autoencoder_3_latent_0_0_6_accuracy.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Model, Sequential\n",
    "from keras.layers import Input, Dense, Conv2D, MaxPooling2D, UpSampling2D, Flatten, Reshape\n",
    "from keras import regularizers\n",
    "from bokeh.plotting import figure, show, output_file, ColumnDataSource\n",
    "from bokeh.io import output_notebook\n",
    "from keras.optimizers import SGD\n",
    "from bokeh.models import HoverTool, Callback\n",
    "from collections import OrderedDict\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from keras.regularizers import l1\n",
    "from keras.layers.normalization import BatchNormalization\n",
    "import numpy as np\n",
    "from keras.optimizers import SGD, Adam"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "train_list = np.load('rnafold_autoencoder_train.np (2).npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(202, 200, 200, 1)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_list.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "adam = Adam(lr=0.01, beta_1=0.9, beta_2=0.999, epsilon=1e-08, decay=0.0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "sgd = SGD(lr=0.1, momentum=0.5, decay=0.0000000001, nesterov=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_287 (Conv2D)          (None, 200, 200, 2)       10        \n",
      "_________________________________________________________________\n",
      "conv2d_288 (Conv2D)          (None, 200, 200, 3)       27        \n",
      "_________________________________________________________________\n",
      "batch_normalization_147 (Bat (None, 200, 200, 3)       12        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_67 (MaxPooling (None, 100, 100, 3)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_289 (Conv2D)          (None, 100, 100, 3)       39        \n",
      "_________________________________________________________________\n",
      "conv2d_290 (Conv2D)          (None, 100, 100, 3)       39        \n",
      "_________________________________________________________________\n",
      "batch_normalization_148 (Bat (None, 100, 100, 3)       12        \n",
      "_________________________________________________________________\n",
      "flatten_27 (Flatten)         (None, 30000)             0         \n",
      "_________________________________________________________________\n",
      "bneck (Dense)                (None, 3)                 90003     \n",
      "_________________________________________________________________\n",
      "dense_21 (Dense)             (None, 30000)             120000    \n",
      "=================================================================\n",
      "Total params: 210,142\n",
      "Trainable params: 210,130\n",
      "Non-trainable params: 12\n",
      "_________________________________________________________________\n",
      "None\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_287 (Conv2D)          (None, 200, 200, 2)       10        \n",
      "_________________________________________________________________\n",
      "conv2d_288 (Conv2D)          (None, 200, 200, 3)       27        \n",
      "_________________________________________________________________\n",
      "batch_normalization_147 (Bat (None, 200, 200, 3)       12        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_67 (MaxPooling (None, 100, 100, 3)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_289 (Conv2D)          (None, 100, 100, 3)       39        \n",
      "_________________________________________________________________\n",
      "conv2d_290 (Conv2D)          (None, 100, 100, 3)       39        \n",
      "_________________________________________________________________\n",
      "batch_normalization_148 (Bat (None, 100, 100, 3)       12        \n",
      "_________________________________________________________________\n",
      "flatten_27 (Flatten)         (None, 30000)             0         \n",
      "_________________________________________________________________\n",
      "bneck (Dense)                (None, 3)                 90003     \n",
      "_________________________________________________________________\n",
      "dense_21 (Dense)             (None, 30000)             120000    \n",
      "_________________________________________________________________\n",
      "reshape_17 (Reshape)         (None, 100, 100, 3)       0         \n",
      "=================================================================\n",
      "Total params: 210,142\n",
      "Trainable params: 210,130\n",
      "Non-trainable params: 12\n",
      "_________________________________________________________________\n",
      "None\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv2d_287 (Conv2D)          (None, 200, 200, 2)       10        \n",
      "_________________________________________________________________\n",
      "conv2d_288 (Conv2D)          (None, 200, 200, 3)       27        \n",
      "_________________________________________________________________\n",
      "batch_normalization_147 (Bat (None, 200, 200, 3)       12        \n",
      "_________________________________________________________________\n",
      "max_pooling2d_67 (MaxPooling (None, 100, 100, 3)       0         \n",
      "_________________________________________________________________\n",
      "conv2d_289 (Conv2D)          (None, 100, 100, 3)       39        \n",
      "_________________________________________________________________\n",
      "conv2d_290 (Conv2D)          (None, 100, 100, 3)       39        \n",
      "_________________________________________________________________\n",
      "batch_normalization_148 (Bat (None, 100, 100, 3)       12        \n",
      "_________________________________________________________________\n",
      "flatten_27 (Flatten)         (None, 30000)             0         \n",
      "_________________________________________________________________\n",
      "bneck (Dense)                (None, 3)                 90003     \n",
      "_________________________________________________________________\n",
      "dense_21 (Dense)             (None, 30000)             120000    \n",
      "_________________________________________________________________\n",
      "reshape_17 (Reshape)         (None, 100, 100, 3)       0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_149 (Bat (None, 100, 100, 3)       12        \n",
      "_________________________________________________________________\n",
      "conv2d_291 (Conv2D)          (None, 100, 100, 5)       140       \n",
      "_________________________________________________________________\n",
      "conv2d_292 (Conv2D)          (None, 100, 100, 5)       230       \n",
      "_________________________________________________________________\n",
      "up_sampling2d_46 (UpSampling (None, 200, 200, 5)       0         \n",
      "_________________________________________________________________\n",
      "batch_normalization_150 (Bat (None, 200, 200, 5)       20        \n",
      "_________________________________________________________________\n",
      "conv2d_293 (Conv2D)          (None, 200, 200, 5)       230       \n",
      "_________________________________________________________________\n",
      "conv2d_294 (Conv2D)          (None, 200, 200, 1)       21        \n",
      "=================================================================\n",
      "Total params: 210,795\n",
      "Trainable params: 210,767\n",
      "Non-trainable params: 28\n",
      "_________________________________________________________________\n",
      "None\n",
      "Train on 180 samples, validate on 22 samples\n",
      "Epoch 1/1000\n",
      "180/180 [==============================] - 6s 32ms/step - loss: 0.4609 - val_loss: 0.2924\n",
      "Epoch 2/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.2307 - val_loss: 0.2183\n",
      "Epoch 3/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1955 - val_loss: 0.2052\n",
      "Epoch 4/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1854 - val_loss: 0.1953\n",
      "Epoch 5/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1819 - val_loss: 0.1924\n",
      "Epoch 6/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1765 - val_loss: 0.1887\n",
      "Epoch 7/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1749 - val_loss: 0.1900\n",
      "Epoch 8/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1717 - val_loss: 0.1868\n",
      "Epoch 9/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1698 - val_loss: 0.1844\n",
      "Epoch 10/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1671 - val_loss: 0.1835\n",
      "Epoch 11/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1676 - val_loss: 0.1829\n",
      "Epoch 12/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1651 - val_loss: 0.1814\n",
      "Epoch 13/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1634 - val_loss: 0.1805\n",
      "Epoch 14/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1628 - val_loss: 0.1806\n",
      "Epoch 15/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1619 - val_loss: 0.1804\n",
      "Epoch 16/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1630 - val_loss: 0.1799\n",
      "Epoch 17/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1614 - val_loss: 0.1788\n",
      "Epoch 18/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1592 - val_loss: 0.1826\n",
      "Epoch 19/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1595 - val_loss: 0.1784\n",
      "Epoch 20/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1616 - val_loss: 0.1778\n",
      "Epoch 21/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1568 - val_loss: 0.1776\n",
      "Epoch 22/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1581 - val_loss: 0.1769\n",
      "Epoch 23/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1590 - val_loss: 0.1770\n",
      "Epoch 24/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1562 - val_loss: 0.1763\n",
      "Epoch 25/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1586 - val_loss: 0.1773\n",
      "Epoch 26/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1563 - val_loss: 0.1801\n",
      "Epoch 27/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1556 - val_loss: 0.1771\n",
      "Epoch 28/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1543 - val_loss: 0.1776\n",
