNzLRLkiSp5RKmmsi0dyd7b9AuSZKk9rM8RpIkSVKbmWmXJElS+014pt2gfUWammLpokXjHoVGNcc/u7j6mnEPoTErr7feuIfQqPWumzfuITRqnS/O3f82AfKpjxv3EBr15zn8/xZpRTJolyRJUrslzWTaO5S8t6ZdkiRJajkz7ZIkSWq/RpZ87A6DdkmSJLVcQk41025HWB4jSZIktZyZdkmSJLXfhC/5aKZdkiRJajkz7ZIkSWq3pJmJqB1K3ptplyRJklrOTLskSZLab8Jr2g3aJUmS1H4THrRbHiNJkiS1nJl2SZIktZ+ZdkmSJEltZqZdkiRJLZcwNdVMux1hpl2SJElqOTPtkiRJarekmZr27iTazbRLkiRJbWemXZIkSe034avHGLRLkiSp/aYmO2i3PEaSJElqOTPtkiRJarUkyax/ycfs0ExUM+2SJElSy5lplyRJUrslzdS0dyfRbqZdkiRJajsz7ZIkSWo/l3yUJEmSWm6q/omoXWJ5jCRJktRyZtolSZLUfhNeHmOmXZIkSWo5M+2SJElqt0yyiZr2DmXvzbRLkiRJLWemXZIkSe3Xoax4EwzaJUmS1H5NPBG1QyyPkSRJklrOTLskSZLaL324kiRJkqQWM2iXJElSu2WSU/VvdU5ujYg1I+KoiLg8IpZExPURcXxEbFJH+wbtkiRJUgURsQZwBnAEsA5wCnAdcADwk4h4eNU+rGmXJElS+7W7pv3twI7A+cDumbkYICIOAT4EHA/sUqUDM+2SJEnSiCJiNeDg8uVBvYAdIDOPBS4Bdo6IJ1Tpx6BdkiRJrddITXs9dgLmAVdm5k/6nP9KuX9OlU4sj5EkSVL7tbc85rHl/uIB53vHt63SiUH7irPZnSzm/Dx93OOQJs+tc/uXirF45XEPoVGZ9457CM268Jxxj6BZuWTcI9AI7mQxwGbjHkdPUzFU+T4XRMSl/c5n5tZDNLN5uf/tgPO941vMbnT3Z9C+4tyZTHEHt1037oGoFgvK/ZVjHYWGs/zkTLc/z9Ymn8aqO5/pnbeNewRd0J3Pc+7YDLhz3IMoXVnGUE21X/XLyTrlftDP645yv26VTgzaV5DMfOi4x6D69L6RD/kNXC3n5zn3+JnOLX6eky0znzvuMbTB3P6dsSRJktSs3moxaw04v3a5v71KJwbtkiRJ0uiuLfebDjjfO35NlU4M2iVJkqTR/azcP37A+d7xS6p0YtAuSZIkje4HwK0Uq9Bs1+f8PuX+1CqdGLRLkiRJI8rMe4CPly+Pi4heDTsRcQjF+uxnZ+ZFVfqJzNqeBiVJkiRNnIhYAzgLWAjcAJxLsS77QuAmYMfMvKpSHwbtkiRJUjURsSZwGLAfxdrvfwK+BRyRmYMevDR8+wbtkiRJUrtZ0y5JkiS1nEG7JEmS1HIG7ZIkSVLLGbRLkiRJLWfQLkmSJLWcQbskSZLUcgbtEsXaqhFxVERcHhFLIuL6iDg+IjaZRRvrRcR+EXFSRPwmIu6JiNsj4kcR8YaIWLXJ96D71PF5Dmj3kRFxV0RkRHy3rvFq+er+TCNifkR8ovy7endE3BwR50fEm+seu/5anZ9nRDw9Ik6LiJsi4t6I+GNEnB4Rz29i7NK4uE67Jl75FLMzgR257ylm84EnMounmEXEu4HDgQR+ClwObAjsBKwOfB94RmbeWfub0F/U9XkOaPtMYGcggO9l5m51jFnLVvdnGhF7AF8B1gQuBq4AHgRsA9yRmY+oc/y6vzo/z4h4I/Bhiv/vng9cR/FQmydR/D09OjMPr/cdSGOSmW5uE70B76b4H/55wDrTjh9SHj9ryHYOA94PbD7j+COBa8q2jh73+53rW12fZ592Dyzv/2S5/+643+ukbHV+psCjgbuAPwBPnnFuJWD7cb/fub7V+P/cDYElwD3AzjPOPa08NwU8fNzv2c2tjs1MuyZaRKxG8Y/3PODxmfmTGed/BmxL8Q/5RRX62Rf4H+DqzHxYhSFrGZr6PCPiIcAvgQuBoymyhGbaV4C6P9OI+AawB/DszPxGA0PWMtT5eUbEnsCpwLcz85l9zp8CPBd4cWZ+qaa3II2NNe2adDtR/ONx5cx/PEpfKffPqdjPz8r9xhXb0bI19Xl+lKKU4rUVxqbR1PaZRsRmwDOAqwzYx6bOv6N3D9nnH4e8Tmq1VcY9AGnMHlvuLx5wvnd824r9PLzc31ixHS1b7Z9nRDwLeDHwjsz8dURsWmF8mr06P9NdKJJV50XEKsDeFEHkysD/A76YmbeMPlQNoc7P88fAIuBvI2LnzDy7dyIinkbxBe0Kipp5qfMM2jXpNi/3vx1wvnd8i4r9vKHcn1KxHS1brZ9nRKwN/DtwGcV8Ba14dX6mW5X7xRSB3I4zzr8nIvbJzDNnN0TNQm2fZ2beGhEHUpQenhkR55X3bwo8GfgB8LLMvKfakKV2sDxGk26dcj9oRZc7yv26o3YQEa8BdqPICL1v1HY0lLo/z3dTBA+v8R/+sanzM12/3L+SYkLqfsADgS2B/yr/fHLVpUG1TLX+Hc3Mr1HMUfgjxW9NXlzubwdOB3438killjFolxoUEU+lqIdO4BWZef2Yh6QhRcT2wOuBz2fmWWMejurR+zdvFeDVmXlSZt6SmZdn5kuBCyjqrZ270BERcSjwXeAcipKadcr9GcBRwNfGNzqpXgbtmnSLy/1aA86vXe5vn23DEfE3FOUwqwFvyMyTZz88zVItn2dZ7/xpit+OvKmWkWlUdf4dXTxt/+U+508o9zsPNzSNoLbPMyJ2AT5I8VyMF2bmzzPzjsz8ObBPefzZ5br8UudZ065Jd225HzS5sHf8mtk0GhEPo/jV7PrAkZn5sdGGp1mq6/PcFNiOYuLwlyNi+rn1yv0TIuIsgMzcZXbD1CzU+Xe0d8212X+946vL/YOHG5pGUOfn+dJyf3JmTk0/kZlLI+JrFH+PnwZ8c5bjlFrHoF2TrrcU4+MHnO8dv2TYBiNiI+A7wEbARzPznaMPT7NU9+f50HLrZz3MyK4IdX6mvSUG1x9w/oHlfvGA86quzs+zF+DfOuB87/igz1vqFMtjNOl+QPE/9gURsV2f8/uU+1OHaSwi1ge+DSyg+FX7P9cwRg2vls8zM6/OzOi3AbuWl31v2jE1p86/o+dRTFh8aERs2ed870tYv/XDVY86P8/eErrbDzi/Q7m/etjBSW1m0K6JVq4I8vHy5XHlEn8ARMQhFBOazp7+ZL6IODgifhUR753eVkSsBZwGbAN8CXjVgF/BqyF1fp5qhzo/08z8M3AsEGVbD5h2z27A/hSTxj/Z0NuZeDX/Hf16uf/78umoTLtnL4rVgaYA5xNpTrA8RiqW9duNYl3fKyLiXIpl/hYCNwGvmHH9BhRLxG004/h7gCcBS4E/A5+dUQsNQGbuX+PY9dfq+jzVHnV+psdQ/LZkN+DyiPhhef2OFA9ZOjwzf9zEm9Bf1PV5fp1iQvELgVMj4kLgN8DDuC/7fnhmXtbAe5BWODPtmniZuYTiH/F3Uawd/DyKf0BOBB6fmVcN2VSvbnJligzPywdsalCNn6daos7PNDPvBZ4FvBW4meKpmdsAZwPPycyj6xy7/lpdn2f5m8wXAwdSLPn4COD5wHzgG8Aefp6aS8Lf3kuSJEntZqZdkiRJajmDdkmSJKnlDNolSZKkljNolyRJklrOoF2SJElqOYN2SZIkqeUM2iVJkqSWM2iXJEmSWs6gXZIkSWo5g3ZJkiSp5QzaJUmSpJYzaJdUi4jIGdu9EXFzRPw8Ik6MiBdExCpD3L8oItYbcM2/lNccOeP4kdPuP7LfveV1SyIiZ/Ge9i/bPHHI688qr99l2D7abNrPdf+29RMR88t7zppxfJd+n9m0z/LIOsYsSSuaQbukun2u3E4CfgCsArwM+Arwy4h44nLunwccUqH/N0bE+hXu1wQpv1DOmS9akuaugVkvSRpFZu4/81hELACOBl4EnBkRO2XmT/vdDtwNvCEiPpyZt8yy+7u4L+g/Ypb3qlt+BzwGuHPI608Gfgjc3NiIJKlBZtolNS4zr8zMFwOfBdYCjh9w6RTwKeABwJtG6OpEYAlF0P/AEe5XR2TmvZn5q8y8dsjrby2vN2iX1EkG7ZJWpEOBO4DHRcRTBlzzPoqM+esi4kGzbP964JPAusCbRx7lmE0v2YiIPSLi+xGxOCJuiYivRcSj+9zzl5rtiHhURHwhIn4fEVMR8bxp1z0rIr5TtrUkIi6LiPcNmkcw7b6FEfHtcs7BbWUbO/a5LiJi37L/yyPijoi4PSJ+HBGvjYhl/rszi3761rQvo92/qmkv5ze8vHx55ow5GfMj4uPln/9xGe1eVv6MHz7MOCRpVAbtklaYzLwV+Gb5ctcB19wAfILRA+9e0H9wRGwwyjhb5IXAacBqwKkUX0qeD/wwIh474J4tgQuAJwJnAt8B7gWIiMPK9nYBLgK+TvGbj7cCP4qIhwxo88nAOcCmFJ/fZcBuwNkRsfuMa1cH/qc8f2M57h8CWwPHMfi3LLPtpw6fA64s//xt7puP8TlgMcUXQIBX9bs5InYGHgV8LzOvamB8kvQXBu2SVrSflvvHLOOa91PUKh8cERvOpvHMvBH4D2Ad4C2jDLBFXgu8OjOfmJn7An9D8bOZR1EK1M9LgM8Dj8zMl2TmMzLztIjYAXg3RTD6lMzcLTNfAjwC+DJF8HncgDZfBRwD/E1m7puZO5RjWw04MSLWnHbtnym+WGyUmU8rx/B0YD5wIfDyiHhaDf1UVs6/+H758n2Zuf+07ebM/DlwHrB9RGw3YLwAn65zXJLUj0G7pBWtV1M8cIWXzPw9ReC9NkUWeLZ6Qf9BEfHgEe5vi/My8y8BYWYmxQTb3wLbDSgxugl4a2YunXH8YIr/538sM380rc27y3N3Ac+PiM36tHkNcGTZf+++/wB+BGwEvGDa8T9n5tcz897pDWTmTcBh5cu9BrzfoftZgT5R7u+XbS9XKHoBxc/76yt4TJImkEG7pBUtyv3y1kt/P0X9+z8to2yjr8z8A0XWuFf60VVfmHmgDIa/Ur58ap97vpuZ/VZU6V37333a/ANwOsW/CTv1ufermfnnPsdPGjSOiNguIt4SEcdFxAnluun/VJ5+ZJ+2RupnBfgy8Efg7yNirWnH/wFYA/hcZt4zhnFJmjAG7ZJWtF6d+Z+WdVGZme0F3v8yQj/HcF/Q/9AR7m+DawYcv7rcb9zn3KDVVHrXXj3gfO/4JlXGERGrRcT/AD+h+OL1WmB/igmfe5eXrTugvVHeb6MycwlFjfs8ijkGPa8s959Z0WOSNJkM2iWtaI8r978Y4tpjKGqwXxMRG82mkzLo/ziwJveVZUyCJSPeN/STYpfjEGBf4OfAHsBDgNUyMygmycJ9v23pik9S/HxeBVA+IGxb4JzMvGycA5M0OQzaJa0wETEPeEb58szlXV+uqf0xijKEUQLvY4DbgX+MiBWepa3BFss5fv0s2updO6jN+eX+dxXH8fxyv29mfisz/zCtvn15yyLW+X5rk5mXU/z3ulNEPIb76ts/NY7xSJpMBu2SVqQPUUwuvSAzz5/FPbcD/0j/0o2BMvOP3Bf0v20297bEi2YeiIhVuG9C5vdnnl+Gc8v9vn3a3JDiy1QCP+hz794RsXKf4y/pM47eBOPf9rn+r95PhX7q0qtHX94TwnsTUg8px3ML8NUGxiNJfRm0S2pcRDw8Ir4IHEhRZ37gsPeWgfe/Uaz/PfR903wIuI2iBnnVEe4fp6dExCtmHHsnsDlwSWae2+eeQY6jeOLs6yNi+97BiFiN4ovNmsDXMvO6PvfOB/51+oHygUNPAn7P/YPXy8v9a2Zcvw/wsuWMcTb91KWXvd9ymVcVK8TcSPHf0TrAf5b17pK0QiwvsyBJs1KuEgJFUuABFOt/P5qijvkKYL9y/evZ+BDwurK9WcnMP0XERymWShzVsyPih8s4//zyoVA9/x4Rtw249uLMfO2Q/f4H8JmIeDXFQ4C2pXhI0W0UkzuHlpk/jogjgPcA55dPEr2ZYrWYzSg+m4MG3P5p4F8iYm/gEoq13XegeGjT/jNWq/kA8EzgfRHxQoog/pHA9sAHgTctY5iz6acupwLvAD4YEU/nviVJ31p+YQSKVXsi4nju+42NpTGSVigz7ZLq9vJy25diib6lFA/72Rt4TGZeONsGM/MW4CMVxnQscGuF+zcAFi5jW33G9Y9ZxrVbzaLfLwHPpfgZ7kXxpNBTgCdl5k9m+yYy82hgT+BsimB4b+BuikB7Ybk+fj/nATtTZJr3pHh/3wN2ycxvzejjHOApwBkUNex7UpSgvIDBD2+adT91ycyLKJZv/AWwO8Vvcw6k/wo3Z5T78zPz0ibGI0mDxLRnWEiSWqD8bcXLgV0z86zxjkY9EfFJirkVB2TmiWMejqQJY6ZdkqTliIgtgJdSlM98cczDkTSBrGmXJGmAiHgzxVyCp1Ou+Z+Zd413VJImkUG7JEmDPZuizv56ipV7/m28w5E0qaxplyRJklrOmnZJkiSp5QzaJUmSpJYzaJckSZJazqBdkiRJajmDdkmSJKnlDNolSZKkljNolyRJklrOoF2SJElqOYN2SZIkqeUM2iVJkqSWM2iXJEmSWs6gXZIkSWo5g3ZJkiSp5f4/v1yggA1PVnUAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 900x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "h_isMIX = ax.hist2d(mix['DNN_LL'].values, mix['DNN_TT'].values, density=True)\n",
    "fig.colorbar(h_isMIX[3], ax=ax)\n",
    "plt.ylabel('DNN TT probability')\n",
