Logistic Regression using TensorFlow.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "import requests\n",
    "from sklearn import datasets\n",
    "from sklearn.preprocessing import normalize\n",
    "from tensorflow.python.framework import ops\n",
    "\n",
    "ops.reset_default_graph()\n",
    "sess = tf.Session()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "bc = datasets.load_breast_cancer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(569, 30)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bc.data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  1.79900000e+01,   1.03800000e+01,   1.22800000e+02, ...,\n",
       "          2.65400000e-01,   4.60100000e-01,   1.18900000e-01],\n",
       "       [  2.05700000e+01,   1.77700000e+01,   1.32900000e+02, ...,\n",
       "          1.86000000e-01,   2.75000000e-01,   8.90200000e-02],\n",
       "       [  1.96900000e+01,   2.12500000e+01,   1.30000000e+02, ...,\n",
       "          2.43000000e-01,   3.61300000e-01,   8.75800000e-02],\n",
       "       ..., \n",
       "       [  1.66000000e+01,   2.80800000e+01,   1.08300000e+02, ...,\n",
       "          1.41800000e-01,   2.21800000e-01,   7.82000000e-02],\n",
       "       [  2.06000000e+01,   2.93300000e+01,   1.40100000e+02, ...,\n",
       "          2.65000000e-01,   4.08700000e-01,   1.24000000e-01],\n",
       "       [  7.76000000e+00,   2.45400000e+01,   4.79200000e+01, ...,\n",
       "          0.00000000e+00,   2.87100000e-01,   7.03900000e-02]])"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bc.data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0,\n",
       "       1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 0, 1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1,\n",
       "       1, 0, 1, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 0, 0, 1, 1, 1, 0,\n",
       "       1, 1, 0, 0, 1, 1, 1, 0, 0, 1, 1, 1, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 0, 0, 0, 1, 0, 0, 1, 1, 1, 0, 0, 1, 0, 1, 0, 0, 1, 0, 0, 1, 1,\n",
       "       0, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1,\n",
       "       0, 0, 1, 0, 1, 1, 0, 0, 1, 1, 0, 0, 1, 1, 1, 1, 0, 1, 1, 0, 0, 0, 1,\n",
       "       0, 1, 0, 1, 1, 1, 0, 1, 1, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 1,\n",
       "       0, 1, 1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0,\n",
       "       0, 1, 1, 0, 1, 1, 0, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 0,\n",
       "       0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1,\n",
       "       1, 0, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1,\n",
       "       1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0,\n",
       "       1, 0, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 0, 1, 0, 1, 0, 1, 1, 1, 0, 1,\n",
       "       1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1,\n",
       "       0, 0, 0, 1, 0, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1,\n",
       "       1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1,\n",
       "       0, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 0, 1, 0, 1,\n",
       "       1, 1, 1, 1, 0, 1, 1, 0, 1, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1,\n",
       "       1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, 1, 1,\n",
       "       1, 1, 1, 0, 1, 1, 0, 1, 0, 1, 0, 0, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
       "       1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bc.target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['mean radius', 'mean texture', 'mean perimeter', 'mean area',\n",
       "       'mean smoothness', 'mean compactness', 'mean concavity',\n",
       "       'mean concave points', 'mean symmetry', 'mean fractal dimension',\n",
       "       'radius error', 'texture error', 'perimeter error', 'area error',\n",
       "       'smoothness error', 'compactness error', 'concavity error',\n",
       "       'concave points error', 'symmetry error', 'fractal dimension error',\n",
       "       'worst radius', 'worst texture', 'worst perimeter', 'worst area',\n",
       "       'worst smoothness', 'worst compactness', 'worst concavity',\n",
       "       'worst concave points', 'worst symmetry', 'worst fractal dimension'],\n",
       "      dtype='<U23')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bc.feature_names"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split data into train & test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_vals = bc.data\n",
    "y_vals = bc.target\n",
    "\n",
    "train_indices = np.random.choice(len(x_vals), round(len(x_vals)*0.8), replace=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_indices = np.array(list(set(range(len(x_vals))) - set(train_indices)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([512,   1, 514,   5, 518,  10,  13,  14,  16,  18, 532,  21, 535,\n",
       "       536, 542, 543,  33, 548,  37, 550,  39, 552,  42, 556,  46,  47,\n",
       "       558,  49, 564, 568,  61,  62,  74,  80,  81,  90, 104, 105, 106,\n",
       "       117, 120, 123, 124, 127, 128, 137, 145, 146, 152, 159, 161, 164,\n",
       "       168, 175, 176, 179, 183, 185, 192, 198, 200, 202, 208, 220, 223,\n",
       "       227, 233, 234, 242, 243, 248, 251, 263, 268, 269, 270, 284, 294,\n",
       "       304, 313, 317, 318, 321, 322, 325, 332, 344, 356, 368, 369, 380,\n",
       "       381, 384, 385, 390, 414, 420, 429, 432, 438, 446, 447, 450, 451,\n",
       "       459, 460, 463, 480, 481, 483, 485, 487, 493, 506])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([118, 283, 494, 310, 457, 395, 421, 160, 153, 305, 327, 326, 504,\n",
       "       508, 489, 131, 537, 113, 195,   6, 245, 116, 400, 502, 455, 525,\n",
       "       265, 528, 436, 224, 366,   8, 517, 469, 214,  53, 408, 258, 343,\n",
       "       374, 132, 289, 527, 491, 516, 412, 563,  67, 250, 405,  87, 348,\n",
       "       398, 392, 474, 484, 462, 279, 173, 540, 236, 276,  11, 456, 156,\n",
       "       331, 166, 136,  23, 407, 342, 387, 498, 360,  82, 178, 281, 413,\n",
       "       187,  88, 174, 323, 239, 298, 154,  32, 328, 199,  69, 129, 273,\n",
       "       373, 240, 181, 352, 490, 221,   9, 100, 354, 388, 371, 461, 402,\n",
       "       557,  38, 253, 102, 266, 246, 510,  44, 189,  66, 138, 365, 167,\n",
       "       519, 190, 479,  52,   4, 286,  77, 544, 437, 222, 158, 529, 430,\n",
       "       415, 111,  57,  64, 204,  34, 193, 309, 303,  60, 280, 108,  98,\n",
       "       411, 316, 140,  40, 391, 162, 377, 182, 333, 148, 428, 565, 230,\n",
       "       353, 290, 561, 336, 497, 539, 513, 526,  17, 291,  26, 134,  94,\n",
       "        54,   0,  86, 141, 119, 492, 389,  97,  93, 341, 231, 541, 495,\n",
       "       482,  75, 302, 216, 133, 329, 549, 439,  59, 433, 477, 382, 300,\n",
       "        24, 299, 545, 435,  35, 314, 417,   7, 285, 337, 401, 509, 566,\n",
       "        41, 567, 358, 260, 264, 209, 319, 191, 228,  91,  70, 507,  58,\n",
       "       555, 257, 522,  30,  63, 422, 277,  50, 165, 464, 247, 262, 441,\n",
       "       101,  84, 534,  89, 501, 367, 560, 383, 423, 143, 177, 255, 307,\n",
       "        73, 533, 170, 125, 531, 465, 296, 301, 427, 530,  45, 350, 226,\n",
       "       500, 271, 546,  99, 442,  72, 295, 130, 109, 274, 444, 338, 237,\n",
       "       434, 562,  51, 359, 194,  78, 476, 340, 363, 320, 225, 551, 275,\n",
       "        19, 188, 364,  79, 155, 150, 346,  95, 410, 397, 349, 520,  22,\n",
       "       122, 426, 425, 547, 180, 372, 205,  29, 449, 440,  43, 394,  55,\n",
       "       157, 468,   2, 345, 112, 169, 386, 399, 393, 315, 213, 357,  31,\n",
       "        85, 252, 334, 503,  83, 297,  65, 197, 376, 370, 293, 229, 475,\n",
       "       211, 443, 362,  56, 361, 184, 256, 244, 403, 287, 142, 471, 144,\n",
       "       355, 306, 259,  27, 339, 505, 215, 453, 419,  36, 523,  76,  68,\n",
       "        48, 172, 278,   3, 375, 488, 186, 478, 379, 312, 538, 335, 282,\n",
       "       206, 472,  96, 454, 272, 121, 416, 218, 496, 473,  20, 207, 232,\n",
       "       424, 470, 486, 515, 311, 466, 217, 110, 396, 499, 521, 235, 559,\n",
       "        12, 212,  25, 139, 351, 554, 524, 238, 467, 249, 135, 203, 308,\n",
       "       103, 406, 219, 210, 458,  15, 347,  71, 292,  92, 431, 151, 254,\n",
       "       409, 196, 553, 288, 324, 147, 418, 114,  28, 511, 452, 261, 126,\n",
       "       163, 201, 445, 149, 267, 404, 107, 241, 115, 330, 171, 378, 448])"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "x_vals_train = x_vals[train_indices]\n",
    "x_vals_test = x_vals[test_indices]\n",
    "y_vals_train = y_vals[train_indices]\n",
    "y_vals_test = y_vals[test_indices]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "###  Creating the placeholders for input data & target data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "batch_size = 25\n",
    "\n",
    "x_data = tf.placeholder(shape=[None,30], dtype=tf.float32)\n",
    "y_target = tf.placeholder(shape=[None,1], dtype=tf.float32)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model parameters should be variables as they keep on changing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "A = tf.Variable(tf.random_normal(shape=[30,1], dtype=tf.float32))\n",
    "b = tf.Variable(tf.random_normal(shape=[1,1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Model Creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model_output = tf.add(tf.matmul(x_data, A), b)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Loss Calculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss = tf.reduce_mean( tf.nn.sigmoid_cross_entropy_with_logits(logits=model_output, labels=y_target))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "init = tf.global_variables_initializer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sess.run(init)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "my_opt = tf.train.GradientDescentOptimizer(0.01)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "train_step = my_opt.minimize(loss)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Create prediction, find accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "prediction = tf.round(tf.sigmoid(model_output))\n",