      "Epoch 29/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1569 - val_loss: 0.1764\n",
      "Epoch 30/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1557 - val_loss: 0.1767\n",
      "Epoch 31/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1541 - val_loss: 0.1759\n",
      "Epoch 32/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1564 - val_loss: 0.1780\n",
      "Epoch 33/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1528 - val_loss: 0.1759\n",
      "Epoch 34/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1520 - val_loss: 0.1761\n",
      "Epoch 35/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1542 - val_loss: 0.1754\n",
      "Epoch 36/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1501 - val_loss: 0.1744\n",
      "Epoch 37/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1507 - val_loss: 0.1770\n",
      "Epoch 38/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1493 - val_loss: 0.1761\n",
      "Epoch 39/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1516 - val_loss: 0.1749\n",
      "Epoch 40/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1472 - val_loss: 0.1751\n",
      "Epoch 41/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1479 - val_loss: 0.1750\n",
      "Epoch 42/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1485 - val_loss: 0.1746\n",
      "Epoch 43/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1504 - val_loss: 0.1751\n",
      "Epoch 44/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1479 - val_loss: 0.1737\n",
      "Epoch 45/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1484 - val_loss: 0.1739\n",
      "Epoch 46/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1471 - val_loss: 0.1739\n",
      "Epoch 47/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1461 - val_loss: 0.1735\n",
      "Epoch 48/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1443 - val_loss: 0.1739\n",
      "Epoch 49/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1459 - val_loss: 0.1728\n",
      "Epoch 50/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1408 - val_loss: 0.1731\n",
      "Epoch 51/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1418 - val_loss: 0.1733\n",
      "Epoch 52/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1427 - val_loss: 0.1726\n",
      "Epoch 53/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1390 - val_loss: 0.1712\n",
      "Epoch 54/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1379 - val_loss: 0.1703\n",
      "Epoch 55/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1378 - val_loss: 0.1706\n",
      "Epoch 56/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1397 - val_loss: 0.1709\n",
      "Epoch 57/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1380 - val_loss: 0.1701\n",
      "Epoch 58/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1346 - val_loss: 0.1707\n",
      "Epoch 59/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1346 - val_loss: 0.1689\n",
      "Epoch 60/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1334 - val_loss: 0.1700\n",
      "Epoch 61/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1347 - val_loss: 0.1684\n",
      "Epoch 62/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1340 - val_loss: 0.1706\n",
      "Epoch 63/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1318 - val_loss: 0.1682\n",
      "Epoch 64/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1318 - val_loss: 0.1678\n",
      "Epoch 65/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1319 - val_loss: 0.1696\n",
      "Epoch 66/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1312 - val_loss: 0.1698\n",
      "Epoch 67/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1275 - val_loss: 0.1664\n",
      "Epoch 68/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1279 - val_loss: 0.1654\n",
      "Epoch 69/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1298 - val_loss: 0.1673\n",
      "Epoch 70/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1271 - val_loss: 0.1661\n",
      "Epoch 71/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1264 - val_loss: 0.1682\n",
      "Epoch 72/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1263 - val_loss: 0.1682\n",
      "Epoch 73/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1254 - val_loss: 0.1666\n",
      "Epoch 74/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1264 - val_loss: 0.1660\n",
      "Epoch 75/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1277 - val_loss: 0.1639\n",
      "Epoch 76/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1231 - val_loss: 0.1674\n",
      "Epoch 77/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1242 - val_loss: 0.1651\n",
      "Epoch 78/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1223 - val_loss: 0.1654\n",
      "Epoch 79/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1232 - val_loss: 0.1644\n",
      "Epoch 80/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1201 - val_loss: 0.1670\n",
      "Epoch 81/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1253 - val_loss: 0.1681\n",
      "Epoch 82/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1203 - val_loss: 0.1657\n",
      "Epoch 83/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1206 - val_loss: 0.1676\n",
      "Epoch 84/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1165 - val_loss: 0.1671\n",
      "Epoch 85/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1198 - val_loss: 0.1707\n",
      "Epoch 86/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1182 - val_loss: 0.1675\n",
      "Epoch 87/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1148 - val_loss: 0.1725\n",
      "Epoch 88/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1173 - val_loss: 0.1816\n",
      "Epoch 89/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1151 - val_loss: 0.1659\n",
      "Epoch 90/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1139 - val_loss: 0.1706\n",
      "Epoch 91/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1127 - val_loss: 0.1799\n",
      "Epoch 92/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1155 - val_loss: 0.1659\n",
      "Epoch 93/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1144 - val_loss: 0.1784\n",
      "Epoch 94/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1130 - val_loss: 0.1784\n",
      "Epoch 95/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1113 - val_loss: 0.1772\n",
      "Epoch 96/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1131 - val_loss: 0.1648\n",
      "Epoch 97/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1175 - val_loss: 0.1648\n",
      "Epoch 98/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1127 - val_loss: 0.1718\n",
      "Epoch 99/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1173 - val_loss: 0.1916\n",
      "Epoch 100/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1107 - val_loss: 0.1655\n",
      "Epoch 101/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1106 - val_loss: 0.1684\n",
      "Epoch 102/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1094 - val_loss: 0.1700\n",
      "Epoch 103/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1130 - val_loss: 0.1771\n",
      "Epoch 104/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1094 - val_loss: 0.1659\n",
      "Epoch 105/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1148 - val_loss: 0.1748\n",
      "Epoch 106/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1097 - val_loss: 0.1669\n",
      "Epoch 107/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1069 - val_loss: 0.1675\n",
      "Epoch 108/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1084 - val_loss: 0.1712\n",
      "Epoch 109/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1073 - val_loss: 0.1802\n",
      "Epoch 110/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1082 - val_loss: 0.1720\n",
      "Epoch 111/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1039 - val_loss: 0.1678\n",
      "Epoch 112/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1102 - val_loss: 0.1675\n",
      "Epoch 113/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1064 - val_loss: 0.1793\n",
      "Epoch 114/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1098 - val_loss: 0.1919\n",
      "Epoch 115/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1105 - val_loss: 0.1715\n",
      "Epoch 116/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1086 - val_loss: 0.1720\n",
      "Epoch 117/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1041 - val_loss: 0.1711\n",
      "Epoch 118/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1102 - val_loss: 0.1808\n",
      "Epoch 119/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1050 - val_loss: 0.1713\n",
      "Epoch 120/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1054 - val_loss: 0.1727\n",
      "Epoch 121/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1025 - val_loss: 0.1729\n",
      "Epoch 122/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1007 - val_loss: 0.1855\n",
      "Epoch 123/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1014 - val_loss: 0.1745\n",
      "Epoch 124/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1035 - val_loss: 0.1731\n",
      "Epoch 125/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1058 - val_loss: 0.1700\n",
      "Epoch 126/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1026 - val_loss: 0.1818\n",
      "Epoch 127/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1017 - val_loss: 0.1838\n",
      "Epoch 128/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1059 - val_loss: 0.1753\n",
      "Epoch 129/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1015 - val_loss: 0.1705\n",
      "Epoch 130/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1018 - val_loss: 0.1743\n",
      "Epoch 131/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1089 - val_loss: 0.1892\n",
      "Epoch 132/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1024 - val_loss: 0.1642\n",
      "Epoch 133/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1070 - val_loss: 0.1712\n",
      "Epoch 134/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1008 - val_loss: 0.1729\n",
      "Epoch 135/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1025 - val_loss: 0.1737\n",
