examples/streamingOutlier/autoencoder_outlier.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import sklearn\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.metrics import confusion_matrix, precision_recall_curve\n",
    "from sklearn.metrics import recall_score, classification_report, auc, roc_curve\n",
    "from sklearn.metrics import precision_recall_fscore_support, f1_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prec_recall(df):\n",
    "    '''Assumes columns of pred and label'''\n",
    "    tp=df[(df['label']==1)&(df['pred']==1)].shape[0]#number of true positives\n",
    "    recall = tp/df['label'].sum()                   #recall\n",
    "    prec =   tp/df['pred'].sum()                    #precision\n",
    "    fp=df[(df['label']!=1)&(df['pred']==1)].shape[0]#number of false positives\n",
    "    fpr=fp/df['pred'].sum()                         #False positive rate\n",
    "    return prec,recall,fpr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "fn=\"creditcard.csv\"\n",
    "df=pd.read_csv(fn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols=[col for col in df.columns ]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "dftrain,dftest=train_test_split(df,test_size=0.15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "30"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "numinps= len(cols)\n",
    "numinps"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<p>Test/Train split for train and eval data.</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "dftrain2,dfeval=train_test_split(dftrain,test_size=0.15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<p>Building the autoencoder</p>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<p>So the Threshold is determined below and then put into the model as hardwired (not used to determine the threshold - no loops here) so that we can export the model at the same time.</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "encodedim=14\n",
    "hiddim=encodedim//2\n",
    "learnrate=0.001\n",
    "droprate=0.2\n",
    "threshold=tf.constant(55.)\n",
    "inputs=tf.placeholder(tf.float32, shape=(None,numinps),name=\"inputs\")\n",
    "#mode=tf.placeholder(tf.string, name=\"mode\")\n",
    "\n",
    "dense0 = tf.layers.dense(inputs=inputs, units=encodedim, activation=tf.nn.tanh)\n",
    "dropout0 = tf.layers.dropout(inputs=dense0, rate=droprate)\n",
    "\n",
    "dense1 = tf.layers.dense(inputs=dropout0, units=hiddim, activation=tf.nn.relu)\n",
    "dropout1 = tf.layers.dropout(inputs=dense1, rate=droprate)\n",
    "\n",
    "dense2 = tf.layers.dense(inputs=dropout1, units=hiddim, activation=tf.nn.tanh)\n",
    "dropout2 = tf.layers.dropout(inputs=dense2, rate=droprate)\n",
    "\n",
    "dense3 = tf.layers.dense(inputs=dropout2, units=numinps, activation=tf.nn.relu)\n",
    "dropout3 = tf.layers.dropout(inputs=dense3, rate=droprate)\n",
    "\n",
    "loss=tf.losses.mean_squared_error(labels=inputs,predictions=dropout3)\n",
    "Yout=tf.cast(tf.math.greater(loss,threshold), tf.int32)\n",
    "trainstep=tf.train.AdamOptimizer(learning_rate=learnrate).minimize(loss)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<p>Training Autoencoder and then exporting saved_model format.</p>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  0 0 0.8786459\n",
      "evalu:  0 0 [1.1967247]\n",
      "train:  0 1 0.82111293\n",
      "train:  0 2 0.8849528\n",
      "train:  0 3 1.006468\n",
      "train:  0 4 1.1138202\n",
      "train:  0 5 1.1362659\n",
      "train:  0 6 0.96332306\n",
      "train:  0 7 1.052875\n",
      "train:  0 8 1.0396622\n",
      "train:  0 9 1.6054584\n",
      "train:  0 10 0.8529419\n",
      "train:  0 11 0.90086263\n",
      "train:  0 12 2.1639411\n",
      "train:  0 13 0.85052085\n",
      "train:  0 14 0.8396027\n",
      "train:  0 15 1.2677279\n",
      "train:  0 16 1.1301173\n",
      "train:  0 17 0.8110599\n",
      "train:  0 18 1.4951175\n",
      "train:  0 19 0.9278629\n",
      "train:  0 20 2.3693\n",
      "evalu:  0 20 [1.1779827]\n",
      "train:  0 21 1.146614\n",
      "train:  0 22 0.987472\n",
      "train:  0 23 0.86754674\n",
      "train:  0 24 0.8092464\n",
      "train:  0 25 1.0109401\n",
      "train:  0 26 0.96182066\n",
      "train:  0 27 3.352582\n",
      "train:  0 28 0.7351525\n",
      "train:  0 29 0.7405391\n",
      "train:  0 30 0.8156986\n",
      "train:  0 31 1.7724568\n",
      "train:  0 32 0.8682583\n",
      "train:  0 33 1.4406427\n",
      "train:  0 34 0.8096988\n",
      "train:  0 35 0.9187361\n",
      "train:  0 36 1.0868374\n",
      "train:  0 37 1.5156949\n",
      "train:  0 38 0.96341634\n",
      "train:  0 39 1.1342824\n",
      "train:  0 40 1.170228\n",
      "evalu:  0 40 [1.1645337]\n",
      "train:  0 41 1.3322325\n",
      "train:  0 42 1.3302653\n",
      "train:  0 43 0.8782261\n",
      "train:  0 44 1.0225474\n",
      "train:  0 45 1.713875\n",
      "train:  0 46 1.8947158\n",
      "train:  0 47 0.93796647\n",
      "train:  0 48 0.8665585\n",
      "train:  0 49 0.7398594\n",
      "train:  0 50 1.58371\n",
      "train:  0 51 0.7865275\n",
      "train:  0 52 0.8918782\n",
      "train:  0 53 0.8931958\n",
      "train:  0 54 1.1094896\n",
      "train:  0 55 1.0826802\n",
      "train:  0 56 0.6655319\n",
      "train:  0 57 0.91976744\n",
      "train:  0 58 1.2017465\n",
      "train:  0 59 0.80555964\n",
      "train:  0 60 0.8589283\n",
      "evalu:  0 60 [1.1519606]\n",
      "train:  0 61 0.6982642\n",
      "train:  0 62 0.78333026\n",
      "train:  0 63 0.8383693\n",
      "train:  0 64 0.80178493\n",
      "train:  0 65 0.91954935\n",
      "train:  0 66 1.2581463\n",
      "train:  0 67 0.85353506\n",
      "train:  0 68 1.3293755\n",
      "train:  0 69 1.057845\n",
      "train:  0 70 1.049903\n",
      "train:  0 71 0.9687894\n",
      "train:  0 72 0.8289356\n",
      "train:  0 73 0.78271174\n",
      "train:  0 74 0.65952206\n",
      "train:  0 75 0.884103\n",
      "train:  0 76 0.99325323\n",
      "train:  0 77 0.8950659\n",
      "train:  0 78 1.4477775\n",
      "train:  0 79 1.5110921\n",
      "train:  0 80 1.0065436\n",
      "evalu:  0 80 [1.1402565]\n",
      "train:  0 81 0.7912518\n",
      "train:  0 82 0.8475591\n",
      "train:  0 83 0.82672185\n",
      "train:  0 84 1.7413579\n",
      "train:  0 85 0.9160288\n",
      "train:  0 86 0.75354916\n",
      "train:  0 87 1.0124801\n",
      "train:  0 88 0.7459381\n",
      "train:  0 89 0.94064695\n",
      "train:  0 90 0.87827927\n",
      "train:  0 91 0.9100233\n",
      "train:  0 92 0.8584767\n",
      "train:  0 93 0.7903019\n",
      "train:  0 94 0.76532763\n",
      "train:  0 95 0.93692225\n",
      "train:  0 96 0.8712811\n",
      "train:  0 97 1.0757513\n",
      "train:  0 98 0.8226978\n",
      "train:  0 99 0.88752264\n",
      "train:  0 100 0.91096467\n",
      "evalu:  0 100 [1.1293503]\n",
      "train:  0 101 0.9023841\n",
      "train:  0 102 1.3845059\n",
      "train:  0 103 0.9847777\n",
      "train:  0 104 2.0605168\n",
      "train:  0 105 1.4036412\n",
      "train:  0 106 0.8645364\n",
      "train:  0 107 1.2815722\n",
      "train:  0 108 1.6567103\n",
      "train:  0 109 1.3640988\n",
      "train:  0 110 0.708163\n",
      "train:  0 111 0.78296775\n",
      "train:  0 112 1.5121171\n",
      "train:  0 113 0.9921733\n",
      "train:  0 114 0.70663893\n",
      "train:  0 115 0.8780782\n",
      "train:  0 116 0.9462412\n",
      "train:  0 117 0.8801483\n",
      "train:  0 118 0.73944443\n",
      "train:  0 119 0.8902887\n",
      "train:  0 120 1.0379455\n",
      "evalu:  0 120 [1.1198945]\n",
      "train:  0 121 1.3311955\n",
      "train:  0 122 0.8633209\n",
      "train:  0 123 1.2295003\n",
      "train:  0 124 0.8616672\n",
      "train:  0 125 0.8997015\n",
      "train:  0 126 0.9452559\n",
      "train:  0 127 0.988758\n",
      "train:  0 128 0.96965444\n",
      "train:  0 129 1.3601874\n",
      "train:  0 130 0.782487\n",
      "train:  0 131 0.7585293\n",
      "train:  0 132 0.887821\n",
      "train:  0 133 0.6719611\n",
      "train:  0 134 1.1487204\n",
      "train:  0 135 1.0370319\n",
      "train:  0 136 1.4794749\n",
      "train:  0 137 2.0768988\n",
      "train:  0 138 0.94972\n",
      "train:  0 139 1.6783444\n",
      "train:  0 140 0.79635847\n",
      "evalu:  0 140 [1.1115302]\n",
      "train:  0 141 0.7875242\n",
      "train:  0 142 0.8059764\n",
      "train:  0 143 0.8403721\n",
      "train:  0 144 0.9530993\n",
      "train:  0 145 0.8262813\n",
      "train:  0 146 0.81196404\n",
      "train:  0 147 3.2706537\n",
      "train:  0 148 0.869231\n",
      "train:  0 149 0.64137495\n",
      "train:  0 150 0.6396983\n",
      "train:  0 151 0.87956613\n",
      "train:  0 152 2.2063558\n",
      "train:  0 153 0.9531912\n",
      "train:  0 154 0.93063414\n",
      "train:  0 155 1.1880989\n",
      "train:  0 156 0.82731676\n",
      "train:  0 157 2.1509695\n",
      "train:  0 158 1.1455362\n",
      "train:  0 159 0.7984915\n",
      "train:  0 160 0.7691895\n",
      "evalu:  0 160 [1.103716]\n",
      "train:  0 161 1.2978697\n",
      "train:  0 162 0.99194825\n",
      "train:  0 163 0.86874986\n",
      "train:  0 164 0.9082941\n",
      "train:  0 165 1.0011395\n",
      "train:  0 166 0.65681636\n",
      "train:  0 167 0.96670043\n",
      "train:  0 168 2.9050713\n",
      "train:  0 169 0.88920313\n",
      "train:  0 170 0.7406721\n",
      "train:  0 171 1.0463173\n",
      "train:  0 172 0.767655\n",
      "train:  0 173 1.0189456\n",
      "train:  0 174 0.86340094\n",
      "train:  0 175 0.73655176\n",
      "train:  0 176 0.86339134\n",
      "train:  0 177 0.81838703\n",
      "train:  0 178 1.0298003\n",
      "train:  0 179 1.0705737\n",
      "train:  0 180 0.6676819\n",
      "evalu:  0 180 [1.0958569]\n",
      "train:  0 181 1.010225\n",
      "train:  0 182 0.83499753\n",
      "train:  0 183 0.71832484\n",
      "train:  0 184 1.1915125\n",
      "train:  0 185 0.7704922\n",
      "train:  0 186 1.266823\n",
      "train:  0 187 1.8062062\n",
      "train:  0 188 0.790487\n",
      "train:  0 189 1.3867736\n",
      "train:  0 190 0.80328953\n",
      "train:  0 191 0.7491766\n",
      "train:  0 192 1.0096444\n",
      "train:  0 193 0.65972674\n",
      "train:  0 194 0.6425807\n",
      "train:  0 195 0.8052865\n",
      "train:  0 196 0.735857\n",
      "train:  0 197 0.80411965\n",
      "train:  0 198 0.80112237\n",
      "train:  0 199 0.7603209\n",
      "train:  0 200 0.7958099\n",
      "evalu:  0 200 [1.088287]\n",
      "train:  0 201 0.7533861\n",
      "train:  0 202 1.5494466\n",
      "train:  0 203 0.86113685\n",
      "train:  0 204 0.6190224\n",
      "train:  0 205 0.8135077\n",
      "train:  0 206 0.97699267\n",
      "train:  0 207 0.746353\n",
      "train:  0 208 1.719679\n",
      "train:  0 209 1.3204966\n",
      "train:  0 210 0.7608874\n",
      "train:  0 211 0.9653693\n",
      "train:  0 212 0.7721651\n",
      "train:  0 213 0.99971664\n",
      "train:  0 214 0.9587483\n",
      "train:  0 215 0.8001646\n",
      "train:  0 216 0.6388235\n",
      "train:  0 217 1.5000343\n",
      "train:  0 218 0.839705\n",
      "train:  0 219 0.76790005\n",
      "train:  0 220 0.77322364\n",
      "evalu:  0 220 [1.0814668]\n",
      "train:  0 221 0.89926195\n",
      "train:  0 222 0.7931491\n",
      "train:  0 223 0.7122148\n",
      "train:  0 224 0.7706041\n",
      "train:  0 225 1.7011803\n",
      "train:  0 226 0.82091504\n",
      "train:  0 227 1.9031638\n",
      "train:  0 228 0.7327557\n",
      "train:  0 229 1.0570639\n",
      "train:  0 230 1.2980881\n",
      "train:  0 231 0.7439114\n",
      "train:  0 232 0.67290026\n",
      "train:  0 233 0.7588477\n",
      "train:  0 234 0.82326245\n",
      "train:  0 235 1.1201707\n",
      "train:  0 236 0.9801693\n",
      "train:  0 237 0.75839484\n",
      "train:  0 238 1.2500335\n",
      "train:  0 239 1.2949768\n",
      "train:  0 240 0.68679476\n",
      "evalu:  0 240 [1.0748649]\n",
      "train:  0 241 1.1902671\n",
      "train:  0 242 0.9927956\n",
      "train:  0 243 1.0297108\n",
      "train:  0 244 0.9225326\n",
      "train:  0 245 0.9938833\n",
      "train:  0 246 0.7007234\n",
      "train:  0 247 0.7365544\n",
      "train:  0 248 0.76070005\n",
      "train:  0 249 1.1063673\n",
      "train:  0 250 0.73858535\n",
      "train:  0 251 0.6858578\n",
      "train:  0 252 0.7167253\n",
      "train:  0 253 0.7747068\n",
      "train:  0 254 1.0941538\n",
      "train:  0 255 0.77507025\n",
      "train:  0 256 0.8192593\n",
      "train:  0 257 0.7541314\n",
      "train:  0 258 1.2279752\n",
      "train:  0 259 1.3157797\n",
      "train:  0 260 1.0550431\n",
      "evalu:  0 260 [1.0687064]\n",
      "train:  0 261 0.9425806\n",
      "train:  0 262 0.72090924\n",
      "train:  0 263 1.7133143\n",
      "train:  0 264 1.210563\n",
      "train:  0 265 1.1843458\n",
      "train:  0 266 0.86327106\n",
      "train:  0 267 1.2699853\n",
      "train:  0 268 0.8748741\n",
      "train:  0 269 1.1373353\n",
      "train:  0 270 0.9242279\n",
      "train:  0 271 0.8674153\n",
      "train:  0 272 0.7519821\n",
      "train:  0 273 0.8764053\n",
      "train:  0 274 0.69699335\n",
      "train:  0 275 0.8435858\n",
      "train:  0 276 0.83194476\n",
      "train:  0 277 1.0390356\n",
      "train:  0 278 1.3672153\n",
      "train:  0 279 0.8484983\n",
      "train:  0 280 0.8128716\n",
      "evalu:  0 280 [1.063185]\n",
      "train:  0 281 0.8113908\n",
      "train:  0 282 0.92668766\n",
      "train:  0 283 0.8568705\n",
      "train:  0 284 0.87570673\n",
      "train:  0 285 0.9473117\n",
      "train:  0 286 0.8001513\n",
      "train:  0 287 0.8823108\n",
      "train:  0 288 0.7313997\n",
      "train:  0 289 0.8372533\n",
      "train:  0 290 0.9362904\n",
      "train:  0 291 0.98094887\n",
      "train:  0 292 0.6851315\n",
      "train:  0 293 1.7433463\n",
      "train:  0 294 0.78673106\n",
      "train:  0 295 1.1236638\n",
      "train:  0 296 3.621517\n",
      "train:  0 297 0.9040646\n",
      "train:  0 298 0.827119\n",
      "train:  0 299 0.8756376\n",
      "train:  0 300 1.4228092\n",
      "evalu:  0 300 [1.0582966]\n",
      "train:  0 301 0.7172874\n",
      "train:  0 302 0.7866486\n",
      "train:  0 303 0.9183886\n",
      "train:  0 304 0.9358815\n",
      "train:  0 305 0.57948273\n",
      "train:  0 306 0.9209262\n",
      "train:  0 307 0.7543754\n",
      "train:  0 308 0.8056868\n",
      "train:  0 309 1.5221672\n",
      "train:  0 310 1.0660933\n",
      "train:  0 311 0.89999926\n",
      "train:  0 312 0.80110496\n",
      "train:  0 313 1.1993567\n",
      "train:  0 314 1.2233632\n",
      "train:  0 315 0.82295734\n",
      "train:  0 316 0.76565665\n",
      "train:  0 317 1.0416024\n",
      "train:  0 318 0.6890207\n",
      "train:  0 319 1.4937522\n",
      "train:  0 320 1.6273428\n",
      "evalu:  0 320 [1.054043]\n",
      "train:  0 321 0.75913787\n",
      "train:  0 322 0.7927592\n",
      "train:  0 323 0.6976379\n",
      "train:  0 324 0.8353806\n",
      "train:  0 325 0.9849617\n",
      "train:  0 326 1.1524205\n",
      "train:  0 327 0.70166373\n",
      "train:  0 328 0.68380404\n",
      "train:  0 329 0.86015934\n",
      "train:  0 330 1.5274762\n",
      "train:  0 331 0.61844456\n",
      "train:  0 332 0.89029974\n",
      "train:  0 333 2.9663906\n",
      "train:  0 334 0.8021683\n",
      "train:  0 335 0.6144588\n",
      "train:  0 336 0.62519467\n",
      "train:  0 337 0.88177156\n",
      "train:  0 338 2.5548306\n",
      "train:  0 339 0.8742125\n",
      "train:  0 340 0.8322325\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evalu:  0 340 [1.0503117]\n",
      "train:  0 341 0.8726758\n",
      "train:  0 342 0.9474022\n",
      "train:  0 343 0.7768831\n",
      "train:  0 344 0.7294697\n",
      "train:  0 345 0.7578459\n",
      "train:  0 346 0.7411533\n",
      "train:  0 347 0.60353196\n",
      "train:  0 348 0.8069346\n",
      "train:  0 349 0.92399675\n",
      "train:  0 350 1.8965787\n",
      "train:  0 351 0.9033384\n",
      "train:  0 352 0.66389716\n",
      "train:  0 353 0.86067903\n",
      "train:  0 354 0.8178473\n",
      "train:  0 355 1.1127583\n",
      "train:  0 356 0.67781526\n",
      "train:  0 357 0.748368\n",
      "train:  0 358 1.007356\n",
      "train:  0 359 0.72736686\n",
      "train:  0 360 0.734645\n",
      "evalu:  0 360 [1.0466993]\n",
      "train:  0 361 1.0393627\n",
      "train:  0 362 1.1794387\n",
      "train:  0 363 0.82918304\n",
      "train:  0 364 0.8529618\n",
      "train:  0 365 0.73344237\n",
      "train:  0 366 0.81827456\n",
      "train:  0 367 2.113022\n",
      "train:  0 368 1.0593925\n",
      "train:  0 369 1.2853152\n",
      "train:  0 370 0.88700765\n",
      "train:  0 371 0.9962818\n",
      "train:  0 372 0.6535548\n",
      "train:  0 373 2.2629473\n",
      "train:  0 374 0.9303637\n",
      "train:  0 375 0.90361166\n",
      "train:  0 376 0.6802382\n",
      "train:  0 377 1.255153\n",
      "train:  0 378 0.7796134\n",
      "train:  0 379 1.1434472\n",
      "train:  0 380 0.71054506\n",
      "evalu:  0 380 [1.0432042]\n",
      "train:  0 381 2.7285893\n",
      "train:  0 382 1.0602177\n",
      "train:  0 383 1.0340631\n",
      "train:  0 384 1.2240564\n",
      "train:  0 385 0.86367434\n",
      "train:  0 386 1.0881746\n",
      "train:  0 387 0.9594395\n",
      "train:  0 388 0.86446494\n",
      "train:  0 389 1.1403459\n",
      "train:  0 390 0.8940154\n",
      "train:  0 391 1.413626\n",
      "train:  0 392 0.89058334\n",
      "train:  0 393 1.4876846\n",
      "train:  0 394 0.676092\n",
      "train:  0 395 0.81019276\n",
      "train:  0 396 1.1569235\n",
      "train:  0 397 1.3827357\n",
      "train:  0 398 1.0770494\n",
      "train:  0 399 0.7067286\n",
      "train:  0 400 0.7709462\n",
      "evalu:  0 400 [1.0395198]\n",
      "train:  0 401 0.8798703\n",
      "train:  0 402 0.5349459\n",
      "train:  0 403 0.74457306\n",
      "train:  0 404 0.61614114\n",
      "train:  0 405 0.5740261\n",
      "train:  0 406 0.7225293\n",
      "train:  0 407 0.7176195\n",
      "train:  0 408 0.8068265\n",
      "train:  0 409 0.7850003\n",
      "train:  0 410 0.84866667\n",
      "train:  0 411 0.6268811\n",
      "train:  0 412 0.65052944\n",
      "train:  0 413 0.68673503\n",
      "train:  0 414 0.7893776\n",
      "train:  0 415 0.69436246\n",
      "train:  0 416 0.72309667\n",
      "train:  0 417 0.6687867\n",
      "train:  0 418 1.0402669\n",
      "train:  0 419 1.7461364\n",
      "train:  0 420 0.78599983\n",
      "evalu:  0 420 [1.0357819]\n",
      "train:  0 421 0.88978255\n",
      "train:  0 422 0.6751086\n",
      "train:  0 423 0.7273344\n",
      "train:  0 424 1.3163543\n",
      "train:  0 425 2.0579824\n",
      "train:  0 426 1.4924064\n",
      "train:  0 427 0.77933496\n",
      "train:  0 428 0.836389\n",
      "train:  0 429 0.6828604\n",
      "train:  0 430 1.1413392\n",
      "train:  0 431 0.71209025\n",
      "train:  0 432 2.3487976\n",
      "train:  0 433 0.7510091\n",
      "train:  0 434 0.8683366\n",
      "train:  0 435 0.7436629\n",
      "train:  0 436 0.75354624\n",
      "train:  0 437 0.7129559\n",
      "train:  0 438 0.90455675\n",
      "train:  0 439 0.7288393\n",
      "train:  0 440 0.77403295\n",
      "evalu:  0 440 [1.0324094]\n",
      "train:  0 441 0.7666951\n",
      "train:  0 442 0.9033637\n",
      "train:  0 443 1.0410918\n",
      "train:  0 444 0.83946615\n",
      "train:  0 445 0.85810536\n",
      "train:  0 446 0.8571996\n",
      "train:  0 447 0.8105168\n",
      "train:  0 448 0.7132011\n",
      "train:  0 449 0.776984\n",
      "train:  0 450 0.7947408\n",
      "train:  0 451 0.8227668\n",
      "train:  0 452 1.3136513\n",
      "train:  0 453 0.8035334\n",
      "train:  0 454 0.84204656\n",
      "train:  0 455 0.52176595\n",
      "train:  0 456 0.6447623\n",
      "train:  0 457 0.7555686\n",
      "train:  0 458 0.99401695\n",
      "train:  0 459 0.6075283\n",
      "train:  0 460 0.6643538\n",
      "evalu:  0 460 [1.0291371]\n",
      "train:  0 461 1.0778222\n",
      "train:  0 462 0.7408291\n",
      "train:  0 463 0.6812197\n",
      "train:  0 464 0.8772645\n",
      "train:  0 465 0.9406442\n",
      "train:  0 466 0.81785995\n",
      "train:  0 467 1.313469\n",
      "train:  0 468 1.0152847\n",
      "train:  0 469 0.79973406\n",
      "train:  0 470 0.9549902\n",
      "train:  0 471 0.77854776\n",
      "train:  0 472 1.4091966\n",
      "train:  0 473 0.823738\n",
      "train:  0 474 1.9654816\n",
      "train:  0 475 0.7905037\n",
      "train:  0 476 0.671459\n",
      "train:  0 477 0.6336877\n",
      "train:  0 478 1.3027909\n",
      "train:  0 479 0.689231\n",
      "train:  0 480 0.74543667\n",
      "evalu:  0 480 [1.0262552]\n",
      "train:  0 481 0.8351035\n",
      "train:  0 482 1.0740597\n",
      "train:  0 483 0.61950594\n",
      "train:  0 484 1.4900277\n",
      "train:  0 485 0.87133706\n",
      "train:  0 486 0.7535324\n",
      "train:  0 487 0.67614985\n",
      "train:  0 488 0.99095786\n",
      "train:  0 489 0.60330504\n",
      "train:  0 490 1.2185444\n",
      "train:  0 491 1.3393443\n",
      "train:  0 492 1.6843047\n",
      "train:  0 493 0.69100016\n",
      "train:  0 494 0.6302052\n",
      "train:  0 495 0.8123503\n",
      "train:  0 496 0.7472949\n",
      "train:  0 497 0.6098838\n",
      "train:  0 498 0.7380019\n",
      "train:  0 499 0.7927661\n",
      "train:  0 500 0.6563656\n",
      "evalu:  0 500 [1.0235506]\n",
      "train:  0 501 0.96716547\n",
      "train:  0 502 1.4126729\n",
      "train:  0 503 0.6474361\n",
      "train:  0 504 1.2605145\n",
      "train:  0 505 0.94074315\n",
      "train:  0 506 0.87557536\n",
      "train:  0 507 1.716793\n",
      "train:  0 508 1.0019953\n",
      "train:  0 509 0.9959002\n",
      "train:  0 510 0.92817\n",
      "train:  0 511 0.87903994\n",
      "train:  0 512 1.1231229\n",
      "train:  0 513 0.6067003\n",
      "train:  0 514 0.82451564\n",
      "train:  0 515 0.9710023\n",
      "train:  0 516 0.7083942\n",
      "train:  0 517 1.0020529\n",
      "train:  0 518 1.1604449\n",
      "train:  0 519 0.572874\n",
      "train:  0 520 0.7254933\n",
      "evalu:  0 520 [1.0209268]\n",
      "train:  0 521 0.6726438\n",
      "train:  0 522 1.227438\n",
      "train:  0 523 1.4992118\n",
      "train:  0 524 1.9512008\n",
      "train:  0 525 0.9805579\n",
      "train:  0 526 0.7847147\n",
      "train:  0 527 0.9201266\n",
      "train:  0 528 0.67464596\n",
      "train:  0 529 0.8730339\n",
      "train:  0 530 0.7918343\n",
      "train:  0 531 0.7915596\n",
      "train:  0 532 0.7636993\n",
      "train:  0 533 0.8059567\n",
      "train:  0 534 0.7944476\n",
      "train:  0 535 1.2513324\n",
      "train:  0 536 2.052772\n",
      "train:  0 537 0.76861906\n",
      "train:  0 538 0.5319318\n",
      "train:  0 539 0.6790205\n",
      "train:  0 540 0.80063754\n",
      "evalu:  0 540 [1.018314]\n",
      "train:  0 541 0.6185992\n",
      "train:  0 542 0.725575\n",
      "train:  0 543 0.6765018\n",
      "train:  0 544 0.7436693\n",
      "train:  0 545 0.50920576\n",
      "train:  0 546 0.6389499\n",
      "train:  0 547 0.6660424\n",
      "train:  0 548 0.9004672\n",
      "train:  0 549 0.5553922\n",
      "train:  0 550 0.71046484\n",
      "train:  0 551 0.85328907\n",
      "train:  0 552 0.86885524\n",
      "train:  0 553 0.6007004\n",
      "train:  0 554 0.7806561\n",
      "train:  0 555 0.76672536\n",
      "train:  0 556 0.6321089\n",
      "train:  0 557 2.4742851\n",
      "train:  0 558 0.6610235\n",
      "train:  0 559 0.73933214\n",
      "train:  0 560 0.60650957\n",
      "evalu:  0 560 [1.0161929]\n",
      "train:  0 561 0.75436705\n",
      "train:  0 562 0.68178874\n",
      "train:  0 563 0.81438535\n",
      "train:  0 564 0.9114826\n",
      "train:  0 565 0.7063934\n",
      "train:  0 566 1.3538276\n",
      "train:  0 567 1.4920328\n",
      "train:  0 568 0.98870265\n",
      "train:  0 569 1.522729\n",
      "train:  0 570 0.9494703\n",
      "train:  0 571 1.3231188\n",
      "train:  0 572 1.3114423\n",
      "train:  0 573 0.76384324\n",
      "train:  0 574 0.9368851\n",
      "train:  0 575 0.5610495\n",
      "train:  0 576 1.1744412\n",
      "train:  0 577 0.744454\n",
      "train:  0 578 1.111788\n",
      "train:  0 579 0.8990162\n",
      "train:  0 580 0.8009821\n",
      "evalu:  0 580 [1.0137225]\n",
      "train:  0 581 0.62088096\n",
      "train:  0 582 1.3404139\n",
      "train:  0 583 0.6959928\n",
      "train:  0 584 0.6949385\n",
      "train:  0 585 0.85173774\n",
      "train:  0 586 0.9994376\n",
      "train:  0 587 0.5753681\n",
      "train:  0 588 0.68604475\n",
      "train:  0 589 0.6711845\n",
      "train:  0 590 0.6709044\n",
      "train:  0 591 0.7610794\n",
      "train:  0 592 1.5015068\n",
      "train:  0 593 1.758835\n",
      "train:  0 594 0.8356911\n",
      "train:  0 595 0.8831162\n",
      "train:  0 596 0.92509407\n",
      "train:  0 597 1.0486767\n",
      "train:  0 598 0.5982809\n",
      "train:  0 599 0.6732391\n",
      "train:  0 600 0.74986345\n",
      "evalu:  0 600 [1.0112406]\n",
      "train:  0 601 0.86446434\n",
      "train:  0 602 1.4558457\n",
      "train:  0 603 0.95193285\n",
      "train:  0 604 1.555319\n",
      "train:  0 605 0.67118984\n",
      "train:  0 606 0.7575481\n",
      "train:  0 607 0.661665\n",
      "train:  0 608 0.8929198\n",
      "train:  0 609 0.90018183\n",
      "train:  0 610 1.1996987\n",
      "train:  0 611 0.7133646\n",
      "train:  0 612 1.3815129\n",
      "train:  0 613 0.8902463\n",
      "train:  0 614 0.7657998\n",
      "train:  0 615 0.58751315\n",
      "train:  0 616 0.6322661\n",
      "train:  0 617 0.6971898\n",
      "train:  0 618 3.487146\n",
      "train:  0 619 0.6366495\n",
      "train:  0 620 0.9310897\n",
      "evalu:  0 620 [1.0090551]\n",
      "train:  0 621 1.2926867\n",
      "train:  0 622 0.6816713\n",
      "train:  0 623 0.60687625\n",
      "train:  0 624 0.91225356\n",
      "train:  0 625 1.0383015\n",
      "train:  0 626 0.80237335\n",
      "train:  0 627 0.7314932\n",
      "train:  0 628 1.2408478\n",
      "train:  0 629 0.8789712\n",
      "train:  0 630 0.8009216\n",
      "train:  0 631 0.6465389\n",
      "train:  0 632 0.6974616\n",
      "train:  0 633 0.69521844\n",
      "train:  0 634 0.7446541\n",
      "train:  0 635 0.7387461\n",
      "train:  0 636 0.7372136\n",
      "train:  0 637 0.9093605\n",
      "train:  0 638 0.88276386\n",
      "train:  0 639 1.0864338\n",
      "train:  0 640 0.7069166\n",
      "evalu:  0 640 [1.0068377]\n",
      "train:  0 641 0.8697412\n",
      "train:  0 642 0.90775913\n",
      "train:  0 643 0.8605726\n",
      "train:  0 644 0.5978155\n",
      "train:  0 645 0.81869775\n",
      "train:  0 646 0.6201991\n",
      "train:  0 647 0.73729867\n",
      "train:  0 648 0.56019336\n",
      "train:  0 649 1.9875787\n",
      "train:  0 650 0.8025383\n",
      "train:  0 651 1.0054572\n",
      "train:  0 652 0.78349113\n",
      "train:  0 653 0.9729624\n",
      "train:  0 654 0.67704475\n",
      "train:  0 655 0.7002041\n",
      "train:  0 656 1.521114\n",
      "train:  0 657 0.6798636\n",
      "train:  0 658 1.0590632\n",
      "train:  0 659 0.9585371\n",
      "train:  0 660 0.61919206\n",
      "evalu:  0 660 [1.0048572]\n",
      "train:  0 661 1.1475219\n",
      "train:  0 662 1.1644342\n",
      "train:  0 663 1.1078001\n",
      "train:  0 664 0.9375835\n",
      "train:  0 665 0.848901\n",
      "train:  0 666 1.823208\n",
      "train:  0 667 0.65444386\n",
      "train:  0 668 0.7988161\n",
      "train:  0 669 0.64785695\n",
      "train:  0 670 0.5931486\n",
      "train:  0 671 0.61264277\n",
      "train:  0 672 0.9070292\n",
      "train:  0 673 0.7689087\n",
      "train:  0 674 0.6181385\n",
      "train:  0 675 1.0695164\n",
      "train:  0 676 1.9492133\n",