    "\n",
    "prediction_correct = tf.cast(tf.equal(prediction, y_target), tf.float32)\n",
    "\n",
    "accuracy = tf.reduce_mean(prediction_correct)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "loss_vec = []\n",
    "train_acc = []\n",
    "test_acc = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "for i in range(1000):\n",
    "    rand_index = np.random.choice(len(x_vals_train), size=batch_size)\n",
    "    rand_x = x_vals_train[rand_index]\n",
    "    rand_y = np.transpose([y_vals_train[rand_index]])\n",
    "    \n",
    "    sess.run(train_step, feed_dict={x_data: rand_x, y_target: rand_y})\n",
    "    \n",
    "    temp_loss = sess.run(loss, feed_dict={x_data: rand_x, y_target: rand_y})\n",
    "    \n",
    "    loss_vec.append(temp_loss)\n",
    "    \n",
    "    temp_acc_train = sess.run(accuracy, feed_dict={x_data: x_vals_train, y_target: np.transpose([y_vals_train])})\n",
    "    \n",
    "    train_acc.append(temp_acc_train)\n",
    "    \n",
    "    temp_acc_test = sess.run(accuracy, feed_dict={x_data: x_vals_test, y_target: np.transpose([y_vals_test])})\n",
    "    \n",
    "    test_acc.append(temp_acc_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.3508772,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.84210527,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.4649123,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.67543858,\n",
       " 0.3508772,\n",
       " 0.89473683,\n",
       " 0.87719297,\n",
       " 0.37719297,\n",
       " 0.64912283,\n",
       " 0.75438595,\n",
       " 0.3508772,\n",
       " 0.88596493,\n",
       " 0.57894737,\n",
       " 0.65789473,\n",
       " 0.3508772,\n",
       " 0.3508772,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.85964912,\n",
       " 0.3508772,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.88596493,\n",
       " 0.89473683,\n",
       " 0.81578946,\n",
       " 0.80701756,\n",
       " 0.3508772,\n",
       " 0.86842108,\n",
       " 0.87719297,\n",
       " 0.78070176,\n",
       " 0.3508772,\n",
       " 0.86842108,\n",
       " 0.3508772,\n",
       " 0.66666669,\n",
       " 0.3508772,\n",
       " 0.67543858,\n",
       " 0.3508772,\n",
       " 0.3859649,\n",
       " 0.64912283,\n",
       " 0.70175439,\n",
       " 0.65789473,\n",
       " 0.3508772,\n",
       " 0.92982459,\n",
       " 0.51754385,\n",
       " 0.66666669,\n",
       " 0.3508772,\n",
       " 0.86842108,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.3859649,\n",
       " 0.64912283,\n",
       " 0.3508772,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.4649123,\n",
       " 0.67543858,\n",
       " 0.3508772,\n",
       " 0.3508772,\n",
       " 0.75438595,\n",
       " 0.71052629,\n",
       " 0.7631579,\n",
       " 0.83333331,\n",
       " 0.89473683,\n",
       " 0.3508772,\n",
       " 0.85087717,\n",
       " 0.40350878,\n",
       " 0.66666669,\n",
       " 0.74561405,\n",
       " 0.7631579,\n",
       " 0.3508772,\n",
       " 0.78947371,\n",
       " 0.3508772,\n",
       " 0.87719297,\n",
       " 0.44736841,\n",
       " 0.74561405,\n",
       " 0.35964912,\n",
       " 0.67543858,\n",
       " 0.3508772,\n",
       " 0.75438595,\n",
       " 0.3508772,\n",
       " 0.74561405,\n",
       " 0.45614034,\n",
       " 0.70175439,\n",
       " 0.3508772,\n",
       " 0.92105263,\n",
       " 0.87719297,\n",
       " 0.52631581,\n",
       " 0.7368421,\n",
       " 0.60526317,\n",
       " 0.70175439,\n",
       " 0.7631579,\n",
       " 0.86842108,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.75438595,\n",
       " 0.85087717,\n",
       " 0.7631579,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.63157892,\n",
       " 0.71052629,\n",
       " 0.3508772,\n",
       " 0.89473683,\n",
       " 0.88596493,\n",
       " 0.75438595,\n",
       " 0.80701756,\n",
       " 0.3508772,\n",
       " 0.69298244,\n",
       " 0.3508772,\n",
       " 0.50877196,\n",
       " 0.89473683,\n",
       " 0.42105263,\n",
       " 0.68421054,\n",
       " 0.92982459,\n",
       " 0.67543858,\n",
       " 0.90350878,\n",
       " 0.92105263,\n",
       " 0.92982459,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.59649122,\n",
       " 0.67543858,\n",
       " 0.3859649,\n",
       " 0.80701756,\n",
       " 0.84210527,\n",
       " 0.89473683,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.42105263,\n",
       " 0.68421054,\n",
       " 0.86842108,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.67543858,\n",
       " 0.80701756,\n",
       " 0.44736841,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.57894737,\n",
       " 0.83333331,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.80701756,\n",
       " 0.89473683,\n",
       " 0.85964912,\n",
       " 0.81578946,\n",
       " 0.45614034,\n",
       " 0.7368421,\n",
       " 0.3508772,\n",
       " 0.89473683,\n",
       " 0.65789473,\n",
       " 0.81578946,\n",
       " 0.89473683,\n",
       " 0.49122807,\n",
       " 0.87719297,\n",
       " 0.7631579,\n",
       " 0.70175439,\n",
       " 0.3508772,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.61403507,\n",
       " 0.74561405,\n",
       " 0.7280702,\n",
       " 0.88596493,\n",
       " 0.85964912,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.92105263,\n",
       " 0.92982459,\n",
       " 0.70175439,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.43859649,\n",
       " 0.75438595,\n",
       " 0.40350878,\n",
       " 0.78947371,\n",
       " 0.59649122,\n",
       " 0.81578946,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.7631579,\n",
       " 0.85964912,\n",
       " 0.89473683,\n",
       " 0.85087717,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.83333331,\n",
       " 0.88596493,\n",
       " 0.78070176,\n",
       " 0.79824561,\n",
       " 0.4649123,\n",
       " 0.68421054,\n",
       " 0.3508772,\n",
       " 0.90350878,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.92105263,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.90350878,\n",
       " 0.89473683,\n",
       " 0.85964912,\n",
       " 0.89473683,\n",
       " 0.71929824,\n",
       " 0.88596493,\n",
       " 0.49122807,\n",
       " 0.75438595,\n",
       " 0.3508772,\n",
       " 0.88596493,\n",
       " 0.92982459,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.83333331,\n",
       " 0.92982459,\n",
       " 0.88596493,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.57894737,\n",
       " 0.78070176,\n",
       " 0.89473683,\n",
       " 0.7719298,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.4649123,\n",
       " 0.68421054,\n",
       " 0.37719297,\n",
       " 0.88596493,\n",
       " 0.57017541,\n",
       " 0.68421054,\n",
       " 0.85087717,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.85964912,\n",
       " 0.88596493,\n",
       " 0.85087717,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.57017541,\n",
       " 0.71929824,\n",
       " 0.3508772,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.80701756,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.92105263,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.7368421,\n",
       " 0.78070176,\n",
       " 0.85087717,\n",
       " 0.86842108,\n",
       " 0.90350878,\n",
       " 0.92105263,\n",
       " 0.88596493,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.89473683,\n",
       " 0.79824561,\n",
       " 0.89473683,\n",
       " 0.84210527,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.90350878,\n",
       " 0.84210527,\n",
       " 0.91228068,\n",
       " 0.86842108,\n",
       " 0.86842108,\n",
       " 0.86842108,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.85087717,\n",
       " 0.90350878,\n",
       " 0.81578946,\n",
       " 0.92982459,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.82456142,\n",
       " 0.78947371,\n",
       " 0.67543858,\n",
       " 0.81578946,\n",
       " 0.91228068,\n",
       " 0.80701756,\n",
       " 0.88596493,\n",
       " 0.7368421,\n",
       " 0.86842108,\n",
       " 0.49122807,\n",
       " 0.68421054,\n",
       " 0.87719297,\n",
       " 0.86842108,\n",
       " 0.84210527,\n",
       " 0.80701756,\n",
       " 0.82456142,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.61403507,\n",
       " 0.85964912,\n",
       " 0.36842105,\n",
       " 0.88596493,\n",
       " 0.74561405,\n",
       " 0.91228068,\n",
       " 0.5,\n",
       " 0.81578946,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.79824561,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.91228068,\n",
       " 0.7368421,\n",
       " 0.89473683,\n",
       " 0.40350878,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.78070176,\n",
       " 0.68421054,\n",
       " 0.57894737,\n",
       " 0.71929824,\n",
       " 0.41228071,\n",
       " 0.88596493,\n",
       " 0.41228071,\n",
       " 0.81578946,\n",
       " 0.88596493,\n",
       " 0.3508772,\n",
       " 0.88596493,\n",
       " 0.75438595,\n",
       " 0.89473683,\n",
       " 0.5,\n",
       " 0.7368421,\n",
       " 0.84210527,\n",
       " 0.7631579,\n",
       " 0.92105263,\n",
       " 0.92982459,\n",
       " 0.88596493,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.92105263,\n",
       " 0.56140351,\n",
       " 0.85964912,\n",
       " 0.92105263,\n",
       " 0.82456142,\n",
       " 0.80701756,\n",
       " 0.91228068,\n",
       " 0.84210527,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.7280702,\n",