    "plt.xlabel('DNN LL probability')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "c51de0ef-f2c1-420c-bf18-f5e2e3cdffb1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n"
     ]
    }
   ],
   "source": [
    "with open('generated_code_dnn_SigVsBkg_emu.h', 'w') as f:\n",
    "\n",
    "\n",
    "    print (\"\"\"\n",
    "    // Select all (Ctrl+A), Copy (Ctrl+C), paste to an empty text file (Ctrl+V) and save that file as 'generated_code.h'\n",
    "    // \n",
    "    // Auto-generated header file. Assumes img to be a floating point array\n",
    "    // of 64 elements (corresponding to an 8x8 b&w image)\n",
    "\n",
    "    #include <math.h>\n",
    "\n",
    "    float norma_VBS;\n",
    "    \"\"\", file=f)\n",
    "    \n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    for iLayer, layer in enumerate(dnn.model1.layers):\n",
    "      print(type(layer))\n",
    "#       if type(layer) is not Dropout and type(layer) is not Input :\n",
    "      if type(layer) is Dense:\n",
    "        kernel, bias = layer.get_weights()\n",
    "        iLayer_tmp+=1\n",
    "        print (\"inline float activation_VBS_%d (float x, float n) \" % iLayer_tmp, file=f);\n",
    "        activation =  layer.get_config()['activation'] \n",
    "        if activation == 'sigmoid':\n",
    "          print (\"{ return 1./(1 + exp(-x)); }\", file=f)\n",
    "        elif activation == 'tanh':\n",
    "          print (\"{ return tanh(x);}\")\n",
    "        elif activation == 'relu':\n",
    "          print (\"{ return x > 0 ? x : 0;}\", file=f)\n",
    "        elif activation == 'linear':\n",
    "          print (\"{ return x;}\")\n",
    "        elif activation == 'softmax':\n",
    "          print (\"{ return exp(x)/n;  }\", file=f)\n",
    "        else:\n",
    "          raise KeyError (\"Unexpected activation %s\"%activation)\n",
    "      \n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    \n",
    "    print (\"\"\"\n",
    "    float guess_VBS (const float *img, int flag)\n",
    "    {\n",
    "    \"\"\", file=f)\n",
    "\n",
    "    max_out = 100\n",
    "    for iLayer, layer in enumerate(dnn.model1.layers):\n",
    "      if type(layer) is Dense:\n",
    "        iLayer_tmp+=1\n",
    "        print (\"  // Declare the arrays in the stack\", file=f)\n",
    "        kernel, bias = layer.get_weights()\n",
    "        \n",
    "        print (type (dnn.means))\n",
    "        print (type(bias))\n",
    "        \n",
    "         \n",
    "        max_out = max(kernel.shape[1], max_out)\n",
    "        #print (\"  //\",bias.shape)\n",
    "        kernel_values = \"{%s}\"%(',\\n   '.join([\"{%s}\"%(','.join([\"%18.13f\"%x for x in row])) for row in kernel]))\n",
    "        bias_values   = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in bias]))\n",
    "        print (\"  const float kernel_%d[%d][%d] = \\n  %s;\" % (iLayer_tmp, kernel.shape[0], kernel.shape[1],kernel_values), file=f)\n",
    "        print (\"  const float bias_%d[%d] = %s;\" % (iLayer_tmp, bias.shape[0], bias_values), file=f)\n",
    "      \n",
    "    print (\"  float buffer_in[%d];\" % max_out, file=f)\n",
    "    print (\"  float buffer_out[%d];\" % max_out, file=f)\n",
    "\n",
    "    print (\"  unsigned int i,j,c; \", file=f)\n",
    "    \n",
    "\n",
    "    mean_values = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in dnn.means]))\n",
    "    sigma_values = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in dnn.sigmas]))\n",