      "train:  0 677 0.72921026\n",
      "train:  0 678 1.1079441\n",
      "train:  0 679 0.9318322\n",
      "train:  0 680 0.62974006\n",
      "evalu:  0 680 [1.0030538]\n",
      "train:  0 681 0.850362\n",
      "train:  0 682 0.5986211\n",
      "train:  0 683 0.81906134\n",
      "train:  0 684 0.62628615\n",
      "train:  0 685 0.8422878\n",
      "train:  0 686 0.63957834\n",
      "train:  0 687 1.3068471\n",
      "train:  0 688 1.7100505\n",
      "train:  0 689 0.6791\n",
      "train:  0 690 1.5488405\n",
      "train:  0 691 0.8063151\n",
      "train:  0 692 0.80490404\n",
      "train:  0 693 0.83208984\n",
      "train:  0 694 0.7883829\n",
      "train:  0 695 0.92383945\n",
      "train:  0 696 0.6767412\n",
      "train:  0 697 0.65929705\n",
      "train:  0 698 0.6597168\n",
      "train:  0 699 0.8549012\n",
      "train:  0 700 0.7868685\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "evalu:  0 700 [1.001342]\n",
      "train:  0 701 0.6035923\n",
      "train:  0 702 0.6480806\n",
      "train:  0 703 0.6183539\n",
      "train:  0 704 0.6052943\n",
      "train:  0 705 0.64395314\n",
      "train:  0 706 0.89964354\n",
      "train:  0 707 0.65537035\n",
      "train:  0 708 0.71801156\n",
      "train:  0 709 0.67213064\n",
      "train:  0 710 0.704797\n",
      "train:  0 711 0.9032457\n",
      "train:  0 712 0.89001995\n",
      "train:  0 713 1.1490571\n",
      "train:  0 714 0.59612846\n",
      "train:  0 715 1.299995\n",
      "train:  0 716 0.8797443\n",
      "train:  0 717 0.60999304\n",
      "train:  0 718 0.68360007\n",
      "train:  0 719 0.86236346\n",
      "train:  0 720 0.9827771\n",
      "evalu:  0 720 [0.9997313]\n",
      "train:  0 721 0.7488405\n",
      "train:  0 722 1.0683244\n",
      "train:  0 723 0.9847111\n",
      "train:  0 724 1.1353264\n",
      "train:  0 725 0.63723683\n",
      "train:  0 726 0.9868592\n",
      "train:  0 727 0.83359945\n",
      "train:  0 728 0.72151816\n",
      "train:  0 729 0.7236063\n",
      "train:  0 730 1.305177\n",
      "train:  0 731 0.7959205\n",
      "train:  0 732 0.6329341\n",
      "train:  0 733 0.8686897\n",
      "train:  0 734 0.9440889\n",
      "train:  0 735 0.8509927\n",
      "train:  0 736 0.8231232\n",
      "train:  0 737 2.1975093\n",
      "train:  0 738 0.83826286\n",
      "train:  0 739 0.7683052\n",
      "train:  0 740 0.6242326\n",
      "evalu:  0 740 [0.998032]\n",
      "train:  0 741 0.9238577\n",
      "train:  0 742 0.85603786\n",
      "train:  0 743 0.72824913\n",
      "train:  0 744 0.727603\n",
      "train:  0 745 0.91825473\n",
      "train:  0 746 1.3385708\n",
      "train:  0 747 0.8304082\n",
      "train:  0 748 0.9956181\n",
      "train:  0 749 1.2600979\n",
      "train:  0 750 0.5645336\n",
      "train:  0 751 1.2626823\n",
      "train:  0 752 0.860726\n",
      "train:  0 753 0.73141396\n",
      "train:  0 754 0.7784974\n",
      "train:  0 755 1.000048\n",
      "train:  0 756 0.73706704\n",
      "train:  0 757 0.7993721\n",
      "train:  0 758 1.0212342\n",
      "train:  0 759 0.79185134\n",
      "train:  0 760 0.66841936\n",
      "evalu:  0 760 [0.9963029]\n",
      "train:  0 761 0.7384517\n",
      "train:  0 762 2.058873\n",
      "train:  0 763 0.72035414\n",
      "train:  0 764 0.7545465\n",
      "train:  0 765 0.72996026\n",
      "train:  0 766 0.62678367\n",
      "train:  0 767 0.5438639\n",
      "train:  0 768 0.8709804\n",
      "train:  0 769 0.91146713\n",
      "train:  0 770 1.0066704\n",
      "train:  0 771 0.8958395\n",
      "train:  0 772 0.64244986\n",
      "train:  0 773 0.8687274\n",
      "train:  0 774 0.8056962\n",
      "train:  0 775 0.6332644\n",
      "train:  0 776 0.92824286\n",
      "train:  0 777 1.0007507\n",
      "train:  0 778 0.7611284\n",
      "train:  0 779 0.8755814\n",
      "train:  0 780 0.6437933\n",
      "evalu:  0 780 [0.9947191]\n",
      "train:  0 781 0.6917392\n",
      "train:  0 782 0.57260495\n",
      "train:  0 783 0.6373361\n",
      "train:  0 784 0.7439245\n",
      "train:  0 785 0.7444852\n",
      "train:  0 786 0.7089463\n",
      "train:  0 787 0.8104032\n",
      "train:  0 788 0.5860404\n",
      "train:  0 789 0.88030976\n",
      "train:  0 790 0.9543304\n",
      "train:  0 791 0.7021766\n",
      "train:  0 792 0.8913813\n",
      "train:  0 793 1.1605659\n",
      "train:  0 794 0.9997791\n",
      "train:  0 795 1.2361687\n",
      "train:  0 796 0.6295754\n",
      "train:  0 797 0.99210835\n",
      "train:  0 798 0.8113617\n",
      "train:  0 799 0.5361918\n",
      "train:  0 800 0.9740547\n",
      "evalu:  0 800 [0.9933377]\n",
      "train:  0 801 0.67435396\n",
      "train:  0 802 0.6454267\n",
      "train:  0 803 0.6027644\n",
      "train:  0 804 0.66335684\n",
      "train:  0 805 0.6913399\n",
      "train:  0 806 0.6016414\n",
      "train:  0 807 1.3173752\n",
      "train:  0 808 0.6667456\n",
      "train:  0 809 0.6791095\n",
      "train:  0 810 0.9542043\n",
      "train:  0 811 0.62027025\n",
      "train:  0 812 0.8291356\n",
      "train:  0 813 1.0331966\n",
      "train:  0 814 3.8360026\n",
      "train:  0 815 1.1075745\n",
      "train:  0 816 0.65235645\n",
      "train:  0 817 0.63499886\n",
      "train:  0 818 0.6252678\n",
      "train:  0 819 0.7295026\n",
      "train:  0 820 0.7875153\n",
      "evalu:  0 820 [0.9918332]\n",
      "train:  0 821 0.7550281\n",
      "train:  0 822 0.7062182\n",
      "train:  0 823 0.78826255\n",
      "train:  0 824 0.691722\n",
      "train:  0 825 0.61700207\n",
      "train:  0 826 1.3520037\n",
      "train:  0 827 2.2734528\n",
      "train:  0 828 1.0983026\n",
      "train:  0 829 0.78834724\n",
      "train:  0 830 1.1078923\n",
      "train:  0 831 1.0745502\n",
      "train:  0 832 0.6565818\n",
      "train:  0 833 0.7406553\n",
      "train:  0 834 0.76463634\n",
      "train:  0 835 0.7385329\n",
      "train:  0 836 0.8832213\n",
      "train:  0 837 0.661734\n",
      "train:  0 838 0.6200519\n",
      "train:  0 839 0.6475119\n",
      "train:  0 840 1.4668424\n",
      "evalu:  0 840 [0.9905453]\n",
      "train:  0 841 0.85377914\n",
      "train:  0 842 0.6815397\n",
      "train:  0 843 1.5412214\n",
      "train:  0 844 0.64361477\n",
      "train:  0 845 0.670827\n",
      "train:  0 846 0.7715837\n",
      "train:  0 847 0.6774985\n",
      "train:  0 848 1.681369\n",
      "train:  0 849 1.0298347\n",
      "train:  0 850 0.6485563\n",
      "train:  0 851 0.6632041\n",
      "train:  0 852 1.0513079\n",
      "train:  0 853 1.4970052\n",
      "train:  0 854 0.6132722\n",
      "train:  0 855 0.7630966\n",
      "train:  0 856 0.64623356\n",
      "train:  0 857 0.86409056\n",
      "train:  0 858 1.6412793\n",
      "train:  0 859 1.0167077\n",
      "train:  0 860 0.6537838\n",
      "evalu:  0 860 [0.9890041]\n",
      "train:  0 861 0.8265467\n",
      "train:  0 862 0.7988996\n",
      "train:  0 863 0.5193696\n",
      "train:  0 864 0.69948256\n",
      "train:  0 865 0.790553\n",
      "train:  0 866 1.0577788\n",
      "train:  0 867 0.5322441\n",
      "train:  0 868 0.6485976\n",
      "train:  0 869 0.85184735\n",
      "train:  0 870 1.2301542\n",
      "train:  0 871 0.83302504\n",
      "train:  0 872 0.8651943\n",
      "train:  0 873 0.764076\n",
      "train:  0 874 0.6436534\n",
      "train:  0 875 0.89278287\n",
      "train:  0 876 0.85879844\n",
      "train:  0 877 1.1548238\n",
      "train:  0 878 1.4824691\n",
      "train:  0 879 0.57421225\n",
      "train:  0 880 0.8110773\n",
      "evalu:  0 880 [0.98758024]\n",
      "train:  0 881 0.8672968\n",
      "train:  0 882 0.9015165\n",
      "train:  0 883 0.69024885\n",
      "train:  0 884 1.23628\n",
      "train:  0 885 0.79498684\n",
      "train:  0 886 0.65129477\n",
      "train:  0 887 1.2463462\n",
      "train:  0 888 1.0655146\n",
      "train:  0 889 0.5792093\n",
      "train:  0 890 0.8818099\n",
      "train:  0 891 0.6810867\n",
      "train:  0 892 0.78828067\n",
      "train:  0 893 0.7796499\n",
      "train:  0 894 0.6260361\n",
      "train:  0 895 1.8882393\n",
      "train:  0 896 0.94880325\n",
      "train:  0 897 0.5943174\n",
      "train:  0 898 0.8825456\n",
      "train:  0 899 0.7370187\n",
      "train:  0 900 0.7622533\n",
      "evalu:  0 900 [0.98646665]\n",
      "train:  0 901 0.6086731\n",
      "train:  0 902 1.1604953\n",
      "train:  0 903 0.78603107\n",
      "train:  0 904 0.71954346\n",
      "train:  0 905 0.78911287\n",
      "train:  0 906 0.7452982\n",
      "train:  0 907 1.181687\n",
      "train:  0 908 0.6526836\n",
      "train:  0 909 0.8578052\n",
      "train:  0 910 0.73018503\n",
      "train:  0 911 0.5828849\n",
      "train:  0 912 0.8820555\n",
      "train:  0 913 0.7978317\n",
      "train:  0 914 0.64471877\n",
      "train:  0 915 0.91911465\n",
      "train:  0 916 0.9222905\n",
      "train:  0 917 0.65452987\n",
      "train:  0 918 1.2342609\n",
      "train:  0 919 1.1326561\n",
      "train:  0 920 0.7339357\n",
      "evalu:  0 920 [0.98508906]\n",
      "train:  0 921 0.98725474\n",
      "train:  0 922 0.8263244\n",
      "train:  0 923 0.5474755\n",
      "train:  0 924 0.93078464\n",
      "train:  0 925 0.7224408\n",
      "train:  0 926 0.7718168\n",
      "train:  0 927 1.2193024\n",
      "train:  0 928 0.5959881\n",
      "train:  0 929 0.573952\n",
      "train:  0 930 1.0955579\n",
      "train:  0 931 0.5224932\n",
      "train:  0 932 0.820965\n",
      "train:  0 933 0.7867147\n",
      "train:  0 934 0.80409634\n",
      "train:  0 935 0.5844756\n",
      "train:  0 936 0.5434071\n",
      "train:  0 937 0.55025136\n",
      "train:  0 938 0.7276132\n",
      "train:  0 939 0.6166391\n",
      "train:  0 940 0.70443606\n",
      "evalu:  0 940 [0.9836175]\n",
      "train:  0 941 0.80511415\n",
      "train:  0 942 0.6698013\n",
      "train:  0 943 0.7136101\n",
      "train:  0 944 0.60489005\n",
      "train:  0 945 0.6497737\n",
      "train:  0 946 0.63568026\n",
      "train:  0 947 0.6489901\n",
      "train:  0 948 1.193939\n",
      "train:  0 949 0.8795959\n",
      "train:  0 950 0.7106885\n",
      "train:  0 951 0.9253094\n",
      "train:  0 952 2.0881138\n",
      "train:  0 953 0.7281539\n",
      "train:  0 954 0.9179407\n",
      "train:  0 955 0.74465233\n",
      "train:  0 956 0.7081511\n",
      "train:  0 957 0.55832624\n",
      "train:  0 958 0.8255061\n",
      "train:  0 959 0.62256277\n",
      "train:  0 960 1.2010882\n",
      "evalu:  0 960 [0.9820085]\n",
      "train:  0 961 0.70674676\n",
      "train:  0 962 2.7608507\n",
      "train:  0 963 1.0899199\n",
      "train:  0 964 0.8430198\n",
      "train:  0 965 0.73992985\n",
      "train:  0 966 1.19039\n",
      "train:  0 967 0.64768845\n",
      "train:  0 968 1.0217346\n",
      "train:  0 969 0.70539385\n",
      "train:  0 970 0.6116512\n",
      "train:  0 971 0.6183348\n",
      "train:  0 972 0.7080053\n",
      "train:  0 973 2.2897651\n",
      "train:  0 974 0.52626425\n",
      "train:  0 975 0.61116105\n",
      "train:  0 976 0.81704646\n",
      "train:  0 977 0.75418025\n",
      "train:  0 978 0.7916852\n",
      "train:  0 979 0.62172663\n",
      "train:  0 980 0.80222356\n",
      "evalu:  0 980 [0.9803077]\n",
      "train:  0 981 0.91949\n",
      "train:  0 982 0.7554724\n",
      "train:  0 983 0.9438077\n",
      "train:  0 984 1.2164454\n",
      "train:  0 985 1.3496834\n",
      "train:  0 986 0.74800634\n",
      "train:  0 987 0.7378592\n",
      "train:  0 988 1.5025929\n",
      "train:  0 989 0.7859087\n",
      "train:  0 990 0.5810523\n",
      "train:  0 991 0.7334266\n",
      "train:  0 992 0.60779816\n",
      "train:  0 993 0.80287784\n",
      "train:  0 994 0.6821447\n",
      "train:  0 995 1.3384004\n",
      "train:  0 996 0.8207496\n",
      "train:  0 997 0.651223\n",
      "train:  0 998 1.1479597\n",
      "train:  0 999 0.64523804\n",
      "train:  0 1000 0.6963645\n",
      "evalu:  0 1000 [0.9783063]\n",
      "train:  0 1001 0.7972867\n",
      "train:  0 1002 0.71586716\n",
      "train:  0 1003 0.7068319\n",
      "train:  0 1004 0.7047456\n",
      "train:  0 1005 0.70520467\n",
      "train:  0 1006 0.6923503\n",
      "train:  0 1007 0.8480074\n",
      "train:  0 1008 1.232641\n",
      "train:  0 1009 0.8262155\n",
      "train:  0 1010 1.1449392\n",
      "train:  0 1011 0.8354492\n",
      "train:  0 1012 1.4042408\n",
      "train:  0 1013 0.7629297\n",
      "train:  0 1014 0.7697155\n",
      "train:  0 1015 1.0470767\n",
      "train:  0 1016 1.1134396\n",
      "train:  0 1017 1.1944932\n",
      "train:  0 1018 0.6074539\n",
      "train:  0 1019 0.6302143\n",
      "train:  0 1020 0.75732195\n",
      "evalu:  0 1020 [0.9765573]\n",
      "train:  0 1021 0.65513295\n",
      "train:  0 1022 0.76853776\n",
      "train:  0 1023 1.3433582\n",
      "train:  0 1024 0.7133735\n",
      "train:  0 1025 0.62937015\n",
      "train:  0 1026 0.7135166\n",
      "train:  0 1027 0.90945196\n",
      "train:  0 1028 1.1318991\n",
      "train:  0 1029 0.7219274\n",
      "train:  0 1030 0.6148565\n",
      "train:  0 1031 0.6524037\n",
      "train:  0 1032 0.8402298\n",
      "train:  0 1033 0.60178757\n",
      "train:  0 1034 0.87780297\n",
      "train:  0 1035 0.6292222\n",
      "train:  0 1036 0.72970355\n",
      "train:  0 1037 0.9115427\n",
      "train:  0 1038 1.0084878\n",
      "train:  0 1039 0.66724855\n",
      "train:  0 1040 0.6634455\n",
      "evalu:  0 1040 [0.97458017]\n",
      "train:  0 1041 0.54759604\n",
      "train:  0 1042 1.145452\n",
      "train:  0 1043 0.800675\n",
      "train:  0 1044 0.6819571\n",
      "train:  0 1045 0.676387\n",
      "train:  0 1046 0.96282715\n",
      "train:  0 1047 0.6401154\n",
      "train:  0 1048 2.1164505\n",
      "train:  0 1049 0.56199664\n",
      "train:  0 1050 1.3940043\n",
      "train:  0 1051 1.5065014\n",
      "train:  0 1052 0.69588023\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  0 1053 0.65626734\n",
      "train:  0 1054 0.67120266\n",
      "train:  0 1055 1.0360036\n",
      "train:  0 1056 0.65548825\n",
      "train:  0 1057 0.5744907\n",
      "train:  0 1058 1.3992527\n",
      "train:  0 1059 0.573332\n",
      "train:  0 1060 0.67490315\n",
      "evalu:  0 1060 [0.9725604]\n",
      "train:  0 1061 0.7137406\n",
      "train:  0 1062 0.67749065\n",
      "train:  0 1063 0.57646304\n",
      "train:  0 1064 1.175021\n",
      "train:  0 1065 0.7015683\n",
      "train:  0 1066 0.6087624\n",
      "train:  0 1067 0.5440642\n",
      "train:  0 1068 0.6689633\n",
      "train:  0 1069 0.64796764\n",
      "train:  0 1070 0.73963606\n",
      "train:  0 1071 0.7409759\n",
      "train:  0 1072 0.6864808\n",
      "train:  0 1073 0.88839406\n",
      "train:  0 1074 0.75387746\n",
      "train:  0 1075 0.68851626\n",
      "train:  0 1076 0.94491464\n",
      "train:  0 1077 0.73860544\n",
      "train:  0 1078 1.047606\n",
      "train:  0 1079 0.9182543\n",
      "train:  0 1080 0.732986\n",
      "evalu:  0 1080 [0.97063243]\n",
      "train:  0 1081 1.3548778\n",
      "train:  0 1082 0.795696\n",
      "train:  0 1083 2.1093514\n",
      "train:  0 1084 0.7736961\n",
      "train:  0 1085 1.3839289\n",
      "train:  0 1086 0.84636515\n",
      "train:  0 1087 2.33511\n",
      "train:  0 1088 0.8208681\n",
      "train:  0 1089 0.6491308\n",
      "train:  0 1090 1.2358099\n",
      "train:  0 1091 0.7158579\n",
      "train:  0 1092 0.87986815\n",
      "train:  0 1093 0.659773\n",
      "train:  0 1094 0.7531226\n",
      "train:  0 1095 0.6264764\n",
      "train:  0 1096 0.6469228\n",
      "train:  0 1097 0.7203832\n",
      "train:  0 1098 1.066\n",
      "train:  0 1099 0.797841\n",
      "train:  0 1100 0.787208\n",
      "evalu:  0 1100 [0.96881944]\n",
      "train:  0 1101 0.8915711\n",
      "train:  0 1102 0.63690704\n",
      "train:  0 1103 0.6744896\n",
      "train:  0 1104 0.6144906\n",
      "train:  0 1105 0.5812819\n",
      "train:  0 1106 0.642227\n",
      "train:  0 1107 0.6682745\n",
      "train:  0 1108 1.3337322\n",
      "train:  0 1109 1.1093701\n",
      "train:  0 1110 0.74915487\n",
      "train:  0 1111 5.657698\n",
      "train:  0 1112 0.96330667\n",
      "train:  0 1113 1.1616412\n",
      "train:  0 1114 0.66220385\n",
      "train:  0 1115 0.5793735\n",
      "train:  0 1116 1.4119303\n",
      "train:  0 1117 0.68052614\n",
      "train:  0 1118 0.91949946\n",
      "train:  0 1119 0.79556876\n",
      "train:  0 1120 0.6293135\n",
      "evalu:  0 1120 [0.96698314]\n",
      "train:  0 1121 1.1505364\n",
      "train:  0 1122 1.1041412\n",
      "train:  0 1123 0.67629105\n",
      "train:  0 1124 0.5550003\n",
      "train:  0 1125 0.96014655\n",
      "train:  0 1126 0.625578\n",
      "train:  0 1127 0.89214516\n",
      "train:  0 1128 1.0513004\n",
      "train:  0 1129 0.56417626\n",
      "train:  0 1130 0.93546605\n",
      "train:  0 1131 0.69804555\n",
      "train:  0 1132 2.2549465\n",
      "train:  0 1133 0.6479356\n",
      "train:  0 1134 0.855561\n",
      "train:  0 1135 0.6681066\n",
      "train:  0 1136 0.66035974\n",
      "train:  0 1137 0.6841348\n",
      "train:  0 1138 0.6701039\n",
      "train:  0 1139 1.689651\n",
      "train:  0 1140 0.53970414\n",
      "evalu:  0 1140 [0.96548444]\n",
      "train:  0 1141 1.0477055\n",
      "train:  0 1142 0.75912625\n",
      "train:  0 1143 0.66346717\n",
      "train:  0 1144 3.0583186\n",
      "train:  0 1145 0.89567614\n",
      "train:  0 1146 1.8869218\n",
      "train:  0 1147 0.7311125\n",
      "train:  0 1148 0.7566782\n",
      "train:  0 1149 0.70703155\n",
      "train:  0 1150 0.725698\n",
      "train:  0 1151 0.8258318\n",
      "train:  0 1152 0.68272054\n",
      "train:  0 1153 0.9606339\n",
      "train:  0 1154 0.6573177\n",
      "train:  0 1155 0.6995366\n",
      "train:  0 1156 0.59233236\n",
      "train:  0 1157 0.89357144\n",
      "train:  0 1158 0.70267576\n",
      "train:  0 1159 0.85617673\n",
      "train:  0 1160 0.72161174\n",
      "evalu:  0 1160 [0.96388024]\n",
      "train:  0 1161 0.7363315\n",
      "train:  0 1162 0.5993565\n",
      "train:  0 1163 0.74571747\n",
      "train:  0 1164 0.9510485\n",
      "train:  0 1165 0.79167336\n",
      "train:  0 1166 0.6850793\n",
      "train:  0 1167 0.93980664\n",
      "train:  0 1168 0.65712005\n",
      "train:  0 1169 0.7233343\n",
      "train:  0 1170 1.4773964\n",
      "train:  0 1171 1.8887408\n",
      "train:  0 1172 0.6911111\n",
      "train:  0 1173 1.2232447\n",
      "train:  0 1174 0.96514195\n",
      "train:  0 1175 0.7357182\n",
      "train:  0 1176 1.4654381\n",
      "train:  0 1177 1.1076915\n",
      "train:  0 1178 0.67829645\n",
      "train:  0 1179 0.6744489\n",
      "train:  0 1180 0.7698901\n",
      "evalu:  0 1180 [0.96229327]\n",
      "train:  0 1181 0.7599541\n",
      "train:  0 1182 0.76762986\n",
      "train:  0 1183 0.635544\n",
      "train:  0 1184 0.73158836\n",
      "train:  0 1185 0.8818841\n",
      "train:  0 1186 0.6713405\n",
      "train:  0 1187 0.6531728\n",
      "train:  0 1188 0.8046146\n",
      "train:  0 1189 0.7101903\n",
      "train:  0 1190 1.18812\n",
      "train:  0 1191 0.6702361\n",
      "train:  0 1192 0.7266959\n",
      "train:  0 1193 0.8332816\n",
      "train:  0 1194 0.80590504\n",
      "train:  0 1195 0.70141894\n",
      "train:  0 1196 0.7235464\n",
      "train:  0 1197 1.1134746\n",
      "train:  0 1198 0.57689023\n",
      "train:  0 1199 0.77134\n",
      "train:  0 1200 0.5100832\n",
      "evalu:  0 1200 [0.96090764]\n",
      "train:  0 1201 0.7002103\n",
      "train:  0 1202 1.1580638\n",
      "train:  0 1203 0.7193846\n",
      "train:  0 1204 0.57843727\n",
      "train:  0 1205 0.7161935\n",
      "train:  0 1206 0.7430288\n",
      "train:  0 1207 0.66122687\n",
      "train:  0 1208 0.8566195\n",
      "train:  0 1209 1.155764\n",
      "train:  0 1210 0.9822578\n",
      "train:  0 1211 0.87200814\n",
      "train:  0 1212 0.47786832\n",
      "train:  0 1213 0.6398275\n",
      "train:  0 1214 0.7296024\n",
      "train:  0 1215 0.7342961\n",
      "train:  0 1216 1.269727\n",
      "train:  0 1217 0.9139563\n",
      "train:  0 1218 0.5910203\n",
      "train:  0 1219 0.58187914\n",
      "train:  0 1220 0.67886513\n",
      "evalu:  0 1220 [0.9594973]\n",
      "train:  0 1221 1.0992836\n",
      "train:  0 1222 0.61051697\n",
      "train:  0 1223 0.7161554\n",
      "train:  0 1224 0.84099555\n",
      "train:  0 1225 0.63680804\n",
      "train:  0 1226 0.89232826\n",
      "train:  0 1227 0.5873022\n",
      "train:  0 1228 0.57043433\n",
      "train:  0 1229 0.857931\n",
      "train:  0 1230 0.7110449\n",
      "train:  0 1231 0.595039\n",
      "train:  0 1232 1.1860416\n",
      "train:  0 1233 0.91731435\n",
      "train:  1 0 0.623372\n",
      "evalu:  1 0 [0.9584623]\n",
      "train:  1 1 0.5908074\n",
      "train:  1 2 0.65745175\n",
      "train:  1 3 0.7870563\n",
      "train:  1 4 0.78391325\n",
      "train:  1 5 0.86966944\n",
      "train:  1 6 0.733353\n",
      "train:  1 7 0.8361082\n",
      "train:  1 8 0.7989992\n",
      "train:  1 9 1.311273\n",
      "train:  1 10 0.6376133\n",
      "train:  1 11 0.6474842\n",
      "train:  1 12 1.9209253\n",
      "train:  1 13 0.62806684\n",
      "train:  1 14 0.647364\n",
      "train:  1 15 0.9865558\n",
      "train:  1 16 0.9138607\n",
      "train:  1 17 0.6175975\n",
      "train:  1 18 1.2532316\n",
      "train:  1 19 0.6740188\n",
      "train:  1 20 2.0935903\n",
      "evalu:  1 20 [0.95696723]\n",
      "train:  1 21 0.8878708\n",
      "train:  1 22 0.7696333\n",
      "train:  1 23 0.6624381\n",
      "train:  1 24 0.60553384\n",
      "train:  1 25 0.7932036\n",
      "train:  1 26 0.7630009\n",
      "train:  1 27 3.1264286\n",
      "train:  1 28 0.53826326\n",
      "train:  1 29 0.5421668\n",
      "train:  1 30 0.5896764\n",
      "train:  1 31 1.5334431\n",
      "train:  1 32 0.6685984\n",
      "train:  1 33 1.2392207\n",
      "train:  1 34 0.6306828\n",
      "train:  1 35 0.68624026\n",
      "train:  1 36 0.8328229\n",
      "train:  1 37 1.2556956\n",
      "train:  1 38 0.7728236\n",
      "train:  1 39 0.926952\n",
      "train:  1 40 0.9337663\n",
      "evalu:  1 40 [0.95556176]\n",
      "train:  1 41 1.1076809\n",
      "train:  1 42 1.0771965\n",
      "train:  1 43 0.67502326\n",
      "train:  1 44 0.80096775\n",
      "train:  1 45 1.4595476\n",
      "train:  1 46 1.6919229\n",
      "train:  1 47 0.7047376\n",
      "train:  1 48 0.66423243\n",
      "train:  1 49 0.54822487\n",
      "train:  1 50 1.3663204\n",
      "train:  1 51 0.5872199\n",
      "train:  1 52 0.71671027\n",
      "train:  1 53 0.67578524\n",
      "train:  1 54 0.9035651\n",
      "train:  1 55 0.87832546\n",
      "train:  1 56 0.49447867\n",
      "train:  1 57 0.712195\n",
      "train:  1 58 0.9994338\n",
      "train:  1 59 0.6118298\n",
      "train:  1 60 0.6592004\n",
      "evalu:  1 60 [0.95416]\n",
      "train:  1 61 0.5191947\n",
      "train:  1 62 0.6110406\n",
      "train:  1 63 0.6494209\n",
      "train:  1 64 0.6185866\n",
      "train:  1 65 0.7092149\n",
      "train:  1 66 1.0851015\n",
      "train:  1 67 0.6702691\n",
      "train:  1 68 1.1213799\n",
      "train:  1 69 0.8646115\n",
      "train:  1 70 0.85607445\n",
      "train:  1 71 0.75054723\n",
      "train:  1 72 0.65827376\n",
      "train:  1 73 0.5966748\n",
      "train:  1 74 0.49848437\n",
      "train:  1 75 0.7035959\n",
      "train:  1 76 0.7669913\n",
      "train:  1 77 0.6986274\n",
      "train:  1 78 1.212577\n",
      "train:  1 79 1.3335693\n",
      "train:  1 80 0.7502433\n",
      "evalu:  1 80 [0.95302653]\n",
      "train:  1 81 0.6283592\n",
      "train:  1 82 0.6460307\n",
      "train:  1 83 0.64950675\n",
      "train:  1 84 1.5066904\n",
      "train:  1 85 0.73605436\n",
      "train:  1 86 0.5959979\n",
      "train:  1 87 0.84047043\n",
      "train:  1 88 0.598919\n",
      "train:  1 89 0.7398304\n",
      "train:  1 90 0.710261\n",
      "train:  1 91 0.70830613\n",
      "train:  1 92 0.67906773\n",
      "train:  1 93 0.63332427\n",
      "train:  1 94 0.60486394\n",
      "train:  1 95 0.7490346\n",
      "train:  1 96 0.683433\n",
      "train:  1 97 0.9180392\n",
      "train:  1 98 0.6397975\n",
      "train:  1 99 0.67380565\n",
      "train:  1 100 0.7395105\n",
      "evalu:  1 100 [0.95193315]\n",
      "train:  1 101 0.74212074\n",
      "train:  1 102 1.2069715\n",
      "train:  1 103 0.81605536\n",
      "train:  1 104 1.8707494\n",
      "train:  1 105 1.2406335\n",
      "train:  1 106 0.6678025\n",
      "train:  1 107 1.0616844\n",
      "train:  1 108 1.499134\n",
      "train:  1 109 1.1665927\n",
      "train:  1 110 0.54797477\n",
      "train:  1 111 0.6213895\n",
      "train:  1 112 1.3081408\n",
      "train:  1 113 0.81998587\n",
      "train:  1 114 0.55740976\n",
      "train:  1 115 0.7006576\n",
      "train:  1 116 0.75398797\n",
      "train:  1 117 0.6865497\n",
      "train:  1 118 0.61667836\n",
      "train:  1 119 0.714821\n",
      "train:  1 120 0.818728\n",
      "evalu:  1 120 [0.9504819]\n",
      "train:  1 121 1.178624\n",
      "train:  1 122 0.70732504\n",
      "train:  1 123 1.0379914\n",
      "train:  1 124 0.7024382\n",
      "train:  1 125 0.7178753\n",
      "train:  1 126 0.7914979\n",
      "train:  1 127 0.8247124\n",
      "train:  1 128 0.82177114\n",
      "train:  1 129 1.195501\n",
      "train:  1 130 0.63996434\n",
      "train:  1 131 0.60113156\n",
      "train:  1 132 0.7339751\n",
      "train:  1 133 0.5272773\n",
      "train:  1 134 0.9646643\n",
      "train:  1 135 0.83271563\n",
      "train:  1 136 1.3277085\n",
      "train:  1 137 1.8754122\n",
      "train:  1 138 0.7511949\n",
      "train:  1 139 1.4843134\n",
      "train:  1 140 0.6691848\n",
      "evalu:  1 140 [0.9492249]\n",
      "train:  1 141 0.6268298\n",
      "train:  1 142 0.67786735\n",
      "train:  1 143 0.67518103\n",
      "train:  1 144 0.7584691\n",
      "train:  1 145 0.68664634\n",
      "train:  1 146 0.68096954\n",
      "train:  1 147 3.0383139\n",
      "train:  1 148 0.71771216\n",
      "train:  1 149 0.52361244\n",
      "train:  1 150 0.52800375\n",
      "train:  1 151 0.70957375\n",
      "train:  1 152 2.029322\n",
      "train:  1 153 0.82434905\n",
      "train:  1 154 0.7961543\n",
      "train:  1 155 1.0202931\n",
      "train:  1 156 0.6772544\n",
      "train:  1 157 1.9343935\n",
      "train:  1 158 0.96922827\n",
      "train:  1 159 0.6325585\n",
      "train:  1 160 0.6623353\n",
      "evalu:  1 160 [0.9480865]\n",
      "train:  1 161 1.165879\n",
      "train:  1 162 0.8132077\n",
      "train:  1 163 0.720555\n",
      "train:  1 164 0.7453602\n",
      "train:  1 165 0.8400861\n",
      "train:  1 166 0.53064126\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  1 167 0.82576334\n",
      "train:  1 168 2.7068844\n",
      "train:  1 169 0.7484001\n",
      "train:  1 170 0.5817629\n",
      "train:  1 171 0.85403997\n",
      "train:  1 172 0.65007126\n",
      "train:  1 173 0.8195813\n",
      "train:  1 174 0.69817823\n",
      "train:  1 175 0.55795807\n",
      "train:  1 176 0.7225907\n",
      "train:  1 177 0.6541641\n",
      "train:  1 178 0.862341\n",
      "train:  1 179 0.91222256\n",
      "train:  1 180 0.5460241\n",
      "evalu:  1 180 [0.9471139]\n",
      "train:  1 181 0.85028285\n",
      "train:  1 182 0.64447916\n",
      "train:  1 183 0.5947928\n",
      "train:  1 184 1.0165747\n",
      "train:  1 185 0.60308295\n",
      "train:  1 186 1.1011122\n",
      "train:  1 187 1.6643904\n",
      "train:  1 188 0.66799474\n",
      "train:  1 189 1.2090278\n",
      "train:  1 190 0.6745075\n",
      "train:  1 191 0.6288793\n",
      "train:  1 192 0.83458465\n",
      "train:  1 193 0.5456392\n",
      "train:  1 194 0.5072552\n",
      "train:  1 195 0.6699589\n",
      "train:  1 196 0.5950868\n",
      "train:  1 197 0.6856896\n",
      "train:  1 198 0.64283615\n",
      "train:  1 199 0.6162569\n",