      "Epoch 136/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0979 - val_loss: 0.1722\n",
      "Epoch 137/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1040 - val_loss: 0.1710\n",
      "Epoch 138/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0993 - val_loss: 0.1718\n",
      "Epoch 139/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1019 - val_loss: 0.1647\n",
      "Epoch 140/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1000 - val_loss: 0.1838\n",
      "Epoch 141/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0975 - val_loss: 0.1957\n",
      "Epoch 142/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0969 - val_loss: 0.1751\n",
      "Epoch 143/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0968 - val_loss: 0.1691\n",
      "Epoch 144/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0999 - val_loss: 0.1776\n",
      "Epoch 145/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0962 - val_loss: 0.1795\n",
      "Epoch 146/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0976 - val_loss: 0.1651\n",
      "Epoch 147/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1002 - val_loss: 0.1903\n",
      "Epoch 148/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0972 - val_loss: 0.1903\n",
      "Epoch 149/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0976 - val_loss: 0.1842\n",
      "Epoch 150/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0957 - val_loss: 0.1781\n",
      "Epoch 151/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0981 - val_loss: 0.1644\n",
      "Epoch 152/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0972 - val_loss: 0.1895\n",
      "Epoch 153/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0963 - val_loss: 0.1806\n",
      "Epoch 154/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0954 - val_loss: 0.1869\n",
      "Epoch 155/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0967 - val_loss: 0.1971\n",
      "Epoch 156/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0974 - val_loss: 0.1733\n",
      "Epoch 157/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0953 - val_loss: 0.1737\n",
      "Epoch 158/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0936 - val_loss: 0.1726\n",
      "Epoch 159/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0947 - val_loss: 0.1749\n",
      "Epoch 160/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0949 - val_loss: 0.2090\n",
      "Epoch 161/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0953 - val_loss: 0.1789\n",
      "Epoch 162/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0969 - val_loss: 0.1752\n",
      "Epoch 163/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0955 - val_loss: 0.1661\n",
      "Epoch 164/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0973 - val_loss: 0.1769\n",
      "Epoch 165/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0952 - val_loss: 0.1984\n",
      "Epoch 166/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0920 - val_loss: 0.1873\n",
      "Epoch 167/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0972 - val_loss: 0.2123\n",
      "Epoch 168/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0948 - val_loss: 0.2078\n",
      "Epoch 169/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0957 - val_loss: 0.2025\n",
      "Epoch 170/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0958 - val_loss: 0.1902\n",
      "Epoch 171/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0948 - val_loss: 0.2118\n",
      "Epoch 172/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0952 - val_loss: 0.2003\n",
      "Epoch 173/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0946 - val_loss: 0.1693\n",
      "Epoch 174/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0949 - val_loss: 0.1935\n",
      "Epoch 175/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0914 - val_loss: 0.1947\n",
      "Epoch 176/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0899 - val_loss: 0.1974\n",
      "Epoch 177/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0900 - val_loss: 0.1943\n",
      "Epoch 178/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0930 - val_loss: 0.1956\n",
      "Epoch 179/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0909 - val_loss: 0.1821\n",
      "Epoch 180/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0884 - val_loss: 0.1973\n",
      "Epoch 181/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0924 - val_loss: 0.1848\n",
      "Epoch 182/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0918 - val_loss: 0.1792\n",
      "Epoch 183/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0990 - val_loss: 0.1813\n",
      "Epoch 184/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0928 - val_loss: 0.1807\n",
      "Epoch 185/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0921 - val_loss: 0.1955\n",
      "Epoch 186/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0930 - val_loss: 0.2000\n",
      "Epoch 187/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.1003 - val_loss: 0.1759\n",
      "Epoch 188/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0930 - val_loss: 0.1830\n",
      "Epoch 189/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0892 - val_loss: 0.1980\n",
      "Epoch 190/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0889 - val_loss: 0.1973\n",
      "Epoch 191/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0923 - val_loss: 0.1772\n",
      "Epoch 192/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0924 - val_loss: 0.2019\n",
      "Epoch 193/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0915 - val_loss: 0.1951\n",
      "Epoch 194/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0904 - val_loss: 0.1770\n",
      "Epoch 195/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0904 - val_loss: 0.1878\n",
      "Epoch 196/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0898 - val_loss: 0.1674\n",
      "Epoch 197/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0910 - val_loss: 0.1711\n",
      "Epoch 198/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0914 - val_loss: 0.1905\n",
      "Epoch 199/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0882 - val_loss: 0.2084\n",
      "Epoch 200/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0925 - val_loss: 0.2109\n",
      "Epoch 201/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0900 - val_loss: 0.1948\n",
      "Epoch 202/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0902 - val_loss: 0.2185\n",
      "Epoch 203/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0914 - val_loss: 0.2289\n",
      "Epoch 204/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0889 - val_loss: 0.1834\n",
      "Epoch 205/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0867 - val_loss: 0.1738\n",
      "Epoch 206/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0885 - val_loss: 0.1922\n",
      "Epoch 207/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0899 - val_loss: 0.1677\n",
      "Epoch 208/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0927 - val_loss: 0.1982\n",
      "Epoch 209/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0877 - val_loss: 0.1946\n",
      "Epoch 210/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0868 - val_loss: 0.2011\n",
      "Epoch 211/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0844 - val_loss: 0.2034\n",
      "Epoch 212/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0857 - val_loss: 0.1955\n",
      "Epoch 213/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0856 - val_loss: 0.1945\n",
      "Epoch 214/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0909 - val_loss: 0.1717\n",
      "Epoch 215/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0888 - val_loss: 0.1917\n",
      "Epoch 216/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0886 - val_loss: 0.2051\n",
      "Epoch 217/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0908 - val_loss: 0.1698\n",
      "Epoch 218/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0906 - val_loss: 0.2035\n",
      "Epoch 219/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0911 - val_loss: 0.1880\n",
      "Epoch 220/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0880 - val_loss: 0.2109\n",
      "Epoch 221/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0914 - val_loss: 0.2117\n",
      "Epoch 222/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0967 - val_loss: 0.1936\n",
      "Epoch 223/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0876 - val_loss: 0.1897\n",
      "Epoch 224/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0868 - val_loss: 0.1897\n",
      "Epoch 225/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0850 - val_loss: 0.1968\n",
      "Epoch 226/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0836 - val_loss: 0.1948\n",
      "Epoch 227/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0827 - val_loss: 0.2097\n",
      "Epoch 228/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0843 - val_loss: 0.2033\n",
      "Epoch 229/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0845 - val_loss: 0.1843\n",
      "Epoch 230/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0905 - val_loss: 0.2227\n",
      "Epoch 231/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0851 - val_loss: 0.2074\n",
      "Epoch 232/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0851 - val_loss: 0.1968\n",
      "Epoch 233/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0901 - val_loss: 0.2291\n",
      "Epoch 234/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0904 - val_loss: 0.2084\n",
      "Epoch 235/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0962 - val_loss: 0.1842\n",
      "Epoch 236/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0871 - val_loss: 0.2052\n",
      "Epoch 237/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0848 - val_loss: 0.1813\n",
      "Epoch 238/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0839 - val_loss: 0.1930\n",
      "Epoch 239/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0884 - val_loss: 0.1912\n",
      "Epoch 240/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0848 - val_loss: 0.1964\n",
      "Epoch 241/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0843 - val_loss: 0.2026\n",