    "    \n",
    "\n",
    "    print (\" float mean[%d] = %s;\" % (max_out, mean_values), file=f)\n",
    "    print (\" float sigma[%d] = %s;\" % (max_out, sigma_values), file=f)\n",
    "\n",
    "    print (\"\\n\\n\\n\", file=f)\n",
    "    print (\"  // Load the input in the buffer\", file=f)\n",
    "    print (\"  for (c = 0; c < 64; ++c) \\n  buffer_in[c] = (img[c]-mean[c])/sigma[c];\", file=f)\n",
    "\n",
    "\n",
    "\n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    for iLayer, layer in enumerate(dnn.model1.layers):\n",
    "      if type(layer) is Dense: \n",
    "        kernel, bias = layer.get_weights()\n",
    "        iLayer_tmp+=1\n",
    "\n",
    "        if iLayer != len(dnn.model1.layers)-1:\n",
    "            \n",
    "            \n",
    "            print ( \"  // Processing layer %i \" % iLayer_tmp , file=f)\n",
    "            print ( \"\"\"\n",
    "            for (c = 0; c < {n_out}; ++c ) \n",
    "              buffer_out[c] = bias_{iLayer_tmp}[c];\n",
    "\n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              for (i = 0; i < {n_in}; ++i)\n",
    "                buffer_out[c] += buffer_in[i] * kernel_{iLayer_tmp}[i][c];\n",
    "\n",
    "            norma_VBS = 0;\n",
    "\n",
    "            // Prepares for next layer \n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              buffer_in[c] = activation_VBS_{iLayer_tmp}(buffer_out[c], norma_VBS);\n",
    "\n",
    "\n",
    "            \"\"\".format (\n",
    "                n_in = kernel.shape[0],\n",
    "                n_out = kernel.shape[1],\n",
    "                iLayer_tmp = iLayer_tmp,\n",
    "            ), file=f)\n",
    "            \n",
    "        else:\n",
    "            \n",
    "            print ( \"  // Processing layer %i \" % iLayer_tmp , file=f)\n",
    "            print ( \"\"\"\n",
    "            for (c = 0; c < {n_out}; ++c ) \n",
    "              buffer_out[c] = bias_{iLayer_tmp}[c];\n",
    "\n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              for (i = 0; i < {n_in}; ++i)\n",
    "                buffer_out[c] += buffer_in[i] * kernel_{iLayer_tmp}[i][c];\n",
    "\n",
    "            norma_VBS = 0;\n",
    "\n",
    "            for(c=0;c<{n_out};++c)\n",
    "              norma_VBS+=exp(buffer_out[c]);\n",
    "\n",
    "            // Prepares for next layer \n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              buffer_in[c] = activation_VBS_{iLayer_tmp}(buffer_out[c], norma_VBS);\n",
    "\n",
    "\n",
    "            \"\"\".format (\n",
    "                n_in = kernel.shape[0],\n",
    "                n_out = kernel.shape[1],\n",
    "                iLayer_tmp = iLayer_tmp,\n",
    "            ), file=f)\n",
    "      \n",
    "    last_kernel, last_bias = dnn.model1.layers[-1].get_weights()\n",
    "    print (\"\"\"\n",
    "      //i = 0;\n",
    "      //for (c = 0; c < {n_out}; ++c)\n",
    "      //  if (buffer_in[c] > buffer_in[i])\n",
    "      //    i = c;\n",
    "      \n",
    "      if (flag==0) //vbf\n",
    "        return buffer_in[0];\n",
    "      else if (flag==1) //bkg\n",
    "        return buffer_in[1];\n",
    "         \n",
    "      \n",
    "      //return i;\n",
    "    \"\"\".format(n_out = last_kernel.shape[1]), file=f)\n",
    "\n",
    "    print (\"}\", file=f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "4c523d50-4f12-4c56-858f-0800889fe9d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n"