       " 0.89473683,\n",
       " 0.7719298,\n",
       " 0.80701756,\n",
       " 0.57894737,\n",
       " 0.78947371,\n",
       " 0.37719297,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.52631581,\n",
       " 0.84210527,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.88596493,\n",
       " 0.86842108,\n",
       " 0.91228068,\n",
       " 0.80701756,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.87719297,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.79824561,\n",
       " 0.84210527,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.84210527,\n",
       " 0.88596493,\n",
       " 0.92105263,\n",
       " 0.82456142,\n",
       " 0.88596493,\n",
       " 0.92982459,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.81578946,\n",
       " 0.84210527,\n",
       " 0.92105263,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.80701756,\n",
       " 0.90350878,\n",
       " 0.89473683,\n",
       " 0.92105263,\n",
       " 0.85087717,\n",
       " 0.92105263,\n",
       " 0.58771932,\n",
       " 0.88596493,\n",
       " 0.47368422,\n",
       " 0.68421054,\n",
       " 0.78947371,\n",
       " 0.88596493,\n",
       " 0.7368421,\n",
       " 0.92982459,\n",
       " 0.88596493,\n",
       " 0.92982459,\n",
       " 0.88596493,\n",
       " 0.7368421,\n",
       " 0.89473683,\n",
       " 0.88596493,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.86842108,\n",
       " 0.84210527,\n",
       " 0.92982459,\n",
       " 0.84210527,\n",
       " 0.81578946,\n",
       " 0.89473683,\n",
       " 0.5,\n",
       " 0.84210527,\n",
       " 0.88596493,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.7368421,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.88596493,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.82456142,\n",
       " 0.80701756,\n",
       " 0.79824561,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.86842108,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.81578946,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.47368422,\n",
       " 0.83333331,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.82456142,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.85964912,\n",
       " 0.92105263,\n",
       " 0.7280702,\n",
       " 0.7368421,\n",
       " 0.47368422,\n",
       " 0.88596493,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.86842108,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.81578946,\n",
       " 0.84210527,\n",
       " 0.92105263,\n",
       " 0.88596493,\n",
       " 0.84210527,\n",
       " 0.89473683,\n",
       " 0.7719298,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.85087717,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.7368421,\n",
       " 0.88596493,\n",
       " 0.81578946,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.82456142,\n",
       " 0.88596493,\n",
       " 0.86842108,\n",
       " 0.81578946,\n",
       " 0.80701756,\n",
       " 0.78947371,\n",
       " 0.79824561,\n",
       " 0.81578946,\n",
       " 0.92105263,\n",
       " 0.90350878,\n",
       " 0.81578946,\n",
       " 0.83333331,\n",
       " 0.54385966,\n",
       " 0.70175439,\n",
       " 0.60526317,\n",
       " 0.71929824,\n",
       " 0.86842108,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.81578946,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.79824561,\n",
       " 0.88596493,\n",
       " 0.64035088,\n",
       " 0.88596493,\n",
       " 0.63157892,\n",
       " 0.85964912,\n",
       " 0.85964912,\n",
       " 0.7368421,\n",
       " 0.7631579,\n",
       " 0.57017541,\n",
       " 0.88596493,\n",
       " 0.79824561,\n",
       " 0.80701756,\n",
       " 0.89473683,\n",
       " 0.57017541,\n",
       " 0.83333331,\n",
       " 0.81578946,\n",
       " 0.90350878,\n",
       " 0.85087717,\n",
       " 0.92982459,\n",
       " 0.92105263,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.88596493,\n",
       " 0.81578946,\n",
       " 0.82456142,\n",
       " 0.81578946,\n",
       " 0.92982459,\n",
       " 0.87719297,\n",
       " 0.85964912,\n",
       " 0.91228068,\n",
       " 0.88596493,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.81578946,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.81578946,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.85087717,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.81578946,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.85087717,\n",
       " 0.90350878,\n",
       " 0.88596493,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.88596493,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.81578946,\n",
       " 0.78947371,\n",
       " 0.63157892,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.87719297,\n",
       " 0.88596493,\n",
       " 0.85964912,\n",
       " 0.85964912,\n",
       " 0.86842108,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.7280702,\n",
       " 0.88596493,\n",
       " 0.57017541,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.92105263,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.82456142,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.89473683,\n",
       " 0.92105263,\n",
       " 0.90350878,\n",
       " 0.88596493,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.85964912,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.85087717,\n",
       " 0.81578946,\n",
       " 0.88596493,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.85087717,\n",
       " 0.86842108,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.85087717,\n",
       " 0.89473683,\n",
       " 0.84210527,\n",
       " 0.92105263,\n",
       " 0.7719298,\n",
       " 0.88596493,\n",
       " 0.5,\n",
       " 0.7368421,\n",
       " 0.7280702,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.93859649,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.85964912,\n",
       " 0.92105263,\n",
       " 0.92982459,\n",
       " 0.81578946,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.58771932,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.82456142,\n",
       " 0.88596493,\n",
       " 0.92982459,\n",
       " 0.87719297,\n",
       " 0.87719297,\n",
       " 0.93859649,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.93859649,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.83333331,\n",
       " 0.90350878,\n",
       " 0.92982459,\n",
       " 0.75438595,\n",
       " 0.93859649,\n",
       " 0.87719297,\n",
       " 0.91228068,\n",
       " 0.63157892,\n",
       " 0.83333331,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.85964912,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.85087717,\n",
       " 0.85087717,\n",
       " 0.90350878,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.85087717,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.89473683,\n",
       " 0.5,\n",
       " 0.89473683,\n",
       " 0.92105263,\n",
       " 0.81578946,\n",
       " 0.85964912,\n",
       " 0.54385966,\n",
       " 0.80701756,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.87719297,\n",
       " 0.91228068,\n",
       " 0.93859649,\n",
       " 0.82456142,\n",
       " 0.88596493,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.85087717,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.87719297,\n",
       " 0.93859649,\n",
       " 0.87719297,\n",
       " 0.91228068,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.79824561,\n",
       " 0.89473683,\n",
       " 0.85964912,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.89473683,\n",
       " 0.7719298,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.82456142,\n",
       " 0.92982459,\n",
       " 0.92105263,\n",
       " 0.87719297,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.85087717,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.86842108,\n",
       " 0.89473683,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.88596493,\n",
       " 0.83333331,\n",
       " 0.86842108,\n",
       " 0.93859649,\n",
       " 0.93859649,\n",
       " 0.79824561,\n",
       " 0.89473683,\n",
       " 0.85964912,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.93859649,\n",
       " 0.82456142,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.85087717,\n",
       " 0.83333331,\n",
       " 0.88596493,\n",
       " 0.89473683,\n",
       " 0.87719297,\n",
       " 0.87719297,\n",
       " 0.90350878,\n",
       " 0.84210527,\n",
       " 0.93859649,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.87719297,\n",
       " 0.93859649,\n",
       " 0.91228068,\n",
       " 0.93859649,\n",
       " 0.91228068,\n",
       " 0.84210527,\n",
       " 0.89473683,\n",
       " 0.87719297,\n",
       " 0.93859649,\n",
       " 0.89473683,\n",
       " 0.89473683,\n",
       " 0.86842108,\n",
       " 0.57894737,\n",
       " 0.80701756,\n",
       " 0.92982459,\n",
       " 0.92105263,\n",
       " 0.91228068,\n",
       " 0.82456142,\n",
       " 0.74561405,\n",
       " 0.42982456,\n",
       " 0.80701756,\n",
       " 0.65789473,\n",
       " 0.85964912,\n",
       " 0.85964912,\n",
       " 0.86842108,\n",
       " 0.62280703,\n",
       " 0.79824561,\n",
       " 0.57894737,\n",
       " 0.80701756,\n",
       " 0.93859649,\n",
       " 0.89473683,\n",
       " 0.92105263,\n",
       " 0.83333331,\n",
       " 0.78070176,\n",
       " 0.57017541,\n",
       " 0.89473683,\n",
       " 0.92982459,\n",
       " 0.90350878,\n",
       " 0.93859649,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.7631579,\n",
       " 0.78070176,\n",
       " 0.89473683,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.82456142,\n",
       " 0.91228068,\n",
       " 0.79824561,\n",
       " 0.87719297,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.90350878,\n",
       " 0.91228068,\n",
       " 0.93859649,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.92982459,\n",
       " 0.83333331,\n",
       " 0.86842108,\n",