      "Epoch 242/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0842 - val_loss: 0.2025\n",
      "Epoch 243/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0837 - val_loss: 0.1877\n",
      "Epoch 244/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0862 - val_loss: 0.1887\n",
      "Epoch 245/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0832 - val_loss: 0.1967\n",
      "Epoch 246/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0844 - val_loss: 0.2044\n",
      "Epoch 247/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0842 - val_loss: 0.2068\n",
      "Epoch 248/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0821 - val_loss: 0.1815\n",
      "Epoch 249/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0859 - val_loss: 0.1887\n",
      "Epoch 250/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0854 - val_loss: 0.1833\n",
      "Epoch 251/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0839 - val_loss: 0.2321\n",
      "Epoch 252/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0834 - val_loss: 0.2110\n",
      "Epoch 253/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0823 - val_loss: 0.2037\n",
      "Epoch 254/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0823 - val_loss: 0.2110\n",
      "Epoch 255/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0810 - val_loss: 0.2041\n",
      "Epoch 256/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0826 - val_loss: 0.1994\n",
      "Epoch 257/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0813 - val_loss: 0.2070\n",
      "Epoch 258/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0823 - val_loss: 0.1768\n",
      "Epoch 259/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0822 - val_loss: 0.2028\n",
      "Epoch 260/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0820 - val_loss: 0.2074\n",
      "Epoch 261/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0914 - val_loss: 0.1811\n",
      "Epoch 262/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0863 - val_loss: 0.2101\n",
      "Epoch 263/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0828 - val_loss: 0.1933\n",
      "Epoch 264/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0816 - val_loss: 0.2304\n",
      "Epoch 265/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0861 - val_loss: 0.2558\n",
      "Epoch 266/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0839 - val_loss: 0.1880\n",
      "Epoch 267/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0818 - val_loss: 0.1997\n",
      "Epoch 268/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0808 - val_loss: 0.1958\n",
      "Epoch 269/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0811 - val_loss: 0.2170\n",
      "Epoch 270/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0813 - val_loss: 0.2102\n",
      "Epoch 271/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0804 - val_loss: 0.2071\n",
      "Epoch 272/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0801 - val_loss: 0.2098\n",
      "Epoch 273/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0840 - val_loss: 0.2049\n",
      "Epoch 274/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0810 - val_loss: 0.2336\n",
      "Epoch 275/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0842 - val_loss: 0.2034\n",
      "Epoch 276/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0837 - val_loss: 0.2116\n",
      "Epoch 277/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0817 - val_loss: 0.1934\n",
      "Epoch 278/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0826 - val_loss: 0.1888\n",
      "Epoch 279/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0827 - val_loss: 0.1771\n",
      "Epoch 280/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0853 - val_loss: 0.1934\n",
      "Epoch 281/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0809 - val_loss: 0.2092\n",
      "Epoch 282/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0807 - val_loss: 0.1849\n",
      "Epoch 283/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0811 - val_loss: 0.2098\n",
      "Epoch 284/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0891 - val_loss: 0.2322\n",
      "Epoch 285/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0870 - val_loss: 0.2160\n",
      "Epoch 286/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0846 - val_loss: 0.1916\n",
      "Epoch 287/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0833 - val_loss: 0.1944\n",
      "Epoch 288/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0820 - val_loss: 0.1954\n",
      "Epoch 289/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0856 - val_loss: 0.1806\n",
      "Epoch 290/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0926 - val_loss: 0.1806\n",
      "Epoch 291/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0870 - val_loss: 0.2152\n",
      "Epoch 292/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0818 - val_loss: 0.1986\n",
      "Epoch 293/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0807 - val_loss: 0.2068\n",
      "Epoch 294/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0804 - val_loss: 0.1881\n",
      "Epoch 295/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0820 - val_loss: 0.1813\n",
      "Epoch 296/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0811 - val_loss: 0.2259\n",
      "Epoch 297/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0801 - val_loss: 0.1917\n",
      "Epoch 298/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0795 - val_loss: 0.1905\n",
      "Epoch 299/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0817 - val_loss: 0.2047\n",
      "Epoch 300/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0831 - val_loss: 0.2101\n",
      "Epoch 301/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0818 - val_loss: 0.2088\n",
      "Epoch 302/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0801 - val_loss: 0.1858\n",
      "Epoch 303/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0823 - val_loss: 0.1784\n",
      "Epoch 304/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0824 - val_loss: 0.1951\n",
      "Epoch 305/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0798 - val_loss: 0.1962\n",
      "Epoch 306/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0797 - val_loss: 0.1994\n",
      "Epoch 307/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0795 - val_loss: 0.2430\n",
      "Epoch 308/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0830 - val_loss: 0.2469\n",
      "Epoch 309/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0793 - val_loss: 0.2087\n",
      "Epoch 310/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0815 - val_loss: 0.2037\n",
      "Epoch 311/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0787 - val_loss: 0.2616\n",
      "Epoch 312/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0822 - val_loss: 0.2185\n",
      "Epoch 313/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0779 - val_loss: 0.2097\n",
      "Epoch 314/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0790 - val_loss: 0.2317\n",
      "Epoch 315/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0775 - val_loss: 0.1949\n",
      "Epoch 316/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0789 - val_loss: 0.2268\n",
      "Epoch 317/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0770 - val_loss: 0.2007\n",
      "Epoch 318/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0814 - val_loss: 0.2065\n",
      "Epoch 319/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0777 - val_loss: 0.2119\n",
      "Epoch 320/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0772 - val_loss: 0.2357\n",
      "Epoch 321/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0810 - val_loss: 0.2316\n",
      "Epoch 322/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0832 - val_loss: 0.2333\n",
      "Epoch 323/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0828 - val_loss: 0.1923\n",
      "Epoch 324/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0799 - val_loss: 0.1934\n",
      "Epoch 325/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0796 - val_loss: 0.2056\n",
      "Epoch 326/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0779 - val_loss: 0.2067\n",
      "Epoch 327/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0776 - val_loss: 0.2073\n",
      "Epoch 328/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0771 - val_loss: 0.2158\n",
      "Epoch 329/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0765 - val_loss: 0.2057\n",
      "Epoch 330/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0775 - val_loss: 0.2205\n",
      "Epoch 331/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0771 - val_loss: 0.2128\n",
      "Epoch 332/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0776 - val_loss: 0.2099\n",
      "Epoch 333/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0769 - val_loss: 0.2271\n",
      "Epoch 334/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0779 - val_loss: 0.1985\n",
      "Epoch 335/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0803 - val_loss: 0.2426\n",
      "Epoch 336/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0813 - val_loss: 0.2021\n",
      "Epoch 337/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0788 - val_loss: 0.2321\n",
      "Epoch 338/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0776 - val_loss: 0.2149\n",
      "Epoch 339/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0778 - val_loss: 0.2236\n",
      "Epoch 340/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0795 - val_loss: 0.2312\n",
      "Epoch 341/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0772 - val_loss: 0.2061\n",
      "Epoch 342/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0760 - val_loss: 0.2166\n",
      "Epoch 343/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0766 - val_loss: 0.2158\n",
      "Epoch 344/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0755 - val_loss: 0.2213\n",
      "Epoch 345/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0772 - val_loss: 0.2233\n",
      "Epoch 346/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0797 - val_loss: 0.2090\n",
      "Epoch 347/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0822 - val_loss: 0.1841\n",
      "Epoch 348/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0834 - val_loss: 0.2309\n",