      "train:  1 200 0.66808176\n",
      "evalu:  1 200 [0.9458729]\n",
      "train:  1 201 0.6451603\n",
      "train:  1 202 1.3886269\n",
      "train:  1 203 0.74966717\n",
      "train:  1 204 0.4873076\n",
      "train:  1 205 0.6793046\n",
      "train:  1 206 0.8416396\n",
      "train:  1 207 0.6197543\n",
      "train:  1 208 1.5263172\n",
      "train:  1 209 1.1406571\n",
      "train:  1 210 0.6458896\n",
      "train:  1 211 0.81689006\n",
      "train:  1 212 0.65288216\n",
      "train:  1 213 0.8846818\n",
      "train:  1 214 0.83622843\n",
      "train:  1 215 0.6765568\n",
      "train:  1 216 0.5383455\n",
      "train:  1 217 1.3494263\n",
      "train:  1 218 0.68372434\n",
      "train:  1 219 0.64884627\n",
      "train:  1 220 0.66454846\n",
      "evalu:  1 220 [0.9448869]\n",
      "train:  1 221 0.7838678\n",
      "train:  1 222 0.6521802\n",
      "train:  1 223 0.59830105\n",
      "train:  1 224 0.63793993\n",
      "train:  1 225 1.529472\n",
      "train:  1 226 0.6912777\n",
      "train:  1 227 1.7464056\n",
      "train:  1 228 0.61451066\n",
      "train:  1 229 0.95283854\n",
      "train:  1 230 1.1331835\n",
      "train:  1 231 0.6098473\n",
      "train:  1 232 0.5561245\n",
      "train:  1 233 0.6436827\n",
      "train:  1 234 0.6890044\n",
      "train:  1 235 0.98241794\n",
      "train:  1 236 0.8733031\n",
      "train:  1 237 0.62700945\n",
      "train:  1 238 1.141583\n",
      "train:  1 239 1.1684866\n",
      "train:  1 240 0.5576308\n",
      "evalu:  1 240 [0.9437981]\n",
      "train:  1 241 1.0235257\n",
      "train:  1 242 0.83647525\n",
      "train:  1 243 0.90459424\n",
      "train:  1 244 0.79101855\n",
      "train:  1 245 0.85249525\n",
      "train:  1 246 0.59583473\n",
      "train:  1 247 0.6174381\n",
      "train:  1 248 0.6648735\n",
      "train:  1 249 0.9853929\n",
      "train:  1 250 0.6469895\n",
      "train:  1 251 0.5743736\n",
      "train:  1 252 0.6277645\n",
      "train:  1 253 0.6289694\n",
      "train:  1 254 0.9462871\n",
      "train:  1 255 0.6543016\n",
      "train:  1 256 0.7054552\n",
      "train:  1 257 0.6265067\n",
      "train:  1 258 1.0499262\n",
      "train:  1 259 1.1623619\n",
      "train:  1 260 0.8923944\n",
      "evalu:  1 260 [0.94278103]\n",
      "train:  1 261 0.8287834\n",
      "train:  1 262 0.62212044\n",
      "train:  1 263 1.6003038\n",
      "train:  1 264 1.0739051\n",
      "train:  1 265 1.0685077\n",
      "train:  1 266 0.74782306\n",
      "train:  1 267 1.1017644\n",
      "train:  1 268 0.74956435\n",
      "train:  1 269 0.987596\n",
      "train:  1 270 0.7986709\n",
      "train:  1 271 0.7722177\n",
      "train:  1 272 0.6491203\n",
      "train:  1 273 0.7676136\n",
      "train:  1 274 0.58965844\n",
      "train:  1 275 0.7042871\n",
      "train:  1 276 0.7283702\n",
      "train:  1 277 0.8939329\n",
      "train:  1 278 1.2136587\n",
      "train:  1 279 0.7309747\n",
      "train:  1 280 0.6906875\n",
      "evalu:  1 280 [0.941707]\n",
      "train:  1 281 0.6665801\n",
      "train:  1 282 0.79005617\n",
      "train:  1 283 0.73558205\n",
      "train:  1 284 0.7554038\n",
      "train:  1 285 0.8042926\n",
      "train:  1 286 0.6805333\n",
      "train:  1 287 0.7500428\n",
      "train:  1 288 0.60746425\n",
      "train:  1 289 0.7283094\n",
      "train:  1 290 0.80318034\n",
      "train:  1 291 0.86538494\n",
      "train:  1 292 0.58538216\n",
      "train:  1 293 1.5930719\n",
      "train:  1 294 0.678056\n",
      "train:  1 295 0.9877508\n",
      "train:  1 296 3.4756074\n",
      "train:  1 297 0.7951905\n",
      "train:  1 298 0.6887824\n",
      "train:  1 299 0.74909556\n",
      "train:  1 300 1.2883912\n",
      "evalu:  1 300 [0.9408388]\n",
      "train:  1 301 0.5974259\n",
      "train:  1 302 0.6793861\n",
      "train:  1 303 0.79583013\n",
      "train:  1 304 0.79359627\n",
      "train:  1 305 0.50360173\n",
      "train:  1 306 0.7975197\n",
      "train:  1 307 0.65865415\n",
      "train:  1 308 0.696875\n",
      "train:  1 309 1.4069669\n",
      "train:  1 310 0.9373276\n",
      "train:  1 311 0.7317291\n",
      "train:  1 312 0.67132217\n",
      "train:  1 313 1.0946939\n",
      "train:  1 314 1.1137263\n",
      "train:  1 315 0.7081535\n",
      "train:  1 316 0.6737829\n",
      "train:  1 317 0.8978488\n",
      "train:  1 318 0.5831166\n",
      "train:  1 319 1.3640304\n",
      "train:  1 320 1.5010226\n",
      "evalu:  1 320 [0.9397602]\n",
      "train:  1 321 0.6494334\n",
      "train:  1 322 0.6617457\n",
      "train:  1 323 0.6023359\n",
      "train:  1 324 0.6843289\n",
      "train:  1 325 0.8679976\n",
      "train:  1 326 1.0239561\n",
      "train:  1 327 0.60766876\n",
      "train:  1 328 0.5773767\n",
      "train:  1 329 0.7969419\n",
      "train:  1 330 1.3861976\n",
      "train:  1 331 0.5407337\n",
      "train:  1 332 0.7480516\n",
      "train:  1 333 2.844353\n",
      "train:  1 334 0.685257\n",
      "train:  1 335 0.53080285\n",
      "train:  1 336 0.5396864\n",
      "train:  1 337 0.7681932\n",
      "train:  1 338 2.4322908\n",
      "train:  1 339 0.78806216\n",
      "train:  1 340 0.74442804\n",
      "evalu:  1 340 [0.93880135]\n",
      "train:  1 341 0.7452199\n",
      "train:  1 342 0.8665315\n",
      "train:  1 343 0.6640795\n",
      "train:  1 344 0.62939554\n",
      "train:  1 345 0.66151804\n",
      "train:  1 346 0.6060286\n",
      "train:  1 347 0.52629226\n",
      "train:  1 348 0.71451384\n",
      "train:  1 349 0.8165751\n",
      "train:  1 350 1.748795\n",
      "train:  1 351 0.7642031\n",
      "train:  1 352 0.55780476\n",
      "train:  1 353 0.75609297\n",
      "train:  1 354 0.7179224\n",
      "train:  1 355 1.0070088\n",
      "train:  1 356 0.5906726\n",
      "train:  1 357 0.6243204\n",
      "train:  1 358 0.9082355\n",
      "train:  1 359 0.6319273\n",
      "train:  1 360 0.62698185\n",
      "evalu:  1 360 [0.93794894]\n",
      "train:  1 361 0.9272076\n",
      "train:  1 362 1.057993\n",
      "train:  1 363 0.74114126\n",
      "train:  1 364 0.7346985\n",
      "train:  1 365 0.63475376\n",
      "train:  1 366 0.6998262\n",
      "train:  1 367 1.9538907\n",
      "train:  1 368 0.9200724\n",
      "train:  1 369 1.1758178\n",
      "train:  1 370 0.7610694\n",
      "train:  1 371 0.8882928\n",
      "train:  1 372 0.57927257\n",
      "train:  1 373 2.1627715\n",
      "train:  1 374 0.8121796\n",
      "train:  1 375 0.8167446\n",
      "train:  1 376 0.5704341\n",
      "train:  1 377 1.1369956\n",
      "train:  1 378 0.6897394\n",
      "train:  1 379 0.9976842\n",
      "train:  1 380 0.61884755\n",
      "evalu:  1 380 [0.9366317]\n",
      "train:  1 381 2.5553114\n",
      "train:  1 382 0.97234523\n",
      "train:  1 383 0.9183042\n",
      "train:  1 384 1.0845337\n",
      "train:  1 385 0.75762874\n",
      "train:  1 386 0.9705594\n",
      "train:  1 387 0.8196872\n",
      "train:  1 388 0.7700618\n",
      "train:  1 389 1.0290856\n",
      "train:  1 390 0.78492814\n",
      "train:  1 391 1.2744057\n",
      "train:  1 392 0.7794483\n",
      "train:  1 393 1.3410441\n",
      "train:  1 394 0.5780042\n",
      "train:  1 395 0.69921076\n",
      "train:  1 396 1.0801297\n",
      "train:  1 397 1.2672739\n",
      "train:  1 398 0.9642082\n",
      "train:  1 399 0.61957705\n",
      "train:  1 400 0.66122043\n",
      "evalu:  1 400 [0.9357312]\n",
      "train:  1 401 0.77485126\n",
      "train:  1 402 0.46867073\n",
      "train:  1 403 0.63924766\n",
      "train:  1 404 0.53835857\n",
      "train:  1 405 0.51004964\n",
      "train:  1 406 0.61790717\n",
      "train:  1 407 0.6219296\n",
      "train:  1 408 0.7171199\n",
      "train:  1 409 0.69549465\n",
      "train:  1 410 0.7475848\n",
      "train:  1 411 0.5496408\n",
      "train:  1 412 0.5749215\n",
      "train:  1 413 0.59841055\n",
      "train:  1 414 0.70151156\n",
      "train:  1 415 0.604739\n",
      "train:  1 416 0.62960106\n",
      "train:  1 417 0.5818243\n",
      "train:  1 418 0.9201242\n",
      "train:  1 419 1.6308677\n",
      "train:  1 420 0.6947097\n",
      "evalu:  1 420 [0.9348869]\n",
      "train:  1 421 0.78040487\n",
      "train:  1 422 0.57816124\n",
      "train:  1 423 0.6666086\n",
      "train:  1 424 1.2331265\n",
      "train:  1 425 1.9062705\n",
      "train:  1 426 1.3973033\n",
      "train:  1 427 0.6741681\n",
      "train:  1 428 0.73556674\n",
      "train:  1 429 0.60098696\n",
      "train:  1 430 1.0458599\n",
      "train:  1 431 0.63375276\n",
      "train:  1 432 2.1663263\n",
      "train:  1 433 0.6535369\n",
      "train:  1 434 0.7275085\n",
      "train:  1 435 0.659512\n",
      "train:  1 436 0.65132916\n",
      "train:  1 437 0.61331415\n",
      "train:  1 438 0.8030752\n",
      "train:  1 439 0.67105687\n",
      "train:  1 440 0.66385823\n",
      "evalu:  1 440 [0.9337839]\n",
      "train:  1 441 0.68171006\n",
      "train:  1 442 0.8120184\n",
      "train:  1 443 0.92430824\n",
      "train:  1 444 0.730763\n",
      "train:  1 445 0.755922\n",
      "train:  1 446 0.72707415\n",
      "train:  1 447 0.73535025\n",
      "train:  1 448 0.60639364\n",
      "train:  1 449 0.6995236\n",
      "train:  1 450 0.71598005\n",
      "train:  1 451 0.7231076\n",
      "train:  1 452 1.2306174\n",
      "train:  1 453 0.71503025\n",
      "train:  1 454 0.72782993\n",
      "train:  1 455 0.45908073\n",
      "train:  1 456 0.55363095\n",
      "train:  1 457 0.6615423\n",
      "train:  1 458 0.8373428\n",
      "train:  1 459 0.53548676\n",
      "train:  1 460 0.5653254\n",
      "evalu:  1 460 [0.9329367]\n",
      "train:  1 461 0.9795969\n",
      "train:  1 462 0.65736455\n",
      "train:  1 463 0.6022375\n",
      "train:  1 464 0.7606105\n",
      "train:  1 465 0.82036066\n",
      "train:  1 466 0.72141033\n",
      "train:  1 467 1.1904414\n",
      "train:  1 468 0.91016275\n",
      "train:  1 469 0.6948135\n",
      "train:  1 470 0.86800504\n",
      "train:  1 471 0.6936854\n",
      "train:  1 472 1.299109\n",
      "train:  1 473 0.70856965\n",
      "train:  1 474 1.8542248\n",
      "train:  1 475 0.71878904\n",
      "train:  1 476 0.59059256\n",
      "train:  1 477 0.55985653\n",
      "train:  1 478 1.1951163\n",
      "train:  1 479 0.5692686\n",
      "train:  1 480 0.66871434\n",
      "evalu:  1 480 [0.9320489]\n",
      "train:  1 481 0.7459988\n",
      "train:  1 482 0.9994874\n",
      "train:  1 483 0.5544484\n",
      "train:  1 484 1.3748734\n",
      "train:  1 485 0.7924448\n",
      "train:  1 486 0.64484257\n",
      "train:  1 487 0.5876368\n",
      "train:  1 488 0.8836355\n",
      "train:  1 489 0.5187545\n",
      "train:  1 490 1.1343396\n",
      "train:  1 491 1.2687745\n",
      "train:  1 492 1.5918493\n",
      "train:  1 493 0.6002426\n",
      "train:  1 494 0.5542491\n",
      "train:  1 495 0.7293971\n",
      "train:  1 496 0.66514504\n",
      "train:  1 497 0.50601715\n",
      "train:  1 498 0.6703466\n",
      "train:  1 499 0.6880484\n",
      "train:  1 500 0.57922226\n",
      "evalu:  1 500 [0.9311323]\n",
      "train:  1 501 0.85936475\n",
      "train:  1 502 1.3207687\n",
      "train:  1 503 0.58446753\n",
      "train:  1 504 1.1601986\n",
      "train:  1 505 0.84420395\n",
      "train:  1 506 0.7933671\n",
      "train:  1 507 1.6432832\n",
      "train:  1 508 0.9032922\n",
      "train:  1 509 0.88135594\n",
      "train:  1 510 0.8168983\n",
      "train:  1 511 0.7973675\n",
      "train:  1 512 1.0234087\n",
      "train:  1 513 0.53819126\n",
      "train:  1 514 0.7145776\n",
      "train:  1 515 0.8764367\n",
      "train:  1 516 0.6145238\n",
      "train:  1 517 0.90412563\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  1 518 1.0349618\n",
      "train:  1 519 0.5093748\n",
      "train:  1 520 0.6353729\n",
      "evalu:  1 520 [0.9305286]\n",
      "train:  1 521 0.5792287\n",
      "train:  1 522 1.1336546\n",
      "train:  1 523 1.3963017\n",
      "train:  1 524 1.843099\n",
      "train:  1 525 0.88945645\n",
      "train:  1 526 0.66695017\n",
      "train:  1 527 0.8283105\n",
      "train:  1 528 0.6089874\n",
      "train:  1 529 0.77555585\n",
      "train:  1 530 0.7097552\n",
      "train:  1 531 0.7096283\n",
      "train:  1 532 0.67994916\n",
      "train:  1 533 0.7210693\n",
      "train:  1 534 0.7110424\n",
      "train:  1 535 1.1468643\n",
      "train:  1 536 1.9526461\n",
      "train:  1 537 0.68273973\n",
      "train:  1 538 0.48234805\n",
      "train:  1 539 0.6033944\n",
      "train:  1 540 0.71323764\n",
      "evalu:  1 540 [0.92972535]\n",
      "train:  1 541 0.55496687\n",
      "train:  1 542 0.66299534\n",
      "train:  1 543 0.6188621\n",
      "train:  1 544 0.64560527\n",
      "train:  1 545 0.44356316\n",
      "train:  1 546 0.56589544\n",
      "train:  1 547 0.5864779\n",
      "train:  1 548 0.8134347\n",
      "train:  1 549 0.481138\n",
      "train:  1 550 0.6340793\n",
      "train:  1 551 0.74675375\n",
      "train:  1 552 0.7995599\n",
      "train:  1 553 0.51602566\n",
      "train:  1 554 0.7060662\n",
      "train:  1 555 0.69919354\n",
      "train:  1 556 0.5711046\n",
      "train:  1 557 2.3729808\n",
      "train:  1 558 0.55457497\n",
      "train:  1 559 0.67855096\n",
      "train:  1 560 0.5209341\n",
      "evalu:  1 560 [0.92896694]\n",
      "train:  1 561 0.65031976\n",
      "train:  1 562 0.60304916\n",
      "train:  1 563 0.71614873\n",
      "train:  1 564 0.8112037\n",
      "train:  1 565 0.6292477\n",
      "train:  1 566 1.2464274\n",
      "train:  1 567 1.3849857\n",
      "train:  1 568 0.8975529\n",
      "train:  1 569 1.4513353\n",
      "train:  1 570 0.8622605\n",
      "train:  1 571 1.2452047\n",
      "train:  1 572 1.2234154\n",
      "train:  1 573 0.7031075\n",
      "train:  1 574 0.8487858\n",
      "train:  1 575 0.48794743\n",
      "train:  1 576 1.0343529\n",
      "train:  1 577 0.6575658\n",
      "train:  1 578 1.0074823\n",
      "train:  1 579 0.81888884\n",
      "train:  1 580 0.7079967\n",
      "evalu:  1 580 [0.9280656]\n",
      "train:  1 581 0.5365188\n",
      "train:  1 582 1.2578325\n",
      "train:  1 583 0.5956354\n",
      "train:  1 584 0.61116725\n",
      "train:  1 585 0.7880183\n",
      "train:  1 586 0.9087848\n",
      "train:  1 587 0.5189468\n",
      "train:  1 588 0.59217817\n",
      "train:  1 589 0.60383713\n",
      "train:  1 590 0.58641976\n",
      "train:  1 591 0.67542666\n",
      "train:  1 592 1.4001786\n",
      "train:  1 593 1.6614741\n",
      "train:  1 594 0.7642635\n",
      "train:  1 595 0.7773909\n",
      "train:  1 596 0.8454158\n",
      "train:  1 597 0.9857243\n",
      "train:  1 598 0.53787655\n",
      "train:  1 599 0.6097584\n",
      "train:  1 600 0.68935984\n",
      "evalu:  1 600 [0.92716205]\n",
      "train:  1 601 0.79399204\n",
      "train:  1 602 1.3337873\n",
      "train:  1 603 0.86765367\n",
      "train:  1 604 1.4633728\n",
      "train:  1 605 0.58685726\n",
      "train:  1 606 0.68059427\n",
      "train:  1 607 0.5957571\n",
      "train:  1 608 0.80485284\n",
      "train:  1 609 0.7873745\n",
      "train:  1 610 1.1269604\n",
      "train:  1 611 0.62270314\n",
      "train:  1 612 1.2879971\n",
      "train:  1 613 0.8190018\n",
      "train:  1 614 0.70257294\n",
      "train:  1 615 0.5090742\n",
      "train:  1 616 0.5406005\n",
      "train:  1 617 0.61961997\n",
      "train:  1 618 3.4112935\n",
      "train:  1 619 0.55902827\n",
      "train:  1 620 0.8000729\n",
      "evalu:  1 620 [0.9264251]\n",
      "train:  1 621 1.1828628\n",
      "train:  1 622 0.6067355\n",
      "train:  1 623 0.5408467\n",
      "train:  1 624 0.84242475\n",
      "train:  1 625 0.9581889\n",
      "train:  1 626 0.7310023\n",
      "train:  1 627 0.65524936\n",
      "train:  1 628 1.1669059\n",
      "train:  1 629 0.7924439\n",
      "train:  1 630 0.7450982\n",
      "train:  1 631 0.5700356\n",
      "train:  1 632 0.6087029\n",
      "train:  1 633 0.6076294\n",
      "train:  1 634 0.67776823\n",
      "train:  1 635 0.6723994\n",
      "train:  1 636 0.66139436\n",
      "train:  1 637 0.8432542\n",
      "train:  1 638 0.80240893\n",
      "train:  1 639 1.0298853\n",
      "train:  1 640 0.6426686\n",
      "evalu:  1 640 [0.9258486]\n",
      "train:  1 641 0.78189176\n",
      "train:  1 642 0.84138995\n",
      "train:  1 643 0.78107274\n",
      "train:  1 644 0.53575796\n",
      "train:  1 645 0.7374567\n",
      "train:  1 646 0.55116767\n",
      "train:  1 647 0.6386014\n",
      "train:  1 648 0.5026837\n",
      "train:  1 649 1.898986\n",
      "train:  1 650 0.7292653\n",
      "train:  1 651 0.9254658\n",
      "train:  1 652 0.7247423\n",
      "train:  1 653 0.8713954\n",
      "train:  1 654 0.6194059\n",
      "train:  1 655 0.59312725\n",
      "train:  1 656 1.4349543\n",
      "train:  1 657 0.6074399\n",
      "train:  1 658 0.9744982\n",
      "train:  1 659 0.8596831\n",
      "train:  1 660 0.5418953\n",
      "evalu:  1 660 [0.92523694]\n",
      "train:  1 661 1.0850211\n",
      "train:  1 662 1.0813142\n",
      "train:  1 663 1.0435781\n",
      "train:  1 664 0.83610904\n",
      "train:  1 665 0.7600703\n",
      "train:  1 666 1.7647505\n",
      "train:  1 667 0.5702666\n",
      "train:  1 668 0.7136012\n",
      "train:  1 669 0.5729889\n",
      "train:  1 670 0.5045609\n",
      "train:  1 671 0.5374772\n",
      "train:  1 672 0.8014939\n",
      "train:  1 673 0.6879095\n",
      "train:  1 674 0.5415835\n",
      "train:  1 675 0.9915778\n",
      "train:  1 676 1.8201703\n",
      "train:  1 677 0.66901714\n",
      "train:  1 678 1.0265578\n",
      "train:  1 679 0.85691714\n",
      "train:  1 680 0.58579636\n",
      "evalu:  1 680 [0.92474324]\n",
      "train:  1 681 0.7901312\n",
      "train:  1 682 0.53511345\n",
      "train:  1 683 0.73190415\n",
      "train:  1 684 0.5613621\n",
      "train:  1 685 0.7388369\n",
      "train:  1 686 0.5827998\n",
      "train:  1 687 1.1984284\n",
      "train:  1 688 1.6266484\n",
      "train:  1 689 0.6143632\n",
      "train:  1 690 1.4780762\n",
      "train:  1 691 0.7510661\n",
      "train:  1 692 0.7330394\n",
      "train:  1 693 0.77358913\n",
      "train:  1 694 0.7062197\n",
      "train:  1 695 0.85892594\n",
      "train:  1 696 0.60741323\n",
      "train:  1 697 0.61225533\n",
      "train:  1 698 0.6034054\n",
      "train:  1 699 0.7861312\n",
      "train:  1 700 0.6887417\n",
      "evalu:  1 700 [0.92418206]\n",
      "train:  1 701 0.51603913\n",
      "train:  1 702 0.5868027\n",
      "train:  1 703 0.5529326\n",
      "train:  1 704 0.5589068\n",
      "train:  1 705 0.53251314\n",
      "train:  1 706 0.82238394\n",
      "train:  1 707 0.5841643\n",
      "train:  1 708 0.64427245\n",
      "train:  1 709 0.60088736\n",
      "train:  1 710 0.63960296\n",
      "train:  1 711 0.8313139\n",
      "train:  1 712 0.8028242\n",
      "train:  1 713 1.0411984\n",
      "train:  1 714 0.5514652\n",
      "train:  1 715 1.2328868\n",
      "train:  1 716 0.7938143\n",
      "train:  1 717 0.54061526\n",
      "train:  1 718 0.60064334\n",
      "train:  1 719 0.7982853\n",
      "train:  1 720 0.9018585\n",
      "evalu:  1 720 [0.9236251]\n",
      "train:  1 721 0.636114\n",
      "train:  1 722 0.95795506\n",
      "train:  1 723 0.9204963\n",
      "train:  1 724 1.0638833\n",
      "train:  1 725 0.5721453\n",
      "train:  1 726 0.8949238\n",
      "train:  1 727 0.7584538\n",
      "train:  1 728 0.6587844\n",
      "train:  1 729 0.6609249\n",
      "train:  1 730 1.2195748\n",
      "train:  1 731 0.7116838\n",
      "train:  1 732 0.545964\n",
      "train:  1 733 0.79719067\n",
      "train:  1 734 0.8526714\n",
      "train:  1 735 0.7747783\n",
      "train:  1 736 0.7569523\n",
      "train:  1 737 2.1156764\n",
      "train:  1 738 0.7539122\n",
      "train:  1 739 0.71568847\n",
      "train:  1 740 0.5558193\n",
      "evalu:  1 740 [0.92295575]\n",
      "train:  1 741 0.8468532\n",
      "train:  1 742 0.7981715\n",
      "train:  1 743 0.65194017\n",
      "train:  1 744 0.6677093\n",
      "train:  1 745 0.851426\n",
      "train:  1 746 1.2554415\n",
      "train:  1 747 0.7431395\n",
      "train:  1 748 0.92565745\n",
      "train:  1 749 1.1771736\n",
      "train:  1 750 0.4969683\n",
      "train:  1 751 1.2020081\n",
      "train:  1 752 0.79248196\n",
      "train:  1 753 0.66335374\n",
      "train:  1 754 0.6898214\n",
      "train:  1 755 0.9424869\n",
      "train:  1 756 0.661882\n",
      "train:  1 757 0.74179995\n",
      "train:  1 758 0.95460093\n",
      "train:  1 759 0.72699827\n",
      "train:  1 760 0.59281373\n",
      "evalu:  1 760 [0.9223121]\n",
      "train:  1 761 0.6617042\n",
      "train:  1 762 1.9928447\n",
      "train:  1 763 0.64636815\n",
      "train:  1 764 0.6860638\n",
      "train:  1 765 0.64619213\n",
      "train:  1 766 0.55433184\n",
      "train:  1 767 0.4738266\n",
      "train:  1 768 0.7797643\n",
      "train:  1 769 0.8291455\n",
      "train:  1 770 0.9249813\n",
      "train:  1 771 0.8118448\n",
      "train:  1 772 0.5649192\n",
      "train:  1 773 0.78451544\n",
      "train:  1 774 0.73496264\n",
      "train:  1 775 0.58001643\n",
      "train:  1 776 0.84623796\n",
      "train:  1 777 0.94157887\n",
      "train:  1 778 0.7031408\n",
      "train:  1 779 0.8088532\n",
      "train:  1 780 0.5830477\n",
      "evalu:  1 780 [0.9216362]\n",
      "train:  1 781 0.6083576\n",
      "train:  1 782 0.5022904\n",
      "train:  1 783 0.5772653\n",
      "train:  1 784 0.6883663\n",
      "train:  1 785 0.68521583\n",
      "train:  1 786 0.6480407\n",
      "train:  1 787 0.7405021\n",
      "train:  1 788 0.52924913\n",
      "train:  1 789 0.82254225\n",
      "train:  1 790 0.9005461\n",
      "train:  1 791 0.6466656\n",
      "train:  1 792 0.8123254\n",
      "train:  1 793 1.0959154\n",
      "train:  1 794 0.9088879\n",
      "train:  1 795 1.1424302\n",
      "train:  1 796 0.58334064\n",
      "train:  1 797 0.8771696\n",
      "train:  1 798 0.7631152\n",
      "train:  1 799 0.48180738\n",
      "train:  1 800 0.8938803\n",
      "evalu:  1 800 [0.92115414]\n",
      "train:  1 801 0.59853786\n",
      "train:  1 802 0.59092873\n",
      "train:  1 803 0.54254305\n",
      "train:  1 804 0.5741203\n",
      "train:  1 805 0.6213572\n",
      "train:  1 806 0.5424814\n",
      "train:  1 807 1.258828\n",
      "train:  1 808 0.59897184\n",
      "train:  1 809 0.6194668\n",
      "train:  1 810 0.8783523\n",
      "train:  1 811 0.5582767\n",
      "train:  1 812 0.76165277\n",
      "train:  1 813 0.9725591\n",
      "train:  1 814 3.7664258\n",
      "train:  1 815 1.0564864\n",
      "train:  1 816 0.61018825\n",
      "train:  1 817 0.5714412\n",
      "train:  1 818 0.54094297\n",
      "train:  1 819 0.65166324\n",
      "train:  1 820 0.7143199\n",
      "evalu:  1 820 [0.9206941]\n",
      "train:  1 821 0.66936994\n",
      "train:  1 822 0.65694994\n",
      "train:  1 823 0.7159979\n",
      "train:  1 824 0.63340855\n",
      "train:  1 825 0.56051886\n",
      "train:  1 826 1.2878036\n",
      "train:  1 827 2.204788\n",
      "train:  1 828 1.0316263\n",
      "train:  1 829 0.7095252\n",
      "train:  1 830 1.0050757\n",
      "train:  1 831 0.994755\n",
      "train:  1 832 0.58990496\n",
      "train:  1 833 0.6800065\n",
      "train:  1 834 0.6750805\n",
      "train:  1 835 0.66317755\n",
      "train:  1 836 0.79636526\n",
      "train:  1 837 0.58801377\n",
      "train:  1 838 0.5663385\n",
      "train:  1 839 0.5859269\n",
      "train:  1 840 1.3867393\n",
      "evalu:  1 840 [0.9202537]\n",
      "train:  1 841 0.7896133\n",
      "train:  1 842 0.6016344\n",
      "train:  1 843 1.4575039\n",
      "train:  1 844 0.5589637\n",
      "train:  1 845 0.6027924\n",
      "train:  1 846 0.70729005\n",
      "train:  1 847 0.6074784\n",
      "train:  1 848 1.5576421\n",
      "train:  1 849 0.9307019\n",
      "train:  1 850 0.57161164\n",
      "train:  1 851 0.6141464\n",
      "train:  1 852 0.96179783\n",
      "train:  1 853 1.4220393\n",
      "train:  1 854 0.5628532\n",
      "train:  1 855 0.6897982\n",
      "train:  1 856 0.57399225\n",
      "train:  1 857 0.7868196\n",
      "train:  1 858 1.5635803\n",
      "train:  1 859 0.96906906\n",
      "train:  1 860 0.597543\n",
      "evalu:  1 860 [0.9195814]\n",
      "train:  1 861 0.766994\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  1 862 0.7446627\n",
      "train:  1 863 0.46817797\n",
      "train:  1 864 0.630106\n",
      "train:  1 865 0.719373\n",
      "train:  1 866 0.9727906\n",
      "train:  1 867 0.48194093\n",
      "train:  1 868 0.5792991\n",
      "train:  1 869 0.7981276\n",
      "train:  1 870 1.1787244\n",
      "train:  1 871 0.7584665\n",
      "train:  1 872 0.79219496\n",
      "train:  1 873 0.7048607\n",
      "train:  1 874 0.57732725\n",
      "train:  1 875 0.81773764\n",
      "train:  1 876 0.79314965\n",
      "train:  1 877 1.0717677\n",
      "train:  1 878 1.3822584\n",
      "train:  1 879 0.53240293\n",
      "train:  1 880 0.7200777\n",
      "evalu:  1 880 [0.91895676]\n",
      "train:  1 881 0.78575706\n",
      "train:  1 882 0.8221199\n",
      "train:  1 883 0.6199007\n",
      "train:  1 884 1.1691167\n",
      "train:  1 885 0.713298\n",
      "train:  1 886 0.58889437\n",
      "train:  1 887 1.1430472\n",
      "train:  1 888 0.98628074\n",
      "train:  1 889 0.522299\n",
      "train:  1 890 0.817152\n",
      "train:  1 891 0.62149465\n",
      "train:  1 892 0.70881444\n",
      "train:  1 893 0.7197404\n",
      "train:  1 894 0.57577384\n",
      "train:  1 895 1.7984803\n",
      "train:  1 896 0.8757447\n",
      "train:  1 897 0.5110891\n",
      "train:  1 898 0.8282769\n",
      "train:  1 899 0.662297\n",
      "train:  1 900 0.6895817\n",
      "evalu:  1 900 [0.9186006]\n",
      "train:  1 901 0.5618622\n",
      "train:  1 902 1.0947782\n",
      "train:  1 903 0.7396149\n",
      "train:  1 904 0.65183127\n",
      "train:  1 905 0.7017522\n",
      "train:  1 906 0.6845503\n",
      "train:  1 907 1.1090913\n",
      "train:  1 908 0.58522093\n",
      "train:  1 909 0.78594726\n",
      "train:  1 910 0.68441033\n",
      "train:  1 911 0.53359556\n",
      "train:  1 912 0.8191655\n",
      "train:  1 913 0.7072303\n",
      "train:  1 914 0.59225446\n",
      "train:  1 915 0.859056\n",
      "train:  1 916 0.8455042\n",
      "train:  1 917 0.6080022\n",
      "train:  1 918 1.1558847\n",
      "train:  1 919 1.062426\n",
      "train:  1 920 0.66972125\n",
      "evalu:  1 920 [0.9180082]\n",
      "train:  1 921 0.93603355\n",
      "train:  1 922 0.7542321\n",
      "train:  1 923 0.49426427\n",
      "train:  1 924 0.8661974\n",
      "train:  1 925 0.6751375\n",
      "train:  1 926 0.70709115\n",
      "train:  1 927 1.1531458\n",
      "train:  1 928 0.5308541\n",
      "train:  1 929 0.53608376\n",
      "train:  1 930 1.0506428\n",
      "train:  1 931 0.46792588\n",
      "train:  1 932 0.74244463\n",
      "train:  1 933 0.7088649\n",
      "train:  1 934 0.7304209\n",
      "train:  1 935 0.52995896\n",
      "train:  1 936 0.48886162\n",
      "train:  1 937 0.50121003\n",
      "train:  1 938 0.6605276\n",
      "train:  1 939 0.5823106\n",
      "train:  1 940 0.632761\n",
      "evalu:  1 940 [0.9175724]\n",
      "train:  1 941 0.7248468\n",
      "train:  1 942 0.6065881\n",
      "train:  1 943 0.6276516\n",
      "train:  1 944 0.54455805\n",
      "train:  1 945 0.59458804\n",
      "train:  1 946 0.5909087\n",
      "train:  1 947 0.5807944\n",
      "train:  1 948 1.1271123\n",
      "train:  1 949 0.8112998\n",
      "train:  1 950 0.6511884\n",
      "train:  1 951 0.8772182\n",
      "train:  1 952 2.006175\n",
      "train:  1 953 0.6596742\n",
      "train:  1 954 0.80649185\n",
      "train:  1 955 0.6931941\n",
      "train:  1 956 0.620985\n",
      "train:  1 957 0.516014\n",
      "train:  1 958 0.7697834\n",