       " 0.4649123,\n",
       " 0.93859649,\n",
       " 0.88596493,\n",
       " 0.90350878,\n",
       " 0.90350878,\n",
       " 0.87719297,\n",
       " 0.87719297,\n",
       " 0.85964912,\n",
       " 0.91228068,\n",
       " 0.92105263,\n",
       " 0.92982459,\n",
       " 0.93859649,\n",
       " 0.89473683,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.88596493,\n",
       " 0.88596493,\n",
       " 0.78947371,\n",
       " 0.92982459,\n",
       " 0.91228068,\n",
       " 0.91228068,\n",
       " 0.87719297,\n",
       " 0.83333331,\n",
       " 0.93859649,\n",
       " 0.91228068,\n",
       " 0.93859649,\n",
       " 0.93859649]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0.37802199,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.80879122,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62417585,\n",
       " 0.37802199,\n",
       " 0.63296705,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.4967033,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62197804,\n",
       " 0.37802199,\n",
       " 0.62637365,\n",
       " 0.37802199,\n",
       " 0.66153848,\n",
       " 0.37802199,\n",
       " 0.90329671,\n",
       " 0.85274726,\n",
       " 0.40000001,\n",
       " 0.62197804,\n",
       " 0.74945056,\n",
       " 0.37802199,\n",
       " 0.91648352,\n",
       " 0.58461541,\n",
       " 0.63296705,\n",
       " 0.37802199,\n",
       " 0.37802199,\n",
       " 0.62637365,\n",
       " 0.37802199,\n",
       " 0.81758243,\n",
       " 0.37802199,\n",
       " 0.87912089,\n",
       " 0.87252748,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.7846154,\n",
       " 0.7846154,\n",
       " 0.37802199,\n",
       " 0.83736265,\n",
       " 0.84835166,\n",
       " 0.77802199,\n",
       " 0.37802199,\n",
       " 0.84175825,\n",
       " 0.37802199,\n",
       " 0.64395607,\n",
       " 0.37802199,\n",
       " 0.66593409,\n",
       " 0.37802199,\n",
       " 0.41538462,\n",
       " 0.62417585,\n",
       " 0.66813189,\n",
       " 0.62857145,\n",
       " 0.37802199,\n",
       " 0.8989011,\n",
       " 0.53186816,\n",
       " 0.64395607,\n",
       " 0.37802199,\n",
       " 0.82637364,\n",
       " 0.91428572,\n",
       " 0.90769231,\n",
       " 0.42637363,\n",
       " 0.62637365,\n",
       " 0.38241759,\n",
       " 0.85494506,\n",
       " 0.90329671,\n",
       " 0.49230769,\n",
       " 0.66593409,\n",
       " 0.37802199,\n",
       " 0.38461539,\n",
       " 0.75384617,\n",
       " 0.69230771,\n",
       " 0.77142859,\n",
       " 0.79340661,\n",
       " 0.90989012,\n",
       " 0.37802199,\n",
       " 0.80219781,\n",
       " 0.42637363,\n",
       " 0.64395607,\n",
       " 0.70769233,\n",
       " 0.76263738,\n",
       " 0.37802199,\n",
       " 0.77802199,\n",
       " 0.37802199,\n",
       " 0.85494506,\n",
       " 0.46813187,\n",
       " 0.74945056,\n",
       " 0.38681319,\n",
       " 0.65494508,\n",
       " 0.37802199,\n",
       " 0.75384617,\n",
       " 0.38241759,\n",
       " 0.74945056,\n",
       " 0.48131868,\n",
       " 0.70329672,\n",
       " 0.37802199,\n",
       " 0.90329671,\n",
       " 0.84615386,\n",
       " 0.55384618,\n",
       " 0.73846155,\n",
       " 0.62197804,\n",
       " 0.70109892,\n",
       " 0.76703298,\n",
       " 0.81978023,\n",
       " 0.84175825,\n",
       " 0.8989011,\n",
       " 0.73406595,\n",
       " 0.81318682,\n",
       " 0.76923078,\n",
       " 0.90769231,\n",
       " 0.87692308,\n",
       " 0.62857145,\n",
       " 0.71868134,\n",
       " 0.37802199,\n",
       " 0.88571429,\n",
       " 0.90769231,\n",
       " 0.73626375,\n",
       " 0.78021979,\n",
       " 0.37802199,\n",
       " 0.69890112,\n",
       " 0.37802199,\n",
       " 0.52747256,\n",
       " 0.90109891,\n",
       " 0.43736264,\n",
       " 0.67032969,\n",
       " 0.90549451,\n",
       " 0.65494508,\n",
       " 0.87472528,\n",
       " 0.90769231,\n",
       " 0.90109891,\n",
       " 0.90989012,\n",
       " 0.85494506,\n",
       " 0.91208792,\n",
       " 0.60439563,\n",
       " 0.66153848,\n",
       " 0.41098902,\n",
       " 0.7824176,\n",
       " 0.80659342,\n",
       " 0.90769231,\n",
       " 0.87252748,\n",
       " 0.90989012,\n",
       " 0.88351649,\n",
       " 0.91428572,\n",
       " 0.88571429,\n",
       " 0.43076923,\n",
       " 0.68791211,\n",
       " 0.83956045,\n",
       " 0.88571429,\n",
       " 0.91208792,\n",
       " 0.65934068,\n",
       " 0.7868132,\n",
       " 0.46373627,\n",
       " 0.85934067,\n",
       " 0.90989012,\n",
       " 0.59120882,\n",
       " 0.80439562,\n",
       " 0.8967033,\n",
       " 0.90549451,\n",
       " 0.91208792,\n",
       " 0.7824176,\n",
       " 0.86153847,\n",
       " 0.83296704,\n",
       " 0.789011,\n",
       " 0.47252747,\n",
       " 0.72967035,\n",
       " 0.37802199,\n",
       " 0.88571429,\n",
       " 0.63736266,\n",
       " 0.79340661,\n",
       " 0.91208792,\n",
       " 0.52527475,\n",
       " 0.84175825,\n",
       " 0.75164837,\n",
       " 0.70109892,\n",
       " 0.37802199,\n",
       " 0.90329671,\n",
       " 0.90329671,\n",
       " 0.61538464,\n",
       " 0.74505496,\n",
       " 0.69010991,\n",
       " 0.84835166,\n",
       " 0.83956045,\n",
       " 0.86593407,\n",
       " 0.84615386,\n",
       " 0.88571429,\n",
       " 0.90989012,\n",
       " 0.66153848,\n",
       " 0.86593407,\n",
       " 0.86593407,\n",
       " 0.46153846,\n",
       " 0.75164837,\n",
       " 0.42417583,\n",
       " 0.77362639,\n",
       " 0.60219783,\n",
       " 0.80000001,\n",
       " 0.8945055,\n",
       " 0.90769231,\n",
       " 0.87912089,\n",
       " 0.73846155,\n",
       " 0.82197803,\n",
       " 0.90769231,\n",
       " 0.81318682,\n",
       " 0.87472528,\n",
       " 0.90769231,\n",
       " 0.79780221,\n",
       " 0.85054946,\n",
       " 0.75824177,\n",
       " 0.78021979,\n",
       " 0.49010989,\n",
       " 0.66813189,\n",
       " 0.37802199,\n",
       " 0.87692308,\n",
       " 0.87252748,\n",
       " 0.91648352,\n",
       " 0.88571429,\n",
       " 0.90549451,\n",
       " 0.90549451,\n",
       " 0.84395605,\n",
       " 0.88571429,\n",
       " 0.91208792,\n",
       " 0.83736265,\n",
       " 0.90989012,\n",
       " 0.6791209,\n",
       " 0.85054946,\n",
       " 0.50989014,\n",
       " 0.74945056,\n",
       " 0.37802199,\n",
       " 0.85054946,\n",
       " 0.8945055,\n",
       " 0.87692308,\n",
       " 0.90989012,\n",
       " 0.90549451,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.91428572,\n",
       " 0.80439562,\n",
       " 0.91208792,\n",
       " 0.87032968,\n",
       " 0.87472528,\n",
       " 0.90329671,\n",
       " 0.90329671,\n",
       " 0.59340662,\n",
       " 0.76923078,\n",
       " 0.90989012,\n",
       " 0.75384617,\n",
       " 0.90109891,\n",
       " 0.91208792,\n",
       " 0.90769231,\n",
       " 0.48791209,\n",
       " 0.69670331,\n",
       " 0.40000001,\n",
       " 0.85274726,\n",
       " 0.58681321,\n",
       " 0.67252749,\n",
       " 0.81538463,\n",
       " 0.90769231,\n",
       " 0.88571429,\n",
       " 0.91428572,\n",
       " 0.84835166,\n",
       " 0.83956045,\n",
       " 0.81758243,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.90989012,\n",
       " 0.88791209,\n",
       " 0.5736264,\n",
       " 0.72527474,\n",
       " 0.37802199,\n",
       " 0.91428572,\n",
       " 0.88791209,\n",
       " 0.91648352,\n",
       " 0.91648352,\n",
       " 0.91648352,\n",
       " 0.77362639,\n",
       " 0.88791209,\n",
       " 0.90329671,\n",
       " 0.88351649,\n",
       " 0.8989011,\n",
       " 0.90989012,\n",
       " 0.88131869,\n",
       " 0.90989012,\n",
       " 0.90549451,\n",
       " 0.90549451,\n",
       " 0.90989012,\n",
       " 0.6857143,\n",
       " 0.76923078,\n",
       " 0.82417583,\n",
       " 0.86153847,\n",
       " 0.90989012,\n",
       " 0.88791209,\n",
       " 0.88571429,\n",
       " 0.87032968,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.88351649,\n",
       " 0.90769231,\n",
       " 0.88351649,\n",
       " 0.90769231,\n",
       " 0.76483518,\n",
       " 0.8967033,\n",
       " 0.80879122,\n",
       " 0.87252748,\n",
       " 0.85494506,\n",
       " 0.90549451,\n",
       " 0.79780221,\n",
       " 0.91648352,\n",
       " 0.86373627,\n",
       " 0.86153847,\n",
       " 0.86153847,\n",
       " 0.91208792,\n",
       " 0.91428572,\n",
       " 0.86373627,\n",
       " 0.85494506,\n",
       " 0.87692308,\n",
       " 0.82417583,\n",
       " 0.90549451,\n",
       " 0.7846154,\n",
       " 0.8945055,\n",
       " 0.85714287,\n",
       " 0.86813188,\n",
       " 0.79340661,\n",
       " 0.76923078,\n",
       " 0.63516486,\n",
       " 0.80659342,\n",
       " 0.91428572,\n",
       " 0.76923078,\n",
       " 0.84395605,\n",
       " 0.6813187,\n",
       " 0.83516484,\n",
       " 0.52307695,\n",
       " 0.69230771,\n",
       " 0.86153847,\n",
       " 0.83296704,\n",
       " 0.81758243,\n",
       " 0.79340661,\n",
       " 0.79340661,\n",
       " 0.87912089,\n",
       " 0.90109891,\n",
       " 0.61098903,\n",
       " 0.83076924,\n",
       " 0.3912088,\n",
       " 0.85274726,\n",
       " 0.70109892,\n",
       " 0.91208792,\n",
       " 0.54065937,\n",
       " 0.80659342,\n",
       " 0.88351649,\n",
       " 0.88351649,\n",
       " 0.76263738,\n",
       " 0.87032968,\n",
       " 0.86153847,\n",
       " 0.90769231,\n",
       " 0.69670331,\n",
       " 0.87252748,\n",
       " 0.41538462,\n",
       " 0.90769231,\n",
       " 0.86373627,\n",
       " 0.88791209,\n",
       " 0.74285716,\n",
       " 0.69450551,\n",
       " 0.59780222,\n",
       " 0.72527474,\n",
       " 0.43076923,\n",
       " 0.85274726,\n",
       " 0.43076923,\n",
       " 0.80439562,\n",
       " 0.83736265,\n",
       " 0.389011,\n",
       " 0.83956045,\n",
       " 0.71208793,\n",