     ]
    }
   ],
   "source": [
    "with open('generated_code_dnn_emu_LLVsOther.h', 'w') as f:\n",
    "\n",
    "\n",
    "    print (\"\"\"\n",
    "    // Select all (Ctrl+A), Copy (Ctrl+C), paste to an empty text file (Ctrl+V) and save that file as 'generated_code.h'\n",
    "    // \n",
    "    // Auto-generated header file. Assumes img to be a floating point array\n",
    "    // of 64 elements (corresponding to an 8x8 b&w image)\n",
    "\n",
    "    #include <math.h>\n",
    "\n",
    "    float norma_LL;\n",
    "    \"\"\", file=f)\n",
    "    \n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    for iLayer, layer in enumerate(dnnLL.model1.layers):\n",
    "      print(type(layer))\n",
    "#       if type(layer) is not Dropout and type(layer) is not Input :\n",
    "      if type(layer) is Dense:\n",
    "        kernel, bias = layer.get_weights()\n",
    "        iLayer_tmp+=1\n",
    "        print (\"inline float activation_LL_%d (float x, float n) \" % iLayer_tmp, file=f);\n",
    "        activation =  layer.get_config()['activation'] \n",
    "        if activation == 'sigmoid':\n",
    "          print (\"{ return 1./(1 + exp(-x)); }\", file=f)\n",
    "        elif activation == 'tanh':\n",
    "          print (\"{ return tanh(x);}\")\n",
    "        elif activation == 'relu':\n",
    "          print (\"{ return x > 0 ? x : 0;}\", file=f)\n",
    "        elif activation == 'linear':\n",
    "          print (\"{ return x;}\")\n",
    "        elif activation == 'softmax':\n",
    "          print (\"{ return exp(x)/n;  }\", file=f)\n",
    "        else:\n",
    "          raise KeyError (\"Unexpected activation %s\"%activation)\n",
    "      \n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    \n",
    "    print (\"\"\"\n",
    "    float guess_LL (const float *img, int flag)\n",
    "    {\n",
    "    \"\"\", file=f)\n",
    "\n",
    "    max_out = 100\n",
    "    for iLayer, layer in enumerate(dnnLL.model1.layers):\n",
    "      if type(layer) is Dense:\n",
    "        iLayer_tmp+=1\n",
    "        print (\"  // Declare the arrays in the stack\", file=f)\n",
    "        kernel, bias = layer.get_weights()\n",
    "        \n",
    "        print (type (dnnLL.means))\n",
    "        print (type(bias))\n",
    "        \n",
    "         \n",
    "        max_out = max(kernel.shape[1], max_out)\n",
    "        #print (\"  //\",bias.shape)\n",
    "        kernel_values = \"{%s}\"%(',\\n   '.join([\"{%s}\"%(','.join([\"%18.13f\"%x for x in row])) for row in kernel]))\n",
    "        bias_values   = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in bias]))\n",
    "        print (\"  const float kernel_%d[%d][%d] = \\n  %s;\" % (iLayer_tmp, kernel.shape[0], kernel.shape[1],kernel_values), file=f)\n",
    "        print (\"  const float bias_%d[%d] = %s;\" % (iLayer_tmp, bias.shape[0], bias_values), file=f)\n",
    "      \n",
    "    print (\"  float buffer_in[%d];\" % max_out, file=f)\n",
    "    print (\"  float buffer_out[%d];\" % max_out, file=f)\n",
    "\n",
    "    print (\"  unsigned int i,j,c; \", file=f)\n",
    "    \n",
    "\n",
    "    mean_values = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in dnnLL.means]))\n",
    "    sigma_values = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in dnnLL.sigmas]))\n",
    "    \n",
    "\n",