      "train:  1 959 0.56468207\n",
      "train:  1 960 1.0963366\n",
      "evalu:  1 960 [0.9169158]\n",
      "train:  1 961 0.6477874\n",
      "train:  1 962 2.6896334\n",
      "train:  1 963 1.0332476\n",
      "train:  1 964 0.79188645\n",
      "train:  1 965 0.6879769\n",
      "train:  1 966 1.1133078\n",
      "train:  1 967 0.60176945\n",
      "train:  1 968 0.9380694\n",
      "train:  1 969 0.6469848\n",
      "train:  1 970 0.5628432\n",
      "train:  1 971 0.5589534\n",
      "train:  1 972 0.6404261\n",
      "train:  1 973 2.1948767\n",
      "train:  1 974 0.4799628\n",
      "train:  1 975 0.5586017\n",
      "train:  1 976 0.7652329\n",
      "train:  1 977 0.7032313\n",
      "train:  1 978 0.71589065\n",
      "train:  1 979 0.5559555\n",
      "train:  1 980 0.74719065\n",
      "evalu:  1 980 [0.91636384]\n",
      "train:  1 981 0.8730729\n",
      "train:  1 982 0.679125\n",
      "train:  1 983 0.8556312\n",
      "train:  1 984 1.1628891\n",
      "train:  1 985 1.2633455\n",
      "train:  1 986 0.70950603\n",
      "train:  1 987 0.67635745\n",
      "train:  1 988 1.4450362\n",
      "train:  1 989 0.72280747\n",
      "train:  1 990 0.5297055\n",
      "train:  1 991 0.67690873\n",
      "train:  1 992 0.55184203\n",
      "train:  1 993 0.72088885\n",
      "train:  1 994 0.5900882\n",
      "train:  1 995 1.2899214\n",
      "train:  1 996 0.7510848\n",
      "train:  1 997 0.58381414\n",
      "train:  1 998 1.0989103\n",
      "train:  1 999 0.595624\n",
      "train:  1 1000 0.6347185\n",
      "evalu:  1 1000 [0.91594195]\n",
      "train:  1 1001 0.7399736\n",
      "train:  1 1002 0.64580363\n",
      "train:  1 1003 0.6512274\n",
      "train:  1 1004 0.64990884\n",
      "train:  1 1005 0.6616958\n",
      "train:  1 1006 0.64836836\n",
      "train:  1 1007 0.75924414\n",
      "train:  1 1008 1.1602513\n",
      "train:  1 1009 0.77694905\n",
      "train:  1 1010 1.0947443\n",
      "train:  1 1011 0.7652579\n",
      "train:  1 1012 1.3528461\n",
      "train:  1 1013 0.7056268\n",
      "train:  1 1014 0.69730276\n",
      "train:  1 1015 0.9808453\n",
      "train:  1 1016 1.0333246\n",
      "train:  1 1017 1.1460376\n",
      "train:  1 1018 0.5488337\n",
      "train:  1 1019 0.57461005\n",
      "train:  1 1020 0.6920115\n",
      "evalu:  1 1020 [0.9152496]\n",
      "train:  1 1021 0.6002756\n",
      "train:  1 1022 0.7153263\n",
      "train:  1 1023 1.265782\n",
      "train:  1 1024 0.6584799\n",
      "train:  1 1025 0.58723056\n",
      "train:  1 1026 0.65981656\n",
      "train:  1 1027 0.81635875\n",
      "train:  1 1028 1.0592175\n",
      "train:  1 1029 0.6468205\n",
      "train:  1 1030 0.57536566\n",
      "train:  1 1031 0.58426106\n",
      "train:  1 1032 0.79172444\n",
      "train:  1 1033 0.5472045\n",
      "train:  1 1034 0.8107948\n",
      "train:  1 1035 0.5858964\n",
      "train:  1 1036 0.6835628\n",
      "train:  1 1037 0.831986\n",
      "train:  1 1038 0.94404554\n",
      "train:  1 1039 0.6131913\n",
      "train:  1 1040 0.592231\n",
      "evalu:  1 1040 [0.9148175]\n",
      "train:  1 1041 0.49610052\n",
      "train:  1 1042 1.0632299\n",
      "train:  1 1043 0.7344638\n",
      "train:  1 1044 0.6408528\n",
      "train:  1 1045 0.6151913\n",
      "train:  1 1046 0.891016\n",
      "train:  1 1047 0.59340227\n",
      "train:  1 1048 2.0614622\n",
      "train:  1 1049 0.50815403\n",
      "train:  1 1050 1.3053789\n",
      "train:  1 1051 1.4498105\n",
      "train:  1 1052 0.66013724\n",
      "train:  1 1053 0.60356337\n",
      "train:  1 1054 0.6224888\n",
      "train:  1 1055 0.9656523\n",
      "train:  1 1056 0.6008572\n",
      "train:  1 1057 0.5258574\n",
      "train:  1 1058 1.3232601\n",
      "train:  1 1059 0.52256244\n",
      "train:  1 1060 0.62488574\n",
      "evalu:  1 1060 [0.9146531]\n",
      "train:  1 1061 0.65472597\n",
      "train:  1 1062 0.61342824\n",
      "train:  1 1063 0.5340556\n",
      "train:  1 1064 1.131446\n",
      "train:  1 1065 0.656577\n",
      "train:  1 1066 0.545606\n",
      "train:  1 1067 0.50027734\n",
      "train:  1 1068 0.5969205\n",
      "train:  1 1069 0.60000324\n",
      "train:  1 1070 0.66978806\n",
      "train:  1 1071 0.66832906\n",
      "train:  1 1072 0.6168782\n",
      "train:  1 1073 0.8313667\n",
      "train:  1 1074 0.701159\n",
      "train:  1 1075 0.6263151\n",
      "train:  1 1076 0.89619124\n",
      "train:  1 1077 0.6917078\n",
      "train:  1 1078 0.9877725\n",
      "train:  1 1079 0.86493635\n",
      "train:  1 1080 0.6723031\n",
      "evalu:  1 1080 [0.91396356]\n",
      "train:  1 1081 1.3122236\n",
      "train:  1 1082 0.7422062\n",
      "train:  1 1083 2.0100915\n",
      "train:  1 1084 0.7213505\n",
      "train:  1 1085 1.33317\n",
      "train:  1 1086 0.78375506\n",
      "train:  1 1087 2.2527714\n",
      "train:  1 1088 0.77915883\n",
      "train:  1 1089 0.59452856\n",
      "train:  1 1090 1.1580997\n",
      "train:  1 1091 0.6596934\n",
      "train:  1 1092 0.82811624\n",
      "train:  1 1093 0.60287267\n",
      "train:  1 1094 0.71021503\n",
      "train:  1 1095 0.5757919\n",
      "train:  1 1096 0.6045152\n",
      "train:  1 1097 0.6788953\n",
      "train:  1 1098 0.9951621\n",
      "train:  1 1099 0.728807\n",
      "train:  1 1100 0.7357537\n",
      "evalu:  1 1100 [0.9133019]\n",
      "train:  1 1101 0.8331844\n",
      "train:  1 1102 0.5814903\n",
      "train:  1 1103 0.61619616\n",
      "train:  1 1104 0.57900614\n",
      "train:  1 1105 0.5381913\n",
      "train:  1 1106 0.5946614\n",
      "train:  1 1107 0.6257631\n",
      "train:  1 1108 1.2819072\n",
      "train:  1 1109 1.0508493\n",
      "train:  1 1110 0.67967397\n",
      "train:  1 1111 5.5916505\n",
      "train:  1 1112 0.8969178\n",
      "train:  1 1113 1.0725213\n",
      "train:  1 1114 0.61159205\n",
      "train:  1 1115 0.5381059\n",
      "train:  1 1116 1.3476276\n",
      "train:  1 1117 0.62864\n",
      "train:  1 1118 0.8754863\n",
      "train:  1 1119 0.7523118\n",
      "train:  1 1120 0.5893853\n",
      "evalu:  1 1120 [0.9127992]\n",
      "train:  1 1121 1.0944905\n",
      "train:  1 1122 1.0278373\n",
      "train:  1 1123 0.6214463\n",
      "train:  1 1124 0.5084863\n",
      "train:  1 1125 0.90112454\n",
      "train:  1 1126 0.5656489\n",
      "train:  1 1127 0.83013463\n",
      "train:  1 1128 0.99592155\n",
      "train:  1 1129 0.53187746\n",
      "train:  1 1130 0.8643846\n",
      "train:  1 1131 0.6563573\n",
      "train:  1 1132 2.1997042\n",
      "train:  1 1133 0.6135169\n",
      "train:  1 1134 0.81159604\n",
      "train:  1 1135 0.6228584\n",
      "train:  1 1136 0.5931818\n",
      "train:  1 1137 0.63370794\n",
      "train:  1 1138 0.63189685\n",
      "train:  1 1139 1.6412393\n",
      "train:  1 1140 0.507674\n",
      "evalu:  1 1140 [0.9124265]\n",
      "train:  1 1141 0.9948234\n",
      "train:  1 1142 0.7141688\n",
      "train:  1 1143 0.63050514\n",
      "train:  1 1144 2.9797854\n",
      "train:  1 1145 0.8454686\n",
      "train:  1 1146 1.8314351\n",
      "train:  1 1147 0.68330127\n",
      "train:  1 1148 0.7230015\n",
      "train:  1 1149 0.6543159\n",
      "train:  1 1150 0.66915476\n",
      "train:  1 1151 0.77614665\n",
      "train:  1 1152 0.62679476\n",
      "train:  1 1153 0.909424\n",
      "train:  1 1154 0.6020909\n",
      "train:  1 1155 0.65104306\n",
      "train:  1 1156 0.5361036\n",
      "train:  1 1157 0.8299502\n",
      "train:  1 1158 0.6588272\n",
      "train:  1 1159 0.8107142\n",
      "train:  1 1160 0.6773008\n",
      "evalu:  1 1160 [0.91205853]\n",
      "train:  1 1161 0.68264884\n",
      "train:  1 1162 0.5647653\n",
      "train:  1 1163 0.71430826\n",
      "train:  1 1164 0.90076464\n",
      "train:  1 1165 0.7225538\n",
      "train:  1 1166 0.6272372\n",
      "train:  1 1167 0.874618\n",
      "train:  1 1168 0.6154536\n",
      "train:  1 1169 0.6560138\n",
      "train:  1 1170 1.4110967\n",
      "train:  1 1171 1.8182412\n",
      "train:  1 1172 0.62731206\n",
      "train:  1 1173 1.187896\n",
      "train:  1 1174 0.899314\n",
      "train:  1 1175 0.6869123\n",
      "train:  1 1176 1.4138433\n",
      "train:  1 1177 1.0564582\n",
      "train:  1 1178 0.6271232\n",
      "train:  1 1179 0.6296363\n",
      "train:  1 1180 0.7167741\n",
      "evalu:  1 1180 [0.91159314]\n",
      "train:  1 1181 0.71712124\n",
      "train:  1 1182 0.7160293\n",
      "train:  1 1183 0.60326564\n",
      "train:  1 1184 0.6872251\n",
      "train:  1 1185 0.8344777\n",
      "train:  1 1186 0.61634076\n",
      "train:  1 1187 0.61081135\n",
      "train:  1 1188 0.74641585\n",
      "train:  1 1189 0.6619198\n",
      "train:  1 1190 1.1263552\n",
      "train:  1 1191 0.6153969\n",
      "train:  1 1192 0.6787172\n",
      "train:  1 1193 0.7900716\n",
      "train:  1 1194 0.7604551\n",
      "train:  1 1195 0.6488257\n",
      "train:  1 1196 0.68005294\n",
      "train:  1 1197 1.0604122\n",
      "train:  1 1198 0.53003156\n",
      "train:  1 1199 0.7220197\n",
      "train:  1 1200 0.4758148\n",
      "evalu:  1 1200 [0.91107464]\n",
      "train:  1 1201 0.65087026\n",
      "train:  1 1202 1.1148875\n",
      "train:  1 1203 0.6899087\n",
      "train:  1 1204 0.53876853\n",
      "train:  1 1205 0.67614883\n",
      "train:  1 1206 0.70271474\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  1 1207 0.603915\n",
      "train:  1 1208 0.8013272\n",
      "train:  1 1209 1.096337\n",
      "train:  1 1210 0.92393166\n",
      "train:  1 1211 0.8197469\n",
      "train:  1 1212 0.43883452\n",
      "train:  1 1213 0.59485465\n",
      "train:  1 1214 0.68729424\n",
      "train:  1 1215 0.6748532\n",
      "train:  1 1216 1.2033545\n",
      "train:  1 1217 0.8544743\n",
      "train:  1 1218 0.5527628\n",
      "train:  1 1219 0.5376373\n",
      "train:  1 1220 0.6476908\n",
      "evalu:  1 1220 [0.9108398]\n",
      "train:  1 1221 1.0380305\n",
      "train:  1 1222 0.5685123\n",
      "train:  1 1223 0.6688236\n",
      "train:  1 1224 0.76843643\n",
      "train:  1 1225 0.5956304\n",
      "train:  1 1226 0.84901637\n",
      "train:  1 1227 0.5433454\n",
      "train:  1 1228 0.5251633\n",
      "train:  1 1229 0.8068374\n",
      "train:  1 1230 0.6684217\n",
      "train:  1 1231 0.5574812\n",
      "train:  1 1232 1.1088084\n",
      "train:  1 1233 0.8751742\n",
      "train:  2 0 0.58555764\n",
      "evalu:  2 0 [0.9105314]\n",
      "train:  2 1 0.54927146\n",
      "train:  2 2 0.61724883\n",
      "train:  2 3 0.74590105\n",
      "train:  2 4 0.7141805\n",
      "train:  2 5 0.8096768\n",
      "train:  2 6 0.6866818\n",
      "train:  2 7 0.77772003\n",
      "train:  2 8 0.7526395\n",
      "train:  2 9 1.2448735\n",
      "train:  2 10 0.59158355\n",
      "train:  2 11 0.58595985\n",
      "train:  2 12 1.8642222\n",
      "train:  2 13 0.58643025\n",
      "train:  2 14 0.61132103\n",
      "train:  2 15 0.9314484\n",
      "train:  2 16 0.87093604\n",
      "train:  2 17 0.5727939\n",
      "train:  2 18 1.1937906\n",
      "train:  2 19 0.62110734\n",
      "train:  2 20 2.0116422\n",
      "evalu:  2 20 [0.91029024]\n",
      "train:  2 21 0.81837636\n",
      "train:  2 22 0.71441567\n",
      "train:  2 23 0.61560255\n",
      "train:  2 24 0.5721927\n",
      "train:  2 25 0.75643075\n",
      "train:  2 26 0.7225511\n",
      "train:  2 27 3.0533724\n",
      "train:  2 28 0.49952272\n",
      "train:  2 29 0.5020496\n",
      "train:  2 30 0.5472281\n",
      "train:  2 31 1.4557049\n",
      "train:  2 32 0.6236632\n",
      "train:  2 33 1.1908072\n",
      "train:  2 34 0.5820974\n",
      "train:  2 35 0.6384741\n",
      "train:  2 36 0.78449905\n",
      "train:  2 37 1.1953579\n",
      "train:  2 38 0.7302308\n",
      "train:  2 39 0.886572\n",
      "train:  2 40 0.87219346\n",
      "evalu:  2 40 [0.9096911]\n",
      "train:  2 41 1.0480413\n",
      "train:  2 42 1.0244827\n",
      "train:  2 43 0.6335398\n",
      "train:  2 44 0.7507928\n",
      "train:  2 45 1.3876214\n",
      "train:  2 46 1.6289526\n",
      "train:  2 47 0.6504174\n",
      "train:  2 48 0.6307996\n",
      "train:  2 49 0.5116508\n",
      "train:  2 50 1.3290157\n",
      "train:  2 51 0.5604448\n",
      "train:  2 52 0.6731969\n",
      "train:  2 53 0.62218165\n",
      "train:  2 54 0.85719955\n",
      "train:  2 55 0.8315439\n",
      "train:  2 56 0.46239904\n",
      "train:  2 57 0.66711956\n",
      "train:  2 58 0.9577176\n",
      "train:  2 59 0.57309\n",
      "train:  2 60 0.62390244\n",
      "evalu:  2 60 [0.9093179]\n",
      "train:  2 61 0.48057747\n",
      "train:  2 62 0.5762754\n",
      "train:  2 63 0.60353595\n",
      "train:  2 64 0.58311415\n",
      "train:  2 65 0.65376467\n",
      "train:  2 66 1.0597285\n",
      "train:  2 67 0.63574517\n",
      "train:  2 68 1.0801623\n",
      "train:  2 69 0.80538046\n",
      "train:  2 70 0.8164697\n",
      "train:  2 71 0.69382334\n",
      "train:  2 72 0.6110811\n",
      "train:  2 73 0.5547537\n",
      "train:  2 74 0.46574625\n",
      "train:  2 75 0.6649592\n",
      "train:  2 76 0.7121346\n",
      "train:  2 77 0.6566368\n",
      "train:  2 78 1.156076\n",
      "train:  2 79 1.2764758\n",
      "train:  2 80 0.6949372\n",
      "evalu:  2 80 [0.9088555]\n",
      "train:  2 81 0.5948462\n",
      "train:  2 82 0.61203694\n",
      "train:  2 83 0.60609204\n",
      "train:  2 84 1.4383192\n",
      "train:  2 85 0.69304067\n",
      "train:  2 86 0.55932474\n",
      "train:  2 87 0.7885023\n",
      "train:  2 88 0.57319266\n",
      "train:  2 89 0.6847039\n",
      "train:  2 90 0.6755613\n",
      "train:  2 91 0.6626195\n",
      "train:  2 92 0.63749266\n",
      "train:  2 93 0.5964535\n",
      "train:  2 94 0.5790061\n",
      "train:  2 95 0.6996359\n",
      "train:  2 96 0.63898844\n",
      "train:  2 97 0.8633371\n",
      "train:  2 98 0.60425556\n",
      "train:  2 99 0.6214914\n",
      "train:  2 100 0.70542824\n",
      "evalu:  2 100 [0.9087028]\n",
      "train:  2 101 0.71155274\n",
      "train:  2 102 1.1758887\n",
      "train:  2 103 0.7769544\n",
      "train:  2 104 1.8086343\n",
      "train:  2 105 1.2065407\n",
      "train:  2 106 0.62465256\n",
      "train:  2 107 0.9999738\n",
      "train:  2 108 1.4493083\n",
      "train:  2 109 1.1231964\n",
      "train:  2 110 0.5133827\n",
      "train:  2 111 0.5869003\n",
      "train:  2 112 1.240827\n",
      "train:  2 113 0.7585337\n",
      "train:  2 114 0.5299721\n",
      "train:  2 115 0.65635204\n",
      "train:  2 116 0.6966512\n",
      "train:  2 117 0.63162184\n",
      "train:  2 118 0.5791033\n",
      "train:  2 119 0.6654054\n",
      "train:  2 120 0.7512614\n",
      "evalu:  2 120 [0.9082242]\n",
      "train:  2 121 1.1449672\n",
      "train:  2 122 0.674859\n",
      "train:  2 123 0.9830652\n",
      "train:  2 124 0.6647787\n",
      "train:  2 125 0.67053866\n",
      "train:  2 126 0.7543587\n",
      "train:  2 127 0.7852985\n",
      "train:  2 128 0.782869\n",
      "train:  2 129 1.1569422\n",
      "train:  2 130 0.599925\n",
      "train:  2 131 0.56937844\n",
      "train:  2 132 0.6986009\n",
      "train:  2 133 0.4886588\n",
      "train:  2 134 0.918944\n",
      "train:  2 135 0.77480996\n",
      "train:  2 136 1.2773665\n",
      "train:  2 137 1.8337666\n",
      "train:  2 138 0.6891947\n",
      "train:  2 139 1.4480212\n",
      "train:  2 140 0.6409066\n",
      "evalu:  2 140 [0.90778697]\n",
      "train:  2 141 0.590715\n",
      "train:  2 142 0.65047354\n",
      "train:  2 143 0.6387963\n",
      "train:  2 144 0.72262144\n",
      "train:  2 145 0.6442058\n",
      "train:  2 146 0.6537308\n",
      "train:  2 147 2.95342\n",
      "train:  2 148 0.68463284\n",
      "train:  2 149 0.49465284\n",
      "train:  2 150 0.49073827\n",
      "train:  2 151 0.6654521\n",
      "train:  2 152 1.9785682\n",
      "train:  2 153 0.7931717\n",
      "train:  2 154 0.7598344\n",
      "train:  2 155 0.9737522\n",
      "train:  2 156 0.64233357\n",
      "train:  2 157 1.8747324\n",
      "train:  2 158 0.92773974\n",
      "train:  2 159 0.5832479\n",
      "train:  2 160 0.6289044\n",
      "evalu:  2 160 [0.90756315]\n",
      "train:  2 161 1.1189834\n",
      "train:  2 162 0.7754707\n",
      "train:  2 163 0.6743079\n",
      "train:  2 164 0.7075685\n",
      "train:  2 165 0.79423046\n",
      "train:  2 166 0.50478494\n",
      "train:  2 167 0.7861932\n",
      "train:  2 168 2.628354\n",
      "train:  2 169 0.7174882\n",
      "train:  2 170 0.54891384\n",
      "train:  2 171 0.800626\n",
      "train:  2 172 0.61672264\n",
      "train:  2 173 0.7620865\n",
      "train:  2 174 0.6709721\n",
      "train:  2 175 0.52658397\n",
      "train:  2 176 0.6866892\n",
      "train:  2 177 0.60692495\n",
      "train:  2 178 0.8203841\n",
      "train:  2 179 0.8619598\n",
      "train:  2 180 0.5119292\n",
      "evalu:  2 180 [0.907281]\n",
      "train:  2 181 0.81755596\n",
      "train:  2 182 0.5956075\n",
      "train:  2 183 0.5607212\n",
      "train:  2 184 0.96787566\n",
      "train:  2 185 0.5607016\n",
      "train:  2 186 1.0403795\n",
      "train:  2 187 1.6191963\n",
      "train:  2 188 0.63728\n",
      "train:  2 189 1.155703\n",
      "train:  2 190 0.64802146\n",
      "train:  2 191 0.5974825\n",
      "train:  2 192 0.7954589\n",
      "train:  2 193 0.5165801\n",
      "train:  2 194 0.4749747\n",
      "train:  2 195 0.6376521\n",
      "train:  2 196 0.5631233\n",
      "train:  2 197 0.6447639\n",
      "train:  2 198 0.6043425\n",
      "train:  2 199 0.58188576\n",
      "train:  2 200 0.6338679\n",
      "evalu:  2 200 [0.90702367]\n",
      "train:  2 201 0.6138636\n",
      "train:  2 202 1.3481529\n",
      "train:  2 203 0.7102071\n",
      "train:  2 204 0.46322617\n",
      "train:  2 205 0.644266\n",
      "train:  2 206 0.8131841\n",
      "train:  2 207 0.5861356\n",
      "train:  2 208 1.4810885\n",
      "train:  2 209 1.078022\n",
      "train:  2 210 0.6106184\n",
      "train:  2 211 0.7828258\n",
      "train:  2 212 0.6193482\n",
      "train:  2 213 0.85342205\n",
      "train:  2 214 0.8054678\n",
      "train:  2 215 0.6525355\n",
      "train:  2 216 0.5093557\n",
      "train:  2 217 1.3077825\n",
      "train:  2 218 0.63928026\n",
      "train:  2 219 0.6220797\n",
      "train:  2 220 0.63199794\n",
      "evalu:  2 220 [0.90654165]\n",
      "train:  2 221 0.7547004\n",
      "train:  2 222 0.6159998\n",
      "train:  2 223 0.5663497\n",
      "train:  2 224 0.6084347\n",
      "train:  2 225 1.4820346\n",
      "train:  2 226 0.65351874\n",
      "train:  2 227 1.6861012\n",
      "train:  2 228 0.5859149\n",
      "train:  2 229 0.919313\n",
      "train:  2 230 1.0852226\n",
      "train:  2 231 0.5778886\n",
      "train:  2 232 0.5344239\n",
      "train:  2 233 0.6118104\n",
      "train:  2 234 0.6603139\n",
      "train:  2 235 0.93887556\n",
      "train:  2 236 0.8356748\n",
      "train:  2 237 0.59268373\n",
      "train:  2 238 1.1070843\n",
      "train:  2 239 1.1195784\n",
      "train:  2 240 0.5218175\n",
      "evalu:  2 240 [0.90611804]\n",
      "train:  2 241 0.9625297\n",
      "train:  2 242 0.7976852\n",
      "train:  2 243 0.863261\n",
      "train:  2 244 0.7502639\n",
      "train:  2 245 0.8053846\n",
      "train:  2 246 0.5707664\n",
      "train:  2 247 0.57642025\n",
      "train:  2 248 0.63577163\n",
      "train:  2 249 0.9435697\n",
      "train:  2 250 0.61864823\n",
      "train:  2 251 0.5487133\n",
      "train:  2 252 0.5945799\n",
      "train:  2 253 0.5824689\n",
      "train:  2 254 0.9077531\n",
      "train:  2 255 0.6263447\n",
      "train:  2 256 0.6734048\n",
      "train:  2 257 0.58296686\n",
      "train:  2 258 1.0120097\n",
      "train:  2 259 1.1162139\n",
      "train:  2 260 0.84234935\n",
      "evalu:  2 260 [0.9057149]\n",
      "train:  2 261 0.7957839\n",
      "train:  2 262 0.5931646\n",
      "train:  2 263 1.5571045\n",
      "train:  2 264 1.0279357\n",
      "train:  2 265 1.0286795\n",
      "train:  2 266 0.71567124\n",
      "train:  2 267 1.0549185\n",
      "train:  2 268 0.71757585\n",
      "train:  2 269 0.93219733\n",
      "train:  2 270 0.7653107\n",
      "train:  2 271 0.7393874\n",
      "train:  2 272 0.6288713\n",
      "train:  2 273 0.72519207\n",
      "train:  2 274 0.5548816\n",
      "train:  2 275 0.6569395\n",
      "train:  2 276 0.70279485\n",
      "train:  2 277 0.85979736\n",
      "train:  2 278 1.1707666\n",
      "train:  2 279 0.7001731\n",
      "train:  2 280 0.6543241\n",
      "evalu:  2 280 [0.9054469]\n",
      "train:  2 281 0.6138013\n",
      "train:  2 282 0.7622954\n",
      "train:  2 283 0.7054906\n",
      "train:  2 284 0.7176843\n",
      "train:  2 285 0.7534678\n",
      "train:  2 286 0.6507628\n",
      "train:  2 287 0.71032715\n",
      "train:  2 288 0.5701292\n",
      "train:  2 289 0.6964821\n",
      "train:  2 290 0.7783185\n",
      "train:  2 291 0.83647865\n",
      "train:  2 292 0.5524261\n",
      "train:  2 293 1.545984\n",
      "train:  2 294 0.64689565\n",
      "train:  2 295 0.95779395\n",
      "train:  2 296 3.450632\n",
      "train:  2 297 0.756971\n",
      "train:  2 298 0.6391191\n",
      "train:  2 299 0.7141684\n",
      "train:  2 300 1.2486818\n",
      "evalu:  2 300 [0.90514886]\n",
      "train:  2 301 0.57145107\n",
      "train:  2 302 0.6416762\n",
      "train:  2 303 0.7683178\n",
      "train:  2 304 0.75525796\n",
      "train:  2 305 0.47640342\n",
      "train:  2 306 0.76147735\n",
      "train:  2 307 0.63399976\n",
      "train:  2 308 0.6559914\n",
      "train:  2 309 1.3584741\n",
      "train:  2 310 0.88850373\n",
      "train:  2 311 0.67269224\n",
      "train:  2 312 0.6255426\n",
      "train:  2 313 1.0556325\n",
      "train:  2 314 1.0760343\n",
      "train:  2 315 0.67166466\n",
      "train:  2 316 0.6501093\n",
      "train:  2 317 0.848944\n",
      "train:  2 318 0.55850995\n",
      "train:  2 319 1.312158\n",
      "train:  2 320 1.4569583\n",
      "evalu:  2 320 [0.9049493]\n",
      "train:  2 321 0.61715394\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  2 322 0.6296251\n",
      "train:  2 323 0.57602125\n",
      "train:  2 324 0.64368016\n",
      "train:  2 325 0.8344209\n",
      "train:  2 326 0.9914121\n",
      "train:  2 327 0.5785963\n",
      "train:  2 328 0.54902446\n",
      "train:  2 329 0.77795345\n",
      "train:  2 330 1.335861\n",
      "train:  2 331 0.51486903\n",
      "train:  2 332 0.6969838\n",
      "train:  2 333 2.7792454\n",
      "train:  2 334 0.6645514\n",
      "train:  2 335 0.50759876\n",
      "train:  2 336 0.5121073\n",
      "train:  2 337 0.72113454\n",
      "train:  2 338 2.3828416\n",
      "train:  2 339 0.7501125\n",
      "train:  2 340 0.7176814\n",
      "evalu:  2 340 [0.90464264]\n",
      "train:  2 341 0.71636635\n",
      "train:  2 342 0.83346677\n",
      "train:  2 343 0.6340996\n",
      "train:  2 344 0.6034167\n",
      "train:  2 345 0.63721246\n",
      "train:  2 346 0.5664722\n",
      "train:  2 347 0.50565684\n",
      "train:  2 348 0.687792\n",
      "train:  2 349 0.77669203\n",
      "train:  2 350 1.7015133\n",
      "train:  2 351 0.7280365\n",
      "train:  2 352 0.52934015\n",
      "train:  2 353 0.73047084\n",
      "train:  2 354 0.6829692\n",
      "train:  2 355 0.9776945\n",
      "train:  2 356 0.55897844\n",
      "train:  2 357 0.5894986\n",
      "train:  2 358 0.8698948\n",
      "train:  2 359 0.6030914\n",
      "train:  2 360 0.5894459\n",
      "evalu:  2 360 [0.9043657]\n",
      "train:  2 361 0.8930866\n",
      "train:  2 362 1.0040966\n",
      "train:  2 363 0.7105791\n",
      "train:  2 364 0.7042016\n",
      "train:  2 365 0.6078101\n",
      "train:  2 366 0.66150475\n",
      "train:  2 367 1.895915\n",
      "train:  2 368 0.86788875\n",
      "train:  2 369 1.1381999\n",
      "train:  2 370 0.7202931\n",
      "train:  2 371 0.8516908\n",
      "train:  2 372 0.5540769\n",
      "train:  2 373 2.1218998\n",
      "train:  2 374 0.76336944\n",
      "train:  2 375 0.7873901\n",
      "train:  2 376 0.5312047\n",
      "train:  2 377 1.0974287\n",
      "train:  2 378 0.6574016\n",
      "train:  2 379 0.95599395\n",
      "train:  2 380 0.592381\n",
      "evalu:  2 380 [0.9039661]\n",
      "train:  2 381 2.4701333\n",
      "train:  2 382 0.9332335\n",
      "train:  2 383 0.88568455\n",
      "train:  2 384 1.0367088\n",
      "train:  2 385 0.7186027\n",
      "train:  2 386 0.93304926\n",
      "train:  2 387 0.77688026\n",
      "train:  2 388 0.74369603\n",
      "train:  2 389 0.9807606\n",
      "train:  2 390 0.7456509\n",
      "train:  2 391 1.235808\n",
      "train:  2 392 0.7463646\n",
      "train:  2 393 1.2896916\n",
      "train:  2 394 0.5502262\n",
      "train:  2 395 0.6595555\n",
      "train:  2 396 1.0580957\n",
      "train:  2 397 1.224895\n",
      "train:  2 398 0.92393196\n",
      "train:  2 399 0.5848317\n",
      "train:  2 400 0.6383843\n",
      "evalu:  2 400 [0.90350884]\n",
      "train:  2 401 0.7344006\n",
      "train:  2 402 0.45381668\n",
      "train:  2 403 0.61071724\n",
      "train:  2 404 0.5138931\n",
      "train:  2 405 0.4898294\n",
      "train:  2 406 0.58604795\n",
      "train:  2 407 0.593559\n",
      "train:  2 408 0.6923068\n",
      "train:  2 409 0.67064524\n",
      "train:  2 410 0.721209\n",
      "train:  2 411 0.5295827\n",
      "train:  2 412 0.5437133\n",
      "train:  2 413 0.5737177\n",
      "train:  2 414 0.672628\n",
      "train:  2 415 0.57557374\n",
      "train:  2 416 0.6058426\n",
      "train:  2 417 0.5556769\n",
      "train:  2 418 0.8838013\n",
      "train:  2 419 1.5649147\n",
      "train:  2 420 0.66714114\n",
      "evalu:  2 420 [0.90328526]\n",
      "train:  2 421 0.74563444\n",
      "train:  2 422 0.54699254\n",
      "train:  2 423 0.63727593\n",
      "train:  2 424 1.2011302\n",
      "train:  2 425 1.8521729\n",
      "train:  2 426 1.3548023\n",
      "train:  2 427 0.65745515\n",
      "train:  2 428 0.7132733\n",
      "train:  2 429 0.58010477\n",
      "train:  2 430 1.026993\n",
      "train:  2 431 0.6141905\n",
      "train:  2 432 2.1046503\n",
      "train:  2 433 0.62664366\n",
      "train:  2 434 0.6896997\n",
      "train:  2 435 0.6351438\n",
      "train:  2 436 0.6133216\n",
      "train:  2 437 0.5898033\n",
      "train:  2 438 0.7744168\n",
      "train:  2 439 0.6486538\n",
      "train:  2 440 0.626142\n",
      "evalu:  2 440 [0.9029373]\n",
      "train:  2 441 0.6537722\n",
      "train:  2 442 0.7828209\n",
      "train:  2 443 0.87308246\n",
      "train:  2 444 0.69249904\n",
      "train:  2 445 0.7192352\n",
      "train:  2 446 0.69406813\n",
      "train:  2 447 0.70868474\n",
      "train:  2 448 0.57682765\n",
      "train:  2 449 0.67385936\n",
      "train:  2 450 0.6841081\n",
      "train:  2 451 0.6886138\n",
      "train:  2 452 1.1996179\n",
      "train:  2 453 0.6933348\n",
      "train:  2 454 0.6959467\n",
      "train:  2 455 0.43919247\n",
      "train:  2 456 0.52933025\n",
      "train:  2 457 0.6275207\n",
      "train:  2 458 0.7921876\n",
      "train:  2 459 0.5076322\n",
      "train:  2 460 0.5262936\n",
      "evalu:  2 460 [0.90273124]\n",
      "train:  2 461 0.93937546\n",
      "train:  2 462 0.6363416\n",
      "train:  2 463 0.5853215\n",
      "train:  2 464 0.71921176\n",
      "train:  2 465 0.7777899\n",
      "train:  2 466 0.6923565\n",
      "train:  2 467 1.1401176\n",
      "train:  2 468 0.8742788\n",
      "train:  2 469 0.6629713\n",
      "train:  2 470 0.83773077\n",
      "train:  2 471 0.6648511\n",
      "train:  2 472 1.2418171\n",
      "train:  2 473 0.6825398\n",
      "train:  2 474 1.8214846\n",
      "train:  2 475 0.69425905\n",
      "train:  2 476 0.56441754\n",
      "train:  2 477 0.5371013\n",
      "train:  2 478 1.1450583\n",
      "train:  2 479 0.5348234\n",
      "train:  2 480 0.6402292\n",
      "evalu:  2 480 [0.90242285]\n",