      "Epoch 349/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0790 - val_loss: 0.2126\n",
      "Epoch 350/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0764 - val_loss: 0.2126\n",
      "Epoch 351/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0767 - val_loss: 0.2112\n",
      "Epoch 352/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0772 - val_loss: 0.2214\n",
      "Epoch 353/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0772 - val_loss: 0.2087\n",
      "Epoch 354/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0801 - val_loss: 0.1902\n",
      "Epoch 355/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0835 - val_loss: 0.2075\n",
      "Epoch 356/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0771 - val_loss: 0.1931\n",
      "Epoch 357/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0816 - val_loss: 0.1925\n",
      "Epoch 358/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0780 - val_loss: 0.2270\n",
      "Epoch 359/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0768 - val_loss: 0.2084\n",
      "Epoch 360/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0766 - val_loss: 0.2036\n",
      "Epoch 361/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0784 - val_loss: 0.2177\n",
      "Epoch 362/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0869 - val_loss: 0.2519\n",
      "Epoch 363/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0802 - val_loss: 0.2130\n",
      "Epoch 364/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0781 - val_loss: 0.1858\n",
      "Epoch 365/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0785 - val_loss: 0.2350\n",
      "Epoch 366/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0776 - val_loss: 0.2113\n",
      "Epoch 367/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0797 - val_loss: 0.2455\n",
      "Epoch 368/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0786 - val_loss: 0.2104\n",
      "Epoch 369/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0771 - val_loss: 0.2000\n",
      "Epoch 370/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.2261\n",
      "Epoch 371/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0785 - val_loss: 0.2287\n",
      "Epoch 372/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0792 - val_loss: 0.2101\n",
      "Epoch 373/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0770 - val_loss: 0.2082\n",
      "Epoch 374/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0783 - val_loss: 0.1887\n",
      "Epoch 375/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0783 - val_loss: 0.2125\n",
      "Epoch 376/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0779 - val_loss: 0.2054\n",
      "Epoch 377/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0766 - val_loss: 0.2232\n",
      "Epoch 378/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0782 - val_loss: 0.2329\n",
      "Epoch 379/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0820 - val_loss: 0.2405\n",
      "Epoch 380/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0772 - val_loss: 0.2164\n",
      "Epoch 381/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0769 - val_loss: 0.2343\n",
      "Epoch 382/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0758 - val_loss: 0.2223\n",
      "Epoch 383/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0797 - val_loss: 0.2087\n",
      "Epoch 384/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0767 - val_loss: 0.2040\n",
      "Epoch 385/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0779 - val_loss: 0.2337\n",
      "Epoch 386/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0762 - val_loss: 0.2109\n",
      "Epoch 387/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0805 - val_loss: 0.2084\n",
      "Epoch 388/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0762 - val_loss: 0.2424\n",
      "Epoch 389/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0779 - val_loss: 0.2257\n",
      "Epoch 390/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0778 - val_loss: 0.2173\n",
      "Epoch 391/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0772 - val_loss: 0.2008\n",
      "Epoch 392/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.1914\n",
      "Epoch 393/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0776 - val_loss: 0.2101\n",
      "Epoch 394/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0800 - val_loss: 0.2012\n",
      "Epoch 395/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0924 - val_loss: 0.1837\n",
      "Epoch 396/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0847 - val_loss: 0.2336\n",
      "Epoch 397/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0785 - val_loss: 0.2304\n",
      "Epoch 398/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0800 - val_loss: 0.1949\n",
      "Epoch 399/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0760 - val_loss: 0.2391\n",
      "Epoch 400/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0759 - val_loss: 0.1951\n",
      "Epoch 401/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0768 - val_loss: 0.2048\n",
      "Epoch 402/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0767 - val_loss: 0.2291\n",
      "Epoch 403/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0752 - val_loss: 0.2218\n",
      "Epoch 404/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0788 - val_loss: 0.1906\n",
      "Epoch 405/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0803 - val_loss: 0.2165\n",
      "Epoch 406/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0756 - val_loss: 0.2153\n",
      "Epoch 407/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0737 - val_loss: 0.2036\n",
      "Epoch 408/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0760 - val_loss: 0.2275\n",
      "Epoch 409/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0764 - val_loss: 0.2267\n",
      "Epoch 410/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0744 - val_loss: 0.2098\n",
      "Epoch 411/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0814 - val_loss: 0.1879\n",
      "Epoch 412/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0807 - val_loss: 0.2496\n",
      "Epoch 413/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0768 - val_loss: 0.2202\n",
      "Epoch 414/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0747 - val_loss: 0.2190\n",
      "Epoch 415/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0761 - val_loss: 0.2037\n",
      "Epoch 416/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0756 - val_loss: 0.2010\n",
      "Epoch 417/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0752 - val_loss: 0.2122\n",
      "Epoch 418/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0809 - val_loss: 0.2002\n",
      "Epoch 419/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0844 - val_loss: 0.2116\n",
      "Epoch 420/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0838 - val_loss: 0.2382\n",
      "Epoch 421/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0793 - val_loss: 0.2389\n",
      "Epoch 422/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0796 - val_loss: 0.2017\n",
      "Epoch 423/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0757 - val_loss: 0.2059\n",
      "Epoch 424/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0785 - val_loss: 0.2037\n",
      "Epoch 425/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0778 - val_loss: 0.1787\n",
      "Epoch 426/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0760 - val_loss: 0.2268\n",
      "Epoch 427/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0797 - val_loss: 0.1995\n",
      "Epoch 428/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0777 - val_loss: 0.2550\n",
      "Epoch 429/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0757 - val_loss: 0.2049\n",
      "Epoch 430/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0746 - val_loss: 0.2018\n",
      "Epoch 431/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0734 - val_loss: 0.1978\n",
      "Epoch 432/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0732 - val_loss: 0.2111\n",
      "Epoch 433/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0739 - val_loss: 0.2351\n",
      "Epoch 434/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0732 - val_loss: 0.2124\n",
      "Epoch 435/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0735 - val_loss: 0.2126\n",
      "Epoch 436/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0768 - val_loss: 0.2361\n",
      "Epoch 437/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.2170\n",
      "Epoch 438/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0814 - val_loss: 0.2131\n",
      "Epoch 439/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0783 - val_loss: 0.1896\n",
      "Epoch 440/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0779 - val_loss: 0.1914\n",
      "Epoch 441/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0785 - val_loss: 0.2142\n",
      "Epoch 442/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0757 - val_loss: 0.2273\n",
      "Epoch 443/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0825 - val_loss: 0.1987\n",
      "Epoch 444/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.2262\n",
      "Epoch 445/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0745 - val_loss: 0.2287\n",
      "Epoch 446/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.2094\n",
      "Epoch 447/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0740 - val_loss: 0.2209\n",
      "Epoch 448/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0738 - val_loss: 0.2251\n",
      "Epoch 449/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.2172\n",
      "Epoch 450/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0758 - val_loss: 0.1958\n",
      "Epoch 451/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0744 - val_loss: 0.2106\n",
      "Epoch 452/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0767 - val_loss: 0.2043\n",
      "Epoch 453/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.2070\n",
      "Epoch 454/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0753 - val_loss: 0.2370\n",
      "Epoch 455/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0777 - val_loss: 0.2090\n",