    "    print (\" float mean[%d] = %s;\" % (max_out, mean_values), file=f)\n",
    "    print (\" float sigma[%d] = %s;\" % (max_out, sigma_values), file=f)\n",
    "\n",
    "    print (\"\\n\\n\\n\", file=f)\n",
    "    print (\"  // Load the input in the buffer\", file=f)\n",
    "    print (\"  for (c = 0; c < 64; ++c) \\n  buffer_in[c] = (img[c]-mean[c])/sigma[c];\", file=f)\n",
    "\n",
    "\n",
    "\n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    for iLayer, layer in enumerate(dnnLL.model1.layers):\n",
    "      if type(layer) is Dense: \n",
    "        kernel, bias = layer.get_weights()\n",
    "        iLayer_tmp+=1\n",
    "\n",
    "        if iLayer != len(dnnLL.model1.layers)-1:\n",
    "            \n",
    "            \n",
    "            print ( \"  // Processing layer %i \" % iLayer_tmp , file=f)\n",
    "            print ( \"\"\"\n",
    "            for (c = 0; c < {n_out}; ++c ) \n",
    "              buffer_out[c] = bias_{iLayer_tmp}[c];\n",
    "\n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              for (i = 0; i < {n_in}; ++i)\n",
    "                buffer_out[c] += buffer_in[i] * kernel_{iLayer_tmp}[i][c];\n",
    "\n",
    "            norma_LL = 0;\n",
    "\n",
    "            // Prepares for next layer \n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              buffer_in[c] = activation_LL_{iLayer_tmp}(buffer_out[c], norma_LL);\n",
    "\n",
    "\n",
    "            \"\"\".format (\n",
    "                n_in = kernel.shape[0],\n",
    "                n_out = kernel.shape[1],\n",
    "                iLayer_tmp = iLayer_tmp,\n",
    "            ), file=f)\n",
    "            \n",
    "        else:\n",
    "            \n",
    "            print ( \"  // Processing layer %i \" % iLayer_tmp , file=f)\n",
    "            print ( \"\"\"\n",
    "            for (c = 0; c < {n_out}; ++c ) \n",
    "              buffer_out[c] = bias_{iLayer_tmp}[c];\n",
    "\n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              for (i = 0; i < {n_in}; ++i)\n",
    "                buffer_out[c] += buffer_in[i] * kernel_{iLayer_tmp}[i][c];\n",
    "\n",
    "            norma_LL = 0;\n",
    "\n",
    "            for(c=0;c<{n_out};++c)\n",
    "              norma_LL+=exp(buffer_out[c]);\n",
    "\n",
    "            // Prepares for next layer \n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              buffer_in[c] = activation_LL{iLayer_tmp}(buffer_out[c], norma_LL);\n",
    "\n",
    "\n",
    "            \"\"\".format (\n",
    "                n_in = kernel.shape[0],\n",
    "                n_out = kernel.shape[1],\n",
    "                iLayer_tmp = iLayer_tmp,\n",
    "            ), file=f)\n",
    "      \n",
    "    last_kernel, last_bias = dnnLL.model1.layers[-1].get_weights()\n",
    "    print (\"\"\"\n",
    "      //i = 0;\n",
    "      //for (c = 0; c < {n_out}; ++c)\n",
    "      //  if (buffer_in[c] > buffer_in[i])\n",
    "      //    i = c;\n",
    "      \n",
    "      if (flag==0) //vbf\n",
    "        return buffer_in[0];\n",
    "      else if (flag==1) //bkg\n",
    "        return buffer_in[1];\n",
    "         \n",
    "      \n",
    "      //return i;\n",
    "    \"\"\".format(n_out = last_kernel.shape[1]), file=f)\n",
    "\n",
    "    print (\"}\", file=f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "72c6e0df-9dac-4ae1-b71f-7a1c7d662fce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'keras.layers.core.dense.Dense'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n",