      "train:  2 481 0.7243055\n",
      "train:  2 482 0.9764159\n",
      "train:  2 483 0.5320743\n",
      "train:  2 484 1.3245599\n",
      "train:  2 485 0.7729357\n",
      "train:  2 486 0.6203891\n",
      "train:  2 487 0.56501895\n",
      "train:  2 488 0.84717953\n",
      "train:  2 489 0.49605587\n",
      "train:  2 490 1.1023109\n",
      "train:  2 491 1.2232382\n",
      "train:  2 492 1.5560052\n",
      "train:  2 493 0.57679844\n",
      "train:  2 494 0.53056127\n",
      "train:  2 495 0.69858164\n",
      "train:  2 496 0.6400481\n",
      "train:  2 497 0.48675984\n",
      "train:  2 498 0.6453399\n",
      "train:  2 499 0.65079683\n",
      "train:  2 500 0.56007874\n",
      "evalu:  2 500 [0.9020689]\n",
      "train:  2 501 0.8197422\n",
      "train:  2 502 1.2850082\n",
      "train:  2 503 0.5633283\n",
      "train:  2 504 1.1200472\n",
      "train:  2 505 0.82109284\n",
      "train:  2 506 0.7642533\n",
      "train:  2 507 1.599441\n",
      "train:  2 508 0.8737645\n",
      "train:  2 509 0.85656315\n",
      "train:  2 510 0.7887625\n",
      "train:  2 511 0.7689873\n",
      "train:  2 512 0.9932331\n",
      "train:  2 513 0.51920086\n",
      "train:  2 514 0.6746542\n",
      "train:  2 515 0.8420266\n",
      "train:  2 516 0.5907341\n",
      "train:  2 517 0.86873394\n",
      "train:  2 518 0.9960143\n",
      "train:  2 519 0.49056107\n",
      "train:  2 520 0.6151451\n",
      "evalu:  2 520 [0.90234894]\n",
      "train:  2 521 0.56039345\n",
      "train:  2 522 1.0946616\n",
      "train:  2 523 1.3749546\n",
      "train:  2 524 1.7942245\n",
      "train:  2 525 0.86178595\n",
      "train:  2 526 0.6280526\n",
      "train:  2 527 0.80811477\n",
      "train:  2 528 0.5931207\n",
      "train:  2 529 0.7362744\n",
      "train:  2 530 0.67489743\n",
      "train:  2 531 0.68651205\n",
      "train:  2 532 0.6622015\n",
      "train:  2 533 0.693197\n",
      "train:  2 534 0.6817004\n",
      "train:  2 535 1.0976597\n",
      "train:  2 536 1.9147106\n",
      "train:  2 537 0.6483865\n",
      "train:  2 538 0.46961528\n",
      "train:  2 539 0.5811103\n",
      "train:  2 540 0.67571443\n",
      "evalu:  2 540 [0.9017614]\n",
      "train:  2 541 0.5361233\n",
      "train:  2 542 0.64562446\n",
      "train:  2 543 0.598138\n",
      "train:  2 544 0.61599946\n",
      "train:  2 545 0.42558187\n",
      "train:  2 546 0.54773724\n",
      "train:  2 547 0.5708386\n",
      "train:  2 548 0.78844863\n",
      "train:  2 549 0.46240804\n",
      "train:  2 550 0.61594754\n",
      "train:  2 551 0.71441007\n",
      "train:  2 552 0.7725327\n",
      "train:  2 553 0.50031936\n",
      "train:  2 554 0.67928606\n",
      "train:  2 555 0.6750218\n",
      "train:  2 556 0.5532698\n",
      "train:  2 557 2.3348517\n",
      "train:  2 558 0.52712214\n",
      "train:  2 559 0.6609935\n",
      "train:  2 560 0.49397498\n",
      "evalu:  2 560 [0.9014187]\n",
      "train:  2 561 0.6266658\n",
      "train:  2 562 0.5860614\n",
      "train:  2 563 0.68653905\n",
      "train:  2 564 0.77603644\n",
      "train:  2 565 0.60983837\n",
      "train:  2 566 1.2157028\n",
      "train:  2 567 1.3540045\n",
      "train:  2 568 0.86514306\n",
      "train:  2 569 1.4264525\n",
      "train:  2 570 0.8292618\n",
      "train:  2 571 1.2127312\n",
      "train:  2 572 1.1838323\n",
      "train:  2 573 0.67336786\n",
      "train:  2 574 0.82607186\n",
      "train:  2 575 0.4707793\n",
      "train:  2 576 0.9928816\n",
      "train:  2 577 0.64069074\n",
      "train:  2 578 0.9617214\n",
      "train:  2 579 0.7954429\n",
      "train:  2 580 0.6785974\n",
      "evalu:  2 580 [0.90110165]\n",
      "train:  2 581 0.52153534\n",
      "train:  2 582 1.2235553\n",
      "train:  2 583 0.572076\n",
      "train:  2 584 0.5850754\n",
      "train:  2 585 0.75954086\n",
      "train:  2 586 0.8762557\n",
      "train:  2 587 0.5056397\n",
      "train:  2 588 0.5624317\n",
      "train:  2 589 0.58401084\n",
      "train:  2 590 0.57319814\n",
      "train:  2 591 0.64260495\n",
      "train:  2 592 1.3638371\n",
      "train:  2 593 1.626779\n",
      "train:  2 594 0.7403674\n",
      "train:  2 595 0.7537256\n",
      "train:  2 596 0.8150443\n",
      "train:  2 597 0.9569219\n",
      "train:  2 598 0.5210287\n",
      "train:  2 599 0.588858\n",
      "train:  2 600 0.6699716\n",
      "evalu:  2 600 [0.9006565]\n",
      "train:  2 601 0.76402086\n",
      "train:  2 602 1.2987115\n",
      "train:  2 603 0.8452412\n",
      "train:  2 604 1.4177868\n",
      "train:  2 605 0.5652426\n",
      "train:  2 606 0.6514083\n",
      "train:  2 607 0.58016336\n",
      "train:  2 608 0.77963036\n",
      "train:  2 609 0.7656413\n",
      "train:  2 610 1.106882\n",
      "train:  2 611 0.60474104\n",
      "train:  2 612 1.2632825\n",
      "train:  2 613 0.79529005\n",
      "train:  2 614 0.6839295\n",
      "train:  2 615 0.48947844\n",
      "train:  2 616 0.52188367\n",
      "train:  2 617 0.59791195\n",
      "train:  2 618 3.3879623\n",
      "train:  2 619 0.5382237\n",
      "train:  2 620 0.75694174\n",
      "evalu:  2 620 [0.90042377]\n",
      "train:  2 621 1.1280143\n",
      "train:  2 622 0.5901844\n",
      "train:  2 623 0.52918285\n",
      "train:  2 624 0.8168821\n",
      "train:  2 625 0.9253759\n",
      "train:  2 626 0.7131408\n",
      "train:  2 627 0.63708496\n",
      "train:  2 628 1.152088\n",
      "train:  2 629 0.7687139\n",
      "train:  2 630 0.7315974\n",
      "train:  2 631 0.55252254\n",
      "train:  2 632 0.58613604\n",
      "train:  2 633 0.57976794\n",
      "train:  2 634 0.66034204\n",
      "train:  2 635 0.648484\n",
      "train:  2 636 0.63867706\n",
      "train:  2 637 0.82095087\n",
      "train:  2 638 0.768843\n",
      "train:  2 639 1.0059459\n",
      "train:  2 640 0.6247806\n",
      "evalu:  2 640 [0.90037036]\n",
      "train:  2 641 0.7536037\n",
      "train:  2 642 0.82526904\n",
      "train:  2 643 0.7560636\n",
      "train:  2 644 0.5169753\n",
      "train:  2 645 0.7148683\n",
      "train:  2 646 0.5276708\n",
      "train:  2 647 0.61604494\n",
      "train:  2 648 0.48744\n",
      "train:  2 649 1.8669896\n",
      "train:  2 650 0.70174444\n",
      "train:  2 651 0.91638815\n",
      "train:  2 652 0.70534456\n",
      "train:  2 653 0.8425288\n",
      "train:  2 654 0.5991849\n",
      "train:  2 655 0.5705185\n",
      "train:  2 656 1.4130905\n",
      "train:  2 657 0.5884707\n",
      "train:  2 658 0.9392457\n",
      "train:  2 659 0.8195936\n",
      "train:  2 660 0.52419114\n",
      "evalu:  2 660 [0.9000832]\n",
      "train:  2 661 1.0601674\n",
      "train:  2 662 1.0647001\n",
      "train:  2 663 1.0251303\n",
      "train:  2 664 0.8119751\n",
      "train:  2 665 0.74088484\n",
      "train:  2 666 1.7502375\n",
      "train:  2 667 0.5537574\n",
      "train:  2 668 0.69088405\n",
      "train:  2 669 0.55710644\n",
      "train:  2 670 0.48194113\n",
      "train:  2 671 0.5105703\n",
      "train:  2 672 0.7677004\n",
      "train:  2 673 0.6621582\n",
      "train:  2 674 0.5194243\n",
      "train:  2 675 0.9679871\n",
      "train:  2 676 1.7680312\n",
      "train:  2 677 0.65367156\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  2 678 0.99519485\n",
      "train:  2 679 0.84625375\n",
      "train:  2 680 0.56733555\n",
      "evalu:  2 680 [0.8997115]\n",
      "train:  2 681 0.76863873\n",
      "train:  2 682 0.5190171\n",
      "train:  2 683 0.70339924\n",
      "train:  2 684 0.54192996\n",
      "train:  2 685 0.70776546\n",
      "train:  2 686 0.5711627\n",
      "train:  2 687 1.1704925\n",
      "train:  2 688 1.5896218\n",
      "train:  2 689 0.59106356\n",
      "train:  2 690 1.4602116\n",
      "train:  2 691 0.73581207\n",
      "train:  2 692 0.71297103\n",
      "train:  2 693 0.7509394\n",
      "train:  2 694 0.6828522\n",
      "train:  2 695 0.83209735\n",
      "train:  2 696 0.5911527\n",
      "train:  2 697 0.5978572\n",
      "train:  2 698 0.59056777\n",
      "train:  2 699 0.7513731\n",
      "train:  2 700 0.67091346\n",
      "evalu:  2 700 [0.89966416]\n",
      "train:  2 701 0.4953383\n",
      "train:  2 702 0.57035285\n",
      "train:  2 703 0.5384105\n",
      "train:  2 704 0.5406341\n",
      "train:  2 705 0.5174014\n",
      "train:  2 706 0.7994968\n",
      "train:  2 707 0.5624466\n",
      "train:  2 708 0.6224085\n",
      "train:  2 709 0.58109534\n",
      "train:  2 710 0.62711936\n",
      "train:  2 711 0.8117952\n",
      "train:  2 712 0.7775436\n",
      "train:  2 713 1.002899\n",
      "train:  2 714 0.5387534\n",
      "train:  2 715 1.2112851\n",
      "train:  2 716 0.7701358\n",
      "train:  2 717 0.5232656\n",
      "train:  2 718 0.57424927\n",
      "train:  2 719 0.77112955\n",
      "train:  2 720 0.8741556\n",
      "evalu:  2 720 [0.89936787]\n",
      "train:  2 721 0.6054352\n",
      "train:  2 722 0.92434174\n",
      "train:  2 723 0.9037135\n",
      "train:  2 724 1.0377568\n",
      "train:  2 725 0.55446804\n",
      "train:  2 726 0.8734728\n",
      "train:  2 727 0.74082977\n",
      "train:  2 728 0.64108115\n",
      "train:  2 729 0.6442792\n",
      "train:  2 730 1.1882067\n",
      "train:  2 731 0.6907986\n",
      "train:  2 732 0.52668136\n",
      "train:  2 733 0.76455116\n",
      "train:  2 734 0.83035296\n",
      "train:  2 735 0.7546696\n",
      "train:  2 736 0.7309673\n",
      "train:  2 737 2.0804458\n",
      "train:  2 738 0.73549366\n",
      "train:  2 739 0.6988704\n",
      "train:  2 740 0.53717893\n",
      "evalu:  2 740 [0.8990689]\n",
      "train:  2 741 0.81718034\n",
      "train:  2 742 0.7746547\n",
      "train:  2 743 0.6388802\n",
      "train:  2 744 0.646718\n",
      "train:  2 745 0.8283672\n",
      "train:  2 746 1.240038\n",
      "train:  2 747 0.72374105\n",
      "train:  2 748 0.89515233\n",
      "train:  2 749 1.1377993\n",
      "train:  2 750 0.4853439\n",
      "train:  2 751 1.1695144\n",
      "train:  2 752 0.76678985\n",
      "train:  2 753 0.65140754\n",
      "train:  2 754 0.6701174\n",
      "train:  2 755 0.9287221\n",
      "train:  2 756 0.63738817\n",
      "train:  2 757 0.7224547\n",
      "train:  2 758 0.93341094\n",
      "train:  2 759 0.7071605\n",
      "train:  2 760 0.5760659\n",
      "evalu:  2 760 [0.8990365]\n",
      "train:  2 761 0.6421708\n",
      "train:  2 762 1.9683174\n",
      "train:  2 763 0.6288828\n",
      "train:  2 764 0.66820204\n",
      "train:  2 765 0.6129648\n",
      "train:  2 766 0.54122293\n",
      "train:  2 767 0.46159694\n",
      "train:  2 768 0.7568696\n",
      "train:  2 769 0.7972545\n",
      "train:  2 770 0.9036725\n",
      "train:  2 771 0.78299993\n",
      "train:  2 772 0.55406266\n",
      "train:  2 773 0.7519699\n",
      "train:  2 774 0.71497613\n",
      "train:  2 775 0.5664111\n",
      "train:  2 776 0.8160482\n",
      "train:  2 777 0.92134523\n",
      "train:  2 778 0.6804913\n",
      "train:  2 779 0.78400165\n",
      "train:  2 780 0.5658982\n",
      "evalu:  2 780 [0.8986738]\n",
      "train:  2 781 0.58974946\n",
      "train:  2 782 0.4862212\n",
      "train:  2 783 0.5555542\n",
      "train:  2 784 0.6758753\n",
      "train:  2 785 0.6690922\n",
      "train:  2 786 0.6252522\n",
      "train:  2 787 0.72349316\n",
      "train:  2 788 0.51803327\n",
      "train:  2 789 0.80741006\n",
      "train:  2 790 0.8773595\n",
      "train:  2 791 0.62689877\n",
      "train:  2 792 0.7933836\n",
      "train:  2 793 1.0726669\n",
      "train:  2 794 0.87738234\n",
      "train:  2 795 1.1050297\n",
      "train:  2 796 0.5654803\n",
      "train:  2 797 0.843883\n",
      "train:  2 798 0.7397741\n",
      "train:  2 799 0.4660211\n",
      "train:  2 800 0.870998\n",
      "evalu:  2 800 [0.8985818]\n",
      "train:  2 801 0.57374805\n",
      "train:  2 802 0.56919247\n",
      "train:  2 803 0.52762973\n",
      "train:  2 804 0.5528945\n",
      "train:  2 805 0.6067573\n",
      "train:  2 806 0.5280032\n",
      "train:  2 807 1.2379552\n",
      "train:  2 808 0.5861093\n",
      "train:  2 809 0.6034564\n",
      "train:  2 810 0.85152817\n",
      "train:  2 811 0.5488054\n",
      "train:  2 812 0.739025\n",
      "train:  2 813 0.9415348\n",
      "train:  2 814 3.7527335\n",
      "train:  2 815 1.0331101\n",
      "train:  2 816 0.58933055\n",
      "train:  2 817 0.553622\n",
      "train:  2 818 0.52419907\n",
      "train:  2 819 0.631556\n",
      "train:  2 820 0.69708335\n",
      "evalu:  2 820 [0.89848274]\n",
      "train:  2 821 0.6524784\n",
      "train:  2 822 0.64592546\n",
      "train:  2 823 0.7013021\n",
      "train:  2 824 0.626836\n",
      "train:  2 825 0.5442402\n",
      "train:  2 826 1.2567412\n",
      "train:  2 827 2.1649065\n",
      "train:  2 828 1.0086514\n",
      "train:  2 829 0.687302\n",
      "train:  2 830 0.9766003\n",
      "train:  2 831 0.9706285\n",
      "train:  2 832 0.57400084\n",
      "train:  2 833 0.6705463\n",
      "train:  2 834 0.6477154\n",
      "train:  2 835 0.63580775\n",
      "train:  2 836 0.7720056\n",
      "train:  2 837 0.5719227\n",
      "train:  2 838 0.54904795\n",
      "train:  2 839 0.56721205\n",
      "train:  2 840 1.3476076\n",
      "evalu:  2 840 [0.89825946]\n",
      "train:  2 841 0.7699869\n",
      "train:  2 842 0.57744753\n",
      "train:  2 843 1.4240957\n",
      "train:  2 844 0.5299664\n",
      "train:  2 845 0.5905035\n",
      "train:  2 846 0.68824816\n",
      "train:  2 847 0.58985275\n",
      "train:  2 848 1.5101722\n",
      "train:  2 849 0.8978011\n",
      "train:  2 850 0.55837727\n",
      "train:  2 851 0.600734\n",
      "train:  2 852 0.9230972\n",
      "train:  2 853 1.3883294\n",
      "train:  2 854 0.55079055\n",
      "train:  2 855 0.66813207\n",
      "train:  2 856 0.55799025\n",
      "train:  2 857 0.76816213\n",
      "train:  2 858 1.5342046\n",
      "train:  2 859 0.94215965\n",
      "train:  2 860 0.58526284\n",
      "evalu:  2 860 [0.8978289]\n",
      "train:  2 861 0.7464934\n",
      "train:  2 862 0.7245679\n",
      "train:  2 863 0.45446435\n",
      "train:  2 864 0.6086198\n",
      "train:  2 865 0.7011808\n",
      "train:  2 866 0.95072126\n",
      "train:  2 867 0.46902058\n",
      "train:  2 868 0.5572213\n",
      "train:  2 869 0.7818924\n",
      "train:  2 870 1.1550906\n",
      "train:  2 871 0.73656297\n",
      "train:  2 872 0.76752174\n",
      "train:  2 873 0.6881976\n",
      "train:  2 874 0.5626722\n",
      "train:  2 875 0.7904248\n",
      "train:  2 876 0.7799904\n",
      "train:  2 877 1.0615923\n",
      "train:  2 878 1.3556232\n",
      "train:  2 879 0.519247\n",
      "train:  2 880 0.69696677\n",
      "evalu:  2 880 [0.89760095]\n",
      "train:  2 881 0.75657123\n",
      "train:  2 882 0.81063527\n",
      "train:  2 883 0.60768116\n",
      "train:  2 884 1.1477395\n",
      "train:  2 885 0.6949878\n",
      "train:  2 886 0.57913816\n",
      "train:  2 887 1.1117691\n",
      "train:  2 888 0.96142304\n",
      "train:  2 889 0.5104284\n",
      "train:  2 890 0.7893447\n",
      "train:  2 891 0.6099856\n",
      "train:  2 892 0.68424284\n",
      "train:  2 893 0.70243824\n",
      "train:  2 894 0.5649873\n",
      "train:  2 895 1.7574039\n",
      "train:  2 896 0.8487085\n",
      "train:  2 897 0.49631137\n",
      "train:  2 898 0.81548667\n",
      "train:  2 899 0.6401579\n",
      "train:  2 900 0.6602915\n",
      "evalu:  2 900 [0.89728034]\n",
      "train:  2 901 0.54979384\n",
      "train:  2 902 1.0729486\n",
      "train:  2 903 0.729377\n",
      "train:  2 904 0.6392277\n",
      "train:  2 905 0.6752298\n",
      "train:  2 906 0.6692737\n",
      "train:  2 907 1.0815208\n",
      "train:  2 908 0.56010944\n",
      "train:  2 909 0.7656165\n",
      "train:  2 910 0.6707601\n",
      "train:  2 911 0.5178151\n",
      "train:  2 912 0.7934671\n",
      "train:  2 913 0.682552\n",
      "train:  2 914 0.579182\n",
      "train:  2 915 0.8440148\n",
      "train:  2 916 0.8226601\n",
      "train:  2 917 0.59567326\n",
      "train:  2 918 1.1205457\n",
      "train:  2 919 1.0403185\n",
      "train:  2 920 0.64399135\n",
      "evalu:  2 920 [0.8970319]\n",
      "train:  2 921 0.91468465\n",
      "train:  2 922 0.7310254\n",
      "train:  2 923 0.48241088\n",
      "train:  2 924 0.8447793\n",
      "train:  2 925 0.6560856\n",
      "train:  2 926 0.6802488\n",
      "train:  2 927 1.1355904\n",
      "train:  2 928 0.5163059\n",
      "train:  2 929 0.52656806\n",
      "train:  2 930 1.0133625\n",
      "train:  2 931 0.45394304\n",
      "train:  2 932 0.7164142\n",
      "train:  2 933 0.69208336\n",
      "train:  2 934 0.700733\n",
      "train:  2 935 0.5173328\n",
      "train:  2 936 0.47529158\n",
      "train:  2 937 0.48700684\n",
      "train:  2 938 0.6443387\n",
      "train:  2 939 0.5686706\n",
      "train:  2 940 0.6132795\n",
      "evalu:  2 940 [0.8968813]\n",
      "train:  2 941 0.7070747\n",
      "train:  2 942 0.5923269\n",
      "train:  2 943 0.6102328\n",
      "train:  2 944 0.52853155\n",
      "train:  2 945 0.57896096\n",
      "train:  2 946 0.5741119\n",
      "train:  2 947 0.5665589\n",
      "train:  2 948 1.1007847\n",
      "train:  2 949 0.788774\n",
      "train:  2 950 0.6313749\n",
      "train:  2 951 0.8537529\n",
      "train:  2 952 1.9650326\n",
      "train:  2 953 0.6460026\n",
      "train:  2 954 0.770709\n",
      "train:  2 955 0.6817047\n",
      "train:  2 956 0.5881042\n",
      "train:  2 957 0.5016971\n",
      "train:  2 958 0.7495718\n",
      "train:  2 959 0.55183756\n",
      "train:  2 960 1.0630659\n",
      "evalu:  2 960 [0.8965511]\n",
      "train:  2 961 0.63442487\n",
      "train:  2 962 2.6545916\n",
      "train:  2 963 1.010424\n",
      "train:  2 964 0.7740324\n",
      "train:  2 965 0.66842955\n",
      "train:  2 966 1.0805317\n",
      "train:  2 967 0.585623\n",
      "train:  2 968 0.906522\n",
      "train:  2 969 0.6319659\n",
      "train:  2 970 0.5497708\n",
      "train:  2 971 0.5452351\n",
      "train:  2 972 0.61952204\n",
      "train:  2 973 2.1581802\n",
      "train:  2 974 0.46888345\n",
      "train:  2 975 0.5443489\n",
      "train:  2 976 0.75124586\n",
      "train:  2 977 0.68553215\n",
      "train:  2 978 0.69822854\n",
      "train:  2 979 0.54077506\n",
      "train:  2 980 0.73038006\n",
      "evalu:  2 980 [0.89650995]\n",
      "train:  2 981 0.8550304\n",
      "train:  2 982 0.6581619\n",
      "train:  2 983 0.82190585\n",
      "train:  2 984 1.1502371\n",
      "train:  2 985 1.2210358\n",
      "train:  2 986 0.6950039\n",
      "train:  2 987 0.65313673\n",
      "train:  2 988 1.4081827\n",
      "train:  2 989 0.7011045\n",
      "train:  2 990 0.5121808\n",
      "train:  2 991 0.6564368\n",
      "train:  2 992 0.53897667\n",
      "train:  2 993 0.6984023\n",
      "train:  2 994 0.5641876\n",
      "train:  2 995 1.2612053\n",
      "train:  2 996 0.730209\n",
      "train:  2 997 0.56767046\n",
      "train:  2 998 1.0735575\n",
      "train:  2 999 0.57905996\n",
      "train:  2 1000 0.61743414\n",
      "evalu:  2 1000 [0.8964313]\n",
      "train:  2 1001 0.7239055\n",
      "train:  2 1002 0.6293745\n",
      "train:  2 1003 0.6391275\n",
      "train:  2 1004 0.63480186\n",
      "train:  2 1005 0.644807\n",
      "train:  2 1006 0.6359659\n",
      "train:  2 1007 0.72237116\n",
      "train:  2 1008 1.1317806\n",
      "train:  2 1009 0.7614515\n",
      "train:  2 1010 1.0615298\n",
      "train:  2 1011 0.73844033\n",
      "train:  2 1012 1.3277119\n",
      "train:  2 1013 0.6895169\n",
      "train:  2 1014 0.67404926\n",
      "train:  2 1015 0.9527728\n",
      "train:  2 1016 1.0047526\n",
      "train:  2 1017 1.1278182\n",
      "train:  2 1018 0.5363189\n",
      "train:  2 1019 0.5639162\n",
      "train:  2 1020 0.6722895\n",
      "evalu:  2 1020 [0.89575446]\n",
      "train:  2 1021 0.5870676\n",
      "train:  2 1022 0.6982558\n",
      "train:  2 1023 1.2473313\n",
      "train:  2 1024 0.6453717\n",
      "train:  2 1025 0.5799941\n",
      "train:  2 1026 0.64525944\n",
      "train:  2 1027 0.78784066\n",
      "train:  2 1028 1.0229309\n",
      "train:  2 1029 0.62812436\n",
      "train:  2 1030 0.55805594\n",
      "train:  2 1031 0.5687828\n",
      "train:  2 1032 0.7762008\n",
      "train:  2 1033 0.533439\n",
      "train:  2 1034 0.79032\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  2 1035 0.5747231\n",
      "train:  2 1036 0.6687074\n",
      "train:  2 1037 0.8074593\n",
      "train:  2 1038 0.9304202\n",
      "train:  2 1039 0.59700817\n",
      "train:  2 1040 0.57616454\n",
      "evalu:  2 1040 [0.89586085]\n",
      "train:  2 1041 0.4806774\n",
      "train:  2 1042 1.025377\n",
      "train:  2 1043 0.7092062\n",
      "train:  2 1044 0.63158315\n",
      "train:  2 1045 0.5928896\n",
      "train:  2 1046 0.8576702\n",
      "train:  2 1047 0.5802187\n",
      "train:  2 1048 2.022455\n",
      "train:  2 1049 0.49444488\n",
      "train:  2 1050 1.2674651\n",
      "train:  2 1051 1.4307767\n",
      "train:  2 1052 0.64584655\n",
      "train:  2 1053 0.58762294\n",
      "train:  2 1054 0.61005443\n",
      "train:  2 1055 0.9379442\n",
      "train:  2 1056 0.5866487\n",
      "train:  2 1057 0.5101515\n",
      "train:  2 1058 1.2852917\n",
      "train:  2 1059 0.5100893\n",
      "train:  2 1060 0.61164373\n",
      "evalu:  2 1060 [0.89581555]\n",
      "train:  2 1061 0.6389728\n",
      "train:  2 1062 0.59430474\n",
      "train:  2 1063 0.5282745\n",
      "train:  2 1064 1.117045\n",
      "train:  2 1065 0.6417199\n",
      "train:  2 1066 0.5319027\n",
      "train:  2 1067 0.48666745\n",
      "train:  2 1068 0.58522016\n",
      "train:  2 1069 0.586408\n",
      "train:  2 1070 0.6563411\n",
      "train:  2 1071 0.64978355\n",
      "train:  2 1072 0.5962788\n",
      "train:  2 1073 0.80880255\n",
      "train:  2 1074 0.6871944\n",
      "train:  2 1075 0.60988826\n",
      "train:  2 1076 0.88066804\n",
      "train:  2 1077 0.6738125\n",
      "train:  2 1078 0.9635729\n",
      "train:  2 1079 0.84514177\n",
      "train:  2 1080 0.6629058\n",
      "evalu:  2 1080 [0.895443]\n",
      "train:  2 1081 1.2965553\n",
      "train:  2 1082 0.71934134\n",
      "train:  2 1083 1.9939977\n",
      "train:  2 1084 0.705478\n",
      "train:  2 1085 1.307786\n",
      "train:  2 1086 0.7669834\n",
      "train:  2 1087 2.2428586\n",
      "train:  2 1088 0.76795363\n",
      "train:  2 1089 0.5763149\n",
      "train:  2 1090 1.126953\n",
      "train:  2 1091 0.6444908\n",
      "train:  2 1092 0.8112448\n",
      "train:  2 1093 0.5960739\n",
      "train:  2 1094 0.6951945\n",
      "train:  2 1095 0.56114674\n",
      "train:  2 1096 0.5913993\n",
      "train:  2 1097 0.6652521\n",
      "train:  2 1098 0.9671448\n",
      "train:  2 1099 0.70476717\n",
      "train:  2 1100 0.71721405\n",
      "evalu:  2 1100 [0.895441]\n",
      "train:  2 1101 0.8196813\n",
      "train:  2 1102 0.56540394\n",
      "train:  2 1103 0.5984601\n",
      "train:  2 1104 0.5646649\n",
      "train:  2 1105 0.52593267\n",
      "train:  2 1106 0.5816669\n",
      "train:  2 1107 0.6108877\n",
      "train:  2 1108 1.2552073\n",
      "train:  2 1109 1.0291544\n",
      "train:  2 1110 0.6563881\n",
      "train:  2 1111 5.5516014\n",
      "train:  2 1112 0.8840855\n",
      "train:  2 1113 1.0472176\n",
      "train:  2 1114 0.59352744\n",
      "train:  2 1115 0.5251824\n",
      "train:  2 1116 1.3412832\n",
      "train:  2 1117 0.60902554\n",
      "train:  2 1118 0.8541727\n",
      "train:  2 1119 0.73832\n",
      "train:  2 1120 0.5800577\n",
      "evalu:  2 1120 [0.89533967]\n",
      "train:  2 1121 1.072742\n",
      "train:  2 1122 0.9998162\n",
      "train:  2 1123 0.5998329\n",
      "train:  2 1124 0.4977077\n",
      "train:  2 1125 0.8749878\n",
      "train:  2 1126 0.54736346\n",
      "train:  2 1127 0.8015638\n",
      "train:  2 1128 0.9743971\n",
      "train:  2 1129 0.5224347\n",
      "train:  2 1130 0.8493293\n",
      "train:  2 1131 0.6475159\n",
      "train:  2 1132 2.1788538\n",
      "train:  2 1133 0.59856004\n",
      "train:  2 1134 0.7946066\n",
      "train:  2 1135 0.6088257\n",
      "train:  2 1136 0.577843\n",
      "train:  2 1137 0.62568057\n",
      "train:  2 1138 0.63009536\n",
      "train:  2 1139 1.6136603\n",
      "train:  2 1140 0.49777764\n",
      "evalu:  2 1140 [0.89514756]\n",
      "train:  2 1141 0.9812219\n",
      "train:  2 1142 0.70200723\n",
      "train:  2 1143 0.6232946\n",
      "train:  2 1144 2.9532273\n",
      "train:  2 1145 0.8320838\n",
      "train:  2 1146 1.8083333\n",
      "train:  2 1147 0.6722626\n",
      "train:  2 1148 0.7144956\n",
      "train:  2 1149 0.64085346\n",
      "train:  2 1150 0.647521\n",
      "train:  2 1151 0.75885516\n",
      "train:  2 1152 0.6093753\n",
      "train:  2 1153 0.8941442\n",
      "train:  2 1154 0.5851785\n",
      "train:  2 1155 0.63095766\n",
      "train:  2 1156 0.5251784\n",
      "train:  2 1157 0.81360424\n",
      "train:  2 1158 0.6428915\n",
      "train:  2 1159 0.7931545\n",
      "train:  2 1160 0.6718471\n",
      "evalu:  2 1160 [0.8951237]\n",
      "train:  2 1161 0.67612934\n",
      "train:  2 1162 0.555241\n",
      "train:  2 1163 0.70147574\n",
      "train:  2 1164 0.89325815\n",
      "train:  2 1165 0.69804037\n",
      "train:  2 1166 0.6173251\n",
      "train:  2 1167 0.8454883\n",
      "train:  2 1168 0.60467225\n",
      "train:  2 1169 0.63913316\n",
      "train:  2 1170 1.378803\n",
      "train:  2 1171 1.7911853\n",
      "train:  2 1172 0.60458195\n",
      "train:  2 1173 1.1732326\n",
      "train:  2 1174 0.87970525\n",
      "train:  2 1175 0.67098707\n",
      "train:  2 1176 1.385375\n",
      "train:  2 1177 1.0399338\n",
      "train:  2 1178 0.60490257\n",
      "train:  2 1179 0.6139674\n",
      "train:  2 1180 0.6972365\n",
      "evalu:  2 1180 [0.89435554]\n",
      "train:  2 1181 0.70628744\n",
      "train:  2 1182 0.7005241\n",
      "train:  2 1183 0.5927436\n",
      "train:  2 1184 0.6714438\n",
      "train:  2 1185 0.8211017\n",
      "train:  2 1186 0.6010448\n",
      "train:  2 1187 0.6010434\n",
      "train:  2 1188 0.7226903\n",
      "train:  2 1189 0.6432866\n",
      "train:  2 1190 1.0978916\n",
      "train:  2 1191 0.5951882\n",
      "train:  2 1192 0.6660691\n",
      "train:  2 1193 0.778111\n",
      "train:  2 1194 0.7426579\n",
      "train:  2 1195 0.63211364\n",
      "train:  2 1196 0.67279476\n",
      "train:  2 1197 1.040399\n",
      "train:  2 1198 0.51248354\n",
      "train:  2 1199 0.70670223\n",
      "train:  2 1200 0.46573964\n",
      "evalu:  2 1200 [0.89404154]\n",
      "train:  2 1201 0.63854146\n",
      "train:  2 1202 1.1004212\n",
      "train:  2 1203 0.6772023\n",
      "train:  2 1204 0.52953273\n",
      "train:  2 1205 0.6635121\n",
      "train:  2 1206 0.691183\n",
      "train:  2 1207 0.5860678\n",
      "train:  2 1208 0.7867946\n",
      "train:  2 1209 1.0850358\n",
      "train:  2 1210 0.90446156\n",
      "train:  2 1211 0.8065379\n",
      "train:  2 1212 0.42653775\n",
      "train:  2 1213 0.5833574\n",
      "train:  2 1214 0.66742164\n",
      "train:  2 1215 0.6546208\n",
      "train:  2 1216 1.1835341\n",
      "train:  2 1217 0.8300742\n",
      "train:  2 1218 0.54092246\n",
      "train:  2 1219 0.5254247\n",
      "train:  2 1220 0.63266945\n",
      "evalu:  2 1220 [0.89393014]\n",
      "train:  2 1221 1.0220503\n",
      "train:  2 1222 0.557999\n",
      "train:  2 1223 0.6594735\n",
      "train:  2 1224 0.74419016\n",
      "train:  2 1225 0.5765553\n",
      "train:  2 1226 0.8340962\n",
      "train:  2 1227 0.5327498\n",
      "train:  2 1228 0.5112042\n",
      "train:  2 1229 0.78746957\n",
      "train:  2 1230 0.6545801\n",
      "train:  2 1231 0.54956526\n",
      "train:  2 1232 1.080324\n",
      "train:  2 1233 0.85499185\n",
      "train:  3 0 0.5769796\n",