      "Epoch 456/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0780 - val_loss: 0.1842\n",
      "Epoch 457/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0805 - val_loss: 0.2376\n",
      "Epoch 458/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0748 - val_loss: 0.2138\n",
      "Epoch 459/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0754 - val_loss: 0.1946\n",
      "Epoch 460/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0728 - val_loss: 0.2298\n",
      "Epoch 461/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0735 - val_loss: 0.2094\n",
      "Epoch 462/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0732 - val_loss: 0.2128\n",
      "Epoch 463/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0731 - val_loss: 0.2200\n",
      "Epoch 464/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0732 - val_loss: 0.1850\n",
      "Epoch 465/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0735 - val_loss: 0.2096\n",
      "Epoch 466/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0771 - val_loss: 0.2486\n",
      "Epoch 467/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0821 - val_loss: 0.2124\n",
      "Epoch 468/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0758 - val_loss: 0.2231\n",
      "Epoch 469/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0749 - val_loss: 0.2047\n",
      "Epoch 470/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0726 - val_loss: 0.2120\n",
      "Epoch 471/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0730 - val_loss: 0.2421\n",
      "Epoch 472/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0774 - val_loss: 0.2156\n",
      "Epoch 473/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0744 - val_loss: 0.2099\n",
      "Epoch 474/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0750 - val_loss: 0.2020\n",
      "Epoch 475/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0750 - val_loss: 0.2024\n",
      "Epoch 476/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0746 - val_loss: 0.2099\n",
      "Epoch 477/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0753 - val_loss: 0.2492\n",
      "Epoch 478/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0735 - val_loss: 0.1993\n",
      "Epoch 479/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0811 - val_loss: 0.1856\n",
      "Epoch 480/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0773 - val_loss: 0.2273\n",
      "Epoch 481/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0769 - val_loss: 0.2267\n",
      "Epoch 482/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0746 - val_loss: 0.2432\n",
      "Epoch 483/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0763 - val_loss: 0.1848\n",
      "Epoch 484/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0761 - val_loss: 0.2086\n",
      "Epoch 485/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0740 - val_loss: 0.2225\n",
      "Epoch 486/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0731 - val_loss: 0.1858\n",
      "Epoch 487/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0838 - val_loss: 0.2095\n",
      "Epoch 488/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0772 - val_loss: 0.2093\n",
      "Epoch 489/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.2132\n",
      "Epoch 490/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0766 - val_loss: 0.2265\n",
      "Epoch 491/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0771 - val_loss: 0.2139\n",
      "Epoch 492/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0737 - val_loss: 0.1956\n",
      "Epoch 493/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0739 - val_loss: 0.2362\n",
      "Epoch 494/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0770 - val_loss: 0.2398\n",
      "Epoch 495/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0737 - val_loss: 0.2213\n",
      "Epoch 496/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0723 - val_loss: 0.2154\n",
      "Epoch 497/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0758 - val_loss: 0.2030\n",
      "Epoch 498/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0727 - val_loss: 0.2040\n",
      "Epoch 499/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0763 - val_loss: 0.1898\n",
      "Epoch 500/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0733 - val_loss: 0.2053\n",
      "Epoch 501/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0715 - val_loss: 0.2147\n",
      "Epoch 502/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0718 - val_loss: 0.2074\n",
      "Epoch 503/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0727 - val_loss: 0.2292\n",
      "Epoch 504/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0742 - val_loss: 0.2049\n",
      "Epoch 505/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0744 - val_loss: 0.2428\n",
      "Epoch 506/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0726 - val_loss: 0.2150\n",
      "Epoch 507/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0745 - val_loss: 0.2408\n",
      "Epoch 508/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0724 - val_loss: 0.2131\n",
      "Epoch 509/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0729 - val_loss: 0.2299\n",
      "Epoch 510/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0707 - val_loss: 0.2169\n",
      "Epoch 511/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0706 - val_loss: 0.2482\n",
      "Epoch 512/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0739 - val_loss: 0.2600\n",
      "Epoch 513/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0813 - val_loss: 0.2274\n",
      "Epoch 514/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0750 - val_loss: 0.1932\n",
      "Epoch 515/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0767 - val_loss: 0.2122\n",
      "Epoch 516/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.1862\n",
      "Epoch 517/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0747 - val_loss: 0.2229\n",
      "Epoch 518/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0735 - val_loss: 0.2117\n",
      "Epoch 519/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0742 - val_loss: 0.2133\n",
      "Epoch 520/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0730 - val_loss: 0.2065\n",
      "Epoch 521/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0713 - val_loss: 0.2510\n",
      "Epoch 522/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0733 - val_loss: 0.2332\n",
      "Epoch 523/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0717 - val_loss: 0.2279\n",
      "Epoch 524/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0785 - val_loss: 0.1957\n",
      "Epoch 525/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0807 - val_loss: 0.1916\n",
      "Epoch 526/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0754 - val_loss: 0.2095\n",
      "Epoch 527/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0758 - val_loss: 0.2171\n",
      "Epoch 528/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0731 - val_loss: 0.2103\n",
      "Epoch 529/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0729 - val_loss: 0.2126\n",
      "Epoch 530/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0743 - val_loss: 0.1871\n",
      "Epoch 531/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0738 - val_loss: 0.2004\n",
      "Epoch 532/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0748 - val_loss: 0.2077\n",
      "Epoch 533/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0728 - val_loss: 0.2413\n",
      "Epoch 534/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0735 - val_loss: 0.2136\n",
      "Epoch 535/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0739 - val_loss: 0.2030\n",
      "Epoch 536/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2327\n",
      "Epoch 537/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0711 - val_loss: 0.2095\n",
      "Epoch 538/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0732 - val_loss: 0.2396\n",
      "Epoch 539/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0730 - val_loss: 0.1982\n",
      "Epoch 540/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2178\n",
      "Epoch 541/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0727 - val_loss: 0.2650\n",
      "Epoch 542/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0728 - val_loss: 0.2126\n",
      "Epoch 543/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0721 - val_loss: 0.2128\n",
      "Epoch 544/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0714 - val_loss: 0.2462\n",
      "Epoch 545/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0742 - val_loss: 0.2206\n",
      "Epoch 546/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0719 - val_loss: 0.2061\n",
      "Epoch 547/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0710 - val_loss: 0.2111\n",
      "Epoch 548/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0789 - val_loss: 0.2022\n",
      "Epoch 549/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0794 - val_loss: 0.2522\n",
      "Epoch 550/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0775 - val_loss: 0.2289\n",
      "Epoch 551/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0796 - val_loss: 0.2029\n",
      "Epoch 552/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0769 - val_loss: 0.2097\n",
      "Epoch 553/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0732 - val_loss: 0.2151\n",
      "Epoch 554/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0764 - val_loss: 0.2415\n",
      "Epoch 555/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0747 - val_loss: 0.2076\n",
      "Epoch 556/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0719 - val_loss: 0.2288\n",
      "Epoch 557/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0711 - val_loss: 0.2159\n",
      "Epoch 558/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0782 - val_loss: 0.2636\n",
      "Epoch 559/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0754 - val_loss: 0.2180\n",
      "Epoch 560/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0727 - val_loss: 0.2100\n",
      "Epoch 561/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0732 - val_loss: 0.2280\n",
      "Epoch 562/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0710 - val_loss: 0.2487\n",