      "<class 'numpy.ndarray'>\n"
     ]
    }
   ],
   "source": [
    "with open('generated_code_dnn_TTVsOther.h', 'w') as f:\n",
    "\n",
    "\n",
    "    print (\"\"\"\n",
    "    // Select all (Ctrl+A), Copy (Ctrl+C), paste to an empty text file (Ctrl+V) and save that file as 'generated_code.h'\n",
    "    // \n",
    "    // Auto-generated header file. Assumes img to be a floating point array\n",
    "    // of 64 elements (corresponding to an 8x8 b&w image)\n",
    "\n",
    "    #include <math.h>\n",
    "\n",
    "    float norma_TT;\n",
    "    \"\"\", file=f)\n",
    "    \n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    for iLayer, layer in enumerate(dnnTT.model1.layers):\n",
    "      print(type(layer))\n",
    "#       if type(layer) is not Dropout and type(layer) is not Input :\n",
    "      if type(layer) is Dense:\n",
    "        kernel, bias = layer.get_weights()\n",
    "        iLayer_tmp+=1\n",
    "        print (\"inline float activation_TT_%d (float x, float n) \" % iLayer_tmp, file=f);\n",
    "        activation =  layer.get_config()['activation'] \n",
    "        if activation == 'sigmoid':\n",
    "          print (\"{ return 1./(1 + exp(-x)); }\", file=f)\n",
    "        elif activation == 'tanh':\n",
    "          print (\"{ return tanh(x);}\")\n",
    "        elif activation == 'relu':\n",
    "          print (\"{ return x > 0 ? x : 0;}\", file=f)\n",
    "        elif activation == 'linear':\n",
    "          print (\"{ return x;}\")\n",
    "        elif activation == 'softmax':\n",
    "          print (\"{ return exp(x)/n;  }\", file=f)\n",
    "        else:\n",
    "          raise KeyError (\"Unexpected activation %s\"%activation)\n",
    "      \n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    \n",
    "    print (\"\"\"\n",
    "    float guess_TT (const float *img, int flag)\n",
    "    {\n",
    "    \"\"\", file=f)\n",
    "\n",
    "    max_out = 100\n",
    "    for iLayer, layer in enumerate(dnnTT.model1.layers):\n",
    "      if type(layer) is Dense:\n",
    "        iLayer_tmp+=1\n",
    "        print (\"  // Declare the arrays in the stack\", file=f)\n",
    "        kernel, bias = layer.get_weights()\n",
    "        \n",
    "        print (type (dnnTT.means))\n",
    "        print (type(bias))\n",
    "        \n",
    "         \n",
    "        max_out = max(kernel.shape[1], max_out)\n",
    "        #print (\"  //\",bias.shape)\n",
    "        kernel_values = \"{%s}\"%(',\\n   '.join([\"{%s}\"%(','.join([\"%18.13f\"%x for x in row])) for row in kernel]))\n",
    "        bias_values   = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in bias]))\n",
    "        print (\"  const float kernel_%d[%d][%d] = \\n  %s;\" % (iLayer_tmp, kernel.shape[0], kernel.shape[1],kernel_values), file=f)\n",
    "        print (\"  const float bias_%d[%d] = %s;\" % (iLayer_tmp, bias.shape[0], bias_values), file=f)\n",
    "      \n",
    "    print (\"  float buffer_in[%d];\" % max_out, file=f)\n",
    "    print (\"  float buffer_out[%d];\" % max_out, file=f)\n",
    "\n",
    "    print (\"  unsigned int i,j,c; \", file=f)\n",
    "    \n",
    "\n",
    "    mean_values = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in dnnTT.means]))\n",
    "    sigma_values = \"{%s}\"% ( \",\".join([\"%18.13f\"%x for x in dnnTT.sigmas]))\n",
    "    \n",
    "\n",