      "evalu:  3 0 [0.89378726]\n",
      "train:  3 1 0.53884625\n",
      "train:  3 2 0.6064674\n",
      "train:  3 3 0.72690725\n",
      "train:  3 4 0.6858676\n",
      "train:  3 5 0.8008958\n",
      "train:  3 6 0.6742104\n",
      "train:  3 7 0.75911057\n",
      "train:  3 8 0.73699415\n",
      "train:  3 9 1.2153924\n",
      "train:  3 10 0.5771483\n",
      "train:  3 11 0.5686059\n",
      "train:  3 12 1.841318\n",
      "train:  3 13 0.57575816\n",
      "train:  3 14 0.59767723\n",
      "train:  3 15 0.9086108\n",
      "train:  3 16 0.85566145\n",
      "train:  3 17 0.56057745\n",
      "train:  3 18 1.1656147\n",
      "train:  3 19 0.60449654\n",
      "train:  3 20 1.9852896\n",
      "evalu:  3 20 [0.8939971]\n",
      "train:  3 21 0.787995\n",
      "train:  3 22 0.69138736\n",
      "train:  3 23 0.60698235\n",
      "train:  3 24 0.5590921\n",
      "train:  3 25 0.74623483\n",
      "train:  3 26 0.71062523\n",
      "train:  3 27 3.0282912\n",
      "train:  3 28 0.4892417\n",
      "train:  3 29 0.48843652\n",
      "train:  3 30 0.5348734\n",
      "train:  3 31 1.4191787\n",
      "train:  3 32 0.6111462\n",
      "train:  3 33 1.1737391\n",
      "train:  3 34 0.5713523\n",
      "train:  3 35 0.6257308\n",
      "train:  3 36 0.76691824\n",
      "train:  3 37 1.1790637\n",
      "train:  3 38 0.7188493\n",
      "train:  3 39 0.87892824\n",
      "train:  3 40 0.84989697\n",
      "evalu:  3 40 [0.8935577]\n",
      "train:  3 41 1.0212132\n",
      "train:  3 42 1.0142338\n",
      "train:  3 43 0.6206971\n",
      "train:  3 44 0.7294824\n",
      "train:  3 45 1.3539951\n",
      "train:  3 46 1.5945387\n",
      "train:  3 47 0.6289137\n",
      "train:  3 48 0.62210166\n",
      "train:  3 49 0.5015572\n",
      "train:  3 50 1.3120954\n",
      "train:  3 51 0.5475275\n",
      "train:  3 52 0.6575882\n",
      "train:  3 53 0.60820943\n",
      "train:  3 54 0.8376022\n",
      "train:  3 55 0.8115833\n",
      "train:  3 56 0.45153686\n",
      "train:  3 57 0.6484493\n",
      "train:  3 58 0.9440241\n",
      "train:  3 59 0.55804425\n",
      "train:  3 60 0.6105794\n",
      "evalu:  3 60 [0.893281]\n",
      "train:  3 61 0.46968848\n",
      "train:  3 62 0.56297356\n",
      "train:  3 63 0.59080213\n",
      "train:  3 64 0.56846946\n",
      "train:  3 65 0.6347728\n",
      "train:  3 66 1.0432404\n",
      "train:  3 67 0.62405133\n",
      "train:  3 68 1.0616847\n",
      "train:  3 69 0.78059405\n",
      "train:  3 70 0.8031621\n",
      "train:  3 71 0.6786641\n",
      "train:  3 72 0.5988005\n",
      "train:  3 73 0.54256415\n",
      "train:  3 74 0.4569768\n",
      "train:  3 75 0.64909256\n",
      "train:  3 76 0.6916566\n",
      "train:  3 77 0.6431047\n",
      "train:  3 78 1.126395\n",
      "train:  3 79 1.2513521\n",
      "train:  3 80 0.6787913\n",
      "evalu:  3 80 [0.8931082]\n",
      "train:  3 81 0.5826788\n",
      "train:  3 82 0.60018414\n",
      "train:  3 83 0.58622813\n",
      "train:  3 84 1.4163449\n",
      "train:  3 85 0.6789305\n",
      "train:  3 86 0.54606706\n",
      "train:  3 87 0.7701136\n",
      "train:  3 88 0.5617622\n",
      "train:  3 89 0.66398317\n",
      "train:  3 90 0.6624144\n",
      "train:  3 91 0.6468038\n",
      "train:  3 92 0.6249952\n",
      "train:  3 93 0.5846678\n",
      "train:  3 94 0.5730688\n",
      "train:  3 95 0.681739\n",
      "train:  3 96 0.62483895\n",
      "train:  3 97 0.8501185\n",
      "train:  3 98 0.58875203\n",
      "train:  3 99 0.61146796\n",
      "train:  3 100 0.6888893\n",
      "evalu:  3 100 [0.89316374]\n",
      "train:  3 101 0.70538974\n",
      "train:  3 102 1.1664287\n",
      "train:  3 103 0.7604769\n",
      "train:  3 104 1.7879461\n",
      "train:  3 105 1.1886595\n",
      "train:  3 106 0.60859495\n",
      "train:  3 107 0.97558105\n",
      "train:  3 108 1.4252671\n",
      "train:  3 109 1.1141288\n",
      "train:  3 110 0.5016173\n",
      "train:  3 111 0.5768973\n",
      "train:  3 112 1.204761\n",
      "train:  3 113 0.74339765\n",
      "train:  3 114 0.5202417\n",
      "train:  3 115 0.6421798\n",
      "train:  3 116 0.67258173\n",
      "train:  3 117 0.609717\n",
      "train:  3 118 0.56705713\n",
      "train:  3 119 0.6426171\n",
      "train:  3 120 0.72340316\n",
      "evalu:  3 120 [0.8925761]\n",
      "train:  3 121 1.1281981\n",
      "train:  3 122 0.66326576\n",
      "train:  3 123 0.9631323\n",
      "train:  3 124 0.6539072\n",
      "train:  3 125 0.65564793\n",
      "train:  3 126 0.74164337\n",
      "train:  3 127 0.772127\n",
      "train:  3 128 0.7674473\n",
      "train:  3 129 1.159324\n",
      "train:  3 130 0.58498985\n",
      "train:  3 131 0.55957437\n",
      "train:  3 132 0.6846649\n",
      "train:  3 133 0.47963372\n",
      "train:  3 134 0.900684\n",
      "train:  3 135 0.74662787\n",
      "train:  3 136 1.2507815\n",
      "train:  3 137 1.8093027\n",
      "train:  3 138 0.6620024\n",
      "train:  3 139 1.4263778\n",
      "train:  3 140 0.62921613\n",
      "evalu:  3 140 [0.892847]\n",
      "train:  3 141 0.5822586\n",
      "train:  3 142 0.639627\n",
      "train:  3 143 0.6299348\n",
      "train:  3 144 0.7136499\n",
      "train:  3 145 0.63393295\n",
      "train:  3 146 0.6456706\n",
      "train:  3 147 2.9038334\n",
      "train:  3 148 0.6734203\n",
      "train:  3 149 0.48560083\n",
      "train:  3 150 0.48079622\n",
      "train:  3 151 0.6520167\n",
      "train:  3 152 1.9611\n",
      "train:  3 153 0.78335065\n",
      "train:  3 154 0.74534136\n",
      "train:  3 155 0.9581247\n",
      "train:  3 156 0.6305354\n",
      "train:  3 157 1.8472158\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  3 158 0.9058029\n",
      "train:  3 159 0.5623884\n",
      "train:  3 160 0.61757696\n",
      "evalu:  3 160 [0.89251167]\n",
      "train:  3 161 1.1080991\n",
      "train:  3 162 0.76064324\n",
      "train:  3 163 0.654488\n",
      "train:  3 164 0.6895654\n",
      "train:  3 165 0.7808381\n",
      "train:  3 166 0.5018384\n",
      "train:  3 167 0.7651692\n",
      "train:  3 168 2.582538\n",
      "train:  3 169 0.7044157\n",
      "train:  3 170 0.53748095\n",
      "train:  3 171 0.7765289\n",
      "train:  3 172 0.6071064\n",
      "train:  3 173 0.7336136\n",
      "train:  3 174 0.66125685\n",
      "train:  3 175 0.51711404\n",
      "train:  3 176 0.67301077\n",
      "train:  3 177 0.5906328\n",
      "train:  3 178 0.8073518\n",
      "train:  3 179 0.8431895\n",
      "train:  3 180 0.5054076\n",
      "evalu:  3 180 [0.89213645]\n",
      "train:  3 181 0.8145495\n",
      "train:  3 182 0.5804156\n",
      "train:  3 183 0.55386305\n",
      "train:  3 184 0.942722\n",
      "train:  3 185 0.5480937\n",
      "train:  3 186 1.0128753\n",
      "train:  3 187 1.5925553\n",
      "train:  3 188 0.624451\n",
      "train:  3 189 1.135463\n",
      "train:  3 190 0.636846\n",
      "train:  3 191 0.5879559\n",
      "train:  3 192 0.7761393\n",
      "train:  3 193 0.5047111\n",
      "train:  3 194 0.46796778\n",
      "train:  3 195 0.62425584\n",
      "train:  3 196 0.54892796\n",
      "train:  3 197 0.6305346\n",
      "train:  3 198 0.5881502\n",
      "train:  3 199 0.56896824\n",
      "train:  3 200 0.6208962\n",
      "evalu:  3 200 [0.89231247]\n",
      "train:  3 201 0.60571057\n",
      "train:  3 202 1.3249972\n",
      "train:  3 203 0.6967911\n",
      "train:  3 204 0.4568515\n",
      "train:  3 205 0.63740844\n",
      "train:  3 206 0.7998301\n",
      "train:  3 207 0.57404816\n",
      "train:  3 208 1.4642415\n",
      "train:  3 209 1.0499326\n",
      "train:  3 210 0.5994085\n",
      "train:  3 211 0.76398504\n",
      "train:  3 212 0.6111891\n",
      "train:  3 213 0.8416788\n",
      "train:  3 214 0.79415315\n",
      "train:  3 215 0.6426399\n",
      "train:  3 216 0.50060284\n",
      "train:  3 217 1.2908024\n",
      "train:  3 218 0.62150055\n",
      "train:  3 219 0.61050034\n",
      "train:  3 220 0.6222627\n",
      "evalu:  3 220 [0.8919716]\n",
      "train:  3 221 0.742011\n",
      "train:  3 222 0.6029923\n",
      "train:  3 223 0.55931914\n",
      "train:  3 224 0.59491855\n",
      "train:  3 225 1.4548357\n",
      "train:  3 226 0.6384873\n",
      "train:  3 227 1.659631\n",
      "train:  3 228 0.5737881\n",
      "train:  3 229 0.9074839\n",
      "train:  3 230 1.0661292\n",
      "train:  3 231 0.565779\n",
      "train:  3 232 0.5265233\n",
      "train:  3 233 0.6017068\n",
      "train:  3 234 0.652657\n",
      "train:  3 235 0.9277166\n",
      "train:  3 236 0.8228882\n",
      "train:  3 237 0.5787833\n",
      "train:  3 238 1.0954959\n",
      "train:  3 239 1.091731\n",
      "train:  3 240 0.5084114\n",
      "evalu:  3 240 [0.8917444]\n",
      "train:  3 241 0.936534\n",
      "train:  3 242 0.7857791\n",
      "train:  3 243 0.8448299\n",
      "train:  3 244 0.73460895\n",
      "train:  3 245 0.78284085\n",
      "train:  3 246 0.556484\n",
      "train:  3 247 0.5645367\n",
      "train:  3 248 0.623917\n",
      "train:  3 249 0.93156755\n",
      "train:  3 250 0.60974085\n",
      "train:  3 251 0.54184824\n",
      "train:  3 252 0.5850118\n",
      "train:  3 253 0.5662732\n",
      "train:  3 254 0.89251846\n",
      "train:  3 255 0.61460805\n",
      "train:  3 256 0.65856856\n",
      "train:  3 257 0.5682214\n",
      "train:  3 258 0.9911102\n",
      "train:  3 259 1.0945511\n",
      "train:  3 260 0.8223198\n",
      "evalu:  3 260 [0.89151454]\n",
      "train:  3 261 0.77730435\n",
      "train:  3 262 0.58096665\n",
      "train:  3 263 1.5347949\n",
      "train:  3 264 1.006664\n",
      "train:  3 265 1.0084189\n",
      "train:  3 266 0.70836437\n",
      "train:  3 267 1.0371919\n",
      "train:  3 268 0.6996507\n",
      "train:  3 269 0.909568\n",
      "train:  3 270 0.75515896\n",
      "train:  3 271 0.7322508\n",
      "train:  3 272 0.6234442\n",
      "train:  3 273 0.70299286\n",
      "train:  3 274 0.54346925\n",
      "train:  3 275 0.64070004\n",
      "train:  3 276 0.6969774\n",
      "train:  3 277 0.84700876\n",
      "train:  3 278 1.1533955\n",
      "train:  3 279 0.68627465\n",
      "train:  3 280 0.64284104\n",
      "evalu:  3 280 [0.89135104]\n",
      "train:  3 281 0.5953145\n",
      "train:  3 282 0.7515962\n",
      "train:  3 283 0.6988387\n",
      "train:  3 284 0.70520836\n",
      "train:  3 285 0.7370011\n",
      "train:  3 286 0.6406567\n",
      "train:  3 287 0.6920404\n",
      "train:  3 288 0.55689627\n",
      "train:  3 289 0.68177235\n",
      "train:  3 290 0.7721992\n",
      "train:  3 291 0.82500553\n",
      "train:  3 292 0.5457694\n",
      "train:  3 293 1.5349789\n",
      "train:  3 294 0.6350248\n",
      "train:  3 295 0.94386965\n",
      "train:  3 296 3.4472253\n",
      "train:  3 297 0.7440472\n",
      "train:  3 298 0.6103679\n",
      "train:  3 299 0.7057214\n",
      "train:  3 300 1.2307401\n",
      "evalu:  3 300 [0.89112335]\n",
      "train:  3 301 0.5625189\n",
      "train:  3 302 0.6283911\n",
      "train:  3 303 0.7571698\n",
      "train:  3 304 0.74457115\n",
      "train:  3 305 0.46776584\n",
      "train:  3 306 0.7508452\n",
      "train:  3 307 0.6267065\n",
      "train:  3 308 0.63485676\n",
      "train:  3 309 1.3362288\n",
      "train:  3 310 0.87305975\n",
      "train:  3 311 0.645026\n",
      "train:  3 312 0.6056358\n",
      "train:  3 313 1.0361713\n",
      "train:  3 314 1.0574986\n",
      "train:  3 315 0.65985453\n",
      "train:  3 316 0.6413959\n",
      "train:  3 317 0.8250518\n",
      "train:  3 318 0.5476373\n",
      "train:  3 319 1.288739\n",
      "train:  3 320 1.4400443\n",
      "evalu:  3 320 [0.8908475]\n",
      "train:  3 321 0.6039796\n",
      "train:  3 322 0.61714035\n",
      "train:  3 323 0.56687486\n",
      "train:  3 324 0.63134974\n",
      "train:  3 325 0.82066685\n",
      "train:  3 326 0.9707641\n",
      "train:  3 327 0.56558347\n",
      "train:  3 328 0.54151976\n",
      "train:  3 329 0.76846737\n",
      "train:  3 330 1.3130484\n",
      "train:  3 331 0.5047628\n",
      "train:  3 332 0.6759189\n",
      "train:  3 333 2.7841308\n",
      "train:  3 334 0.6535922\n",
      "train:  3 335 0.50002056\n",
      "train:  3 336 0.5000639\n",
      "train:  3 337 0.7040133\n",
      "train:  3 338 2.3824346\n",
      "train:  3 339 0.73510337\n",
      "train:  3 340 0.709361\n",
      "evalu:  3 340 [0.8907382]\n",
      "train:  3 341 0.70452523\n",
      "train:  3 342 0.8147792\n",
      "train:  3 343 0.6219617\n",
      "train:  3 344 0.589564\n",
      "train:  3 345 0.63459224\n",
      "train:  3 346 0.5527252\n",
      "train:  3 347 0.49528846\n",
      "train:  3 348 0.6750808\n",
      "train:  3 349 0.7590194\n",
      "train:  3 350 1.6823157\n",
      "train:  3 351 0.7096017\n",
      "train:  3 352 0.51826733\n",
      "train:  3 353 0.7182065\n",
      "train:  3 354 0.67095274\n",
      "train:  3 355 0.96577406\n",
      "train:  3 356 0.5502199\n",
      "train:  3 357 0.5767874\n",
      "train:  3 358 0.86601806\n",
      "train:  3 359 0.59352666\n",
      "train:  3 360 0.56994134\n",
      "evalu:  3 360 [0.8906756]\n",
      "train:  3 361 0.8756139\n",
      "train:  3 362 0.9812379\n",
      "train:  3 363 0.69822395\n",
      "train:  3 364 0.6958386\n",
      "train:  3 365 0.5963484\n",
      "train:  3 366 0.64208865\n",
      "train:  3 367 1.8739336\n",
      "train:  3 368 0.8468283\n",
      "train:  3 369 1.124093\n",
      "train:  3 370 0.69862264\n",
      "train:  3 371 0.83320653\n",
      "train:  3 372 0.545192\n",
      "train:  3 373 2.0863426\n",
      "train:  3 374 0.7467277\n",
      "train:  3 375 0.77367693\n",
      "train:  3 376 0.5166596\n",
      "train:  3 377 1.0801319\n",
      "train:  3 378 0.65062183\n",
      "train:  3 379 0.94240534\n",
      "train:  3 380 0.582511\n",
      "evalu:  3 380 [0.89029723]\n",
      "train:  3 381 2.4254978\n",
      "train:  3 382 0.9169598\n",
      "train:  3 383 0.8723929\n",
      "train:  3 384 1.0155644\n",
      "train:  3 385 0.7021513\n",
      "train:  3 386 0.918404\n",
      "train:  3 387 0.76004004\n",
      "train:  3 388 0.7360871\n",
      "train:  3 389 0.9594219\n",
      "train:  3 390 0.7325618\n",
      "train:  3 391 1.220701\n",
      "train:  3 392 0.73202944\n",
      "train:  3 393 1.2666174\n",
      "train:  3 394 0.5417371\n",
      "train:  3 395 0.6437732\n",
      "train:  3 396 1.0521467\n",
      "train:  3 397 1.2155094\n",
      "train:  3 398 0.9128906\n",
      "train:  3 399 0.5710349\n",
      "train:  3 400 0.629309\n",
      "evalu:  3 400 [0.8900912]\n",
      "train:  3 401 0.72368914\n",
      "train:  3 402 0.4481561\n",
      "train:  3 403 0.59989595\n",
      "train:  3 404 0.50506955\n",
      "train:  3 405 0.4848625\n",
      "train:  3 406 0.5744796\n",
      "train:  3 407 0.5832531\n",
      "train:  3 408 0.68289065\n",
      "train:  3 409 0.6585922\n",
      "train:  3 410 0.70796615\n",
      "train:  3 411 0.5231991\n",
      "train:  3 412 0.5338358\n",
      "train:  3 413 0.5672122\n",
      "train:  3 414 0.6606438\n",
      "train:  3 415 0.5657311\n",
      "train:  3 416 0.5979664\n",
      "train:  3 417 0.54417926\n",
      "train:  3 418 0.87259424\n",
      "train:  3 419 1.532939\n",
      "train:  3 420 0.656021\n",
      "evalu:  3 420 [0.88992864]\n",
      "train:  3 421 0.73120636\n",
      "train:  3 422 0.53551155\n",
      "train:  3 423 0.62558043\n",
      "train:  3 424 1.1866279\n",
      "train:  3 425 1.8221842\n",
      "train:  3 426 1.3332818\n",
      "train:  3 427 0.6500543\n",
      "train:  3 428 0.6980321\n",
      "train:  3 429 0.5708692\n",
      "train:  3 430 1.0263575\n",
      "train:  3 431 0.6068043\n",
      "train:  3 432 2.0603917\n",
      "train:  3 433 0.6158884\n",
      "train:  3 434 0.6795783\n",
      "train:  3 435 0.6267566\n",
      "train:  3 436 0.59676635\n",
      "train:  3 437 0.57622373\n",
      "train:  3 438 0.75717187\n",
      "train:  3 439 0.6419022\n",
      "train:  3 440 0.6124537\n",
      "evalu:  3 440 [0.88985324]\n",
      "train:  3 441 0.64263135\n",
      "train:  3 442 0.7641811\n",
      "train:  3 443 0.84930533\n",
      "train:  3 444 0.67696613\n",
      "train:  3 445 0.70298684\n",
      "train:  3 446 0.6833877\n",
      "train:  3 447 0.69554555\n",
      "train:  3 448 0.56777376\n",
      "train:  3 449 0.6643811\n",
      "train:  3 450 0.6693612\n",
      "train:  3 451 0.67492825\n",
      "train:  3 452 1.1839726\n",
      "train:  3 453 0.6847644\n",
      "train:  3 454 0.6804012\n",
      "train:  3 455 0.43232107\n",
      "train:  3 456 0.5210659\n",
      "train:  3 457 0.61266106\n",
      "train:  3 458 0.77164924\n",
      "train:  3 459 0.49637696\n",
      "train:  3 460 0.5092002\n",
      "evalu:  3 460 [0.8899257]\n",
      "train:  3 461 0.91550946\n",
      "train:  3 462 0.6298374\n",
      "train:  3 463 0.58112526\n",
      "train:  3 464 0.70093733\n",
      "train:  3 465 0.7626871\n",
      "train:  3 466 0.6792885\n",
      "train:  3 467 1.1125898\n",
      "train:  3 468 0.8607572\n",
      "train:  3 469 0.6505338\n",
      "train:  3 470 0.82896906\n",
      "train:  3 471 0.65606195\n",
      "train:  3 472 1.2141029\n",
      "train:  3 473 0.673461\n",
      "train:  3 474 1.7957822\n",
      "train:  3 475 0.6820147\n",
      "train:  3 476 0.5498851\n",
      "train:  3 477 0.5290177\n",
      "train:  3 478 1.1175265\n",
      "train:  3 479 0.5197024\n",
      "train:  3 480 0.63112134\n",
      "evalu:  3 480 [0.8896188]\n",
      "train:  3 481 0.71216863\n",
      "train:  3 482 0.9643691\n",
      "train:  3 483 0.52305967\n",
      "train:  3 484 1.2979356\n",
      "train:  3 485 0.7671585\n",
      "train:  3 486 0.60793734\n",
      "train:  3 487 0.5577333\n",
      "train:  3 488 0.82783616\n",
      "train:  3 489 0.48720542\n",
      "train:  3 490 1.0848743\n",
      "train:  3 491 1.2080587\n",
      "train:  3 492 1.5511017\n",
      "train:  3 493 0.5677374\n",
      "train:  3 494 0.51812994\n",
      "train:  3 495 0.6882054\n",
      "train:  3 496 0.6298666\n",
      "train:  3 497 0.47828704\n",
      "train:  3 498 0.6350964\n",
      "train:  3 499 0.63368356\n",
      "train:  3 500 0.55138606\n",
      "evalu:  3 500 [0.8891575]\n",
      "train:  3 501 0.8000288\n",
      "train:  3 502 1.2680099\n",
      "train:  3 503 0.5558343\n",
      "train:  3 504 1.1033417\n",
      "train:  3 505 0.8101452\n",
      "train:  3 506 0.766189\n",
      "train:  3 507 1.5802541\n",
      "train:  3 508 0.8692123\n",
      "train:  3 509 0.8497976\n",
      "train:  3 510 0.77569175\n",
      "train:  3 511 0.7556315\n",
      "train:  3 512 0.97403\n",
      "train:  3 513 0.5118648\n",
      "train:  3 514 0.6575065\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  3 515 0.8329925\n",
      "train:  3 516 0.58029026\n",
      "train:  3 517 0.84857404\n",
      "train:  3 518 0.98105013\n",
      "train:  3 519 0.4824671\n",
      "train:  3 520 0.6032677\n",
      "evalu:  3 520 [0.8893118]\n",
      "train:  3 521 0.5516407\n",
      "train:  3 522 1.0848157\n",
      "train:  3 523 1.3506078\n",
      "train:  3 524 1.7607247\n",
      "train:  3 525 0.8496009\n",
      "train:  3 526 0.61147636\n",
      "train:  3 527 0.79597753\n",
      "train:  3 528 0.5876392\n",
      "train:  3 529 0.71334946\n",
      "train:  3 530 0.66194\n",
      "train:  3 531 0.6739556\n",
      "train:  3 532 0.6543192\n",
      "train:  3 533 0.6773732\n",
      "train:  3 534 0.66478467\n",
      "train:  3 535 1.0799217\n",
      "train:  3 536 1.9004033\n",
      "train:  3 537 0.6344784\n",
      "train:  3 538 0.4609073\n",
      "train:  3 539 0.577139\n",
      "train:  3 540 0.66127527\n",
      "evalu:  3 540 [0.8888527]\n",
      "train:  3 541 0.52601403\n",
      "train:  3 542 0.64035475\n",
      "train:  3 543 0.5903221\n",
      "train:  3 544 0.603236\n",
      "train:  3 545 0.41695946\n",
      "train:  3 546 0.5388743\n",
      "train:  3 547 0.5634549\n",
      "train:  3 548 0.77652556\n",
      "train:  3 549 0.45598975\n",
      "train:  3 550 0.60667056\n",
      "train:  3 551 0.7002842\n",
      "train:  3 552 0.7614867\n",
      "train:  3 553 0.49291545\n",
      "train:  3 554 0.6674637\n",
      "train:  3 555 0.670229\n",
      "train:  3 556 0.54771495\n",
      "train:  3 557 2.3216462\n",
      "train:  3 558 0.51707387\n",
      "train:  3 559 0.65250075\n",
      "train:  3 560 0.48569205\n",
      "evalu:  3 560 [0.8887114]\n",
      "train:  3 561 0.6167639\n",
      "train:  3 562 0.5800568\n",
      "train:  3 563 0.6744333\n",
      "train:  3 564 0.75959814\n",
      "train:  3 565 0.5978252\n",
      "train:  3 566 1.2038013\n",
      "train:  3 567 1.3355727\n",
      "train:  3 568 0.84875\n",
      "train:  3 569 1.4164428\n",
      "train:  3 570 0.812936\n",
      "train:  3 571 1.1932069\n",
      "train:  3 572 1.1644331\n",
      "train:  3 573 0.6603964\n",
      "train:  3 574 0.819632\n",
      "train:  3 575 0.46566007\n",
      "train:  3 576 0.96932894\n",
      "train:  3 577 0.6307103\n",
      "train:  3 578 0.94346076\n",
      "train:  3 579 0.7822546\n",
      "train:  3 580 0.66626734\n",
      "evalu:  3 580 [0.88855326]\n",
      "train:  3 581 0.51291984\n",
      "train:  3 582 1.2110113\n",
      "train:  3 583 0.56184083\n",
      "train:  3 584 0.5744058\n",
      "train:  3 585 0.74443847\n",
      "train:  3 586 0.8618625\n",
      "train:  3 587 0.49704418\n",
      "train:  3 588 0.5463146\n",
      "train:  3 589 0.57442033\n",
      "train:  3 590 0.56483865\n",
      "train:  3 591 0.6255644\n",
      "train:  3 592 1.3459965\n",
      "train:  3 593 1.6179911\n",
      "train:  3 594 0.7307468\n",
      "train:  3 595 0.7426152\n",
      "train:  3 596 0.79569924\n",
      "train:  3 597 0.94614536\n",
      "train:  3 598 0.5129805\n",
      "train:  3 599 0.5829393\n",
      "train:  3 600 0.6654883\n",
      "evalu:  3 600 [0.88848954]\n",
      "train:  3 601 0.75359625\n",
      "train:  3 602 1.2742767\n",
      "train:  3 603 0.83843064\n",
      "train:  3 604 1.3997519\n",
      "train:  3 605 0.5552181\n",
      "train:  3 606 0.6382856\n",
      "train:  3 607 0.5752742\n",
      "train:  3 608 0.76915365\n",
      "train:  3 609 0.75400805\n",
      "train:  3 610 1.0942489\n",
      "train:  3 611 0.59802383\n",
      "train:  3 612 1.2567068\n",
      "train:  3 613 0.78627443\n",
      "train:  3 614 0.67707884\n",
      "train:  3 615 0.48072517\n",
      "train:  3 616 0.51374304\n",
      "train:  3 617 0.5894267\n",
      "train:  3 618 3.3838747\n",
      "train:  3 619 0.5285617\n",
      "train:  3 620 0.737618\n",
      "evalu:  3 620 [0.8883843]\n",
      "train:  3 621 1.0997952\n",
      "train:  3 622 0.5836594\n",
      "train:  3 623 0.5255743\n",
      "train:  3 624 0.8021884\n",
      "train:  3 625 0.9121652\n",
      "train:  3 626 0.700625\n",
      "train:  3 627 0.62197125\n",
      "train:  3 628 1.1393801\n",
      "train:  3 629 0.75707567\n",
      "train:  3 630 0.7243813\n",
      "train:  3 631 0.5453993\n",
      "train:  3 632 0.5769035\n",
      "train:  3 633 0.5718586\n",
      "train:  3 634 0.6490682\n",
      "train:  3 635 0.6403343\n",
      "train:  3 636 0.62911546\n",
      "train:  3 637 0.80789304\n",
      "train:  3 638 0.75112516\n",
      "train:  3 639 0.9908231\n",
      "train:  3 640 0.61586314\n",
      "evalu:  3 640 [0.8881394]\n",
      "train:  3 641 0.7381873\n",
      "train:  3 642 0.8157037\n",
      "train:  3 643 0.7450675\n",
      "train:  3 644 0.51135623\n",
      "train:  3 645 0.70794606\n",
      "train:  3 646 0.52068484\n",
      "train:  3 647 0.60829794\n",
      "train:  3 648 0.48124537\n",
      "train:  3 649 1.8485143\n",
      "train:  3 650 0.68540347\n",
      "train:  3 651 0.901782\n",
      "train:  3 652 0.6955612\n",
      "train:  3 653 0.8227372\n",
      "train:  3 654 0.58840567\n",
      "train:  3 655 0.56400186\n",
      "train:  3 656 1.3950502\n",
      "train:  3 657 0.57902634\n",
      "train:  3 658 0.91608965\n",
      "train:  3 659 0.7976097\n",
      "train:  3 660 0.51457596\n",
      "evalu:  3 660 [0.8877935]\n",
      "train:  3 661 1.0457165\n",
      "train:  3 662 1.0489419\n",
      "train:  3 663 1.017032\n",
      "train:  3 664 0.79861045\n",
      "train:  3 665 0.73077285\n",
      "train:  3 666 1.7267915\n",
      "train:  3 667 0.5469742\n",
      "train:  3 668 0.6827103\n",
      "train:  3 669 0.5482089\n",
      "train:  3 670 0.474677\n",
      "train:  3 671 0.50380474\n",
      "train:  3 672 0.7433898\n",
      "train:  3 673 0.64710253\n",
      "train:  3 674 0.5112205\n",
      "train:  3 675 0.95152265\n",
      "train:  3 676 1.7394246\n",
      "train:  3 677 0.64801633\n",
      "train:  3 678 0.9762936\n",
      "train:  3 679 0.8384928\n",
      "train:  3 680 0.5586268\n",
      "evalu:  3 680 [0.8878274]\n",
      "train:  3 681 0.7602936\n",
      "train:  3 682 0.5106443\n",
      "train:  3 683 0.6966278\n",
      "train:  3 684 0.5324964\n",
      "train:  3 685 0.6953862\n",
      "train:  3 686 0.56433284\n",
      "train:  3 687 1.1527847\n",
      "train:  3 688 1.5729703\n",
      "train:  3 689 0.583625\n",
      "train:  3 690 1.4463942\n",
      "train:  3 691 0.73138636\n",
      "train:  3 692 0.70345867\n",
      "train:  3 693 0.7347504\n",
      "train:  3 694 0.6745338\n",
      "train:  3 695 0.8224118\n",
      "train:  3 696 0.57963777\n",
      "train:  3 697 0.58948946\n",
      "train:  3 698 0.57993215\n",
      "train:  3 699 0.73898196\n",
      "train:  3 700 0.6591435\n",
      "evalu:  3 700 [0.8876231]\n",
      "train:  3 701 0.48868033\n",
      "train:  3 702 0.561135\n",
      "train:  3 703 0.5280978\n",
      "train:  3 704 0.5345584\n",
      "train:  3 705 0.5107803\n",
      "train:  3 706 0.78516406\n",
      "train:  3 707 0.55285853\n",
      "train:  3 708 0.6115468\n",
      "train:  3 709 0.5708347\n",
      "train:  3 710 0.62276787\n",
      "train:  3 711 0.7993447\n",
      "train:  3 712 0.75930226\n",
      "train:  3 713 0.986158\n",
      "train:  3 714 0.52915883\n",
      "train:  3 715 1.1941453\n",
      "train:  3 716 0.7592186\n",
      "train:  3 717 0.5167541\n",
      "train:  3 718 0.5619339\n",
      "train:  3 719 0.75918573\n",
      "train:  3 720 0.8599738\n",
      "evalu:  3 720 [0.8875149]\n",
      "train:  3 721 0.59232754\n",
      "train:  3 722 0.9049577\n",
      "train:  3 723 0.89542806\n",
      "train:  3 724 1.0207957\n",
      "train:  3 725 0.54381025\n",
      "train:  3 726 0.860056\n",
      "train:  3 727 0.73204315\n",
      "train:  3 728 0.627459\n",
      "train:  3 729 0.64008063\n",
      "train:  3 730 1.1702234\n",
      "train:  3 731 0.6788486\n",
      "train:  3 732 0.5155111\n",
      "train:  3 733 0.7475474\n",
      "train:  3 734 0.82197845\n",
      "train:  3 735 0.74370384\n",
      "train:  3 736 0.7204076\n",
      "train:  3 737 2.0818257\n",
      "train:  3 738 0.72435653\n",
      "train:  3 739 0.68799037\n",
      "train:  3 740 0.5293753\n",
      "evalu:  3 740 [0.88702905]\n",
      "train:  3 741 0.8023787\n",
      "train:  3 742 0.7633239\n",
      "train:  3 743 0.63292706\n",
      "train:  3 744 0.63313675\n",
      "train:  3 745 0.8214271\n",
      "train:  3 746 1.2306403\n",
      "train:  3 747 0.7148767\n",
      "train:  3 748 0.8841278\n",
      "train:  3 749 1.125764\n",
      "train:  3 750 0.479728\n",
      "train:  3 751 1.1619408\n",
      "train:  3 752 0.7536726\n",
      "train:  3 753 0.6409661\n",
      "train:  3 754 0.661176\n",
      "train:  3 755 0.92217463\n",
      "train:  3 756 0.6242723\n",
      "train:  3 757 0.71412355\n",
      "train:  3 758 0.92214715\n",
      "train:  3 759 0.6937356\n",
      "train:  3 760 0.5688096\n",
      "evalu:  3 760 [0.88693976]\n",
      "train:  3 761 0.6301479\n",
      "train:  3 762 1.9514241\n",