      "Epoch 563/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2256\n",
      "Epoch 564/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2220\n",
      "Epoch 565/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2104\n",
      "Epoch 566/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0735 - val_loss: 0.2538\n",
      "Epoch 567/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0747 - val_loss: 0.2340\n",
      "Epoch 568/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0748 - val_loss: 0.1974\n",
      "Epoch 569/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2371\n",
      "Epoch 570/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0725 - val_loss: 0.2336\n",
      "Epoch 571/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0707 - val_loss: 0.2029\n",
      "Epoch 572/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0730 - val_loss: 0.2412\n",
      "Epoch 573/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0757 - val_loss: 0.2143\n",
      "Epoch 574/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0765 - val_loss: 0.1926\n",
      "Epoch 575/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0762 - val_loss: 0.2211\n",
      "Epoch 576/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0735 - val_loss: 0.2187\n",
      "Epoch 577/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0736 - val_loss: 0.1864\n",
      "Epoch 578/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0736 - val_loss: 0.2176\n",
      "Epoch 579/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0707 - val_loss: 0.2446\n",
      "Epoch 580/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2319\n",
      "Epoch 581/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0750 - val_loss: 0.2004\n",
      "Epoch 582/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0722 - val_loss: 0.2204\n",
      "Epoch 583/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0749 - val_loss: 0.2378\n",
      "Epoch 584/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0737 - val_loss: 0.2504\n",
      "Epoch 585/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0718 - val_loss: 0.2018\n",
      "Epoch 586/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0703 - val_loss: 0.2335\n",
      "Epoch 587/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0739 - val_loss: 0.2292\n",
      "Epoch 588/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0736 - val_loss: 0.2196\n",
      "Epoch 589/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0740 - val_loss: 0.2717\n",
      "Epoch 590/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0725 - val_loss: 0.2167\n",
      "Epoch 591/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0740 - val_loss: 0.2488\n",
      "Epoch 592/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0727 - val_loss: 0.2265\n",
      "Epoch 593/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0710 - val_loss: 0.2290\n",
      "Epoch 594/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0704 - val_loss: 0.2385\n",
      "Epoch 595/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0703 - val_loss: 0.2254\n",
      "Epoch 596/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0711 - val_loss: 0.2012\n",
      "Epoch 597/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0781 - val_loss: 0.1905\n",
      "Epoch 598/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0779 - val_loss: 0.2541\n",
      "Epoch 599/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0730 - val_loss: 0.2096\n",
      "Epoch 600/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0754 - val_loss: 0.2817\n",
      "Epoch 601/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0778 - val_loss: 0.2145\n",
      "Epoch 602/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0727 - val_loss: 0.2187\n",
      "Epoch 603/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2394\n",
      "Epoch 604/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2279\n",
      "Epoch 605/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0715 - val_loss: 0.2193\n",
      "Epoch 606/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0725 - val_loss: 0.2053\n",
      "Epoch 607/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0779 - val_loss: 0.2293\n",
      "Epoch 608/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0738 - val_loss: 0.2242\n",
      "Epoch 609/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0706 - val_loss: 0.2162\n",
      "Epoch 610/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0717 - val_loss: 0.2252\n",
      "Epoch 611/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0702 - val_loss: 0.2139\n",
      "Epoch 612/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2252\n",
      "Epoch 613/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0780 - val_loss: 0.1923\n",
      "Epoch 614/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0780 - val_loss: 0.2457\n",
      "Epoch 615/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0742 - val_loss: 0.1985\n",
      "Epoch 616/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0724 - val_loss: 0.2337\n",
      "Epoch 617/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0708 - val_loss: 0.1975\n",
      "Epoch 618/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0737 - val_loss: 0.2127\n",
      "Epoch 619/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0709 - val_loss: 0.2334\n",
      "Epoch 620/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2234\n",
      "Epoch 621/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0716 - val_loss: 0.1969\n",
      "Epoch 622/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2527\n",
      "Epoch 623/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0716 - val_loss: 0.2298\n",
      "Epoch 624/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0694 - val_loss: 0.2113\n",
      "Epoch 625/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0707 - val_loss: 0.2256\n",
      "Epoch 626/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2389\n",
      "Epoch 627/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0741 - val_loss: 0.2008\n",
      "Epoch 628/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0705 - val_loss: 0.2138\n",
      "Epoch 629/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2036\n",
      "Epoch 630/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2318\n",
      "Epoch 631/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2323\n",
      "Epoch 632/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0712 - val_loss: 0.2070\n",
      "Epoch 633/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0773 - val_loss: 0.2265\n",
      "Epoch 634/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2083\n",
      "Epoch 635/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0698 - val_loss: 0.2333\n",
      "Epoch 636/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0708 - val_loss: 0.2270\n",
      "Epoch 637/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0698 - val_loss: 0.2121\n",
      "Epoch 638/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0726 - val_loss: 0.2293\n",
      "Epoch 639/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0731 - val_loss: 0.2263\n",
      "Epoch 640/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2279\n",
      "Epoch 641/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0723 - val_loss: 0.2092\n",
      "Epoch 642/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0709 - val_loss: 0.2272\n",
      "Epoch 643/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0695 - val_loss: 0.2279\n",
      "Epoch 644/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0751 - val_loss: 0.2373\n",
      "Epoch 645/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0723 - val_loss: 0.2113\n",
      "Epoch 646/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0700 - val_loss: 0.1992\n",
      "Epoch 647/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0697 - val_loss: 0.2185\n",
      "Epoch 648/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0701 - val_loss: 0.2082\n",
      "Epoch 649/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0699 - val_loss: 0.2463\n",
      "Epoch 650/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2277\n",
      "Epoch 651/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0712 - val_loss: 0.1941\n",
      "Epoch 652/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0718 - val_loss: 0.2396\n",
      "Epoch 653/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2097\n",
      "Epoch 654/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0708 - val_loss: 0.2048\n",
      "Epoch 655/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0737 - val_loss: 0.2389\n",
      "Epoch 656/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0745 - val_loss: 0.2541\n",
      "Epoch 657/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0743 - val_loss: 0.2200\n",
      "Epoch 658/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0749 - val_loss: 0.2110\n",
      "Epoch 659/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0746 - val_loss: 0.2143\n",
      "Epoch 660/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0748 - val_loss: 0.1941\n",
      "Epoch 661/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0825 - val_loss: 0.2676\n",
      "Epoch 662/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0764 - val_loss: 0.2105\n",
      "Epoch 663/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0772 - val_loss: 0.2202\n",
      "Epoch 664/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0747 - val_loss: 0.2450\n",
      "Epoch 665/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0709 - val_loss: 0.2036\n",
      "Epoch 666/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0690 - val_loss: 0.2332\n",
      "Epoch 667/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0703 - val_loss: 0.2315\n",
      "Epoch 668/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0696 - val_loss: 0.2274\n",
      "Epoch 669/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2469\n",
      "Epoch 670/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0705 - val_loss: 0.2124\n",
      "Epoch 671/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0692 - val_loss: 0.2248\n",
      "Epoch 672/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0682 - val_loss: 0.2130\n",