    "    print (\" float mean[%d] = %s;\" % (max_out, mean_values), file=f)\n",
    "    print (\" float sigma[%d] = %s;\" % (max_out, sigma_values), file=f)\n",
    "\n",
    "    print (\"\\n\\n\\n\", file=f)\n",
    "    print (\"  // Load the input in the buffer\", file=f)\n",
    "    print (\"  for (c = 0; c < 64; ++c) \\n  buffer_in[c] = (img[c]-mean[c])/sigma[c];\", file=f)\n",
    "\n",
    "\n",
    "\n",
    "    iLayer_tmp=-1\n",
    "\n",
    "    for iLayer, layer in enumerate(dnnTT.model1.layers):\n",
    "      if type(layer) is Dense: \n",
    "        kernel, bias = layer.get_weights()\n",
    "        iLayer_tmp+=1\n",
    "\n",
    "        if iLayer != len(dnnTT.model1.layers)-1:\n",
    "            \n",
    "            \n",
    "            print ( \"  // Processing layer %i \" % iLayer_tmp , file=f)\n",
    "            print ( \"\"\"\n",
    "            for (c = 0; c < {n_out}; ++c ) \n",
    "              buffer_out[c] = bias_{iLayer_tmp}[c];\n",
    "\n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              for (i = 0; i < {n_in}; ++i)\n",
    "                buffer_out[c] += buffer_in[i] * kernel_{iLayer_tmp}[i][c];\n",
    "\n",
    "            norma_TT = 0;\n",
    "\n",
    "            // Prepares for next layer \n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              buffer_in[c] = activation_TT_{iLayer_tmp}(buffer_out[c], norma_TT);\n",
    "\n",
    "\n",
    "            \"\"\".format (\n",
    "                n_in = kernel.shape[0],\n",
    "                n_out = kernel.shape[1],\n",
    "                iLayer_tmp = iLayer_tmp,\n",
    "            ), file=f)\n",
    "            \n",
    "        else:\n",
    "            \n",
    "            print ( \"  // Processing layer %i \" % iLayer_tmp , file=f)\n",
    "            print ( \"\"\"\n",
    "            for (c = 0; c < {n_out}; ++c ) \n",
    "              buffer_out[c] = bias_{iLayer_tmp}[c];\n",
    "\n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              for (i = 0; i < {n_in}; ++i)\n",
    "                buffer_out[c] += buffer_in[i] * kernel_{iLayer_tmp}[i][c];\n",
    "\n",
    "            norma_TT = 0;\n",
    "\n",
    "            for(c=0;c<{n_out};++c)\n",
    "              norma_TT+=exp(buffer_out[c]);\n",
    "\n",
    "            // Prepares for next layer \n",
    "            for (c = 0; c < {n_out}; ++c )\n",
    "              buffer_in[c] = activation_TT_{iLayer_tmp}(buffer_out[c], norma_TT);\n",
    "\n",
    "\n",
    "            \"\"\".format (\n",
    "                n_in = kernel.shape[0],\n",
    "                n_out = kernel.shape[1],\n",
    "                iLayer_tmp = iLayer_tmp,\n",
    "            ), file=f)\n",
    "      \n",
    "    last_kernel, last_bias = dnnTT.model1.layers[-1].get_weights()\n",
    "    print (\"\"\"\n",
    "      //i = 0;\n",
    "      //for (c = 0; c < {n_out}; ++c)\n",
    "      //  if (buffer_in[c] > buffer_in[i])\n",
    "      //    i = c;\n",
    "      \n",
    "      if (flag==0) //vbf\n",
    "        return buffer_in[0];\n",
    "      else if (flag==1) //bkg\n",
    "        return buffer_in[1];\n",
    "         \n",
    "      \n",
    "      //return i;\n",
    "    \"\"\".format(n_out = last_kernel.shape[1]), file=f)\n",
    "\n",
    "    print (\"}\", file=f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "11d4d6ba-3206-4c98-89ec-24dffc5708cb",
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}