       " 0.90549451,\n",
       " 0.53846157,\n",
       " 0.73406595,\n",
       " 0.81318682,\n",
       " 0.71648353,\n",
       " 0.90989012,\n",
       " 0.8923077,\n",
       " 0.85714287,\n",
       " 0.91208792,\n",
       " 0.91208792,\n",
       " 0.91648352,\n",
       " 0.88571429,\n",
       " 0.90549451,\n",
       " 0.58021981,\n",
       " 0.83296704,\n",
       " 0.88351649,\n",
       " 0.79560441,\n",
       " 0.79560441,\n",
       " 0.90769231,\n",
       " 0.81098902,\n",
       " 0.90989012,\n",
       " 0.91428572,\n",
       " 0.90769231,\n",
       " 0.91428572,\n",
       " 0.68791211,\n",
       " 0.86813188,\n",
       " 0.71868134,\n",
       " 0.7824176,\n",
       " 0.59780222,\n",
       " 0.77362639,\n",
       " 0.40879121,\n",
       " 0.91868132,\n",
       " 0.91868132,\n",
       " 0.91868132,\n",
       " 0.91868132,\n",
       " 0.55604398,\n",
       " 0.81758243,\n",
       " 0.91868132,\n",
       " 0.91868132,\n",
       " 0.87252748,\n",
       " 0.85274726,\n",
       " 0.90769231,\n",
       " 0.76703298,\n",
       " 0.90549451,\n",
       " 0.91648352,\n",
       " 0.86593407,\n",
       " 0.91868132,\n",
       " 0.91648352,\n",
       " 0.91648352,\n",
       " 0.91428572,\n",
       " 0.76263738,\n",
       " 0.81758243,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.88131869,\n",
       " 0.90549451,\n",
       " 0.81318682,\n",
       " 0.85934067,\n",
       " 0.90769231,\n",
       " 0.7912088,\n",
       " 0.84395605,\n",
       " 0.91208792,\n",
       " 0.88791209,\n",
       " 0.91428572,\n",
       " 0.87252748,\n",
       " 0.87252748,\n",
       " 0.7824176,\n",
       " 0.81758243,\n",
       " 0.90769231,\n",
       " 0.86813188,\n",
       " 0.86373627,\n",
       " 0.86373627,\n",
       " 0.75164837,\n",
       " 0.90769231,\n",
       " 0.88571429,\n",
       " 0.90549451,\n",
       " 0.81538463,\n",
       " 0.90989012,\n",
       " 0.59560442,\n",
       " 0.84395605,\n",
       " 0.51208794,\n",
       " 0.6835165,\n",
       " 0.74285716,\n",
       " 0.83956045,\n",
       " 0.6769231,\n",
       " 0.90989012,\n",
       " 0.87912089,\n",
       " 0.90989012,\n",
       " 0.85934067,\n",
       " 0.68791211,\n",
       " 0.90549451,\n",
       " 0.87472528,\n",
       " 0.90549451,\n",
       " 0.91648352,\n",
       " 0.86153847,\n",
       " 0.81978023,\n",
       " 0.91428572,\n",
       " 0.83296704,\n",
       " 0.7868132,\n",
       " 0.88351649,\n",
       " 0.52747256,\n",
       " 0.82857144,\n",
       " 0.85274726,\n",
       " 0.91648352,\n",
       " 0.91428572,\n",
       " 0.68791211,\n",
       " 0.84835166,\n",
       " 0.90769231,\n",
       " 0.87472528,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.80219781,\n",
       " 0.79560441,\n",
       " 0.75164837,\n",
       " 0.86373627,\n",
       " 0.8901099,\n",
       " 0.86153847,\n",
       " 0.91868132,\n",
       " 0.91868132,\n",
       " 0.76043957,\n",
       " 0.87472528,\n",
       " 0.86813188,\n",
       " 0.87692308,\n",
       " 0.90989012,\n",
       " 0.91868132,\n",
       " 0.91868132,\n",
       " 0.51208794,\n",
       " 0.81538463,\n",
       " 0.86373627,\n",
       " 0.90329671,\n",
       " 0.90769231,\n",
       " 0.80219781,\n",
       " 0.84395605,\n",
       " 0.87912089,\n",
       " 0.8901099,\n",
       " 0.91428572,\n",
       " 0.83956045,\n",
       " 0.90769231,\n",
       " 0.66593409,\n",
       " 0.72967035,\n",
       " 0.51208794,\n",
       " 0.83736265,\n",
       " 0.8989011,\n",
       " 0.90989012,\n",
       " 0.84835166,\n",
       " 0.90989012,\n",
       " 0.90549451,\n",
       " 0.77362639,\n",
       " 0.81758243,\n",
       " 0.90769231,\n",
       " 0.87912089,\n",
       " 0.83076924,\n",
       " 0.86813188,\n",
       " 0.72087914,\n",
       " 0.90769231,\n",
       " 0.91428572,\n",
       " 0.8967033,\n",
       " 0.8967033,\n",
       " 0.8901099,\n",
       " 0.8967033,\n",
       " 0.82857144,\n",
       " 0.88131869,\n",
       " 0.88131869,\n",
       " 0.88131869,\n",
       " 0.6791209,\n",
       " 0.85274726,\n",
       " 0.77142859,\n",
       " 0.91428572,\n",
       " 0.90549451,\n",
       " 0.8967033,\n",
       " 0.90549451,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.92087913,\n",
       " 0.91648352,\n",
       " 0.88351649,\n",
       " 0.80659342,\n",
       " 0.85934067,\n",
       " 0.86813188,\n",
       " 0.76923078,\n",
       " 0.7912088,\n",
       " 0.74065936,\n",
       " 0.7824176,\n",
       " 0.77362639,\n",
       " 0.90329671,\n",
       " 0.91208792,\n",
       " 0.76923078,\n",
       " 0.81538463,\n",
       " 0.56703299,\n",
       " 0.69450551,\n",
       " 0.60000002,\n",
       " 0.72087914,\n",
       " 0.85934067,\n",
       " 0.91868132,\n",
       " 0.91868132,\n",
       " 0.78021979,\n",
       " 0.91428572,\n",
       " 0.90329671,\n",
       " 0.74945056,\n",
       " 0.85274726,\n",
       " 0.62857145,\n",
       " 0.83956045,\n",
       " 0.62197804,\n",
       " 0.83296704,\n",
       " 0.83296704,\n",
       " 0.6835165,\n",
       " 0.74505496,\n",
       " 0.58901101,\n",
       " 0.84835166,\n",
       " 0.74285716,\n",
       " 0.7912088,\n",
       " 0.86813188,\n",
       " 0.58241761,\n",
       " 0.81538463,\n",
       " 0.77582419,\n",
       " 0.91208792,\n",
       " 0.82857144,\n",
       " 0.91428572,\n",
       " 0.90549451,\n",
       " 0.90769231,\n",
       " 0.90989012,\n",
       " 0.8967033,\n",
       " 0.86813188,\n",
       " 0.77142859,\n",
       " 0.81538463,\n",
       " 0.76263738,\n",
       " 0.90769231,\n",
       " 0.87252748,\n",
       " 0.83516484,\n",
       " 0.91428572,\n",
       " 0.87032968,\n",
       " 0.90769231,\n",
       " 0.90109891,\n",
       " 0.91648352,\n",
       " 0.789011,\n",
       " 0.91648352,\n",
       " 0.91648352,\n",
       " 0.91428572,\n",
       " 0.90989012,\n",
       " 0.78021979,\n",
       " 0.88351649,\n",
       " 0.8989011,\n",
       " 0.90769231,\n",
       " 0.90989012,\n",
       " 0.91648352,\n",
       " 0.90769231,\n",
       " 0.81758243,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.75604397,\n",
       " 0.90769231,\n",
       " 0.91648352,\n",
       " 0.90989012,\n",
       " 0.80879122,\n",
       " 0.90769231,\n",
       " 0.87692308,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.88791209,\n",
       " 0.91868132,\n",
       " 0.91428572,\n",
       " 0.88351649,\n",
       " 0.91648352,\n",
       " 0.8901099,\n",
       " 0.91648352,\n",
       " 0.91648352,\n",
       " 0.87032968,\n",
       " 0.90769231,\n",
       " 0.91428572,\n",
       " 0.7846154,\n",
       " 0.77362639,\n",
       " 0.61098903,\n",
       " 0.8989011,\n",
       " 0.91648352,\n",
       " 0.90329671,\n",
       " 0.90769231,\n",
       " 0.8967033,\n",
       " 0.88351649,\n",
       " 0.88351649,\n",
       " 0.85714287,\n",
       " 0.90769231,\n",
       " 0.8989011,\n",
       " 0.91428572,\n",
       " 0.86373627,\n",
       " 0.90329671,\n",
       " 0.84175825,\n",
       " 0.83076924,\n",
       " 0.86153847,\n",
       " 0.91428572,\n",
       " 0.8989011,\n",
       " 0.8923077,\n",
       " 0.91648352,\n",
       " 0.6769231,\n",
       " 0.86153847,\n",
       " 0.5758242,\n",
       " 0.91208792,\n",
       " 0.91648352,\n",
       " 0.91868132,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.87692308,\n",
       " 0.8901099,\n",
       " 0.76923078,\n",
       " 0.87032968,\n",
       " 0.88571429,\n",
       " 0.8923077,\n",
       " 0.90109891,\n",
       " 0.90769231,\n",
       " 0.87472528,\n",
       " 0.91428572,\n",
       " 0.91648352,\n",
       " 0.91648352,\n",
       " 0.91648352,\n",
       " 0.90109891,\n",
       " 0.84175825,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.81758243,\n",
       " 0.7868132,\n",
       " 0.85494506,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.84175825,\n",
       " 0.85054946,\n",
       " 0.85054946,\n",
       " 0.90549451,\n",
       " 0.81318682,\n",
       " 0.90769231,\n",
       " 0.81098902,\n",
       " 0.90329671,\n",
       " 0.71868134,\n",
       " 0.85274726,\n",
       " 0.52307695,\n",
       " 0.73186815,\n",
       " 0.6769231,\n",
       " 0.88791209,\n",
       " 0.88791209,\n",
       " 0.88791209,\n",
       " 0.90549451,\n",
       " 0.90549451,\n",
       " 0.90989012,\n",
       " 0.91648352,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.87692308,\n",
       " 0.91208792,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.82857144,\n",
       " 0.91648352,\n",
       " 0.90989012,\n",
       " 0.79780221,\n",
       " 0.90989012,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.90549451,\n",
       " 0.91648352,\n",
       " 0.90109891,\n",
       " 0.59780222,\n",
       " 0.91428572,\n",
       " 0.91208792,\n",
       " 0.91648352,\n",
       " 0.7868132,\n",
       " 0.86373627,\n",
       " 0.91428572,\n",
       " 0.86813188,\n",
       " 0.87032968,\n",
       " 0.90989012,\n",
       " 0.8967033,\n",
       " 0.91428572,\n",
       " 0.8901099,\n",
       " 0.90109891,\n",
       " 0.90989012,\n",
       " 0.90109891,\n",
       " 0.91208792,\n",
       " 0.80659342,\n",
       " 0.90109891,\n",
       " 0.91208792,\n",
       " 0.69230771,\n",
       " 0.91428572,\n",
       " 0.86593407,\n",
       " 0.91428572,\n",
       " 0.61978024,\n",
       " 0.81758243,\n",
       " 0.87692308,\n",
       " 0.90329671,\n",
       " 0.84395605,\n",
       " 0.90109891,\n",
       " 0.90109891,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.91208792,\n",
       " 0.91428572,\n",
       " 0.91648352,\n",
       " 0.91428572,\n",
       " 0.91208792,\n",
       " 0.91428572,\n",
       " 0.84395605,\n",
       " 0.82637364,\n",
       " 0.8967033,\n",
       " 0.91208792,\n",
       " 0.91428572,\n",
       " 0.84395605,\n",