      "train:  3 763 0.61927044\n",
      "train:  3 764 0.6612868\n",
      "train:  3 765 0.5920418\n",
      "train:  3 766 0.5336984\n",
      "train:  3 767 0.4552372\n",
      "train:  3 768 0.7386457\n",
      "train:  3 769 0.78321433\n",
      "train:  3 770 0.8958546\n",
      "train:  3 771 0.77261835\n",
      "train:  3 772 0.547568\n",
      "train:  3 773 0.734223\n",
      "train:  3 774 0.70405877\n",
      "train:  3 775 0.55566347\n",
      "train:  3 776 0.7984185\n",
      "train:  3 777 0.9131768\n",
      "train:  3 778 0.6707695\n",
      "train:  3 779 0.7674156\n",
      "train:  3 780 0.559107\n",
      "evalu:  3 780 [0.88662374]\n",
      "train:  3 781 0.57736903\n",
      "train:  3 782 0.48156747\n",
      "train:  3 783 0.5464332\n",
      "train:  3 784 0.6691329\n",
      "train:  3 785 0.6608348\n",
      "train:  3 786 0.61295784\n",
      "train:  3 787 0.7150983\n",
      "train:  3 788 0.5080184\n",
      "train:  3 789 0.79874575\n",
      "train:  3 790 0.86448634\n",
      "train:  3 791 0.6169307\n",
      "train:  3 792 0.7845202\n",
      "train:  3 793 1.0556076\n",
      "train:  3 794 0.86293405\n",
      "train:  3 795 1.084522\n",
      "train:  3 796 0.5561528\n",
      "train:  3 797 0.8203675\n",
      "train:  3 798 0.7260098\n",
      "train:  3 799 0.45921907\n",
      "train:  3 800 0.85632163\n",
      "evalu:  3 800 [0.88687706]\n",
      "train:  3 801 0.56192976\n",
      "train:  3 802 0.5562093\n",
      "train:  3 803 0.5208052\n",
      "train:  3 804 0.5406793\n",
      "train:  3 805 0.60035247\n",
      "train:  3 806 0.51988316\n",
      "train:  3 807 1.227619\n",
      "train:  3 808 0.5782193\n",
      "train:  3 809 0.5952168\n",
      "train:  3 810 0.8368238\n",
      "train:  3 811 0.5412991\n",
      "train:  3 812 0.72677857\n",
      "train:  3 813 0.92642593\n",
      "train:  3 814 3.7452617\n",
      "train:  3 815 1.0191073\n",
      "train:  3 816 0.5789342\n",
      "train:  3 817 0.5464418\n",
      "train:  3 818 0.51508075\n",
      "train:  3 819 0.622286\n",
      "train:  3 820 0.6871827\n",
      "evalu:  3 820 [0.8868605]\n",
      "train:  3 821 0.6447589\n",
      "train:  3 822 0.63920355\n",
      "train:  3 823 0.69330996\n",
      "train:  3 824 0.62701917\n",
      "train:  3 825 0.5344835\n",
      "train:  3 826 1.2455206\n",
      "train:  3 827 2.1451569\n",
      "train:  3 828 0.9958387\n",
      "train:  3 829 0.6753196\n",
      "train:  3 830 0.95668733\n",
      "train:  3 831 0.95513004\n",
      "train:  3 832 0.5642246\n",
      "train:  3 833 0.665466\n",
      "train:  3 834 0.63080084\n",
      "train:  3 835 0.62497675\n",
      "train:  3 836 0.755072\n",
      "train:  3 837 0.5636488\n",
      "train:  3 838 0.5415222\n",
      "train:  3 839 0.5585832\n",
      "train:  3 840 1.328832\n",
      "evalu:  3 840 [0.8869761]\n",
      "train:  3 841 0.75955194\n",
      "train:  3 842 0.564574\n",
      "train:  3 843 1.4055898\n",
      "train:  3 844 0.5131815\n",
      "train:  3 845 0.58243144\n",
      "train:  3 846 0.6760394\n",
      "train:  3 847 0.5805961\n",
      "train:  3 848 1.4803789\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  3 849 0.86865884\n",
      "train:  3 850 0.5538558\n",
      "train:  3 851 0.5909898\n",
      "train:  3 852 0.9048284\n",
      "train:  3 853 1.3734599\n",
      "train:  3 854 0.5453544\n",
      "train:  3 855 0.65350705\n",
      "train:  3 856 0.5478584\n",
      "train:  3 857 0.7580716\n",
      "train:  3 858 1.5182058\n",
      "train:  3 859 0.9302705\n",
      "train:  3 860 0.5852078\n",
      "evalu:  3 860 [0.8865688]\n",
      "train:  3 861 0.7351407\n",
      "train:  3 862 0.71377504\n",
      "train:  3 863 0.44678938\n",
      "train:  3 864 0.6005643\n",
      "train:  3 865 0.6922539\n",
      "train:  3 866 0.93375194\n",
      "train:  3 867 0.46520883\n",
      "train:  3 868 0.54555947\n",
      "train:  3 869 0.7770092\n",
      "train:  3 870 1.1488984\n",
      "train:  3 871 0.7255047\n",
      "train:  3 872 0.7556367\n",
      "train:  3 873 0.6784524\n",
      "train:  3 874 0.5545763\n",
      "train:  3 875 0.7748529\n",
      "train:  3 876 0.77307534\n",
      "train:  3 877 1.051692\n",
      "train:  3 878 1.3446921\n",
      "train:  3 879 0.5096072\n",
      "train:  3 880 0.6812435\n",
      "evalu:  3 880 [0.8860244]\n",
      "train:  3 881 0.74310154\n",
      "train:  3 882 0.79707754\n",
      "train:  3 883 0.6002063\n",
      "train:  3 884 1.1373583\n",
      "train:  3 885 0.68498015\n",
      "train:  3 886 0.5727555\n",
      "train:  3 887 1.096686\n",
      "train:  3 888 0.9483491\n",
      "train:  3 889 0.50238305\n",
      "train:  3 890 0.76973647\n",
      "train:  3 891 0.6004434\n",
      "train:  3 892 0.66963917\n",
      "train:  3 893 0.69811606\n",
      "train:  3 894 0.55949026\n",
      "train:  3 895 1.7391412\n",
      "train:  3 896 0.836113\n",
      "train:  3 897 0.48955786\n",
      "train:  3 898 0.809875\n",
      "train:  3 899 0.6271648\n",
      "train:  3 900 0.6423456\n",
      "evalu:  3 900 [0.88587815]\n",
      "train:  3 901 0.54286474\n",
      "train:  3 902 1.0632522\n",
      "train:  3 903 0.72786826\n",
      "train:  3 904 0.63332236\n",
      "train:  3 905 0.6580579\n",
      "train:  3 906 0.66199464\n",
      "train:  3 907 1.0623854\n",
      "train:  3 908 0.5467498\n",
      "train:  3 909 0.7537338\n",
      "train:  3 910 0.663912\n",
      "train:  3 911 0.5087114\n",
      "train:  3 912 0.781103\n",
      "train:  3 913 0.66637766\n",
      "train:  3 914 0.56978524\n",
      "train:  3 915 0.83403254\n",
      "train:  3 916 0.8106015\n",
      "train:  3 917 0.58658975\n",
      "train:  3 918 1.1043345\n",
      "train:  3 919 1.0303115\n",
      "train:  3 920 0.6337128\n",
      "evalu:  3 920 [0.8857192]\n",
      "train:  3 921 0.90527606\n",
      "train:  3 922 0.7205644\n",
      "train:  3 923 0.47284895\n",
      "train:  3 924 0.8409723\n",
      "train:  3 925 0.6459177\n",
      "train:  3 926 0.6676731\n",
      "train:  3 927 1.1258683\n",
      "train:  3 928 0.5083745\n",
      "train:  3 929 0.52153695\n",
      "train:  3 930 0.99770766\n",
      "train:  3 931 0.44636622\n",
      "train:  3 932 0.7006293\n",
      "train:  3 933 0.6849443\n",
      "train:  3 934 0.6827841\n",
      "train:  3 935 0.5120654\n",
      "train:  3 936 0.47171608\n",
      "train:  3 937 0.4765162\n",
      "train:  3 938 0.6377236\n",
      "train:  3 939 0.55940306\n",
      "train:  3 940 0.6011257\n",
      "evalu:  3 940 [0.88569885]\n",
      "train:  3 941 0.6971663\n",
      "train:  3 942 0.58112144\n",
      "train:  3 943 0.60281837\n",
      "train:  3 944 0.52192855\n",
      "train:  3 945 0.56818205\n",
      "train:  3 946 0.5618056\n",
      "train:  3 947 0.55907977\n",
      "train:  3 948 1.0865154\n",
      "train:  3 949 0.7770441\n",
      "train:  3 950 0.6216085\n",
      "train:  3 951 0.8382156\n",
      "train:  3 952 1.9389176\n",
      "train:  3 953 0.63844097\n",
      "train:  3 954 0.75236195\n",
      "train:  3 955 0.67297816\n",
      "train:  3 956 0.5708716\n",
      "train:  3 957 0.49222723\n",
      "train:  3 958 0.7361133\n",
      "train:  3 959 0.54318124\n",
      "train:  3 960 1.0457234\n",
      "evalu:  3 960 [0.8854025]\n",
      "train:  3 961 0.6250966\n",
      "train:  3 962 2.6561842\n",
      "train:  3 963 0.99763095\n",
      "train:  3 964 0.7582233\n",
      "train:  3 965 0.6624735\n",
      "train:  3 966 1.0583259\n",
      "train:  3 967 0.5791401\n",
      "train:  3 968 0.8859694\n",
      "train:  3 969 0.625879\n",
      "train:  3 970 0.5424069\n",
      "train:  3 971 0.5360747\n",
      "train:  3 972 0.6067062\n",
      "train:  3 973 2.1239538\n",
      "train:  3 974 0.46440428\n",
      "train:  3 975 0.53702855\n",
      "train:  3 976 0.74278355\n",
      "train:  3 977 0.6740309\n",
      "train:  3 978 0.6849489\n",
      "train:  3 979 0.5331605\n",
      "train:  3 980 0.72238576\n",
      "evalu:  3 980 [0.88538855]\n",
      "train:  3 981 0.8507918\n",
      "train:  3 982 0.6491111\n",
      "train:  3 983 0.805082\n",
      "train:  3 984 1.1484191\n",
      "train:  3 985 1.19405\n",
      "train:  3 986 0.68664455\n",
      "train:  3 987 0.64350605\n",
      "train:  3 988 1.3946613\n",
      "train:  3 989 0.6919242\n",
      "train:  3 990 0.5030603\n",
      "train:  3 991 0.6461876\n",
      "train:  3 992 0.53442746\n",
      "train:  3 993 0.680626\n",
      "train:  3 994 0.5513351\n",
      "train:  3 995 1.2443062\n",
      "train:  3 996 0.7184279\n",
      "train:  3 997 0.5568662\n",
      "train:  3 998 1.0661145\n",
      "train:  3 999 0.5717621\n",
      "train:  3 1000 0.6119782\n",
      "evalu:  3 1000 [0.88531524]\n",
      "train:  3 1001 0.7110781\n",
      "train:  3 1002 0.6146355\n",
      "train:  3 1003 0.63394344\n",
      "train:  3 1004 0.6264775\n",
      "train:  3 1005 0.6351085\n",
      "train:  3 1006 0.6264563\n",
      "train:  3 1007 0.7003926\n",
      "train:  3 1008 1.1165161\n",
      "train:  3 1009 0.75259817\n",
      "train:  3 1010 1.0416532\n",
      "train:  3 1011 0.724735\n",
      "train:  3 1012 1.3111588\n",
      "train:  3 1013 0.6815285\n",
      "train:  3 1014 0.6659488\n",
      "train:  3 1015 0.9394848\n",
      "train:  3 1016 0.98237205\n",
      "train:  3 1017 1.1176668\n",
      "train:  3 1018 0.53057444\n",
      "train:  3 1019 0.55721605\n",
      "train:  3 1020 0.6629055\n",
      "evalu:  3 1020 [0.8849877]\n",
      "train:  3 1021 0.5820087\n",
      "train:  3 1022 0.6877334\n",
      "train:  3 1023 1.2345375\n",
      "train:  3 1024 0.63983554\n",
      "train:  3 1025 0.5748676\n",
      "train:  3 1026 0.63135743\n",
      "train:  3 1027 0.7731447\n",
      "train:  3 1028 1.0035075\n",
      "train:  3 1029 0.615462\n",
      "train:  3 1030 0.54899025\n",
      "train:  3 1031 0.55864435\n",
      "train:  3 1032 0.76970637\n",
      "train:  3 1033 0.52591246\n",
      "train:  3 1034 0.7742339\n",
      "train:  3 1035 0.56611705\n",
      "train:  3 1036 0.6597517\n",
      "train:  3 1037 0.7937443\n",
      "train:  3 1038 0.91906184\n",
      "train:  3 1039 0.5837521\n",
      "train:  3 1040 0.5677523\n",
      "evalu:  3 1040 [0.88496375]\n",
      "train:  3 1041 0.4734612\n",
      "train:  3 1042 0.9989716\n",
      "train:  3 1043 0.69564617\n",
      "train:  3 1044 0.6247745\n",
      "train:  3 1045 0.58372426\n",
      "train:  3 1046 0.84096354\n",
      "train:  3 1047 0.5742354\n",
      "train:  3 1048 2.0004525\n",
      "train:  3 1049 0.4864873\n",
      "train:  3 1050 1.2471107\n",
      "train:  3 1051 1.4207134\n",
      "train:  3 1052 0.6387681\n",
      "train:  3 1053 0.5803876\n",
      "train:  3 1054 0.60631984\n",
      "train:  3 1055 0.92044914\n",
      "train:  3 1056 0.5770137\n",
      "train:  3 1057 0.50336385\n",
      "train:  3 1058 1.2538605\n",
      "train:  3 1059 0.5060766\n",
      "train:  3 1060 0.603006\n",
      "evalu:  3 1060 [0.8850535]\n",
      "train:  3 1061 0.63227785\n",
      "train:  3 1062 0.5839647\n",
      "train:  3 1063 0.5236681\n",
      "train:  3 1064 1.1084336\n",
      "train:  3 1065 0.63196665\n",
      "train:  3 1066 0.5258146\n",
      "train:  3 1067 0.4799895\n",
      "train:  3 1068 0.5838452\n",
      "train:  3 1069 0.58038825\n",
      "train:  3 1070 0.64817667\n",
      "train:  3 1071 0.6412039\n",
      "train:  3 1072 0.5882393\n",
      "train:  3 1073 0.7994064\n",
      "train:  3 1074 0.6751621\n",
      "train:  3 1075 0.6002189\n",
      "train:  3 1076 0.87275136\n",
      "train:  3 1077 0.6591781\n",
      "train:  3 1078 0.9523779\n",
      "train:  3 1079 0.83381194\n",
      "train:  3 1080 0.6582542\n",
      "evalu:  3 1080 [0.8848266]\n",
      "train:  3 1081 1.2949235\n",
      "train:  3 1082 0.70955193\n",
      "train:  3 1083 1.9503642\n",
      "train:  3 1084 0.6922666\n",
      "train:  3 1085 1.301543\n",
      "train:  3 1086 0.75784665\n",
      "train:  3 1087 2.2191687\n",
      "train:  3 1088 0.76187265\n",
      "train:  3 1089 0.566453\n",
      "train:  3 1090 1.1112661\n",
      "train:  3 1091 0.6340043\n",
      "train:  3 1092 0.8007158\n",
      "train:  3 1093 0.59186584\n",
      "train:  3 1094 0.68861574\n",
      "train:  3 1095 0.55076414\n",
      "train:  3 1096 0.5851167\n",
      "train:  3 1097 0.6533841\n",
      "train:  3 1098 0.949041\n",
      "train:  3 1099 0.6954275\n",
      "train:  3 1100 0.7037581\n",
      "evalu:  3 1100 [0.88470536]\n",
      "train:  3 1101 0.81322443\n",
      "train:  3 1102 0.55504763\n",
      "train:  3 1103 0.58916235\n",
      "train:  3 1104 0.555354\n",
      "train:  3 1105 0.52298665\n",
      "train:  3 1106 0.57477456\n",
      "train:  3 1107 0.59546036\n",
      "train:  3 1108 1.2450266\n",
      "train:  3 1109 1.01769\n",
      "train:  3 1110 0.6426656\n",
      "train:  3 1111 5.5175586\n",
      "train:  3 1112 0.8727941\n",
      "train:  3 1113 1.039927\n",
      "train:  3 1114 0.5795686\n",
      "train:  3 1115 0.5165729\n",
      "train:  3 1116 1.3109156\n",
      "train:  3 1117 0.59683007\n",
      "train:  3 1118 0.8445156\n",
      "train:  3 1119 0.7311971\n",
      "train:  3 1120 0.57269305\n",
      "evalu:  3 1120 [0.8843397]\n",
      "train:  3 1121 1.0537759\n",
      "train:  3 1122 0.98080415\n",
      "train:  3 1123 0.5892521\n",
      "train:  3 1124 0.49202493\n",
      "train:  3 1125 0.8619179\n",
      "train:  3 1126 0.5376871\n",
      "train:  3 1127 0.78328973\n",
      "train:  3 1128 0.9644383\n",
      "train:  3 1129 0.518033\n",
      "train:  3 1130 0.8425459\n",
      "train:  3 1131 0.63870656\n",
      "train:  3 1132 2.1727605\n",
      "train:  3 1133 0.59004265\n",
      "train:  3 1134 0.7865799\n",
      "train:  3 1135 0.6020247\n",
      "train:  3 1136 0.5709311\n",
      "train:  3 1137 0.61648554\n",
      "train:  3 1138 0.6235914\n",
      "train:  3 1139 1.6028208\n",
      "train:  3 1140 0.49117577\n",
      "evalu:  3 1140 [0.88436544]\n",
      "train:  3 1141 0.972519\n",
      "train:  3 1142 0.6924783\n",
      "train:  3 1143 0.6184691\n",
      "train:  3 1144 2.9355202\n",
      "train:  3 1145 0.82063735\n",
      "train:  3 1146 1.8109404\n",
      "train:  3 1147 0.66466194\n",
      "train:  3 1148 0.7078086\n",
      "train:  3 1149 0.63035554\n",
      "train:  3 1150 0.63429683\n",
      "train:  3 1151 0.7462938\n",
      "train:  3 1152 0.5971012\n",
      "train:  3 1153 0.88256794\n",
      "train:  3 1154 0.5781756\n",
      "train:  3 1155 0.62282956\n",
      "train:  3 1156 0.520871\n",
      "train:  3 1157 0.8036875\n",
      "train:  3 1158 0.6342392\n",
      "train:  3 1159 0.7834811\n",
      "train:  3 1160 0.6688937\n",
      "evalu:  3 1160 [0.8842834]\n",
      "train:  3 1161 0.672091\n",
      "train:  3 1162 0.54780775\n",
      "train:  3 1163 0.6918172\n",
      "train:  3 1164 0.8807612\n",
      "train:  3 1165 0.681396\n",
      "train:  3 1166 0.6092055\n",
      "train:  3 1167 0.8316164\n",
      "train:  3 1168 0.59737086\n",
      "train:  3 1169 0.62961966\n",
      "train:  3 1170 1.3586661\n",
      "train:  3 1171 1.7856891\n",
      "train:  3 1172 0.59519595\n",
      "train:  3 1173 1.1681902\n",
      "train:  3 1174 0.8741366\n",
      "train:  3 1175 0.66062194\n",
      "train:  3 1176 1.3728125\n",
      "train:  3 1177 1.0326467\n",
      "train:  3 1178 0.5921328\n",
      "train:  3 1179 0.6057432\n",
      "train:  3 1180 0.6814156\n",
      "evalu:  3 1180 [0.8839649]\n",
      "train:  3 1181 0.6983916\n",
      "train:  3 1182 0.6907728\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train:  3 1183 0.5893039\n",
      "train:  3 1184 0.66121715\n",
      "train:  3 1185 0.8125016\n",
      "train:  3 1186 0.5929029\n",
      "train:  3 1187 0.5909613\n",
      "train:  3 1188 0.71107095\n",
      "train:  3 1189 0.63737136\n",
      "train:  3 1190 1.0832665\n",
      "train:  3 1191 0.58441186\n",
      "train:  3 1192 0.6588402\n",
      "train:  3 1193 0.7734965\n",
      "train:  3 1194 0.7333313\n",
      "train:  3 1195 0.6245362\n",
      "train:  3 1196 0.6662344\n",
      "train:  3 1197 1.0304805\n",
      "train:  3 1198 0.505826\n",
      "train:  3 1199 0.7008331\n",
      "train:  3 1200 0.4601569\n",
      "evalu:  3 1200 [0.88370097]\n",
      "train:  3 1201 0.62796617\n",
      "train:  3 1202 1.0919532\n",
      "train:  3 1203 0.66912025\n",
      "train:  3 1204 0.5216104\n",
      "train:  3 1205 0.6582994\n",
      "train:  3 1206 0.691079\n",
      "train:  3 1207 0.5771231\n",
      "train:  3 1208 0.78046274\n",
      "train:  3 1209 1.0760444\n",
      "train:  3 1210 0.8916748\n",
      "train:  3 1211 0.79514176\n",
      "train:  3 1212 0.4215648\n",
      "train:  3 1213 0.5758271\n",
      "train:  3 1214 0.6562874\n",
      "train:  3 1215 0.64541936\n",
      "train:  3 1216 1.1757239\n",
      "train:  3 1217 0.81537044\n",
      "train:  3 1218 0.53170425\n",
      "train:  3 1219 0.5197373\n",
      "train:  3 1220 0.624975\n",
      "evalu:  3 1220 [0.883531]\n",
      "train:  3 1221 1.0137684\n",
      "train:  3 1222 0.5502723\n",
      "train:  3 1223 0.6531421\n",
      "train:  3 1224 0.7269906\n",
      "train:  3 1225 0.5655535\n",
      "train:  3 1226 0.82540196\n",
      "train:  3 1227 0.52902925\n",
      "train:  3 1228 0.49976298\n",
      "train:  3 1229 0.7713602\n",
      "train:  3 1230 0.6475392\n",
      "train:  3 1231 0.5410522\n",
      "train:  3 1232 1.0598212\n",
      "train:  3 1233 0.84531397\n",
      "Running Test data...\n",
      "WARNING:tensorflow:From /anaconda3/lib/python3.6/site-packages/tensorflow/python/saved_model/simple_save.py:85: calling SavedModelBuilder.add_meta_graph_and_variables (from tensorflow.python.saved_model.builder_impl) with legacy_init_op is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Pass your op to the equivalent parameter main_op instead.\n",
      "INFO:tensorflow:Assets added to graph.\n",
      "INFO:tensorflow:No assets to write.\n",
      "INFO:tensorflow:SavedModel written to: ./m/saved_model.pb\n"
     ]
    }
   ],
   "source": [
    "init = tf.global_variables_initializer()\n",
    "batchsize=100\n",
    "#xbatch=dftrain2.loc[:,cols].head(10).values\n",
    "evalbatch=dfeval[cols]\n",
    "numepochs=4\n",
    "\n",
    "treerr=[]\n",
    "everr=[]\n",
    "\n",
    "with tf.Session() as sess:\n",
    "    sess.run(init)\n",
    "    for epoch in range(numepochs):\n",
    "        for i in range(dftrain2.shape[0]//batchsize):\n",
    "            xbatch=dftrain2.iloc[i*batchsize:(i+1)*batchsize][cols]\n",
    "            _,ltr=sess.run([trainstep,loss],feed_dict={inputs:xbatch})\n",
    "            print(\"train: \",epoch,i,ltr)\n",
    "            if i%20==0:\n",
    "                l=sess.run([loss],feed_dict={inputs:evalbatch})\n",
    "                print(\"evalu: \",epoch,i,l)\n",
    "                treerr.append(ltr)\n",
    "                everr.append(l)\n",
    "     \n",
    "    print(\"Running Test data...\")\n",
    "    out=[]\n",
    "    for i in range(dftest.shape[0]):\n",
    "        testbatch=np.array([dftest.iloc[i][cols].values])\n",
    "        l=sess.run([loss,Yout],feed_dict={inputs:testbatch})\n",
    "        out.append([dftest.iloc[i]['Class'],l[0]])\n",
    "    tf.saved_model.simple_save(sess,\"./m\", inputs=\n",
    "                           {\n",
    "                               \"inputs\":inputs#,'labels':labels,\"learnrate\":learnrate,\"droprate\":droprate,\"mode\":mode\n",
    "                           },\n",
    "                           outputs={\"Yout\":Yout}#,\"batchloss\":batchloss}\n",
    "                          )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1a3383a780>]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAD8CAYAAABw1c+bAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzsfXmcHMV59lPdc+99SatdaXWAJC6BBDKHcQAnNgYTg00cA7GNE4dg+7O/L3acwzl8H3Hi2I4JcTCxsfEBvjHYYEAGjEAHoAshJKH7WO1q73Pu7q7vj+rqqZ7t2Z2Z7jk0W8/vt7+Z7Znprqmpfuut533etwilFBISEhIS8wdKpRsgISEhIVFeSMMvISEhMc8gDb+EhITEPIM0/BISEhLzDNLwS0hISMwzSMMvISEhMc8gDb+EhITEPIM0/BISEhLzDNLwS0hISMwz+CrdACe0t7fTZcuWVboZEhISEmcMtm/fPkwp7cjnvVVp+JctW4Zt27ZVuhkSEhISZwwIIcfzfa+keiQkJCTmGaThl5CQkJhnkIZfQkJCYp5BGn4JCQmJeQZp+CUkJCTmGaThl5CQkJhnkIZfQkJCYp6h5gz/nlMT2HVyvNLNkJCQkKha1Jzh/48nX8MXH91b6WZISEhIVC1qzvCndQMpXW4gLyEhIZELNWf4DQMwDGn4JSQkJHKh9gw/pTCoNPwSEhISuVBzhp9SQJcev4SEhERO1JzhNyiFdPglJCQkcqMmDb8uLb+EhIRETtSg4Yfk+CUkJCRmQc0ZfkqpVPVISEhIzII5DT8hZAkh5BlCyD5CyKuEkL92eM+7CSG7zb/NhJCLhNeOEUJeIYTsIoSUfFst5vGX+ioSEhISZy7y2XpRA/BxSukOQkgDgO2EkA2UUjE99iiAqymlY4SQ6wHcC+Ay4fU3UkqHvWt2bhiUSlWPhISExCyY0/BTSvsB9JvPpwgh+wB0A9grvGez8JGtABZ73M68oRsUVHL8EhISEjlREMdPCFkGYB2AF2Z5218C+K3wPwXwJCFkOyHkzkIbWCgohVT1SEhISMyCfKgeAAAhpB7ALwB8lFI6meM9bwQz/G8QDl9JKe0jhCwAsIEQsp9SutHhs3cCuBMAenp6CvgKdjCqp+iPS0hISNQ88vL4CSF+MKP/I0rpL3O850IA3wZwE6V0hB+nlPaZj4MAHgJwqdPnKaX3UkrXU0rXd3R0FPYtBLAELunxS0hISORCPqoeAuA7APZRSr+W4z09AH4J4L2U0gPC8TozIAxCSB2AawHs8aLhuSCpHgkJCYnZkQ/VcyWA9wJ4hRCyyzz2TwB6AIBSeg+ATwFoA/BNNk9Ao5SuB7AQwEPmMR+AByilj3v6DbJgSB2/hISExKzIR9XzPAAyx3vuAHCHw/EjAC6a+YnSQer4JSQkJGZHzWXuyrLMEhISErOj5gy/LMssISEhMTtqzvDLsswSEhISs6MmDb9U9UhISEjkRg0aflmWWUJCQmI21JzhpybVI5O4JCQkJJxRc4afx3VlgFdCQkLCGTVo+Kn5WOGGSEhISFQpas/wG9zwS8svISEh4YSaM/zc3kvDLyEhIeGMmjP83OBLjj8/HByYwlu+vhETsXSlmyIhIVEm1KDhtz9KzI4DA9N4bWAKp8bjlW6KhIREmVCDht/k+KXlzwuZYLjsLwmJ+YKaM/yS4y8M0vBLSMw/1Jzhtzh+acjyApV5DxIS8w41Z/i5wZd2Pz/IvAcJifmHmjL8VKjMKT3Y/GBIakxCYt4hnz13lxBCniGE7COEvEoI+WuH9xBCyF2EkEOEkN2EkIuF195HCDlo/r3P6y8gQrRd0vDnBxkMl5CYf8jH49cAfJxSei6AywF8mBByXtZ7rgew0vy7E8D/AAAhpBXApwFcBuBSAJ8mhLR41PYZEL1W6cDmB27wZUwkf2w5PCLlrxJnNOY0/JTSfkrpDvP5FIB9ALqz3nYTgO9Thq0AmgkhiwC8BcAGSukopXQMwAYA13n6DQSITqs0ZPnBonqMyrbjTML/fXAH7nv+aKWbISFRNAri+AkhywCsA/BC1kvdAE4K//eax3Iddzr3nYSQbYSQbUNDQ4U0y4Lo8UvOOj9IOWfhSGoGkppe6WZISBSNvA0/IaQewC8AfJRSOpn9ssNH6CzHZx6k9F5K6XpK6fqOjo58m5V1jsxzyVnnByrlrwWD7etc6VacOdh+fAxfffK1SjdDQkBehp8Q4gcz+j+ilP7S4S29AJYI/y8G0DfL8ZLA7vGX6iq1hQzVIzssX7B9nWV/5Ysn957GPc8ernQzJATko+ohAL4DYB+l9Gs53vYIgNtNdc/lACYopf0AngBwLSGkxQzqXmseKwlEwy9VPflB6vgLh25QOb4KAFshyf6qJvjyeM+VAN4L4BVCyC7z2D8B6AEASuk9AB4D8FYAhwDEAPyF+dooIeTzAF4yP/c5Sumod823QxxbkrPOD3LHssJBqaTGCoFhUBjmdqjMj5SoNOY0/JTS5+HM1YvvoQA+nOO1+wDcV1TrCgSVwd2CwftMUhf5wxASBSXmBvcpKAWk3a8O1FTmrt3jr1w7ziTI2kaFw6CS6ikEcoxVH2rM8EuOv1BIqqdwGJLqKQhyc6TqQ80afkn15AfDonoq3JAzBJIaKxxyjFUfasrwSx1/4ZBF7QqDXCEVDqvPpOWvGtSU4bdRPXKQ5QVZq6cwcIMvE7jyB5VUT9Whxgx/5rm0Y/lBl9RFQTBkfxUMXgdK9ln1oLYMv2D5pXeRHzLURWXbcaaAStqiYMjgbvWhpgy/jeOXN2Ze4F6Y7K/8II1Y4eCTpJwsqwc1ZfilqqdwyOqchUH2V+HgXSW7rHpQu4ZfUhd5QapUCoPcv6BwyFVS9aHGDH/muVxW5gdZpK0wyDLWhUM6F9WHmjL8ompAKgjyA7U8WNlf+YAbL9lf+UMmcFUfasrwi16YVKnkB6njLwwW1SP7K2/IVVL1oaYMv8i7ykGWH6QhKwwZI1bhhpxByCS9yU6rFtSW4ZdUT8GwOH55U+YFuWNZ4ciUZZZ9Vi2oKcMvjivpXeQHKoO7BUHKOQuHpHqqDzVl+OWeu4VDKi4Kg5QmFg45xqoPc+7ARQi5D8AfAxiklF7g8PrfAXi3cL5zAXSY2y4eAzAFQAegUUrXe9VwJ9h1/HKQ5QPpwRYGSwUl+ytvSFVP9SEfj/97AK7L9SKl9CuU0rWU0rUA/hHAs1n76r7RfL2kRh+Qe+4WA2n4C4Ml55TdlTekx199mNPwU0o3Ash3g/TbADzoqkUuIAaPJJ+YH7gSSspf84MMhhcOKRmuPnjG8RNCImArg18IhymAJwkh2wkhd87x+TsJIdsIIduGhoaKaoPcc7dwSI+/MMhNRQqHnCyrD14Gd98GYFMWzXMlpfRiANcD+DAh5KpcH6aU3kspXU8pXd/R0VFUAyTHXzikPLEwyGqmhUOWBak+eGn4b0UWzUMp7TMfBwE8BOBSD683A7I6Z+GQUrvCIIu0FQ7J8VcfPDH8hJAmAFcDeFg4VkcIaeDPAVwLYI8X18sFqeMvHFJxURiknLNwyFVS9SEfOeeDAK4B0E4I6QXwaQB+AKCU3mO+7R0AnqSURoWPLgTwECGEX+cBSunj3jV9JqTHXzikN1YYZEykcMiyINWHOQ0/pfS2PN7zPTDZp3jsCICLim1YMZDB3cJhSKqnIHCKRxqx/CFr9VQfajZzVw6y/JDZHUn2Vz6QVE/hkFRP9aGmDL+sx184pCErDFKhUjhkQLz6UFOG31aWWQ6yvCANWWGQ8tfCIenE6kNtGX4Z3C0Y0pAVBil/LRxyjFUfaszwi8/lIMsH0pAVBqlQKRyy9Hf1oaYMP5Uef8GQRccKQ6b8QIUbcgbBUvXIe7JqUFOGXzRebjn+WErDZ3/9KqJJzd2JqhxyGV4YZMGxwiFr9VQfaszwe6fq2X58DN/ddAw7T4y7bVZVQ6p6CoOkegqH3MOg+lCzht+tIZtKME+/1j07eVMWBrHEhZQM5wfpXFQfatfwu7wppxJpdp4aJ3NlCYLCIJMEC4dcJVUfasvwCzba7RjjHr+m1/ZglTr+wiCOK9ln+UGOsepDbRl+D72xSU711PholUXaCoPMFSkcVkBcjrGqQU0Zfrs35hHVU+M3t6yjUhjsyjHZZ/lAUj3Vh5oy/F56Y1PzzOOXN2V+EMeD7LP8IOWc1YcaM/zskRD3CTbT84zjr/UJzivYkgRrO+7vGTL7FFe2HRIZ1JjhZyPLryjuVT1Jruqp7dEqKycWBhvVIz3+vEClx191qCnDzweYTyWeUT1ajQ9WyfEXBhncLRxSMlx9mNPwE0LuI4QMEkIc98slhFxDCJkghOwy/z4lvHYdIeQ1QsghQsgnvGy4E7iNVhXi2rvIcPy17QrLkrmFwWb4a9wp8AqyVk/1IR+P/3sArpvjPc9RSteaf58DAEKICuC/AVwP4DwAtxFCznPT2LlgUT2q4lozzFU9te7xZ7YSrGw7zhRQSfUUDCs7XA6yqsGchp9SuhHAaBHnvhTAIUrpEUppCsCPAdxUxHnyhujxu70p54+O3zv+9fE9p7HjxJjr81Qz7FRPBRtyBiEjIKhwQyQseMXxX0EIeZkQ8ltCyPnmsW4AJ4X39JrHHEEIuZMQso0Qsm1oaKioRlAruOuO6klqOlIaG6XzxfB78T2/9Ng+3Pf8UdfnqWbY9nyo8bHhFaRkuPrgheHfAWAppfQiAP8F4FfmceLw3py/PKX0Xkrpekrp+o6OjqIawgeW6jK4y/l9YB5QPR7elFOJdO3LX4XxUOtOgVeQwd3qg2vDTymdpJROm88fA+AnhLSDefhLhLcuBtDn9nqzgd+HfkVxtawUDX+t39xe3ZSUUkwntXkwUUpVT6Hg3VTr99KZBNeGnxDSSQgh5vNLzXOOAHgJwEpCyHJCSADArQAecXu92WB5/ApxVTKXB3aBynn8pycSWPOZJ7D/9GRJr+PVTZnUDKR1Og9UUOJzacjygW45FxVuiISFfOScDwLYAmA1IaSXEPKXhJAPEkI+aL7lnQD2EEJeBnAXgFspgwbgIwCeALAPwE8ppa+W5msw8PvQp7pL4LJ7/JUxZH0TcUwlNJwYiZX0OmJ9eTeYL3kP9kKA7s61Ye8Arvr3Z6x4Uq1CUj3VB99cb6CU3jbH63cDuDvHa48BeKy4phUOzr/6VeLKu7AbfretKg6cKy/1zeKVjn86OT9UUF7u67y3bxInRmOIpTQEfAG3TatKUEo9pXqODkfR1RxC0Ke6Ptd8Rk1l7nqVwCVSPZXy+DXzuqX2oDM6fpeGf954/Jnn7nd5q/1cES8r5sZTOq77z434+fZel62SqDHDb5ZsUM58VQ/3+EvtQWfqqLg7T2bHsto1YoC3wd35sEryMtN5PJ5CUjMwHkvP/WaJWVFThp9SCoUAhBBXN1MsxW7IuoBasZuSX7fU8kivNmKZSs4Pj99eltndueZDXMTLonbT8ySpshyoKcNvUEAhBCohroKVadPYhgNqxW7KtF6eBDKvAm/T86S2ka1kg+td3sxVUg3nPniZ6Tw5DybKcqHGDD+FQggUxZ13oRkGVIXApygVuynLVdjKqwQui6+uYSMGlIbq0Wp4svSS6slQY7XbX+VCjRl+tgmL4pLq0XQKn0I8qflTLNKc6ikTx+/We60Gvvp7m47i4V2nSnoNL0s2zDuqx6tgeI07F+VAjRl+0+Mn7hK40jqFX1WY4a8Yx29SPSXWk3qVXFMNW1X++KWTeHhXSZPD7Tp+uUqaE15SPfNFOVYO1JbhN1hw162nrhkGfCqBTyEV5PjL4/Fzr9W191oFwV3NoGVbIQHebe9Zy8FKKvSR3Ae7elBbht8M7iou99xN6xQ+hXv8leET+eAudQKXVSu9BhQXukFLXjHTq5INukERTekA5g/H713eQ+32V7lQY4afWhy/m5tS0w34VcbxV2oZrullSuDyKHOX35TpChZdT+tGyY2CaLzc9Nl0FRQCPDESw+d/s7ekk6XuYTB8qgriSLWCmjL8lFIoZlDWleE3KKN61Mpx/Nzgl1pV5NVm69UQ3NUNWraEN8AdPTaVrHwhwN8fGMR3nj+Koelkya7hpQrKCobXcEykXKgpw5+hetwZ7LRuwK8oUEnlVD1auTh+j3T81aBQSeul5/jtVE/x56mG0t/pMmSHe5n3UOngLqUUn/31q9hZA7vMzVmk7UyCYWbuKoq7BC5NZx5/JVU9fHCXi+N3T/VUg8dvlC3hjV3PBdWTrHxZEEs5VsLre6nq4aukyvUXxXc3HUNjyI91PS0VaYNXqDmPn5jBXdeqHkWBT1HmDcdPKVxJYKshGUnTadlKXLDnLqgecc+HCsVFyqEc8zLvodLZ4Rb9WgMxhpoy/LxWj+oyuMt0/NXh8ZfTgy32Unz3LaCyN4VWbo6/ZgoBls6Q2raqPMM5/nSZnLFyoKYMP0/gIoS4ClYyHb8Cn0oq5sFaZZnL4MH6FLY9crFGM57WoRsUAVWpsI7fKEOJC2+onmrg+PUylP4uRa2eijljZZgoy4V8duC6jxAySAjZk+P1dxNCdpt/mwkhFwmvHSOEvEII2UUI2eZlw51gFWlT3A2OtFiyoYY5fu69qqbhL/ZaMVOP3hDygVL3S/piUQ6PX/coIakaPH6rLEgJnQtPqZ4Kc/z8uvPF4/8egOtmef0ogKsppRcC+DyAe7NefyOldC2ldH1xTcwf3ur4maqn0svwUq44+Ffzq4r5f3Hflbc15Ge7IlWiz3SD7fRU6hWavehY8eepis1+ylAB1qsVUlo3kEjzFUplV+G1wPHns/XiRkLIslle3yz8uxXAYvfNKg6Uyzk90vFXtlZP6Tl+a+Ma1R3Vw7nPkF9xdR43sG7KElNjIsfvKoFLVPVUjLMuvQfrWX8lKt9f5ZJYlwNec/x/CeC3wv8UwJOEkO2EkDs9vtYMGJRCVYgZ3C3+PLxkQyUTuKxAUgkHOf9uPovqKe48/EbIePzl98jKl/cgPHfJ8buNrbhFeZyLzHM3qrFqiIlYe2TUQAKZZzp+QsgbwQz/G4TDV1JK+wghCwBsIITsp5RuzPH5OwHcCQA9PT1FtYHJOcHknC4GR6Zkg1L5m7KkHD979Ckm1VPkd9Usj58Z/sp4/GeOCgoAokkNjWE/RqOpCnLWpadOPAuGV0GmM79uej4Ed/MBIeRCAN8GcBOldIQfp5T2mY+DAB4CcGmuc1BK76WUrqeUru/o6CiqHZmNWLygepSqqM5ZVqqnyD5LWxw/G06V6LPy5T1knrvLFaEIV3CiBMozxuy1jYo/D/f4Q/7KOWPl2ge7HHBt+AkhPQB+CeC9lNIDwvE6QkgDfw7gWgCOyiCvwHX8CiGuluGsZEOlOf7ySe18LlU93GMM+Wrf46eUyVYBd1RPWjcQNCfKShW2s/Z1LlPJBjdUT9xUjtUH/VWwQjrzDf+cVA8h5EEA1wBoJ4T0Avg0AD8AUErvAfApAG0AvkkIAQDNVPAsBPCQecwH4AFK6eMl+A4WDIPLOd1x/LxkA0HldPxcaldKPpH3kc8yZMWdJ51F9VTE4y+T4dfNwH9KdyfnTOtGRSdK3gag1GPMO1UPAIQDSsUznecFx08pvW2O1+8AcIfD8SMALpr5idKhMT2IOhAQL0o2qFyh4lXrCgMfXLm+x+ZDw6gP+XDh4uair0E98vj5DcE92ErcGOWQJgJsslQ9CMpqOkU4ULmJkreh1NfPSIZdbofKBQQ+tYJUT+14/LWTuRsfw6d778D74/dD9WLrRYXtwFW5uiCzG7IvPLoPdz11yNU1Mh6/O0M2U8dfAVWPRVuU9tqUUtd5DwBb0VVS/gqUZ5WUoRMVV4UTxVVlJYu0sUcZ3K0ehFuwue5a3JB8DIundrtW9fjMPXcrrSDIdf2EpiOR1l1dQ7wpxf8LBVc5VDJYyScfg5Y2c9ig1LX8FWBjjFM9leesS2fIqCAgcLMK56vKsL9yHn96jnvyTELtGH4Av2j+cwwqHXjroU+jjY4V7fWnTR7XV8mSDXPUBUnrBlKauxt2ZnC3uPNo1aDqEfqplBJYg7qvbQRk4kgVXVWWRdXDHv2q4mpCtiTDAbVymbtlohPLgZoy/DESxr83/jMi6THcH/h36LHxos6j8Y1YlMoVHZuL6klrFEnNncdv6fhdUz1VoOoR4gqlpi58Xqh6jGpYVZZPOeZ2VzzubYd8lZNzliPTuVyoKcNPKcWRwCpsWPMfOJv0gvz4NiAVK+gchkGZV6e6L/bmBnN5Y2ndQNIjj98tZ23dlFWg6in19SnNTJRulWNWHKnCJQjKwfH7FXdUj5gkmK5Qf1ny1wruK+0Vasrw8wSu0c434OPpD0E5uRX40TuBxGTe5+B8tV9VrMxdN4HiYjEXx5/yhOphj651/MIyHKhM8Eu8GUudkGRRPS4NWaU9/nJw1ryLVNVlqXSBTqxcMFxSPVUJruMP+VX82ng9Rq77JnDyBeD7NwGx0bzOwQeYz/TGAPd1xIvBnFSPFx6/wQNv7tQl1k3pU2z/lxO2DNESe7CKucubG6onZW7241OVinHW1taLJfRgxVWlO6pHVPXU7o5l5UJtGX6zLDOnHMaWvw141w+AgT3At98E9O+e8xyW4Te9MaCyRcdyGbGUZrjm+DM3Jfuexd6XqSqo1ZO2UT2lLWWdSRJ0o0s3S39XgYCgLDp+l3Wv0lrlVT0yuFul4GWZuawwntaBc94K3P4IkI4x4//Sd2a1cBmqh1S0euJsGmvdjEO45/jZo6q49fgrn7kr0kul/L0opVAUtrezO6rHrACrkJreSpCviryYKAkBgr5KCi6kx1+VMMybkmdE8voeWHoF8MHngWVvAB79G+AHbwdGjzieI0P1iB5/5agLp2vzG9az4K5LzjpTlrlyCUliwK+0O0oxqkf1oh7UPNrzwa+6L5XOlXaV2uVNevxVCn5Tcs8zIRrGunbg3T8HbvgqcGoH8M0rgOe+Buhp2zm4UeUbsQCVKUGQnmWQcWolpRmuAs/ZWy8WnfdQJTtwOT3nODESw78+ts+1wTAo8/Zd14Myc0X8Fdyn2OKsy1QPyn1SJbEUVZUojVyu7PByoMYMP7spLaonlcWBKwrwujuAD78ArHwz8NRngW9dDZx8yXoL/3FFqqeShszpZkkLE1rKRWBOvCnZtYo7z0wdf/lvDLHCpdPv9fT+AXxr4xEMTiVdXccwK8ASF3s+UEpNdVCFOX5LQFD64K6quFshaUZmH2ygsqvKWijSVlOGn5dl5pRDzpIGjV3ALT8Ebn0AiI8B33kT8PO/BEaPWEbMZy4rgdJueJ4Ls9VKF2kNN3RPdnC36K0Xs6ieSuis5/L4eZu8KHPhNrjL28Kdi8qpesoh52TnDrhV9Zj7YFfSGaulIm2e7cBVDeCKC4vjn+smP+cGYNkfAJv+E9jyTWDvr9B+9s04h1wMv3pxhT3+3HVURO/WjZafn9ot1aPpBpO/upSFusFcmbspr+IiBqByjt/l/gW+Cqt65tqIpW88jrb6AILmSq4YZFaVbmv1ZFRQQIUqwJYhJlIu1JTHz5fhOakeJ4QagT/6FPDXuzC4+s/QePgRPB78BC579nacffJnaMFkhcoM5x5kovHyxuM3DbaL4K6vwtSYNofHz/vJC4+fEFPVU+z+BRoXELA+q1Qm6mwerG5QvOXrG/HACydcXUPc19ltApdPzTgXlawHVamNc7xEzXn8xBbcLeAmb+jEW157G9a03oDz+h/C/0tsxcW7P4uXggrSv7wfOP8GYNV1QNvZjOAtMWbzLsSBl3RhyDJ77rpLVEtbtY04ZVSJwJvh+JyD95lbw08pWyGpSvHKEnt2eCU5/tmcCx1TSQ1DHsREAEaduq3VI1I9lVxV5rr2ydEYBqcSuGRpazmbVRRqyuPnHH/Qp4AQICF4/Lt7x/GDLcdm/fxUQsOxWBD36Ddi542/w7N/+BDu0d8GJTEOPPkvwN3rgbvWAb/9BHD4GUBLley7zFZAy0b1uAruZrwxwM1m61Xg8c9F9XCP3wMJrKLAHdVjcfwKfErlMndnU6nw/pqTLp0DYiFA1yUuhOBuJbzuuTJ3v/n7Q/h/D+4qZ5OKRl6GnxByHyFkkBDiuGcuYbiLEHKIELKbEHKx8Nr7CCEHzb/3edVwJ/DAGyEEIZ9qG7QPvHACX3psf87ParoBzaCYiDN5p09VEGs5F/+h3YIj73wS+OgeJgVtXwlsu4/lAvz7CuDBPwM2fQM4sRVIJzz7Lvl7/B4YfpfcfFqoO+PmPG5g9/hz95mbFRJgH2PFb05vlwxXor+4sgiYPSaScDG+ADudSKk7ybCvwh6/PkcZlWhSx3RSK2eTika+VM/3ANwN4Ps5Xr8ewErz7zIA/wPgMkJIK9gevesBUADbCSGPUErH3DQ6F3hwF2BJXKLhH5pKIp7WoRvUMlAi+ECfTGQMv82QNS9hUtDX3QGkosDRjcBrvwWOPQe89ig7iRoAFq0Fui8BFpzL/jrOYXGEAsAKw7HnTkYspXml6mGPmWqTxd+UrNJk5Wr1zMXxe+fxCzr+YldIomRYJUhole0vp9+L95f7iZI9inSiWgRTqhkGAkJuTSVWlWJRO0opSBblm9IM1yukciEvw08p3UgIWTbLW24C8H3KpvOthJBmQsgisE3aN1BKRwGAELIBwHUAHnTT6FzggTeABXjjKQMvHh3FOYsaMDTNuMpoSkNjyD/js9xz5rbPLySLzDAkgTpg9fXsDwCmh4DeF1lBuBMvADu+D6Sjmfc3LWETAJ8MFpwLtK8GAhHH7yF6r07GxStVD/e+/C5lq3yPYtWlLNQNROPlOFl6xvEzOtFNApcoGa7UZj/5U2Me7fImrCqdHK+5wFeVPpflRdxArADrNIHxirmGQaEU8R3LCa+Cu90ATgr/95rHch0vCXitHoBpysdiKdz2v1vxN29eZQWpYknd0fBnD3BRxz+nd1HfwaSh59zA/jcMYOIEMLgfGNwLDO4DhvaxVYLOg2UEaFkGLDgP6FjNnrcsA1qWQgt1AmCxiqSZnSt6Fzaqx8WNaSkuLINd3Hmqg+OffbL0yoPVKYVKCEuTjnjuAAAgAElEQVTg8kDHrypKRVZI6TmoMc+oHkHVA7hbVTLJcOULJ/Lrq4pd5ipOlpFAdetmvGqd0/RGZzk+8wSE3AngTgDo6ekpqhFczgkwqqd3LAbdoDgyFLUMfy4OLpsrd1WkTVEyhnz1dZnjugaMHWUTAZ8MBvcBB58AjEy7IkTFc4FWDCjtOE7aQJ/aAdK8mK0cmpdAT2QGnCsd/4xlePGcdeVVPXN4/Nzwe6DjJy5r9Vg6/gp6/PocHj+XnLqXv7JHL3Z5iwR8VVE4Mdf1LcOfNhAJlK1ZRcErw98LYInw/2IAfebxa7KO/97pBJTSewHcCwDr168v6lflgTeAlQ84Nsx233q1b8L60WKpHIY/yyD4VMU6l2fehepjweH2lcB5N+LHL57AJm0E//XBNcDkKWD8ODB2HPHBw9i+6QUsJWO4TNkPsmkLQDM34LUAdgbrcZwuQOsLFwDjaxl1xFcO6swVjRNo1jK8eKqHefwqqaDHb7spSyfnNASqx00wHDCDu2plMnfTc8hfUzrrJy/6C3C/y5s1xixVT/UJCJK6N0qocsArw/8IgI8QQn4MFtydoJT2E0KeAPAlQkiL+b5rAfyjR9ecAe6NAczj5979awNT1ntyevwzqJ5ZOH6PsPXICJ47OMQmhJalQMtS/MeB17D5yArsSL8OSxrCODkax77PvAnh5BAw0QtM9OKVvXuw65VXsIycxqrhrUDvI5mTKn6gdTnQehbQZv7x5w1dbDViwlOPX1WgKGyDkkpz1qWkLjidSEjxHH/aJueszA5cc3H8SUvO6ba/siXDxZ0nbZWxrt4KsGmPkgTLgbwMPyHkQTDPvZ0Q0gum1PEDAKX0HgCPAXgrgEMAYgD+wnxtlBDyeQC8CtrneKC3FKAC1cOTuNjxzHtiSecfJdsg+MsgT5xMaDMopn39k9h1km0Sz1PlNShA02L2B2BP/FJ8cucrAIDPX30+3ruuFRg5CAwdAIZfA4YPsrLTR54BNEFiqgaBpm7zXEuwRGvBLWoSXSPTWEnGQZIrWGcVmKCmmbtJATB16ZX1xmajLrzYvEZRwBK4PNDxqxWqx6/nS415RvW4zA7XDQR8oqqnEhz/XKukGjP8lNLb5nidAvhwjtfuA3Bf4U0rHAbN1J0J+53ri0RzUj1ZHn8ZNmKZSqSR0HRb8Dae1q2bJVd9e3tw12By0e5L2J8Iw2AU0uhhYOQwiy9MnGIrh8PPoGeqH//mp8CW/8WGIIDHATxVBzQuAhoWAY3d5vMuoGEhEGkDIu3sMdzCViowVT2mJ1YpXfpc8sSkh7p0zvEXX9QuQ/VUiuMXx1Cpi9oBmT0f3KwqfYpS8lX4bMif468Rw3+mQBcMaC7Dn3dwtwwbsUzGNVDKbrKAj11LHDSWx591/VS+tXoUheUfNC8BVlwz4+Vn957Cv/xgA756bSt+tOEF/MWFQaxrjgOTfezv+CZgqt8WeLYh1AzUteNzE37EfM3Aw2fjb5VJ9JxaAuw8j+2BEGkDIq3sMdhYsnIXc2227tUy3E71uPT4TeVYpWMis+n43eY98N+CS32LLnNRFcqxPPvMpXNRDtSU4bdTPcwDJSTDXlCam+qZGdwlnvGJzx0cgkoIXn92u+04TxZLaDoCPl5KOtOOoHks+2bxqiyzTnzopR2YXLgejxgqLlt+AdZdttT+JsMAokNAdBCIjQDRYbZxfWzE/BtGcuIwFugDwKFjuJ0Mwn9KB045XFDxmxNBG1DXlnlurSRahf9bgUA9y5nII1itGYxuSut0do7frSGjFCpxt5WgJgR3/SqpjApqzmqm3gR3rbwYxSXVYzDlmHVPVji4O1u2c17FISuMmjL8YuZuyCzNvLy9DkeGouhqCuPUeDynx589wJmCgD1361189ckDCKjKDMM/lWBtSaYNIGS2QxM9fuc8Au5Z+BTiirPmp+WrI8fyD4rCaJ6GhTnP8w9f34ilbRHce/t6XPH5J3HjuY341B8uFCaIYWGiGAGi5uPpPewxPoYcKl8GNQD4I5mJIGA+90fM/+twY18M5/gMjMOPlUdfBvyLzdfY+1akDoASikicAPFlgL8O8BWuuRO3XnS7fwHferHSMZHZavUk0rpjlmq+yCRwuSsEyHNFKsnxp+cSEHiU9FYO1JjhF3T8pjHrbAwhmtSwsDGI0WgqbzmnX0jgcuuRTcbTCGZRT7pBrUlInHTEwnI8QO3E8SuEKZfc6fjZeeuCee5fkANpw7DkeqqqIEYiprJoeX4n0DUgMS5MDMNAfBRIxVh5jHSUPaZiQGraPBYDpk9b77k0Ook3kDh8fh3YD/Yn4F4ACAI4AeDfzIOK32ESqWfH/BHAF2KTgy/EJh9fCLenTmHRSBOGEkAafmDXUes1+ILCc/FzQds5NNMwVFLHz41Yrp3E+LgyKPNki63Jnx3cLZ7qMTdiUStH9cy12Y9XyrFyoLYMvzGT42+tC6CndQHa6gM4MRrHdE6qJ3NcIYCiiMFdd+2aiKcRyZo8phOZCUicdEQqIpfHz2+CoE91RfVwqV3Qp0JVSM5JcS5wbwwoUtWj+lg8oK591rfpBsV3Nx3Fuy9bam22w/HR72/Dq6cmMDQxhU9euxTvvbidTQ6paSAVw0d/8Dz0xDTO6/DhQ1d0mhNJVHhPNDPRRIfYo5YE9BRTRmns8a+oDgwKF/5VYV8VAG4GcEPQD/+9YXzUUPHnRAXuaspMGGrQnCzMP/6/GgAUH6Co7FH1m8f85nPzf9VvHguwvs1+j+JHaDCG88gxKD4/FqSngNGjwuf8oMkpBJGCDgWJlO7C8NuDu0UncJlUXmW3Xsy9SqKUelbRtByoKcMvlmzghqGtLoDP3nQBAOA3u/sRzUn1ZH5I7p14kYlKKav4mc0Hc36fXVt3fB70OXv8Kd1AQFVYSQdX1TnZo6oQRPwqYgVyk4eHprG7d9wsmVt6Vc/u3nF84dF96GmN4NrzWVmLiVgaX3h0LybiaQQDKlLwI6Y2soC2gKeMGKYMDScCzfjQ5VcW3YYrv/gk3riyCcPjk0glE7jvPWusSSEzSSTNSSPp+Nru44PY/Fof3nt+Jw72DeNQ3wje2dWR9bkkkJiYMfGA6izuYmjsNSM9d6MdcD6Ax4LmPxMA7rK/fjuA2036Ef8OgAiTDZ94FB9AFACEPRLCXg/UMSpN9eNt40lc4E9g1bYW3OufRsdjPwTCQfOzqjCRZZ2TPzePv58ewiUDHWh5uR3vU4+i5/A+QG9nE5U4GWa3UbwGUTPXsh1X7OcgKqM4sz6v6YZVRmXmKlyIu0nDX16IVA/f+LulLsPj1gV8s1A9mR+LeydeKAhiKR2aQRHNWmnw8s/s2rxAHLUb/lnknAEfM/xe1OPntFGuwHcuPPjCCdy/5RiaIwFBx186zprHRESPavuJUfxsey98CsHKhQ0ASqtLT0OB7gsj5iOYSEdYpnSB2LHpKL786l68601vxjObj+Ebxw/i5pvfWlxhL0ozk4CeZn9G2vxfmBz4a+b/r5wcwd0b9qEtTFAfAP7pLSuFz2rYuL8fmw+ehg86/urKHjQFCXvN0M33aabaiwLUYCEaarBJKxVj9Jyhw6cn0ESiCCV1LCbT8I1PA5M083k+iRmaOanp5v965hiAv1XBaLoTwGf9AF4x/8qI3wHQfAp0VYH/u3yCYROEj6jYGkxDg4qGTUFgZ/bkpuSehKz3qUwmfdPdJf8uNWb4Yd08IcHj56gLqrNk7s7m8RdvyLiBT+kGklpmyTwlUj2mMUrr1Bb84hx/9rIyrbHdiAI+xZUh499LUQgiARWxAs8VT+tI6xTTCc2iepjHXxqOk/924iqHr9Q0g+bMewC8LNnAa/W42LhGKI5n0YmUQnEsbTUHuJedQ/n0i+29aAj5rBUSABwcmMKrdWN4wqjH0lAECiH4p7XX2D73wsR+3LP/MADgxkuuQpM5qRaKn2w4gG88dRB3/9E6fOSBnXjiXVdhdWcB56IUmqbhvE8+io/94XL8ybpFuParT+NTb12Fmy9aKEwgujkxCf/raWEy0e0Ty4zjWtZ7so4bOu577hBAdSRTKbz9gk4saghY50olU/j9jhNQYeC8+giaOuuF6xgzJzZqsBWc7ToGEC5JxfoZqCnD//VbLkJPax0AkeMPWq/XBX0YjTrvmpVMG6gzjZ9fMGKAu/ryomcfTWYMv43qMVcb2dwg5/idPH6/j1jLzmLB2Se2Qb0P8QI5fm5042k9E9wtYSbqlNln4upMlM4FcuzHqumGNaF6kcClErb3a7G7QKVFHb+QkJQj9cQVvrXxMDqbwjbD/4EfbseYeR+EfCpi6Zm/u0hduJksZ5RsKFTOSQg0KEjBDxqogxJuwjgaMO1vZVnoHsMwKAankuhsCs147d7NTyHoV3A8FsNFF1+GRWdlYlLjEwl84sWnAADvX7oc57/tPM/b5iVqauvF6y5YhPO62KYnzRHmAXU22Q1/TjmnpiMcUBHxqza+GnDn8U8Khl8M6No9fmcaIpfhTwrBXS9UPQoBm/QK5PhFg8ANv08tHcfP+0w03vGsNjhtkGLPe/CmSFsx/cVh0/HnW/q7SEwltBmGeyKWxliMjcugX3HUxIvjys1kybPpueiimLHBJ9iAsANXqZyLx/b046qvPIOJWOa+ffHoKD78wA6kdcOikGfE3WwCjern+GvK8ItYv7QFD9xxGS7uabGO1c3CYyfTTLJWF/TZFCqAu5tS9Pinkpnn4oTAB0r2DZYruJvWzOCuX/FEx888fneGn9+QpcxEdZS/Cs9VxVkXL+Y9eFFfnhCCuqAvp1BgNjx3cMii1MQ9ZEuVkDSV0GY4FOJkGfKpzpUmNefJtVDwidJN+RNu5G39VaIx1jsWR0ozMB7PMAObDw/j0d39GIulLDpxxhjTxYmy+g1/TVE9IohDpuxsN2tS0xH0K1YGLZDx+IvN0ARmUj0cTh5/tqeQa5BxOWdAVTDqgcdPCBAJqBicTM7xCTvE9vK4SCl16ZbHn4Pq8Vu1b+x9kjSzUBvDfozFUq4SkrhyrC7oy1n3KRdOjsbw3u+8iMUtYfhML7iUG4vwXBFxsqOU2gx50K/MquMH3BkyHhPhKrtiVkmZMtaKJ87YbOAOmdOq0qCw8nGyJ2qv+qtcqFmP3wl1AXazOm34nNS4x69a3onPLDNcrL4dsBv+adHjd5BzZg+YQE6OnzJVj1/xZOtFhRBEAj5Hrnc2iDeHX8nERUqVVTntQPWIk0Auj59TPQ0hn1UbqVhwDzYSUJFIz5T1zQb+mw9MJmzBcKA0Hqy1QhL6iO3olnlPMIfHn/bIg+X9VR9kPmahq6TTEwmcHIsDyN4OtTRjzEk5JjoXObPpdW+osXJhfhn+oA8Gdf5hEmkdQZ/CdvkxvVdFIehsDKF/PDHj/fnCxvELHv9kPI3GELsZrNrn5gDj8QlO9cRSum3wpXQDfpW4TuDKpnrEa4xGU3jz157Fvv7JnJ+3UT1l8PidqJ54yq7Gcro+nxwbzP52w8Fy5Rg3ZIU4BXzcsc3pM/0FlMaD5cHwXNQYwFaVuTx+/h1d5YoY1HIsgNzVcXPhnx56BR/7yS4A5naoJd7sx6nP4nlk04sO2JlQq2eeGX72ozkFeJnHr2B5Wx26mzMR/e6WMHrH43lfg1JqedJJTbd7/FnB3fYGFnhOWFQPe1xgHudUz1//eCc++MPt1mdFqser4G52Atfu3nEcHJzGy+beAE6wB3czHmypdkea4nLOHPwz44Bnxhi499oQ9M9od6FgZZmRMWQF5D7YJ8pMTAQoTbByrmA4wAxZrtryjV5NlIQIHn9h5xqNpnDKvP98KrE2+ymdcszBuUjPNPzZfcbvw6BPkcHdakNdwL7c3HliDA/vYmUkk5qBkF/FF95xAf7nPZm69t3NYZway9/wf/LhPbjj/m04NhzF+Z96As8dHEarmUuQTfW0RgK2Qmt8sC1oYBOP6PFvPjxsvZ42M3fDAbVgD0oEt4+EmDr+lG4pYo4MRQEAIznkr6y9grct0GNuPf5tx0Zx2Zd+h/GY/dpzebA+RYGq5OZfG8MeeLBmkTbuRBTS/44qqBIWHXM0YlneaNCX2+NvDHszUSok43QVSvUk0rrVvoDVZ6UTEDhNlglbMNyZfuXVTJvCfkn1VBvOWlAPAHju0DAA4N6NR/Cxn+zCocFpJE2qx68q1k0JAItbIjg9mbAkeJpu4NvPHcGx4ajjNbYeGcWBwSkcH41BMyiODEfR2cgM+XRWcLch5EPIr2Y8fm74G5nHHxQCzWmdYucJ5n2nNObxdzQEMZXQHJeW8ZSOTz+8ByPTuQO21Ja5a/fujprfb2wWw590CO56oerZeWIcA5NJnBiN2Y47cfwzg7szr8/514aQf0a7C4VBAVWgLgrJds6WngKl5fgzeQ+GNaE7e/zOBcc4NSbSaYWCUmomCLJz5ZJT50LcgU4sZZLgpEOuiLgSDuZS9Zg7vDWG/bUT3CWEXEcIeY0QcogQ8gmH179OCNll/h0ghIwLr+nCa49kf7acuGhxEy5c3ITvbjoKw6AYmkrCoMDXNrzGqB6HDJruljB0g+L0JOP5v/LEa/jCo/vwrm9twZGhadt7U5qBY8NRTCU0G7ffHPGjPuizUT3j8RSawn4zCYsNFO6J3rBmEf7ssh601tvLBr9wdASAyRH7FHSZlFTfxMwVyebDw7h/y3E8b05yTuDGQDE9fiAzyLnhz5XwBmQFd1XR43d3U/ZPsL4eE7TUgJC5K9yUTsHdbBXWDI7ftcc/O22YC/Z6UN6VBcmFKYdCgE4cP6Uzs5BT5grYrxJPqB5VIQj71YI9fnFit/pMLUNZEOG62fJXYCbVxJ2Lplox/IQQFcB/A7gewHkAbiOE2NLSKKUfo5SupZSuBfBfAH4pvBznr1FKb/Sw7QWDEIL3X7kcR4aieP7QMIZMb/ixV06jfyJuLeNEdDeHAQCnxuLYfnwU39p4BDesWYR4Wsc3njpoe++xkSg0g2IynrapdprCftQF7YN+PJpGcyRg8/j5ALtoSTO+9I41tpUHwBJJAO7xEyxqYm1zCj7zfXtFj/3w0DTe/LVnMWBOYmJwlxt+PuD5pDYam83w22kWgO205Pam5O3LXm3MRV3k2rycc/yNIXfUBYvfsHHEacNCgrvZ8QhA2Iu2hB4/kPnO2d77bLu8BX0KQj7VlSHThfpZTALrIklQKb2AYHKO4C6XpWY7NxadGPKdEdU58/H4LwVwiFJ6hFKaAvBjADfN8v7bADzoReNKgWvPZxuKvHJqAsNTSaw06Z9E2rCWcSIWt5iGfzyOrUeY4f3SzWuwprsJJ7OoiIMDzFgalMnQOJrCpsdvGv60bmAqqaG1LmDz+Plg4wEkbhwANqC2Hx/DdFJD2qwS2NXE2xbDPc8extBUhtaxDL/gNX/1yddwcHAazx4YMttp6viVTLCSK4j6JpyNL4emGzZj4fd5d1P2myuYMWHSMYT9C5I5NNPc4891U7r1+MUSFxmOP/+bPDkrx++uzx7edcpapXFMCh6/U1kQhWTa4VgB1qcgKDgmxUDMmagPFu7xO62SSpUkKI6xRA4BQSa4myuOVCMeP4BuACeF/3vNYzNACFkKYDmAp4XDIULINkLIVkLI24tuqUeIBHxoifhxaHAa0ZSOS5ZmMnudao53CR7/K70TWNYWQVPYj86mEE5PJJDSDLx0jE0IBwenrM+JAWFu+LkqhRu0lojfdmPxR77yUIQko3dfvhRJzcBDO09Zqp6FZjmKp/YN4su/3Y9HXu4DwG62ly3DnzGeu3snAMBSa1AHj/++54/iPd95AQCrd5QruJu9haFNx+9ScTFgJpKJk5YYRM1WXGS8Z+J4/Yw35o7jt5W44HJOwZD980Ov4IuP7s35eSfawovS30lNx8d+sgv3bz5mOy5SPdmrStYGJWdwOWVmh4f87goBGgYsCWYkUFi2s6YbNn28rQKs200yHMByfNjzXB5/rjIqKXNMNYZYcNcpV6iakI/hd0pxzPWtbgXwc0qpOFJ6KKXrAfwZgP8khJzleBFC7jQniG1DQ0N5NKt4dDWH8XIvM4xrlzRbN1/IweMP+VW01wfROxbHnr4JXNDdBABY1BTC4FQSP9/eiz+9ZwtOjsYsjx8AesfjaIn48ffXrcZNa7tRH8oM+nHToLVke/yabiapZOrecFyzqgPndzXiR1uPW8HdoE9Fe30AGw+y/uIrkGMjMcvb48bTMCh6zcmIS+r4/qeqkFn5q12nsP04qxC4rqc5p8efHVDmbXZbOM4wqCPVI3Lp2fJELn/1KcSxVhA3HlxdJdJwIo6PRC3poGPb+ETpEKxMaQZ+saMXzx3MHVOxZTpn6/hdTJa9Y3EYFLYVH+BM9Yg7vPH+AnIUAlQVRkW64vgzVE99gdnOM5wLsRCgS4//0OA0frO7z3bMaaIEkJXpnMPj53Ri2J6bU63Ix/D3Algi/L8YQF+O996KLJqHUtpnPh4B8HsA65w+SCm9l1K6nlK6vqOjI49mFY+u5rAlV1zUHLZ4/Fy7DC1ri2DLkRH0jsUtw9/ZFIZmUGw+zG70YyNRHBycsjznU2NxNIb9+D/XnI3zuhpRF8gEd3nAtCUSMD2qTAJXSGiDKlA9rXUBvPfypdh/egrRVKYa5qKmsDVIuWHfdZIZ7oagz5JE7j+dWY1wXjq7ZAPABuw5nQ3489cvw2XL2xBN6Y5L1wwtZZ+k2uuDGI0mi6Z7hqNJ66YSVyv8pmwI+rLknAYWmKopn7ldZq5l+JLWCADg9ISz0ulDP9yBT/5qT862OfUXD4bvPDGGRNqwJi0niPz6jAqwLgzZiRE24Q9NZxv+maskbsR85g5zuaimlMaonrBfdZWQxEs2ADDjXPZzzea5z9gHW8mMNbd04v/8/jD+5qcv2zxzp/7Kfp5LzslzV9zGkcqFfAz/SwBWEkKWE0ICYMZ9hjqHELIaQAuALcKxFkJI0HzeDuBKALnXwmUCN/QA0FEfxNI2ZhCCDsFdALj10h5LWriGG37T2Gw9wpQ2R4aiODocxUWLmwEwnpoPAgCoD2U4fm6MmyN+M/vWVPVouk1ZxAc6e28AN67tsnjqgGk4FgnlY3vHWBtfPjmBSEDFxUtbrEmG01FAxlg5UT0AcPXqDnzmxvPRYXrSYw4BXt5mnnPAA28LGkMwKGaVkc4GMTYyLlA9/KbsaAjO2KO40zL8rE7+ydEYPv7Tl4WYCvuizRE/GoI+R+OcSOt4bWDKdn0A2Ns3aa0CDGGFxPdD4B7s5sNsHIzF0jlvetFz9met6twY/uMjzIkZnmH4RY/fTvUsbAyZ1UxzURfM8DdH/BiNOa+QYikNT7x6eta2MTkne55dK2vD3gFc+Nknc46V7AnHy81+Dg1OsWJssdnLqKR1A2ld2ODJ76zqSQocP1BcTaJyYk7DTynVAHwEwBMA9gH4KaX0VULI5wghokrnNgA/pnZy61wA2wghLwN4BsCXKaUVN/xdQmZue0PAMvyhHAXR3762Cz2mt3i+WfaZG9zhaWYUNx4YQlqnVszAoJllH8A81emknX5prWMev8jxhwNCkTiB42+O+BEJ+PAnFy8GAMHjFw1/HJRS7Do5jgu6m9BWH7AGdt9E3LpxeEAyI+fM6PgBWN+VUyMj0zMNP28znxy4AVto/j+QZ8G3ff2TuPjzG9A3HsfWIyP43b5B67yjDlRPe31wRuBtUXMIAVVBQ8gHn6LgyHAUv9jRi1/tZMl5nH8NqAoWmrGZbBwcmIZuUNs1Nd3AW+96Djd/cxMAuwoKMKkLs11bTMMPwPH8QBbNYlFjdjVVMThuOiXDM6gezVqRWcHdVCZXRJ3F40+awd2FjSEM5Pg+D7xwAh/4wfYZIgcRupnwBrAESpGye/DFE4ildBwcnHb87AyPX0jgclPNlFJqXXNQ6DNHFVRWUqVfVcwN6p1jIm2z3DPVhLx0/JTSxyilqyilZ1FKv2ge+xSl9BHhPZ+hlH4i63ObKaVrKKUXmY/f8bb5xYEHbBUCtNUFsdTcvCWXx+9TFXzxHRfgA1evQHOE/bDZGzVsMimfS5ZlgsW8RADAArxTiTTSumGjekSPP5HOonoEjp8b+vdc3gOArSAARlUBbCUyndQwNJXE3r5JrFvSjJZIwPLWJ+MamsIBlqGb5FQPzH5ge+5yLGmxG34nj5/fGB31QbN9rK2cdhG9ak03cpaWeOnYKEajKRwdjuLWe7fiLlMie+6iRlvmLr8p2xsCSGkseJY2lUWtkQAe+vDr8a7XLbHRYz95iWkSrI1PfAo6G0NWToaIvf0s8D0aTVnLfx7r4JOYSPUAMPtSx09fOoltx0exdglb7TmdH8jy+M128okz21svBJzqmUxotsD1VEKzDBYP0CbSOkJ+BS2RgLV/AQDsOTVh1aDnfRtQWX8NTTtTd9uOsf4RjWc0qVmyYyCj4weYx8894fFYChtNdVlvjsz4bFkkn6RCLjPWT43HrXYMTmV+K76q9KuZ8t18sl7UHLJec1px8BWSJbF2yK1J6wau+coz+PGLJ4puu1eYV5m7HNzwt9Yxr6eHUz0OwV2OP1jZgX+8/lzr/9ZIwEohBzIe8Drz5gfsHv/ilggMCvSNxzFu1vUO+dUsj1+3rTpEOSfH2Qsa8Mv/83r86XoWdlm9sAGqQnDjRV0AgCf3DiClG7hoSTNaIn7ETI5+Mp5GU9jHlBXc4xcMWVigerI9fqckrhkev7meX2hmHQ8IN9Tf/uxl/Om3tjgqHXisRUx48ykEKxfU21Q9PD7Sbk40Sc2wyV/P72oyC+xl+uyVUxPYc2rCCrwFVNODnUzgvueP4u6nM3kYe/tYMbqUblhe6VP72eqDr/KoOXeJHuy242P4+1/sxpVnt+NzN50PILfHH09ldnfj7WytC4CQmYHZQnBc8LhFT3MqkUa7mQQoUgyNB58AACAASURBVD1hv4qzF9SjqzlkteMDP9iOz/2GLcY1g+Ur8BWSbtAZExOlFNtPjJnXzLx2493P413f2mKLI2WCu6pVHffXu/st45lrxZAtI+XVatvrAq48anGFIZYin0yIq0q7x89X1j5zspwZ3NXZCslU2jlN/rt7x3FsJIZXTk0U3XavMC8NP+f4udE6t7MRqkLQ2Rie7WM2KAqxfuQLuplhWNwSRnMkYK0cRI6fTy4nRmMYi7E6PQBsHn/c9MY4VAfDDwAX97RYRa+uWd2BLZ/4Q7z+7DYAwKO7+wGwJDC+0fx4LI2JeBqNYT8iARXxlIZdJ8ext38ShLDgW9CnmDsliRPjbIaftfmqVR1465pOnG3mQ7TXB0FI5oaaiKXx2Cun8fLJcUtOKuKwmSg2mUhbN/YfnrMArXUBxNOZwDJPJFtorigSad26KUOBmQHxrqYQGoI+/Nvj+y3+1a8SdDYFMTiVxPc2H8PPtvdan9srVCEdi7IJ53f7BgBkPHxRzgmwYCWP/XzpHWtwVgfrg5wef9qwPHBOW/hVBa2RwIzA7GyIp3Ro5gT1+J5+nByNYXk7W7WKBnoqoVljPJPAxQz/P1x3Dh74q8ttY+zJvaeR0jKrs4C5QgJmTma9Y3FrsuKS38HJBA6bEzn3nqlI9QRZWeyfb+/F53+9F2u6m7CwMViwx99WH8BItPiJ8pCgvnOiejoaghYlxlcG53Y2QiEstudTFGw9MoL/eOI167M8qbK9LgifQhwn/02HGB0o/kYT8TT+7mcv49W+8k4G89Lwd9QH4VeJdVP0tEWw9R//CJevaC3oPIvMieLqVUyFxJPBeICnQTD8PI5wfCSGsWjKooy4x//sgSFMxjWbxy9y/LlACMGCxhAWm/TM1qMj6GgIoqsphJZIhqqZiKfRZBr+aErHv/zqFWzYO2DdlMSkexY1hiwD3BT2g5DMTU8pxcYDQ+ifiFse0dK2CL757kssXbtfVdBWF7SW0I/t6UdKN6AqBD/ZxqiXF46MWF4i9/iHppJIaQb+7i2rce/t62fQTP3jCTSF/dZ3SqQNJEyVTNhhlbSiox4fe/MqPHdwGL95uQ8BVQEhrMy2blCcGI2hfzwBw6AwDIp9/VNWst5INInh6aTVNq5EsQy/kjFk/PrdzWHUBX1oCPlye/xp3RpzfsHgttcHbfw8pRSHhJyQbLz9vzfhqxsO4GfbTuKDP9yBpGZYO81xo2IYFNMpB8Of1hEKqFAVYmU6c0wlNGw+POxo+LcfH8MHfrDNWg1xGgzIePzfE/II+PsMI7NCipj99amHX8Wqznr84C8vxZKWiCVKmNFfKa6NZ5/jk2VbfdBGyRWCkekkdvWOo70+iLqAOoPq8auElV3Q7MHwCxY3Yfdn3oLVnWyFvbt3Anc/c8iq2cWpHkUhWNjoHEfaZJZP4RNmIq3jPd9+AT/b3ovH98weJPca89LwKwrB6s4GnG16aACb5QvdlWlRcwhNYb+l5Fm1sAFAJkNUpHoWNjCDyjz+FFrqMjX342kd77vvReztn7QZfiWHx++EprAfDSEfFELwqT8+D4QQq67/WDSFyUTaLB3hQyylYdRcKovcbTigWpJHgHnPa7qb8K2NR3D30wex48QYbr/vRVzxr0/jQZOnDDlIYBc2Bi1e/Fc7T2FFRx1uuqgLv97Vh1hKw3u+8wL+97mjiKd0SzHDH5vMSbPFbDtfbfRPxNHVHM4EKwWPXzT83INd0BDE7VcsxfL2OhwZjlqTGV8xAIzWGZpO4rWBKUwnNfzROQusa/LrigFc3aLGeEISu+7KhfXWb9XZGHLkd3mb+YrIJ9CEHQ1Bm8f/m939eNPXNtqUWBzDZnuPDUctKmzlgnq87aJF7PUp1u7RWAqUwuKcuSFLmFRPpr/sJuCxV/qtEhdikuB3Nx/FE68OYLeZGLjjxBjqAirqgz5L4CBSGJya42WsWV+agey0jhvWdKE5EsCS1khOj59PVjyexmmytroA0jq1ZSYnNT2nxPPAwBTuefYwUpqBq7/yezy6ux+rO+uxoDFk8/gn42k0hPwI+9VMTMScfCJ+1Vpl+wU6ka8K0zq1qN/OphD6JxL49MN78PgetgKPp3SryCLvr02Hhq0+K1UJilyYl4YfAH5y5xX4xPXnuDrHR954Nu66bZ1V9fPcRYzy4RSPSPUoCsGSljCOj0QxHktbnmt20pib+vr/csO5uO/PX4e3mXx/xmtO2zz+WErHeHymRO/cRY24dLl91fOjOy7D65a14Ofbe203KA/gOSW9cR59LJrCS8dG8cdrFuHys9owldSw4/g40jrFseGorcTAyVF2br5a4isirko6NZ5Ad3PIUsEkNSNj+AUlFPcuOxqD8KmK1RfcA80Oyp8aj1uKnLeuYcZzJJqyrtvdHEY0qbFVghlEVQWOH4BFc/Hzn86haEqkdUQCKprCftsWnx0NQRvHzzNwN+wdmHGOPaahmIinrc18NvzN1bhsOaP6+ATC6zetMCkg0eN3WiEBwBvObscvd5zCi+aEE/ApaDfjYPz34WNg+/ExrO1pRkdD0FpljMfS1rmjSZHjt/cXAJzTyZykxS1h9E/EbTt+if0FZCZrHkdqq+fKGXZdTTdw/Teewxcf3TfjHADrzy//dj8GJhOYTmp4x7pufP6mC1i/C7/VeIzdI6x+lp3qCTvQiUDmN0pqBgLm2OxsCuHAwBTu33Icvzap150nxpDSDazoqLP6y0m1Vi7MW8NfF/TZbr5isHJhA65e1YGzOurxiw9dYRkZbrz4I8fStjqcGI1jNJayDD83ZDwucGTYLm378s1rsOFjV+XVnlte12PRTgCsa4xGk6aRYIZ/PJZ21Bnf//5L8fFrV9uONYTYimZgMmnx9t3NYUsp41TRlHv8zx8ahkGBa85ZgMVm3IB7sSdGY7bvypf7fFmfHV/on4hjUZPd47eCu8Kqg99AnEu//oJOW9u44ee//akxJiPtaY1YyXlj0ZSlKOpqDiGW1vGb3X145z0sRUUsOgZkVnoACwL2jccdaYhE2kDYr+Lf/uRC/MXrl1nHueGnlGJv3yS2HR+DTyF4xgwunxyNWYmCr5pBaMvwm2MsbHnf7DfiFVsXt0RsKpV4Ss9pxL5+y1osbAzhIw/sBMDGpKIQKzMaYBnp0aSGff2TuKSnBW1CoHUinrak0tOCcowvKrjHDACrBMOfXduKI55l+DMeP2sPjy38ds9pHBmK4olXTzv2O48t8SDyNas7sKKjHgsagjaq59R4HF3NIZvgwmlVySeygE/BS8dGcXBgyqptBLBVH28bnyh3miulN5+3ELGUjmhSszZpagj5pOE/U3HJ0lbrJuJUD3/k6GmN4NhwFBPxtEVlULP6xQ1rFuEzbzsP/3nLWttnbr20BysFw1IIONXDU/qbwn7UBXzoK2BHMYDdePG0jsND0wj6FKxamPFwnTz+BQ0hjESTeGrfgEWFdZv8+bbjzPCfHI3h8CDz+Fe011lUDzdkXCY6NJVENKlhPJY2qR52A+YK7nIjxI0V9yw5ePDtqpXtVt+8cHQUl69oRSSgIuBTMBpNWSui7pYwKGUp/hwZzppdV+yPdT0tGJpK4rO/3gtKKb7yxH7cdu9W3L/5mBW8f8v5nbbftL0+gKTGgrUP7exFQFXwoWvOwsHBaZwcjeG/nzmEO7+/HZRSm8fPV3HieTiNwH/jRc0hW4XNeNrIqRzraAjiMzeeb/3PqQuRHusdi+Hlk+MwKLBuaYst0DoeS6HbjDVFUxq+8buD2N07nqnVYxr+hqAPXeYEzKXDTsoe/vvefHE3Pnj1WRbFJnr8lFJ8+7kjAJjhPjIchWFQ/GDLMcRTOpKajv2n2WTJ1U98jC1osFM9feNxdJtjLJ62q3rEPuNG/fbLl6Ip7MeNd2/C1sMjCDrk1pwyHZqdJ8axor3OopeHp5OYiKeZmKIpbCvZXg745n6LRKFwonoAFgjlA2l5B1uC95lL8stWtOKW1/V42o6gT0VD0GcpZ5rCfkSChe/TyzeGeeXUBBY0Bq3cAUJgk7RycGP52J7TePN5C6EqrIQ0IcCO48zzmUpq2HhwCCs66tBRH8QRk/ZpsqgexrX2jcctzpx7YwDjrDn/KnpjfIXADT8hBF+/5SKcGGHnUBSCL//JhVjT3YQXjo7id/sGMBFP4/IVbSCEMA82mkK7RfUwwyTW8OGcNacuVi7IGPFbX7cE+/sn8b3Nx/Cu9Uvws229GJxKWrJSpyRBHoAdmkpi06ERXLK0Be9Y143/evoQNh4cwqnxOKaTGgYmk9hjqj946W9xjLXXBzFkerD9EwkEfCyhKOi354o4xUQ43rg6s2LkuSM8wFsf9KF3LI4dpozz4iUt2LB3ANuOjUE3KKaSmqWYG5lO4eu/OwAAlvCAc/yrOhssI37WgnooBPivpw9hbU8zYikdP9x6HBPxtJVZfsWKNrz+rHbb9wQYV35ocBov907g/Vcux32bjuK5A0OYXJLGJx9+FXVBH87qqLdWp1yBxcfYgsYgYikd00kW1B2cSqKrOYy4UKaEP4qZ7ZyOvXJlO+74gxW44a7nMBJNOcaRhqdTiKd07Do5jqtWtlvbrXLD3xjyWx7/vz62D88eGMLjH81vhe8G0vCXADyoKwZ3gUy5hw9dcxbevpYVOP3g1Wch6FfwjnWLS9KW7pawRQ80mjp+jg9ctcKme88FfuMfGJjChYubscj8P+RTHQPiN17UhS2HR/DQzlO49jxWBjvgU7CgIWjL6N1+fAw3X9xtq5HCDRkhBF3NIZwaj1uTY1dz2KLGcgV3uexzgXDzZfftOy9h/3c3h7H9+BhUheAqkyJrrQuYHn8KqkKsvASx2ir3+N9yfidiKd1SA/F2v/OSJbh/y3GcGI1ieDqJgKpYlVkdDX89799p7O2fxN9euwrL2+sQ8is4Nhy1KLadJ8ZwcjSOSEDFVFLDWCxtEygsbApZ+QinxuPoagqBEGKnLlJZHL/5+1vbGqoKLl/Riq1HRi0HYUkro9iuXtWBXSfHsf34GFYuqEdTxM9qM8UYNUZppox5r62/2COnxlYLq7CFjSF89V0X4eM/fRnfeOogWiIB/OfvDlq/T8ivzBhjnMIcmU5ZpTL+4spleGr/ADYeHLbovKPDUVvZ7BmG3zTCg5OJjAy4OYzesTiSZpKgE8fPsaQljM6mEG5c24Xvbjpm5S5wj7+7OYxT43G8cHQEw9NJrOtpFlayGaVdfciH0WgKpycTZavlL6meEqA5zAZmNse/flkrdn7yzfiH686xBnNnUwj/eP25ruMNudDTGrF2tOI6fo4/vrALf/eWuQPc3INJ6xQLGjIevxPNw46r+Pota/Hc37/RSiwDMvkT4ufWLmm2ea0iPdbVHEbfeNyiLXKqehy8MZGXzgXenqtXdVhepGX4zUAf56XFXc44Z726swGfuP6cGYaJTxZ7+6dgUGDN4iZb32SDe/y/NqtFXnFWuznxhdE3nrDyAn5l7g99xYo2UMoLAWb6ixsaSin6x+NWPkbYr9qDu4GZqp52Ybe3r9+yFn90zgJcZsqbP3TN2fj5B1+PFR116J+I48Wjo5YIoL0+AEpZkUL23UNQiH2FxI1nSySA1roAXn9Wm+37v2PdYqxa2IDDg9M2rr9vIm6bpDgCPgVNYT9Go0lsOTyC7uYwlrRGsH5pK17tm7DomyPDUew+OW4ZdR6c54afTxCnJxJWexcLYyypGZak1Em9xvuXl1HZYap2Vi5swDmdDbjjD5YDYCotAFi7pEVYrSQzht8s5zIWS1uihlJDGv4S4E/XL8bdf7ZuBtUDwEqqKhd6BHkm5/g5eAxgLnCqB2AGlXs0uWobcSxpjdiMIud/ueYcAC5a3GwZr6BPsZ2TGbIE+iYSUAirA8Q9/mcPDOHhnX0z2sFXJ3XBuRezPO5w88WZ7SXaLI8/jWbB8Iu7nPEEr1xoqw9CIRkFzoWC4Q87TJbc6P5u7wDqAqr1/u7mMA4PTVtBwKf3D4IQ4A1mfCKe1m0cf3dzGCnNwPB0Cv0TCUvKGcoy/GJ/cRlhuzBRLmoK4zt//jrbZHhBdxO6m1kgNprSLQUUD7TyeE2zOcbEFRKvDBvyq9j2z2/CDeZnRTA1VAIDkwmc1VGHoI9tCelk+Nl1WdLb1qMjuMKcSLpbwhicSlpG/MhQFDtPjuPiHia35h4/vy+t3fWyVpUhYVWZMPfidpJW89Xz+V2NCPoU3HYpo2qbwn48/tGrrD76ze4+NIR8OGdRgxWfGJrKGP6GEKvcOxFLoTmc3z3pFpLqKQHa64P44wu75n5jGcATxwA2IEVvrylPwx8JsMSkqYSGBY0hy/Dnuilzgd9oZy+ox4GBaUzG0zh3USOeNtUrTVmDvqs5jOHpJI4OR7GgIQSfmpkYfrnjlPU+cQXxqw9fiVPjuYuGibhseRtePDqKN5270DrWVs/kieOxFJoifisgKabo59Lpc6gKSw7kapIL5/D4WyIBNJgb9dxyxVKLW+9uDtv2TE7rFKsW1lueJmCPI/HjJ0ZjGJhMWAobTvXoBkVKM2y/Gy+V0dlol7k6YbFQw+ky0+PnhuywqdBqjrBckVz7GuTKTelsDGHPqQn4VQWdTSFoBsXxkVhO56KtPoDnDg5jKqHhihWm4W8OgdLMhHt4cBop3cDfXrsK24+PYSLO5KaW+sYcx33jCStu09kUsq4ZN1eVEQeaRwQhBPs+d92M79ZRH0RAZX1/w5ou63dtifgtj7+rKWwVrxuPp7HMlN+WGtLw1zjEhKxGc+9fgBmnhjy8Yo6FjSFMJabR0RC0PEknKeds4B52V3MYZy+og6ZTBHxKTvkrnyie3jdgUQuikV/T3QS/SmwB5s6m0Aytfi7ccOEi3HCh3fvsaY0gltJxaHAa5y1qtAKSAItdLGwM4s6rHPcSsmFhY8gy/Dy2AzhzxYpC8Mj/fQP8KrGMK8D6iQc4m8J+TMTTWLekxTZBipM3N/Q7T4zBoJmJIORXsbt3Arfeu8VsQ6a//n975x4jV1kF8N+Zx85rZ2Z3O93tdrfdR7dKS6XQLlAtrWgRSqu2ihKCIMRHRUFEYwxYEo0Yo6gkGokBlKQaAyQ+MfHVGB8xCrQoFkoFykMplBYKpaUL7bb9/OPe7+6d2ZntzO7M3O3M+SWbmblz784597v33O873znnWz6U4/JlfVy7av4JdbL++9WLZnkJaHZUYBchsj7rUmUrStGVifPSa0cQEVbMzzF6bGLDHwmFvOqjK9wRkL0utz3rnHdbo2m4v4NsIsorI/lRULFImM50zHMl5lpjxKNh7/yMHDnGSMGcCMAtFy/OS8CD4g+0UEjoaU/w9EuHeLfvOsu5nYtXR5xw3Na4U7xu32tjYd61Rg1/g2NdPSGB1paxyV2nHEP5mcFdmRg7975GZzpGoiVMWzJa0sdfChvL352Nc8vFY2GrXqZzQfirNVyHjhxjxXxn8tVvCDZ95Cwv3r9a2B7XngOHedu8ljyX0cx0jI1rF5b1f5w8AscA9bYnaU86hqfUYj8DRXp6/p79WQMdbH50D6fPbcszXv4ef68bgWST6+zILBYJ8+rro2xxq2lGfUarJRLipvWLytJpbkeSa985xAeWjq3L1NvuRGttc1e0y7o9fssdHx4e167FsLK+ePAwXZm4F4lT7EEJTjv946l93Hb5sDeRb8/XwcNHSbmlSSIhYXFvW1HDb495bv/riIx1TKyrZ9W3/wLglVq2vH9J+YEYczqSvHzoCMuHxqKSbGav38cPTu5DoXy1Qg1/g9Pj3piZRJRQSDwff6W+RDvBaxOjurOJil09S/vbWXtaN2+bl/MmNGHMeBVe9P4Fc1a+yblx/HHn1Tb6AAMzxgywk+nsmxOp4JzZCV6bCdqVifPKyGhJQ1YMv/7nvnkmmx/dw3Bfe55h9U/uZhIRUi1h/vyYU+7YJqTZB/TSvnbmzUx5EUyVEgoJnytI8ItHw8zOJvJKbthRUjQsnLegs6wORpdvlNaVjnkrc5XqXNyw5hQ+vmKAQV9Uk3+djeH+Dv7y+IucOjtDws2WtvL56WlLsGP3AY4Z41VhLRxlRIuELJfLFy54M/tHRvOCN/pmJPnpg7s4etw4iwP5HoztZbpfp4oa/gYnFnEKr0XdC88annL9+xbP8LsG7ca1Cyq+ITLxKLdeumT8di/8NV+mrqxT16YrHfcqX4oIG1YOcu4kjdeJ6GlPEA0Lo8ecmzLlM9TlToaD/0HpnK/ubJz/vHDQW7qvLFlcw59sCXPJmXNZ3NvG/K50Xpan35DZSKAn9r7Gop6M54axoZXvO6OHy5b1lf375TLgJuAlomFikbDXucgmWsoeVfrnGKyPH0rPI2Xi0XHBE8mWCG3JKPtHRjmtN8v251/1JsKzyeKRdrPb4vxu+wscO2740NnO5Ky/PPs9G5blPfwrZZHPzWfpn5Hywmtt/SxLvaJ61PA3Af25lBf6aH38lfb4V586iwOvj3rDXv/QdaqUSniLRcIM5FIsd8MbLV9cs4BaEQ4JczuSPPniIdoSUXdC2ZmgKzQaE2F7/PZBOcv1P1fS45+VjTsPvkyccEg8I5Jyq2seO27GnbOedsfwr5w/9mC0hn/Vgs6yf7sS+nNJ/rZz7CFkXReVPCj9hr8zM2b4TxQ5Vkh3NsH+kVE6M3F+d91Krzddqsc/uy3hRTadMzTenXj2YH7oaTUYnFk4qpxc52IqlGX4RWQ18B0gDPzAGPP1gu+vBL4J2FCL7xljfuB+dwVwo7v9q8aYTVWQW6mAm9Yv8tYI9Vw9FfYsFs9pY7FvkZlqYm/GwoQ3gF98cvmEC+TUgoFcyjH87jlKtUR4Y/RIReesc5xrrLwQWD826a0rk5+TICJuHPuRooYM8OZEAG67fAkPPP2KN/lZbfpd95g1WrYHW4nbwll/OsTho8fpcktnw2Qix+Ls2H2AznTMG/EAZBP5DwCLPV+51havvIddP7pY2Gk16C9wJ/rdWdOmxy8iYeBW4F3ALmCLiNxbZO3ce4wx1xQc2wF8CRgGDPCge+wrKHVjns8P6rl66jSJVA5tyagTAtk6PumqUpdUNbA3ZtZnyPYdqizGuiud7+oZyKUIh6TiHt1lZ/fl5VFYMm62Z+EoZOncdu57cp+39jM4daSW9lW21kQl2IlpK0trfMzVUy4iwqxsnP/uG6EzHfPWg55Mjx/GJ/DZpMpS80jnDOW8yJxFPRlu/sBpvKdGIdlzOpLeiC2biOa5TKeTj/8sYKcx5ikAEbkbWAeUs2j6BcBmY8zL7rGbgdXAXZMTV5kqrbEIIrWZGJ0s6XiUn1711rxU/iCxdZSsobc92Eoelj1tzlyBXXlt7Vu6WdCd8UYA5fLpEmGWtrRzoWG8aGkvFy2tTfmPUthIqLYpuHrAcWkdOnyUaDjkPewqjRyzPfjOgryEUqPK/lyKbrfsgkVEuHh4DrUiGg4xpz3BM/tGvMWOLG0VPCynQjmGvwd41vd5F3B2kf0uEpGVwOPAZ40xz5Y4tqfIsUqdiEfD3HbZUs7wZc9OB6aTPOct6OJf/9vvra9gJ3grMWTZZJTfXLuCPnf0EApJXt3+qZJJjJ/cDIo57UlCMmZcvfNV4ahyxVDOC/mNRcLcuHaBl5VbLhct6SERDXnVPy129Fb48G6NRfjHDasq+o1q0J9LOYY/GfWt5Ty+om+tKOdXik3LFxa9/jVwlzHmsIhcBWwC3lnmsc6PiGwANgDMnVvdKpVKPuefOuvEOzUxXZk43/rgYu9zyh0lpSs0tJMtp10Og7mUV4YhaFoiIT517hBnukl2no+/wlFl4ejmYysGK5alMxPnyuUD47aXmtwNinkzW/n7zn20tkS8JT2zbsh1PSjH8O8C/OOeXuB5/w7GmH2+j3cA3/Ade27BsX8u9iPGmNuB2wGGh4fruw6ZokxAayxCOhYZV8I4SDauXegZjOnA5y8Yi+9vnYRrrNacMivNzHQsb9GcIPnE2wdZdUonoZAQQkhEw3XL2oXyDP8WYL6IDOBE7VwCXOrfQUS6jTG73Y/vBewaaL8HviYidhx/PnDDlKVWlDqycHaGA29MXJit3tSqmms1sJO79QpNLIe+GSm2bDwvaDE8OtPxvPmeVCxS10CGExp+Y8xREbkGx4iHgTuNMdtF5CvAVmPMvcC1IvJe4CjwMnCle+zLInITzsMD4Ct2oldRThaufscQV79jKGgxThr6ZzjVNf0L1CgTk45H6laZE0CKrVEZNMPDw2br1q1Bi6EoilIXfvmv55jR2pKXf1EpIvKgMWa4nH01c1dRFCVg1p9R32DH6esoVBRFUWqCGn5FUZQmQw2/oihKk6GGX1EUpclQw68oitJkqOFXFEVpMtTwK4qiNBlq+BVFUZqMaZm5KyIvAv+d5OE54KUqinMy0Iw6g+rdTDSjzlCZ3n3GmLJSf6el4Z8KIrK13LTlRqEZdQbVO2g56kkz6gy101tdPYqiKE2GGn5FUZQmoxEN/+1BCxAAzagzqN7NRDPqDDXSu+F8/IqiKMrENGKPX1EURZmAhjH8IrJaRB4TkZ0icn3Q8tQSEXlGRB4WkYdEZKu7rUNENovIE+5r+4n+z3RHRO4Ukb0i8ohvW1E9xeG7bvtvE5ElwUk+eUro/GURec5t74dEZI3vuxtcnR8TkQuCkXrqiMgcEfmTiOwQke0i8hl3e8O29wQ61769jTEn/R/OkpBPAoNAC/BvYGHQctVQ32eAXMG2m4Hr3ffXA98IWs4q6LkSWAI8ciI9gTXAbwEBlgH3By1/FXX+MvD5IvsudK/1GDDg3gPhoHWYpN7dwBL3fRp43NWvYdt7Ap1r3t6N0uM/C9hpjHnKGHMEuBtYF7BM9WYdsMl9vwlYH6AsVcEY81ecNZz9lNJzHfAj43Af0CYi3fWRtHqU0LkU64C7jTGHbiH0GQAAAg1JREFUjTFPAztx7oWTDmPMbmPMP933B4EdQA8N3N4T6FyKqrV3oxj+HuBZ3+ddTHwCT3YM8AcReVBENrjbuowxu8G5oIDOwKSrLaX0bPRr4BrXpXGnz43XkDqLSD9wBnA/TdLeBTpDjdu7UQy/FNnWyOFKy40xS4ALgatFZGXQAk0DGvka+D4wDzgd2A18293ecDqLSCvwM+A6Y8yBiXYtsu2k1L2IzjVv70Yx/LuAOb7PvcDzAclSc4wxz7uve4Ff4Az39tihrvu6NzgJa0opPRv2GjDG7DHGHDPGHAfuYGx431A6i0gUxwD+xBjzc3dzQ7d3MZ3r0d6NYvi3APNFZEBEWoBLgHsDlqkmiEhKRNL2PXA+8AiOvle4u10B/CoYCWtOKT3vBT7sRnssA161LoKTnQLf9ftw2hscnS8RkZiIDADzgQfqLV81EBEBfgjsMMbc4vuqYdu7lM51ae+gZ7arOEO+BmdW/ElgY9Dy1FDPQZyZ/X8D262uwAzgj8AT7mtH0LJWQde7cIa6ozi9nY+W0hNnGHyr2/4PA8NBy19FnX/s6rTNvfm7fftvdHV+DLgwaPmnoPc5OG6LbcBD7t+aRm7vCXSueXtr5q6iKEqT0SiuHkVRFKVM1PAriqI0GWr4FUVRmgw1/IqiKE2GGn5FUZQmQw2/oihKk6GGX1EUpclQw68oitJk/B+peW6OrtENtQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(treerr)\n",
    "plt.plot(everr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "out=np.array(out)\n",
    "dfout=pd.DataFrame(data={\"label\":out[:,0],\"err\":out[:,1]})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "false_pos_rate, true_pos_rate, thresholds = roc_curve(dfout.label, dfout.err)\n",
    "roc_auc = auc(false_pos_rate, true_pos_rate,)\n",
    "\n",
    "plt.plot(false_pos_rate, true_pos_rate, linewidth=5, label='AUC = %0.3f'% roc_auc)\n",
    "plt.plot([0,1],[0,1], linewidth=5)\n",
    "\n",
    "plt.xlim([-0.01, 1])\n",
    "plt.ylim([0, 1.01])\n",
    "plt.legend(loc='lower right')\n",
    "plt.title('Receiver operating characteristic curve (ROC)')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "precision_rt, recall_rt, threshold_rt = precision_recall_curve(dfout.label, dfout.err)\n",
    "plt.plot(recall_rt, precision_rt, linewidth=5, label='Precision-Recall curve')\n",
    "plt.title('Recall vs Precision')\n",
    "plt.xlabel('Recall')\n",
    "plt.ylabel('Precision')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(threshold_rt, precision_rt[1:], label=\"Precision\",linewidth=5)\n",
    "plt.plot(threshold_rt, recall_rt[1:], label=\"Recall\",linewidth=5)\n",
    "plt.title('Precision and recall for different threshold values')\n",
    "plt.xlabel('Threshold')\n",
    "plt.ylabel('Precision/Recall')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "correct inlier 0.998%, num pred outlier 18,  true positive 5, actual num outliers 52.0\n",
      "prec 0.2777777777777778, recall 0.09615384615384616\n"
     ]
    }
   ],
   "source": [
    "###Selecting threshold\n",
    "thresh=55\n",
    "print(\"correct inlier {:5.3f}%, num pred outlier {},  true positive {}, actual num outliers {}\".format(\n",
    "                             dfout[(dfout['err']<thresh)&dfout['label']!=1].shape[0]/dfout.shape[0],\n",
    "                             dfout[dfout['err']>=thresh].shape[0],\n",
    "                             dfout[(dfout['err']>=thresh)&dfout['label']==1].shape[0],\n",
    "                             dfout['label'].sum()))\n",
    "recall = dfout[(dfout['err']>=thresh)&dfout['label']==1].shape[0]/dfout['label'].sum()\n",
    "prec = dfout[(dfout['err']>=thresh)&dfout['label']==1].shape[0]/dfout[dfout['err']>=thresh].shape[0]\n",
    "print(\"prec {}, recall {}\".format(prec,recall))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Defining Prediction from loss\n",
    "dfout['pred']=0\n",
    "dfout.loc[dfout['err']>thresh,'pred']=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "autoencoder   prec 0.278, recall 0.096, false pos rate 0.722\n"
     ]
    }
   ],
   "source": [
    "prec2,recall2,fpr2=prec_recall(dfout)\n",
    "print(\"autoencoder   prec {:5.3f}, recall {:5.3f}, false pos rate {:5.3f}\".format(prec2,recall2,fpr2))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}