       " 0.87912089,\n",
       " 0.91208792,\n",
       " 0.84395605,\n",
       " 0.91208792,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.91208792,\n",
       " 0.88131869,\n",
       " 0.53186816,\n",
       " 0.87692308,\n",
       " 0.90329671,\n",
       " 0.74505496,\n",
       " 0.84175825,\n",
       " 0.54945058,\n",
       " 0.80439562,\n",
       " 0.91648352,\n",
       " 0.91648352,\n",
       " 0.90549451,\n",
       " 0.88791209,\n",
       " 0.87692308,\n",
       " 0.91208792,\n",
       " 0.86593407,\n",
       " 0.91648352,\n",
       " 0.91208792,\n",
       " 0.77582419,\n",
       " 0.86153847,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.91208792,\n",
       " 0.83516484,\n",
       " 0.90549451,\n",
       " 0.8967033,\n",
       " 0.8967033,\n",
       " 0.85714287,\n",
       " 0.91208792,\n",
       " 0.87252748,\n",
       " 0.91428572,\n",
       " 0.87692308,\n",
       " 0.90769231,\n",
       " 0.90549451,\n",
       " 0.91208792,\n",
       " 0.92087913,\n",
       " 0.73626375,\n",
       " 0.87912089,\n",
       " 0.84615386,\n",
       " 0.8945055,\n",
       " 0.8945055,\n",
       " 0.90109891,\n",
       " 0.86593407,\n",
       " 0.69670331,\n",
       " 0.87692308,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.91428572,\n",
       " 0.91648352,\n",
       " 0.80659342,\n",
       " 0.91208792,\n",
       " 0.90329671,\n",
       " 0.85714287,\n",
       " 0.90989012,\n",
       " 0.91208792,\n",
       " 0.83736265,\n",
       " 0.90329671,\n",
       " 0.8967033,\n",
       " 0.85274726,\n",
       " 0.8923077,\n",
       " 0.86373627,\n",
       " 0.87692308,\n",
       " 0.90769231,\n",
       " 0.91648352,\n",
       " 0.90549451,\n",
       " 0.91428572,\n",
       " 0.87912089,\n",
       " 0.77142859,\n",
       " 0.85494506,\n",
       " 0.91428572,\n",
       " 0.91648352,\n",
       " 0.73626375,\n",
       " 0.88571429,\n",
       " 0.84615386,\n",
       " 0.86593407,\n",
       " 0.8967033,\n",
       " 0.91208792,\n",
       " 0.77582419,\n",
       " 0.90769231,\n",
       " 0.91428572,\n",
       " 0.83296704,\n",
       " 0.82197803,\n",
       " 0.87912089,\n",
       " 0.90549451,\n",
       " 0.85274726,\n",
       " 0.85274726,\n",
       " 0.8989011,\n",
       " 0.82197803,\n",
       " 0.91208792,\n",
       " 0.88571429,\n",
       " 0.91428572,\n",
       " 0.85934067,\n",
       " 0.91428572,\n",
       " 0.92087913,\n",
       " 0.91428572,\n",
       " 0.92087913,\n",
       " 0.81318682,\n",
       " 0.90549451,\n",
       " 0.85054946,\n",
       " 0.91428572,\n",
       " 0.8901099,\n",
       " 0.8901099,\n",
       " 0.85054946,\n",
       " 0.59780222,\n",
       " 0.789011,\n",
       " 0.91208792,\n",
       " 0.90329671,\n",
       " 0.91428572,\n",
       " 0.74285716,\n",
       " 0.72747254,\n",
       " 0.45274726,\n",
       " 0.79560441,\n",
       " 0.64615387,\n",
       " 0.84175825,\n",
       " 0.84615386,\n",
       " 0.85494506,\n",
       " 0.61758244,\n",
       " 0.77582419,\n",
       " 0.60000002,\n",
       " 0.80000001,\n",
       " 0.91648352,\n",
       " 0.88791209,\n",
       " 0.91208792,\n",
       " 0.80439562,\n",
       " 0.76923078,\n",
       " 0.58461541,\n",
       " 0.88351649,\n",
       " 0.91428572,\n",
       " 0.8945055,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.91208792,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.90769231,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.69450551,\n",
       " 0.76483518,\n",
       " 0.88791209,\n",
       " 0.8989011,\n",
       " 0.91648352,\n",
       " 0.91208792,\n",
       " 0.74505496,\n",
       " 0.91648352,\n",
       " 0.73186815,\n",
       " 0.86593407,\n",
       " 0.90989012,\n",
       " 0.91868132,\n",
       " 0.8945055,\n",
       " 0.90989012,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.90329671,\n",
       " 0.91208792,\n",
       " 0.789011,\n",
       " 0.85274726,\n",
       " 0.4989011,\n",
       " 0.91208792,\n",
       " 0.87692308,\n",
       " 0.8989011,\n",
       " 0.8989011,\n",
       " 0.86373627,\n",
       " 0.86373627,\n",
       " 0.84615386,\n",
       " 0.90549451,\n",
       " 0.91428572,\n",
       " 0.91208792,\n",
       " 0.91428572,\n",
       " 0.88131869,\n",
       " 0.90989012,\n",
       " 0.90989012,\n",
       " 0.87692308,\n",
       " 0.87692308,\n",
       " 0.72967035,\n",
       " 0.90989012,\n",
       " 0.90769231,\n",
       " 0.90769231,\n",
       " 0.85714287,\n",
       " 0.80439562,\n",
       " 0.91208792,\n",
       " 0.91428572,\n",
       " 0.91428572,\n",
       " 0.91428572]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEWCAYAAACe8xtsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJztvXmcFMX9//98yyVR8IiIAiIeqCBBVES8EhOjEDWiCUSM\nxuOHike8AA8SVFT8wkfjEcSogETiASpeKKKCMcYLcD0QQQigICwIi4LLcix7vH9/dPXQO9sz07M7\ns7M7vJ+PRz+mu7qq+l3dPfXqd1V1tagqhmEYhpEOO+XaAMMwDKPhYeJhGIZhpI2Jh2EYhpE2Jh6G\nYRhG2ph4GIZhGGlj4mEYhmGkjYmHYRh5g4g8KiK35tqOHQETjzxFRP4oIgUiUiIiq0VkuoicmEN7\nnhCRbc4ef5kbMe1wEXkq2zZGRUSWicivc21HNhCRjiIyWUSKRKRYRBaLyEMi0i7XtsUjIheLyPvB\nMFW9QlXvypVNOxImHnmIiAwCHgT+H9AaaA88DJyVIH7jOjLtHlXdNbAckYlMxcPu5TQIu+YicjAw\nG1gFHKmqLYETgKVAnT541OE9adQUVbUljxZgN6AE6JckznBgCvAUUAxcCjTDE5xVbnkQaObi7wW8\nBmwAfgDeA3Zy+24GCoGNwCLglATHfAIYkWBfB0CBi4BvgXXAX92+3sA2oMyVa64L/w9wN/ABsAU4\nGGgDTHU2LgEuCynzs87WT4Ej3L4bgRfibBoN/D2BvcuAXyfYd5k79g/OljYuXIAHgLXunM8Durh9\npwMLnF2FwJAEeV/syjsG+BFYGDzf7to/Dqx2+YwAGsWlfQD4PuxauPvh1Qj32JnA5+5++BDoGndu\nhgBfOBufBXZOI+3NLm0p0Bi4BU+8NrpzdI6L2wnYClS4+2JD2H2W6Hq4fQpcASx29jwMSK7/ww1l\nybkBtmT4gnqVbTnQOEmc4XiV8dl43mdz4E5gFrA30Mr9se9y8UcCjwJN3HKSqwwPBVYEKsgOwEEJ\njlnlTx23r4P7I49zthzhKo9OAXufikvzHzyhOdxVMk2A/wL/AHYGugFFwK/iytzXxR0CfOPW9wU2\nAbu7uI3xKvmjE9i7jBDxAH6FJ3xH4YnxQ8B/3b5ewCfA7u7cdQL2dftWAye59T2AoxIc92J3bW9w\ndp+LV0Hv6fa/BDwG7OKu4xxgYFzaa1z5mofk/x1wcYr760h3bo4FGuEJ/jK2P2gsc8dtA+wJfAVc\nkUbaz4H9fPuAfi6vnVx5NwXO28XA+4nus2TXw+1XvIei3fG88yKgd67/ww1lMVc///gpsE5Vy1PE\n+0hVX1bVSlXdApwP3Kmqa1W1CLgD+JOLW4ZXwe6vqmWq+p56/74KvD9lZxFpoqrLVHVpkmMOEZEN\ngWVi3P47VHWLqs4F5uKJSDKeUNX5rqz74DWx3KyqW1X1c2A8cGEg/ieqOkVVy4D78USmp6quxhOe\nfi5eb7xz+EmK48dzPjBBVT9V1VJgKHCciHTAO4ctgMPwnm6/csfF7essIi1Vdb2qfprkGGuBB911\neBbP2ztDRFrjeTDXq+omVV2L52X0D6RdpaoPqWq5u+bx7IUnIACIyJ/ddSoRkXEu+HLgMVWdraoV\nqjoRT+h7BvIZraqrVPUH4FU8IU8n7QrfPlV93uVV6cq7GOiR5PwESXY9fEap6gZV/RZ4J2CrkQIT\nj/zje2CvCG3GK+K22wDLA9vLXRjAvXiu/1si8rWI3AKgqkuA6/Ge6te6jtY2JOZvqrp7YLkobv93\ngfXNwK5plKEN8IOqbowrQ9uw+KpaCawMlHEicIFbvwB4MsWxw6hyDlW1BO96tFXVf+M1Nz2Md67G\nikhLF/X3eBX/chF5V0SOS3KMQifcPv512h/PG1ntizOeF7J3IG78NY/ne7yHBN/+Maq6O14TZhMX\nvD8wOPgQgOcpBK97ousYJW0VG0XkQhH5PBC/C57IRSHh9Yhgq5ECE4/84yO8p7mzU8SLn055Fd6f\n26e9C0NVN6rqYFU9EK/TfZCInOL2PaOqJ7q0Cvxf7YuQ0taw8FXAniLSIhDWHq/t32c/f8V1sLdz\n6QBeBrqKSBe8dvmna2BnlXMoIrvgeYKFAKo6WlWPBjoDh+D1taCqH6tqH7yK/mXguSTHaCsiElfG\nVXiVbimwV0CcW6rq4YG4qabQfhv4XYo4K4C74x4CfqKqk1Kki5o2ZqOI7I/XlPln4KdOyL7Ea/aL\nUp6k18OoHSYeeYaq/gjcBjwsImeLyE9EpImI/EZE7kmSdBIwTERaicheLo+nAETkTBE52FVaP+I1\nV1WKyKEi8isRaYbXebkFqMxCsdYAHZKNqFLVFXj9NCNFZGcR6QoM8MvgOFpEfue8suvxKttZLv1W\nvA71Z4A5rhkjGU3ccfylMd45vEREurlz8v+A2aq6TESOEZFjRaQJXrv9Vrxz2FREzheR3VxzWjHJ\nz+HewLXumvbD6zt53TWBvQXcJyItRWQnETlIRH6RohxBhgMnicj9ItIWwN0LnQJxxgFXuLKIiOwi\nImfEiXYi0k27C55AFDlbLsHzPHzWAO1EpGmC9AmvRwRbjRSYeOQhqnofMAgYhvfHW4H39PZykmQj\ngAK8kS7z8EYjjXD7OgIz8Ua1fAT8Q1XfwevvGIXXKfkdXsU2NMkxbpKq73msi1ik593v9yKSrD/g\nPLzO91V4nce3q+rMwP5X8Dpd1+P15/zOVdg+E4GfEa3J6nU8sfSX4e5YtwIv4HWCH8T2PoeWeJXn\nerymlO/xmgNxtiwTkWK80T/nJznubLzrsQ5vtFlfVf3e7bsQaIo3Kmk9nhjuG5ZJGKr6P7zO7HbA\nXBHZiDdCa5UrF6pagDeCaYw7xhK8juso+aeVVlUXAPfh3XNr8K7NB4Eo/wbmA9+F3UsprodRS6Rq\n86lh5CciMhw4WFUvSBKnPd7w131UtbiubIuKiFwMXOqaCQ0jp5jnYRjE+kAGAZPro3AYRn3D3uI0\ndnhcR+oavOak3jk2xzAaBNZsZRiGYaSNNVsZhmEYaZO3zVZ77bWXdujQIddmGIZhNCg++eSTdara\nKlW8vBWPDh06UFBQkGszDMMwGhQisjx1LGu2MgzDMGqAiYdhGIaRNiYehmEYRtqYeBiGYRhpY+Jh\nGIZhpI2Jh2EYhpE2Jh6GYRhG2ph4ZJh58+bxwQcfpI5oGIbRgMnblwRzRdeuXQGwOcMMw8hnzPMw\nDMMw0sbEwzAMw0ibrImH+67zHBGZKyLzReQOFz5cRApF5HO3nB5IM1RElojIIhHpFQg/WkTmuX2j\n3be0DcMwjByRzT6PUuBXqloiIk2A90Vkutv3gKr+LRhZRDrjfV/4cKANMFNEDlHVCuARvG8fz8b7\ndnRvYDqGYRhGTsia56EeJW6ziVuS9SL3wfsEaKmqfgMsAXqIyL5AS1WdpV4v9L+As7Nlt2EYhpGa\nrPZ5iEgjEfkcWAvMUNXZbtc1IvKFiEwQkT1cWFtgRSD5ShfW1q3HhxuGYRg5IqvioaoVqtoNaIfn\nRXTBa4I6EOgGrAbuy9TxRORyESkQkYKioqJMZWsYhmHEUSejrVR1A/AO0FtV1zhRqQTGAT1ctEJg\nv0Cydi6s0K3Hh4cdZ6yqdlfV7q1apfwQlmEYhlFDsjnaqpWI7O7WmwOnAgtdH4bPOcCXbn0q0F9E\nmonIAUBHYI6qrgaKRaSnG2V1IfBKtuw2DMMwUpPN0Vb7AhNFpBGeSD2nqq+JyJMi0g2v83wZMBBA\nVeeLyHPAAqAcuNqNtAK4CngCaI43yspGWhmGYeQQyddpNLp37665+Ia5/wpKvp5XwzDyGxH5RFW7\np4pnb5gbhmEYaWPiYRiGYaSNiYdhGIaRNiYehmEYRtqYeBiGYRhpY+JhGIZhpI2Jh2EYhpE2Jh6G\nYRhG2ph4GIZhGGlj4mEYhmGkjYmHYRiGkTYmHoZhGEbamHgYhmEYaWPiYRiGYaSNiYdhGIaRNiYe\nhmEYRtqYeBiGYRhpY+JhGIZhpI2Jh2EYhpE2WRMPEdlZROaIyFwRmS8id7jwPUVkhogsdr97BNIM\nFZElIrJIRHoFwo8WkXlu32jxPxRuGIZh5IRseh6lwK9U9QigG9BbRHoCtwBvq2pH4G23jYh0BvoD\nhwO9gX+ISCOX1yPAZUBHt/TOot2GYRhGCrImHupR4jabuEWBPsBEFz4RONut9wEmq2qpqn4DLAF6\niMi+QEtVnaWqCvwrkMYwDMPIAVnt8xCRRiLyObAWmKGqs4HWqrraRfkOaO3W2wIrAslXurC2bj0+\nPOx4l4tIgYgUFBUV1dju8vJypk2bhqdVhmEYRjxZFQ9VrVDVbkA7PC+iS9x+xfNGMnW8saraXVW7\nt2rVqsb5jBw5kjPPPJPp06dnyjTDMIy8ok5GW6nqBuAdvL6KNa4pCve71kUrBPYLJGvnwgrdenx4\n1vj6668BWLNmTTYPYxiG0WDJ5mirViKyu1tvDpwKLASmAhe5aBcBr7j1qUB/EWkmIgfgdYzPcU1c\nxSLS042yujCQJuNcd911PPHEE9nK3jAMIy9onMW89wUmuhFTOwHPqeprIvIR8JyIDACWA38AUNX5\nIvIcsAAoB65W1QqX11XAE0BzYLpbssLMmTOzlbVhGEbekDXxUNUvgCNDwr8HTkmQ5m7g7pDwAqBL\n9RSZJ/gKiXWYG4ZhhGNvmBuGYRhpY+IRR9DzsBfZDcMwwjHxiMMEwzAMIzUmHoZhGEbamHjEYZ6H\nYRhGakw84rDRVoZhGKkx8TAMwzDSxsQjDmu2MgzDSI2JRxw2VNcwDCM1Jh6GYRhG2ph4xGHehmEY\nRmpMPOKw0VaGYRipMfEwDMMw0sbEIw5rtjIMw0hNSvEQkXtEpKWINBGRt0WkSEQuqAvjcoGJh2EY\nRmqieB6nqWoxcCawDDgYuDGbRtUXTEgMwzDCiSIe/gejzgCeV9Ufs2hPzjHBMAzDSE2ULwm+JiIL\ngS3AlSLSCtiaXbNyh422MgzDSE1Kz0NVbwGOB7qrahmwCeiTKp2I7Cci74jIAhGZLyLXufDhIlIo\nIp+75fRAmqEiskREFolIr0D40SIyz+0bLeYeGIZh5JQoHeb9gDJVrRCRYcBTQJsIeZcDg1W1M9AT\nuFpEOrt9D6hqN7e87o7TGegPHA70Bv4hIo1c/EeAy4CObukduYRpYrpkGIaRmih9Hreq6kYRORH4\nNfA4XmWeFFVdraqfuvWNwFdA2yRJ+gCTVbVUVb8BlgA9RGRfoKWqzlKvHelfwNkR7K4RJh6GYRip\niSIeFe73DGCsqk4DmqZzEBHpABwJzHZB14jIFyIyQUT2cGFtgRWBZCtdWFu3Hh8edpzLRaRARAqK\niorSMdEwDMNIgyjiUSgijwHnAq+LSLOI6QAQkV2BF4Dr3ZDfR4ADgW7AauC+tK1OgKqOVdXuqtq9\nVatWNcrDPA/DMIzURBGBPwBvAr1UdQOwJxHf8xCRJnjC8bSqvgigqmtUtUJVK4FxQA8XvRDYL5C8\nnQsrdOvx4VnBpmQ3DMNITZTRVpuBpUAvEfkzsLeqvpUqnRsR9TjwlareHwjfNxDtHOBLtz4V6C8i\nzUTkALyO8TmquhooFpGeLs8LgVeiFa922FBdwzCMcFK+5+GG2F4GvOiCnhKRsar6UIqkJwB/AuaJ\nyOcu7C/AeSLSDVC8N9YHAqjqfBF5DliAN1LralX1+1uuAp4AmgPT3ZIVzNswDMNITZSXBAcAx6rq\nJgAR+T/gIyCpeKjq+0BYTfx6kjR3A3eHhBcAXSLYWmtMPAzDMFITpc9D2D7iCreetzWsiYdhGEZq\nonge/wRmi8hLbvtsYEL2TDIMwzDqOynFQ1XvF5H/ACe6oEtU9bOsWpVDzPMwDMNITRTPA/em+Kf+\ntoh8q6rts2ZVDokiHgsXLqRdu3bsuuuudWCRYRhG/aOmXxLcoR/PO3XqxFlnnZVrMwzDMHJGTcUj\nb1+ASOV5+O9+vPPOO3VhjmEYRr0kYbOViAxKtAvI2/aaoHgMGDCAfv360aJFi1iYvThoGIaR3PNo\nkWDZFfh79k2rHyxcuLDKdlA8BgwYYGJiGMYOSULPQ1XvqEtD6gtRm60AJkyYwAMPPEDLli2zbZZh\nGEa9oqZ9HnlLvHjEb8d7GuZ5GIaxI2LikSa1EY9WrVoxevToTJtkGIZR50T5DG2jVHHyiXSarcK2\nk7Fu3Tquu+66GtllGIZRn4jieSwWkXsD3x/Pa7IpHoZhGPlCFPE4AvgfMF5EZrlPve6wPcQmHoZh\nGNE+BrVRVcep6vHAzcDtwGoRmSgiB2fdwjomU56HiYphGPlMpD4PETnLzar7IN43xw8EXiXJtzka\nKqnEo7Kyssq2iYRhGDsiUSZGXAy8A9yrqh8GwqeIyM+zY1b9JR3PIyhEJjKGYeQTUcSjq6qWhO1Q\n1WszbE/OsQ5zwzCM1ETpMN9bRF4VkXUislZEXhGRA1MlEpH9ROQdEVkgIvPdt9ARkT1FZIaILHa/\newTSDBWRJSKySER6BcKPFpF5bt9oyeJHN9IVj/hmrETxTGQMw8gnoojHM8BzwD5AG+B5YFKEdOXA\nYFXtDPQErnbDfW8B3lbVjsDbbhu3rz9wONAb+EfgHZNHgMuAjm7pHal0WcDEwzAMI5p4/ERVn1TV\ncrc8BeycKpGqrnYfkUJVNwJfAW2BPsBEF20i3mdtceGTVbVUVb8BlgA9RGRfoKWqzlKvBv5XIE3G\nyVazlYmHYRj5RBTxmC4it4hIBxHZX0RuAl53zU97RjmIiHQAjgRmA61VdbXb9R3Q2q23BVYEkq10\nYW3denx4VqiLZqvFixfX0DrDMIz6QZQO8z+434Fx4f3xPgqVtP9DRHYFXgCuV9XiuBFIKiIZeyQX\nkcuBywHat8/OV3KjikeydIcccgjFxcVVvhNiGIbRkIjykuABSZZUwtEETzieVtUXXfAa1xSF+13r\nwguB/QLJ27mwQrceHx5m61hV7a6q3Vu1apWqaIlsrratqowYMYLFixfX2POIZ+vWrTWyzzAMoz4Q\n5SXBJiJyrYhMccufnSikSifA48BXqnp/YNdU4CK3fhHwSiC8v4g0E5ED8DrG57gmrmIR6enyvDCQ\nJuOENVsVFRVx6623cuqpp1qfh2EYBtGarR4BmgD/cNt/cmGXpkh3gos7T0Q+d2F/AUYBz4nIAGA5\nrllMVeeLyHPAAryRWleraoVLdxXwBNAcmO6WOsP3LrZu3Zqx0VZZHG1sGIaRdaKIxzGqekRg+98i\nMjdVIlV9H+9752GckiDN3cDdIeEFQJcIttaasEo9WPHbUF3DMIxoo60qROQgf8O9IFiRJH6DJplH\n4Pd/BKlJh7lhGEZDJ4rncSPwjoh8jedJ7A9cklWr6hnJPI+os+qaeBiGkU8kFQ8R2QnYgtd5fagL\nXqSqpdk2LFdky/NI5ziGYRj1naTioaqVIvKwqh4JfFFHNuWUREN1fazPwzAMI1qfx9si8vtsTkbY\nUBAR2rVrVyXMxMMwjB2RKOIxEG8yxFIRKRaRjSJSnGW7ckaq0Vbp7EsWz7TYMIyGTMoOc1XdoebQ\nSCYeq1atqrYvlecR/xu/3zAMoyES5Q3zt6OE5SuqmrSiTyQeU6dOZdu2bey0004MGzbMxMMwjLwi\noXiIyM5u1ty9RGQPfxZdN0Nu1ma1zTXxnkeqSn7t2rXcdNNNlJeXVwk///zz2bJlCwBjxozJrJGG\nYRg5Jlmz1UDgerwPQH3C9rfFi4G8rQ3T7Yu4+uqrWb58OSeccAJ9+vSpsq8m74cYhmE0BBJ6Hqr6\nd1U9ABiiqgcGZtI9QlXzVjziWbJkSdKKfvPmzUDy9z3C3g8Jbl988cWMGzeulpYahmHUHVGmZH9I\nRI4XkT+KyIX+UhfG5YJ4z+O8885LKgz+vkWLFvHII48kjJdMPCZOnMjll19eE3MNwzByQsrRViLy\nJHAQ8Dnb57TyPwebd4Q1Wx14YOLPlvjiMXTo0KT5WjOVYRj5RJS5rboDndVqv1DWr1+fMk7YiK10\nTufQoUNZsGABr7yStc+YGIZhpEUU8fgS2AdYnSpiPlBXL++lIx6jRo3KoiWGYRjpE0U89gIWiMgc\nIDYhoqqelTWrckg2xCOsw9wwDKMhE0U8hmfbiPpEtjwPG6prGEY+kVA8ROQwVV2oqu+KSLPgNOwi\n0rNuzGvY2HsehmHkK8mG6j4TWP8obt8/yFMy6XmYeBiGka8kEw9JsB62XT2xyAQRWSsiXwbChotI\noYh87pbTA/uGisgSEVkkIr0C4UeLyDy3b3S2p4bPlnikw5YtW3j00UezKjDTp09HRFixYkXWjmEY\nRv6STDw0wXrYdhhPAL1Dwh9Q1W5ueR1ARDoD/YHDXZp/iEgjF/8R4DK8rxl2TJBnvaSmnsdf/vIX\nrrzySl5++eWUx1i6dCkFBQVp2zZ+/HgA5syZk3baTLB8+fKcHNcwjMyQrMO8nYiMxvMy/HXcdsqJ\nEVX1v24SxSj0ASa7fpVvRGQJ0ENElgEtVXUWgIj8CzgbmB4x37TJhueRanqSeIqKigDYtGlTymMc\nfPDBKfNLZlsueOmll/jd737HtGnTOP3001MnMAyj3pFMPG4MrMc/2qb/qLuda9z0JgXAYFVdjydG\nswJxVrqwMrceHx6KiFwOXA7Qvn37Ghnnz1WVCZJ5HplGVWskfLn4KJXvKX322WcmHobRQEk2MeLE\nZEsNj/cIcCDQDe+lw/tqmE8oqjpWVburavdWrVrVKI8pU6Zk0p7Q9bDtqPsSMWzYsLTiW4e9YRi1\nIcpnaDOGqq5R1QpVrQTGAT3crkJgv0DUdi6s0K3HhzcI0hWP0047jd/85jex7XS8ggkTJtTItlx4\nHiZchtHwqVPxEJF9A5vn4E19AjAV6C8izUTkALyO8TmquhooFpGebpTVhUCDmeApXe9ixowZvPHG\nG9k0qRoiwtKlS3n77br/OKR9x90wGi5R3jCvESIyCTgZ70uEK4HbgZNFpBveaK1leB+cQlXni8hz\nwAKgHLhaVf0ZfK/CG7nVHK+jPGud5ZkmWYd5lHTZJHiMmna6G4ax4xJlSvZ7gBHAFuANoCtwg6o+\nlSydqp4XEvx4kvh3A3eHhBcAXVLZWR9J1mz1/fffc8ABByRNXxdP5pk4Rs+ePRERPvoo/l1SwzDy\nlSjNVqepajFwJp63cDBVR2IZCfAFI2xK9mOOOYbp06fnxdP+7NmzmTVrVuqIhmHkDVHEw/dOzgCe\nV9Ufs2hP3hImEh988EHoVwrrutnKMAwjXaL0ebwmIgvxmq2uFJFWwNbsmpUfpHrPI1VfSENptjIM\nY8cjyjfMbwGOB7qrahmwCe+NcCMFqZ7ud9ppp6Sex+LFixkzZkxObMsm5vUYRsMnSod5P+ANVa0Q\nkWHAUXgd6N9l27iGTqrRViISKh4+t99+e9ZsC9qQK8zrMYyGS5Q+j1tVdaOInAj8Gm/E1CPZNSs/\niDI9Saaewq0iNgyjLokiHv77FmcAY1V1GtA0eyblD6nEI1WzVTaxpiPDMGpDFPEoFJHHgHOB10Wk\nWcR0OzxROsyTNVvVBeaxGIZRE6KIwB+AN4FeqroB2BN7zyMSUTyP2ngAixcvrnHa+uB51AcbDMOo\nGVFGW20GlgK9ROTPwN6q+lbWLcsDUk0+mMjziFKpzpo1i0MOOSRrthmGYSQjpXiIyHXA08DebnlK\nRK7JtmH5QDabrWrjdcTbkA4jRoygV69eqSNm4diGYdQforwkOAA4VlU3AYjI/wEfAQ9l07B8oLYv\nCUbNuy7T33rrrbU6rmEY+UGUPg9h+4gr3Lo9Mkagpp5HXfYF2NO/YRg1IYrn8U9gtoi85LbPJsns\nuMZ26vNoK+usNgyjNqQUD1W9X0T+A5zogi5R1c+yalUecMghh9R4tJV5HoZh1HeSioeINALmq+ph\nwKd1Y1J+EJyGPd3RVoZhGPWdpH0e7mt+i0SkfR3ZkzfEex3btm2rFieReJSUlGTVNt+mXGFNZobR\n8InS57EHMF9E5uDNqAuAqp6VNavygPgPQHXv3r1anESjrd58882s2hZvQ66wJjPDaLhEEY8ajc0U\nkQl4Xx9cq6pdXNiewLNAB7yvEv5BVde7fUPxhgVXANeq6psu/Gi2f8P8deA6bSCPrqnMzGSzVboV\ncQM5hYZh1FMSNluJyMEicoKqvhtc8Cr3lRHyfgLoHRd2C/C2qnYE3nbbiEhnoD9wuEvzD9ffAt4M\nvpcBHd0Sn2e95Icffqjx9zyiEJ93TcXAnv4Nw6gJyfo8HgSKQ8J/dPuSoqr/BX6IC+4DTHTrE/GG\n/frhk1W1VFW/AZYAPURkX6Clqs5y3sa/AmnqNevXr2fcuHGAJyRh1OYlQcMwjFySTDxaq+q8+EAX\n1qGGx2utqqvd+ndAa7feFlgRiLfShbWlqpfjh4ciIpeLSIGIFBQVFdXQxMzx+uuvJ91fm2areI/B\nmq0Mw6hLkonH7kn2Na/tgZ0nkdEaTFXHqmp3Ve3eqlWrTGZdU3uS7s+W5zFp0iReeumlpHFy+X6J\nCZdhNHySdZgXiMhlqjouGCgilwKf1PB4a0RkX1Vd7Zqk1rrwQmC/QLx2LqzQrceHNwjqssM8yB//\n+MeUx/f3xQ8prss+EOtvMYyGSzLxuB54SUTOZ7tYdMf7iuA5NTzeVOAiYJT7fSUQ/oyI3A+0wesY\nn+O+m14sIj2B2cCF5NGEjNnuMK+srGSnnZJPXxZMFyW+YRgGJBEPVV0DHC8ivwS6uOBpqvrvKBmL\nyCTgZGAvEVkJ3I4nGs+JyABgOd6HplDV+SLyHLAAKAeudi8oAlzF9qG6093SIKjLZquwfLZt28bO\nO+8cOZ01JxmGEZUoc1u9A7yTbsaqel6CXackiH83cHdIeAHbxatBUZfNVmHHqqioCIlZNX62xePD\nDz9k5MiRvPzyyzRq1Ch1AsMwGgTWRpFFsvmeRzxh+ZSXl6dMl0g8pk6dyq677srmzZtrZVe/fv14\n7bXXWL16derIhmE0GEw8aklBQUHCfanE46233uKII47IiB2Z9jyGDh3Kpk2b+Oabb2pllz+nV9Om\nTZPaahhxw0XbAAAgAElEQVRGw8LEo5bsvnuyEc3Jef755zNmR7riEZYuG01YvnjUZUf8F198QVlZ\nWZ0dzzB2REw8akmy4aZ1+YQd1mxVG/Hwqe1wWl88wuzLxlDdr7/+miOOOIIhQ4ZkPG/DMLZj4hHH\nhx9+mFb8+iIeYcdK1udRVx3mvnjU1bnwZxaYNWtWrfN6+umnmTFjRq3zMYx8JMqsujsUxx13XFrx\ncyUeUd7zyITnUVt8j6MhfvTqggsuAKyPxjDCMM+jltSXt6Qz3WyV6QqzvlTAH3/8MV9++WWuzTCM\nBo95HrWkPjdbZeI9j0yJY33xPHr06AHUHzEzjIaKeR61JNkoorp8ek+3zyMsXTYnSwyKh1Xcxo5M\naWkpGzduzLUZtcbEo5bUpeeRLL90m61qKxTpli2bXo1hNCSOPfZYWrZsmWszao2JRy2pL+JR19OT\npFu2umq2Mq/GqO/MnTs31yZkBBOPWtJQxSMsXaKZedO1JYywfHJZ0X/99dc5O7Zh5AMmHrWkPneY\n1+Y9D389lXhELWNdC0XwuowdO5bXX3+9yvk46KCD6tSeumb06NH897//zbUZRh5jo61qSa48j/i8\ns/WGeSLx8MNHjx6d8hjx+dR1X8fAgQMBOPTQQ+v0uLnkuuuuAxpeM96qVato0aIFLVq0yLUpRgrM\n8wjh/fff57XXXosUty4rwvgPN6Ui3T6PM844g61bt1aJl8rzmD492udV6sNoq0WLFuXkuEZ02rZt\nS8+ePXNthhEBE48QTjjhBI466qhIcfPJ85gzZw6zZ8+uEp5KPKKKZ0N7AjZyx4IFC3JtghEBE48E\n7Lvvvnz77bcp4+VKPOIr9Uy85wHVy5Oq2SqqeNTVxIiGYdQNJh5J2G+//VLGqS+eRyaG6kJ08aiv\nnod5OKkpKiqql0/3UTxlo/6QE/EQkWUiMk9EPheRAhe2p4jMEJHF7nePQPyhIrJERBaJSK9c2FwT\n6tLzCCP4Z5wzZw6FhYVJ84TsiUc23vN49913qw25rS/ToNRnunTpwuGHH55rM6pRWlqaaxOMNMil\n5/FLVe2mqt3d9i3A26raEXjbbSMinYH+wOFAb+AfItIgPoadafG44IILks5Su3bt2irbQfE49thj\nqwxPTeV5+OGJngb9/VE/8pSNSv3kk0+uNuQ2nadXEQkV1Hwn/j6pL2zZsiXXJhhpUJ+arfoAE936\nRODsQPhkVS1V1W+AJUCPHNgXSl02W7300kt89913CfP2pxD3ie/zCD7Zff/996H5ZKvPI5vzZgVJ\nV6Rsht36g4lHwyJX4qHATBH5REQud2GtVXW1W/8OaO3W2wIrAmlXurAdEr+yDqskly1bVmW7oqKC\nRYsW8dVXX1WLu3jxYiCxeCQ7TjBdTZqt0n3BMB3848yePZtJkyaljN+oUXac2KKionohTK+++ipd\nu3aNNHgi1+yI4vHVV1+xZs2aXJtRI3IlHieqajfgN8DVIvLz4E71apW0axYRuVxECkSkwP+iXLZJ\nd8qQTB0varPVYYcdRufOnWNhUW/UTA/Vzeb0JBUVFWzYsKHacf74xz+mTNu4cfh7srW1rUuXLvzs\nZz+rVR41Jdh0N3DgQObNmxfzWIP06NGDX/ziFwBs3LiRVq1a8fbbb9eZnfHEv2OUDUpLS6mruiEK\nnTt35uCDD861GTUiJ+KhqoXudy3wEl4z1BoR2RfA/fo1YSEQHPbUzoWF5TtWVburavdWrVply/yc\n4lcMYZXbjz/+GBo3yD777FPFQ6lts1VUstlsdfPNN7PHHnuwefPmtO1K5HnUto8ml/0Kl156aWx9\nzz33BOCHH36oFu/jjz+OTWHy5Zdfsm7dOv7617/WjZEh1IV49OvXj7333jvrx0mHkpKSXJtQI+pc\nPERkFxFp4a8DpwFfAlOBi1y0i4BX3PpUoL+INBORA4COwJy6tbpmxFeO48aNq3WeZWVlQLTKLVFT\nxfLly2PrqYb81nRixHbt2sWmBUmUT6bE4/HHHwdg/fr1aVf6iTyPhjxq64knnoit++Lh93Elwh/4\nkMuhzv737rPJq6++mvVjROXTTz8NDd+6dSvHHHMMH3zwQR1blB658DxaA++LyFw8EZimqm8Ao4BT\nRWQx8Gu3jarOB54DFgBvAFerar0ZEP6Tn/wk4b74P2KiiiodfEFI9CcPhicaeeQLUFg+fqUZ1my1\nfv36aukSNVsVFhYyduzY2HbYk2+mKir/nPz4449pV/qJRotl6p2DbFfGW7du5f3330+4f/fddwdg\n3bp1SfNJ1cdVFwTvy2xTHx4Ojj322NDwBQsWUFBQwDXXXFPHFqVHnYuHqn6tqke45XBVvduFf6+q\np6hqR1X9tar+EEhzt6oepKqHqmq0yZTqiObNm3P99deH7suGeKTyPKKIR9Ajic8nPk1w/29+85tq\nx4laOZ5xxhlJba0Nfnk2bNiQdqWQ6BxlqnIJjnBbvnx5xt9luPbaaznppJP43//+F7rfb5ZLVTHH\nD9HOBXUpHrl4IXHTpk18/PHHse1ELQPpjmTMFfVpqG6DZbfddgsNj/8jNmnSpNbH8m+4KH0RUcQj\nPk4y8fDnvUp2/ChkYrRVWDnDxCPVMRL9gTNVufhisW3bNjp06MCFF16YkXx9PvnkEwCKi4tD9/sV\nUKrRVvVBPOqi2cqnLoXK5/zzz6dHj9RvGaT7DlWuqN/W1SOuvvrqhPsSPSFk0/NI9CcPVp6JKoxg\neHwcP/2SJUuq5bfXXnvF1tMdbRWfVzAP8Eb7iAiPPfZYpLyCdvui/OOPP6YUw3iieB4iwtKlSyPZ\nFY/fCexXjFFnaw6ybdu2hNfbtz/VvZUr8Rg/fjzPPPNMpLhRKnQRYfjw4bW0Kjdf84vah+FfUxOP\nPGHMmDEJ99WleGTC8wg+4YVVtsHx9sH89tlnn9h6skomkW3xxwrm4Y9OuvfeexPmmygv/3hlZWUp\nm+FS2eQT/yT/xhtvRLIrHt/zqKmnVlFRQbNmzRg8eHDofv9+SDRqLKrnEd/XlSkuu+wyzj///Ehx\nU4mHb9sdd9xRa7uOP/74SEO5M0nUd21SNVutXbu2Rg8hmcbEI03ChvlFFY9MNFv5I5hqIx5BcYgf\n/19ZWRlaMUPVjykla3pKVAkkEw//3EQdrhnmPWVSPPr27Vtlu6ZNKn55olYcpaWl3H///bH4mzZt\nAkjokfn2p/IAUx3fz6c+d5hn2rYoL5FmkmT2B/usUjVb9erVi9/+9rfcdNNNOZ0PzMQjDTZs2FCr\nb19nwvP44osvgMRv46YrHiNHjqyyr6KiokpFExzi2bRp09j6119/jYjw3nvvVcs/0Q0dP1IsKB7+\nvqjiESybv15b8bjsssu46aabAG8iySA1FQ//XEQVj3vvvZfBgwfHhh9v3LgRgGbNmoXG9+1PVM6o\nnkeqUXx1Qapz3BDekh8/fjwiElqWZPfioYceymeffQZs/w8n8iZ9obn33nuTtohkGxOPNNhtt93Y\nZZddqoXXpefhs3nz5tDwKH0eyaaBiBeP4LsawSfDmTNnAuHvD9SFeARt9PPZtm0bTz31VJV46YjH\n+PHjEzab1fQJL97z2Lx5c6w/KQz/RU+/2cwXj5133jk0vm9/omsdLx4TJ04MjZfs5dO6IpXnkWnx\nCD4MZYphw4YB4f+LVPb7gx+SeZPLli2r8t9PVA/UBSYeGSCReMQ/BWfC8/BJdNNEGaqb7IarrKxM\n2IkefJoK+yMEK/EwkomHX3HUxPMI5hHfFpyqqSOsHDfeeGO1sHTEI/gRsTDPo2PHjrFyjh07liOP\nPDK2L77juraeh2+Lf/yLL744NF5NPY/NmzfTvn17XnjhhZRxn3322aTT49S1eGRjKKz/Hw8rS6oH\nGX+GZz9tWLNVfD9NLsXexCMD5EI8/LbwZMeM0mwVT7znAdtv5uAfIpl4pPI8wmYG9vdFGSI7e/Zs\nTjnllGrhNfnDhu3/29/+Vi0snWar4MisRH0e/sSUAwcO5PPPP4+FJxKPmnoefvNbsopXVVMOxEjE\n4sWLWbFiBX379k16rtetW0f//v05++yzE8apa/HwKSkpSfoG/ooVK6pNOpoIv3Uh7D+Q6l70X+RM\nJh7x+Zp4NHCiPsHUdbNVopvVr5DC6Nu3b7XZYP2KM6p4JKpo586dy+zZs2P5h3keUbjwwguZP39+\ntfDaiEfwHZYw0vE8gveDP4os/nzF953FD332t/3hnS1btgSqVnRFRUWsXLkyNP94kp3fsrKyGjdb\nBY+bzGv0529atWpVwjjB+ybMjmyJx+GHH15lGHo87du354ADDoiUl/+AGPb/THVu/ck9/WsVVq8k\nG3RS15h4ZICo4pHJ6b8TeR6pKnhILh5AtWGh/p86arNVooq2d+/e9OzZM9auH+Z5hOUZ/8nUKNOu\n+ASnSAnDz6tnz55J46XjeQTjnn/++Sxfvrxa+eIr2vgKwz83ixYtArymLoBOnTrFKjo/DKJ1iCeq\naEpLS2vcbBWcjNO/7sXFxfzhD3+o0kTlV6ZBD2rFihWISGwm3+D1C7vGyco4c+ZM7rzzzoRv2icj\n2MwYlW3btlFeXs7cuXPp378/K1Z4X43wxaMm08vHi0fYPZfuS7DZxMQjA0QVj7ro83jnnXdi6zXx\nPKB6JZyu55HqKd3vDI7ieTz77LMcfvjhVfoyElUid911V7Ww22+/PaktUd8kT8fziI+7dOnSajbH\nVwz+drx4+Hn558f3NKBqxZ1KPD777DO6dOkSum/Lli0pPY9169bx73//u1q4X+EFbX3qqad4/vnn\nq5x7/5oH+27+85//AN5ABah6D8TPv3bXXXdVKXs8p556KrfffjuHHnpoTHDXrVuHiKR8gEhGoma8\nZs2aceKJJ3Leeefx7LPP0r59eyZNmhRrXciEeLz//vsJ+x99TDx2EDIhHv7c/4k8j+AEeIkqxlRT\nQCeq2FKJR1jHehh+RVJZWckDDzzADz/8kFA8Vq/2vg8WfEkvk/MS1UQ8SktL+eSTT1i3bh2zZ8+u\ncs5HjhwZG04dPEZ8+eIFJpV4xJ/TGTNmpFWOadOmVfPgfDZv3pzS8xg8eDCnnHJKtSf7MM+jRYsW\n1faF9d34U7X4FW6w3yF4f82ZM4fbbrst4cuG8f8F3+Px+5X8Yc9Q9UEv/qEvrOyJ/mfgNXUGWxM+\n/PDDpM1WqQjz8ONFyMQjz4haASXzUK677rqE+4I3aEVFBSUlJQlvzmBzSE09D//m/d3vfhfb3rhx\nY62breL54IMPGDRoEFdeeWXCJ+fWrb0PSvoiArkRj+Li4li5LrroIrp3706rVq3o2bMn/rdjSkpK\n+Mtf/hIbrhk8RqY8D5/46TVq0x8QFI9EIu5X7P67CD7BBxHfVt/24Fv6YZ6Hjy8ewYEDYQ8q8TMz\nl5SUUFZWxjnnnFMl3Bco/5wGh+QG/4Px1z7soSfVvRzs1G7SpElCzyPR3GNhxwqWPT6fsD6PLVu2\ncO2111bxAusCE49a4LvxUSugZHPV+DdufMfdX//6V+65557Y9jfffEOLFi0Stu364rHLLrvUuM/D\nr6j96eYvvfRSWrZsWWUUUViZ16xZw8KFC6tM3Z4Mv+LZsGFDwkrLPy9R3l9JxW677cb9999fJezx\nxx9n1qxZKdNOmzaNXXfdlU2bNjFlypRq+5cuXZqwWSVYOftE9TyCc2MFZzWOT18b8Qg2W4VVoCUl\nJbGPRsVXUMHKrbS0lOXLl3PRRd5neYL3mX9PhH3CwH9aD852ELwfEr3o2KJFC84555yEXphv2+zZ\nsxERPv300yriUVZWVuXJ3b8fVbXa+ffx52DzCT7YNW3aNKHnkcyD8fHtTdfzePLJJ3nooYdSNtFm\nGhOPWvDLX/4SiP7HTSYefmUwcuTIKhXDiBEjEs7aG8arr75K48aNadasWY09Dx//j+5XHJs2bYo9\nOYZV9ldeeSWdOnWiX79+kfL382jUqFHCc+ifi7A3ytOlcePG1ZoOP/roI4477rhI6cvLy+nWrVvo\n8Q8++GA6deoUmm7jxo0pxeP+++/niCOOYMSIEUC45xFsuou/hpnyPFatWsULL7xAeXl5LGzAgAGx\n48V/rTJePBL1yfieiz9qLMjYsWN58803KS4ujv1HwjyPsCfyadOmVcvPP2f+8f10M2bMqPYfDF7L\nkpISKisr2WmnnRg+fDhbt26tViHHD9mNF49EnkeU95f8OMG0qTyPb775Jlbeun5h0MQjA8Rf0H33\n3Tc0XhTxaNasWa1eXvr444/ZeeedadSoUeRJ/xIR9pToNyOFPaGm+21ov0Jq1KhRlYqhoKAgtn7F\nFVcA28/x6tWrQz8sFYVGjRrVesRbsrfDE1FSUpKy2ervf/97lb6S22+/HRGJPbHHV9rx17A2TXk/\n//nPq6Tv27cvzZo1iw1PDQ6L3rBhA8uXL+fBBx9EVWOd0+Ddw0FRDJbHF49E/X69e/dmzZo1/PSn\nPwW2V/jz5s2LvTwXrICTfYTtl7/8JVOmTKnmJd1yyy3VHnqC1+X111+PeTF33nknt956KxMmTKgS\nP74yD95PlZWVCT2PsP6m0aNHh+YdPEZ8PvH30aRJk7j22muBup9m3sQjA8Rf0KhTUAcJikejRo1o\n3rw5Dz/8cGj+qaioqKBRo0YJ20C///772J80GWF/UL99P2xOq3RHmPjNFI0bN66Stnfv3tXifvXV\nVxx88MG0adMmNK8oU02kKx7JKqh0CPM8br/99kgeoN88Gd/XEN806Odfk1E+UF3MKisrWblyJSJS\nZSjrhg0bOPfcc7nhhhsYPnx4lckFS0tLq4mcjz+oYPLkycyaNSthM47/2Vx/nrKuXbvWaPbb4cOH\nV3tfKYxghXvVVVdVuffiXxZdtWpVtf9U8EHv22+/jXXWb9myhccee4wHH3yQ0tJSzjzzzGrHbty4\ncZWy+dcuKJJXX3018+bNA2DUqFGxIcFhLFiwgOeee67GD1dp47fv5dty9NFHayYA1DtNicMGDx4c\nCwN0zpw52qFDhyphw4YN0/Xr11cJCy5nnXWWAvryyy9Xs+Ghhx5SQH/xi1+Ept1pp52qhbVt2zbh\nsQA97LDDku4HdMSIEdXCevTokTD+gQcemDLPKEvjxo1VVfW9996LnGb//fdPGaddu3Y6btw4BfRn\nP/tZyvht2rSpVTnef/99BXS//fYL3d++ffsa592lS5cq2//85z9VVfVPf/pTpPQnnnhile0777wz\nUrrzzjuv2rH95ZVXXtEJEyaE7kt23wSX448/XgGdP3++fvvttxm5n5Ita9euzfoxEi2PPvpolevV\nqFEjVVUdNGhQtbgffvhh5HxPPfVULS8vr02dV6AR6ljzPDJAfPt1+/btqzVRXX755ey+++6hbbTB\nPMJGo/hPqIleZAtrRw6bbiE41NH3IJLRvHnzamFDhgxJGL82Mw4HKS8vZ+HChZx00kmR0yR7ye+Y\nY44BYI899oh5HkcddVTCSQLj09WERo0acfzxx9OkSZOET4s1eTnNJ/6p+oorrqCkpIQnn3wSqHpP\nXHDBBVXi3nfffdW+iRHv2SRi0qRJCZ/o+/Tpk9Drjp+lOBH+gJEbb7yRAw88MFKadNlnn3247LLL\ngPBPLNQV8Z5wRUUFmzdvrjaoA7zvj0RlxowZkeeIqw0NRjxEpLeILBKRJSJyS67tCRI/5cLee+9d\n7TsZ/tj3RBPcjRkzht///vecfPLJ1fb5QuC79ECVeFdeeSUAt912Wyws/ub5+c9/XmXG2MMOOyxR\ncWKEddSfdtpp1SbXC/ZRZIpEHz9KRLJmuBNOOAGAtm3b0q9fP377299y1113JU3Tq1evKt8vSZfj\njz8eEamzdujS0tLYKCfwrv8hhxzCE088UU0EBw0aRNeuXQFivy+99FJG7PBnW45KcFJIgA4dOgBe\n/0O2piM56KCDUs4okC1atGjBz372MyC8GfXyyy/PyHHCZv/OOFHck1wvQCNgKXAg0BSYC3ROliab\nzVYTJkzQUaNGxbaPOeYYBXTq1Km6cOHCKuni07/xxhuhrmYyLrnkEgV0/PjxVeL76xUVFbpx40Yt\nLCwMzXv//ffX8vJyLS4uDnXX/aaC+GXOnDkKaKdOnXT//ffXRx99VFVVb7jhhmq2h6VPtjRt2rTW\nbr/fNHjBBRfoAw88EBrntttu0yuuuEIBHTBgQJXzumHDhoR5z58/X2+55ZYa23b99deHnpd33nlH\nL7300rTy8u+vdevW6ejRo7V79+6R0vmMHj06NHzDhg06Y8aMWPiDDz6oXbt2rfV1Ofnkk/Wmm26K\nFHfLli1V7oUnn3yyWpzy8nJt2bJl0ny6deuWlo1vvvlmrcuZatltt92qhb333nuxpqqJEyfqr3/9\nawW0ZcuWetRRR1WLP3fu3BoduzYQsdkq58IQyUg4DngzsD0UGJosTSbF44gjjkgaZ+bMmdqxY0fd\nvHlzLOzaa69V8Pop2rdvHwsvKyvTIUOGaP/+/RXQe++9VydOnJg0//PPP1/Ba1MO3hxffPGFPvnk\nk7F4W7ZsUahaMf/0pz/Vd999NxanvLxcKysrEwrJqFGjYv0lc+fO1YULF2pRUVEVe0pKSmLxu3fv\nrqqqvXr1qnLzdu/eXQcOHKhNmjTRV155RYcMGVJlf79+/UJv+hdffDFy38kVV1yhGzZs0LKyMi0r\nK9P77rtPoWp/wD//+U897rjjFNCZM2dWO7fTpk2LxV21apXecccdumrVKlXVWH6dO3fWCy64QM85\n55xqNpx++umhtvl9EP72jz/+qIWFhbHjFhUVxfbtscceCcs4ZswYraysrGJzsMKPUoH4wnrVVVfp\n6tWrq+S1adOmWPzFixfH7kt/id8OW1q3bq3FxcXapEmTWAVZWVmpAwYM0KeffjoW79xzz42tT5ky\nRR9++GFVVW3Xrl2VaxBWjj59+lQJO/vss6tsf/bZZ7pgwQIFtHfv3rHwY489NlbWxo0bx8I//fRT\nBfSkk07SRx55pEpeO++8c1oV9bnnnquTJ09WQMeOHRsL79u3b+zBz1/effddnTt3rrZu3VrXrFmj\ne+21lwJ6ww03VBPOIUOG6A8//BDZjuBDSW0gz8SjLzA+sP0nYExIvMuBAqAgWGHXhuLiYt26dWva\n6SorK2OdVvF/fj+stLQ0Ul7fffedDh06VMvLy/W9997T8ePHJ4w7atQo/eKLL3T27Nn697//PWm+\nI0eO1EWLFqmq6t/+9jedO3euqnqV08CBA5Pa98wzz+iUKVN027Ztqqq6detWPe200/S0007TSZMm\nxcro7/d56qmntFOnTlpUVKRDhgzRwYMH66hRo/SFF16IiejKlSv1pJNO0lNOOUUHDBign332mT7/\n/PN6zjnn6JQpU/TWW2/VDz74oFreqqqFhYVaWlqq//vf//SZZ57RyspKLSgo0JEjR4aWo6SkRK+/\n/vqYYARZv369XnLJJfr222/HwtauXatDhgzRgQMH6osvvqiFhYV6880364033qgLFy7Uvn376k03\n3aRlZWWqqvrSSy+FDoJQ9R4kbr75Zl2xYoUOGTJEH3vsMb3sssv0L3/5i95zzz1aUVGR8PyvX79e\nly1bppdccok+88wzesMNN+iFF16oDz/8sD7xxBP65ptvVinj4MGDdePGjaF5TZ8+XR988EFVVf3+\n++910KBB+sUXX+iIESNi99w111yjt9xyi7Zp00anTJmixcXFeskll2iPHj10+fLlsXzGjBmT0GZV\n1a+++kq//fbbKmFvvfVWlftt2LBh2qZNGx02bFhM8D/99FO95JJLdM2aNfrcc8/punXrtKysTMeN\nG6fDhg2LnauNGzfG1gsKCqqUefLkyXrOOefod999pxUVFXrrrbfqsmXLtLKyUm+88Ua98sorddCg\nQfr5559r165ddcqUKXrbbbdpnz599Le//W3sONu2bdNRo0bp2rVrdebMmdX+38XFxTpo0CD9+uuv\nVdX7byxYsCD2Hw7y0Ucfxc59WVmZ3n333XrmmWfq5MmTY3Hmzp2rH374od566606ZMgQLSkp0V/9\n6ld65513xv53gwcPVlXvgXLBggVJr0EqooqHeHHrNyLSF+itqpe67T8Bx6rqnxOl6d69u2ajLd4w\nDCOfEZFPVLV7qngNpcO8ENgvsN3OhRmGYRg5oKGIx8dARxE5QESaAv2BqTm2yTAMY4clcx+YyCKq\nWi4ifwbexBt5NUFVq39KzjAMw6gTGoR4AKjq68DrubbDMAzDaDjNVoZhGEY9wsTDMAzDSBsTD8Mw\nDCNtTDwMwzCMtGkQLwnWBBEpApbXMPlewLoMmtMQsDLvGFiZdwxqU+b9VTXltNt5Kx61QUQKorxh\nmU9YmXcMrMw7BnVRZmu2MgzDMNLGxMMwDMNIGxOPcMbm2oAcYGXeMbAy7xhkvczW52EYhmGkjXke\nhmEYRtqYeBiGYRhpY+IRQER6i8giEVkiIrfk2p5MISL7icg7IrJAROaLyHUufE8RmSEii93vHoE0\nQ915WCQivXJnfe0QkUYi8pmIvOa287rMIrK7iEwRkYUi8pWIHLcDlPkGd19/KSKTRGTnfCuziEwQ\nkbUi8mUgLO0yisjRIjLP7RstIlJjo6J8bnBHWPCmel8KHAg0BeYCnXNtV4bKti9wlFtvAfwP6Azc\nA9ziwm8B/s+td3blbwYc4M5Lo1yXo4ZlHwQ8A7zmtvO6zMBE4FK33hTYPZ/LDLQFvgGau+3ngIvz\nrczAz4GjgC8DYWmXEZgD9AQEmA78pqY2meexnR7AElX9WlW3AZOBPjm2KSOo6mpV/dStbwS+wvvT\n9cGrbHC/Z7v1PsBkVS1V1W+AJXjnp0EhIu2AM4DxgeC8LbOI7IZXyTwOoKrbVHUDeVxmR2OguYg0\nBn4CrCLPyqyq/wV+iAtOq4wisi/QUlVnqack/wqkSRsTj+20BVYEtle6sLxCRDoARwKzgdaqutrt\n+jFT/ccAAASbSURBVA5o7dbz5Vw8CNwEVAbC8rnMBwBFwD9dU914EdmFPC6zqhYCfwO+BVYDP6rq\nW+RxmQOkW8a2bj0+vEaYeOxAiMiuwAvA9apaHNznnkTyZty2iJwJrFXVTxLFybcy4z2BHwU8oqpH\nApvwmjNi5FuZXTt/HzzhbAPsIiIXBOPkW5nDyEUZTTy2UwjsF9hu58LyAhFpgiccT6vqiy54jXNl\ncb9rXXg+nIsTgLNEZBleE+SvROQp8rvMK4GVqjrbbU/BE5N8LvOvgW9UtUhVy4AXgePJ7zL7pFvG\nQrceH14jTDy28zHQUUQOEJGmQH9gao5tyghuRMXjwFeqen9g11TgIrd+EfBKILy/iDQTkQOAjngd\nbQ0GVR2qqu1UtQPetfy3ql5Afpf5O2CFiBzqgk4BFpDHZcZrruopIj9x9/kpeH16+Vxmn7TK6Jq4\nikWkpztXFwbSpE+uRxHUpwU4HW8k0lLgr7m2J4PlOhHPpf0C+NwtpwM/Bd4GFgMzgT0Daf7qzsMi\najEioz4swMlsH22V12UGugEF7lq/DOyxA5T5DmAh8CXwJN4oo7wqMzAJr0+nDM/DHFCTMgLd3Xla\nCozBzTJSk8WmJzEMwzDSxpqtDMMwjLQx8TAMwzDSxsTDMAzDSBsTD8MwDCNtTDwMwzCMtDHxMAyH\niLQWkWdE5GsR+UREPhKRc3Jky8kicnxg+woRuTAXthhGGI1zbYBh1AfcS1MvAxNV9Y8ubH/grCwe\ns7GqlifYfTJQAnwIoKqPZssOw6gJ9p6HYQAicgpwm6r+ImRfI2AUXoXeDHhYVR8TkZOB4cA6oAvw\nCXCBqqqIHA3cD+zq9l+sqqtF5D94L2meiPfi1/+AYXjTp38PnA80B2YBFXgTHV6D9+Z0iar+TUS6\nAY/izSC7FPj/VHW9y3s28Eu8qdgHqOp7mTtLhrEda7YyDI/DgU8T7BuAN1vrMcAxwGVu2gfwZii+\nHu8bCgcCJ7h5xB4C+qrq0cAE4O5Afk1Vtbuq3ge8D/RUbyLDycBNqroMTxweUNVuIQLwL+BmVe0K\nzANuD+xrrKo9nE23YxhZwpqtDCMEEXkYzzvYBiwHuopIX7d7N7z5grbhzRm00qX5HOgAbMDzRGa4\nD7U1wptawufZwHo74Fk3sV1TvA8bJbNrN2B3VX3XBU0Eng9E8Se9/MTZYhhZwcTDMDzmA7/3N1T1\nahHZC2+eqG+Ba1T1zWAC12xVGgiqwPtPCTBfVY9LcKxNgfWHgPtVdWqgGaw2+Pb4thhGVrBmK8Pw\n+Dews4hcGQj7ift9E7jSNUchIoe4jywlYhHQSkSOc/GbiMjhCeLuxvZpsS8KhG/E+2RwFVT1R2C9\niJzkgv4EvBsfzzCyjT2ZGAbex3RE5GzgARG5Ca+jehNwM16zUAfgUzcqq4gkn+9U1W2uiWu0a2Zq\njPdVw/kh0YcDz4vIejwB8/tSXgWmiEgfvA7zIBcBj4rIT4CvgUvSL7Fh1A4bbWUYhmGkjTVbGYZh\nGGlj4mEYhmGkjYmHYRiGkTYmHoZhGEbamHgYhmEYaWPiYRiGYaSNiYdhGIaRNv8/S62NHLFNH2MA\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x23a6ede8160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEWCAYAAACJ0YulAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\naDQajQdtFDQajUbjQRsFjUaj0Xj4/xRTvRGXndUmAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x23a6ede8da0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(loss_vec, 'k-')\n",
    "plt.title('Cross Entropy Loss per Generation')\n",
    "plt.xlabel('Generation')\n",
    "plt.ylabel('Cross Entropy Loss')\n",
    "plt.show()\n",
    "plt.plot(train_acc, 'k-', label='Train Set Accuracy')\n",
    "plt.plot(test_acc, 'r--', label='Test Set Accuracy')\n",
    "plt.title('Train and Test Accuracy')\n",
    "plt.xlabel('Generation')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend(loc='lower right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}