      "Epoch 673/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0687 - val_loss: 0.2087\n",
      "Epoch 674/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0762 - val_loss: 0.2254\n",
      "Epoch 675/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0725 - val_loss: 0.2048\n",
      "Epoch 676/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0724 - val_loss: 0.2078\n",
      "Epoch 677/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2389\n",
      "Epoch 678/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0700 - val_loss: 0.2263\n",
      "Epoch 679/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0701 - val_loss: 0.2210\n",
      "Epoch 680/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2046\n",
      "Epoch 681/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0741 - val_loss: 0.2311\n",
      "Epoch 682/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0725 - val_loss: 0.1890\n",
      "Epoch 683/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0738 - val_loss: 0.2249\n",
      "Epoch 684/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0708 - val_loss: 0.2321\n",
      "Epoch 685/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0702 - val_loss: 0.2351\n",
      "Epoch 686/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0717 - val_loss: 0.2132\n",
      "Epoch 687/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0708 - val_loss: 0.2044\n",
      "Epoch 688/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0708 - val_loss: 0.2295\n",
      "Epoch 689/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2042\n",
      "Epoch 690/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0691 - val_loss: 0.2376\n",
      "Epoch 691/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0714 - val_loss: 0.2188\n",
      "Epoch 692/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0696 - val_loss: 0.2168\n",
      "Epoch 693/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0707 - val_loss: 0.2093\n",
      "Epoch 694/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0697 - val_loss: 0.2062\n",
      "Epoch 695/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0690 - val_loss: 0.2344\n",
      "Epoch 696/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0713 - val_loss: 0.2330\n",
      "Epoch 697/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0716 - val_loss: 0.2064\n",
      "Epoch 698/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0713 - val_loss: 0.2244\n",
      "Epoch 699/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0696 - val_loss: 0.2490\n",
      "Epoch 700/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0704 - val_loss: 0.2092\n",
      "Epoch 701/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0710 - val_loss: 0.2175\n",
      "Epoch 702/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2221\n",
      "Epoch 703/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0775 - val_loss: 0.1954\n",
      "Epoch 704/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0746 - val_loss: 0.2296\n",
      "Epoch 705/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0759 - val_loss: 0.1913\n",
      "Epoch 706/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0726 - val_loss: 0.2120\n",
      "Epoch 707/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0715 - val_loss: 0.2227\n",
      "Epoch 708/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0694 - val_loss: 0.2190\n",
      "Epoch 709/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0695 - val_loss: 0.2537\n",
      "Epoch 710/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0684 - val_loss: 0.2248\n",
      "Epoch 711/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0713 - val_loss: 0.2125\n",
      "Epoch 712/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0747 - val_loss: 0.2560\n",
      "Epoch 713/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0735 - val_loss: 0.1983\n",
      "Epoch 714/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0701 - val_loss: 0.2172\n",
      "Epoch 715/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0688 - val_loss: 0.2068\n",
      "Epoch 716/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0677 - val_loss: 0.2143\n",
      "Epoch 717/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2339\n",
      "Epoch 718/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0689 - val_loss: 0.1997\n",
      "Epoch 719/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0699 - val_loss: 0.2535\n",
      "Epoch 720/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0712 - val_loss: 0.1929\n",
      "Epoch 721/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0731 - val_loss: 0.2120\n",
      "Epoch 722/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0696 - val_loss: 0.2320\n",
      "Epoch 723/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0695 - val_loss: 0.2479\n",
      "Epoch 724/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2317\n",
      "Epoch 725/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0696 - val_loss: 0.2406\n",
      "Epoch 726/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0698 - val_loss: 0.2564\n",
      "Epoch 727/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2260\n",
      "Epoch 728/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0675 - val_loss: 0.2264\n",
      "Epoch 729/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2240\n",
      "Epoch 730/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0711 - val_loss: 0.2333\n",
      "Epoch 731/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0700 - val_loss: 0.2295\n",
      "Epoch 732/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2613\n",
      "Epoch 733/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0690 - val_loss: 0.2260\n",
      "Epoch 734/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0759 - val_loss: 0.1958\n",
      "Epoch 735/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0744 - val_loss: 0.2362\n",
      "Epoch 736/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0695 - val_loss: 0.2019\n",
      "Epoch 737/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0697 - val_loss: 0.2624\n",
      "Epoch 738/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0755 - val_loss: 0.2008\n",
      "Epoch 739/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0719 - val_loss: 0.2175\n",
      "Epoch 740/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0697 - val_loss: 0.2350\n",
      "Epoch 741/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0695 - val_loss: 0.2107\n",
      "Epoch 742/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0705 - val_loss: 0.2404\n",
      "Epoch 743/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2292\n",
      "Epoch 744/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0734 - val_loss: 0.2114\n",
      "Epoch 745/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0692 - val_loss: 0.2132\n",
      "Epoch 746/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0714 - val_loss: 0.3056\n",
      "Epoch 747/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0731 - val_loss: 0.2332\n",
      "Epoch 748/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0689 - val_loss: 0.2173\n",
      "Epoch 749/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0701 - val_loss: 0.2210\n",
      "Epoch 750/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0698 - val_loss: 0.2204\n",
      "Epoch 751/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0680 - val_loss: 0.2376\n",
      "Epoch 752/1000\n",
      "180/180 [==============================] - 1s 5ms/step - loss: 0.0685 - val_loss: 0.2372\n",
      "Epoch 753/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2423\n",
      "Epoch 754/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0682 - val_loss: 0.2218\n",
      "Epoch 755/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0686 - val_loss: 0.2100\n",
      "Epoch 756/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0705 - val_loss: 0.2054\n",
      "Epoch 757/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0682 - val_loss: 0.2285\n",
      "Epoch 758/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0679 - val_loss: 0.2112\n",
      "Epoch 759/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0705 - val_loss: 0.2138\n",
      "Epoch 760/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0699 - val_loss: 0.2206\n",
      "Epoch 761/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0687 - val_loss: 0.2393\n",
      "Epoch 762/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0694 - val_loss: 0.2232\n",
      "Epoch 763/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0681 - val_loss: 0.2388\n",
      "Epoch 764/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0679 - val_loss: 0.2116\n",
      "Epoch 765/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0735 - val_loss: 0.2261\n",
      "Epoch 766/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0727 - val_loss: 0.2298\n",
      "Epoch 767/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0703 - val_loss: 0.2137\n",
      "Epoch 768/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0725 - val_loss: 0.1995\n",
      "Epoch 769/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0696 - val_loss: 0.2488\n",
      "Epoch 770/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2044\n",
      "Epoch 771/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0727 - val_loss: 0.2159\n",
      "Epoch 772/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0692 - val_loss: 0.2406\n",
      "Epoch 773/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0690 - val_loss: 0.2132\n",
      "Epoch 774/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0669 - val_loss: 0.2417\n",
      "Epoch 775/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0721 - val_loss: 0.2321\n",
      "Epoch 776/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0697 - val_loss: 0.2181\n",
      "Epoch 777/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0689 - val_loss: 0.2181\n",
      "Epoch 778/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0697 - val_loss: 0.2057\n",
      "Epoch 779/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0696 - val_loss: 0.2444\n",
      "Epoch 780/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2051\n",
      "Epoch 781/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0720 - val_loss: 0.2512\n",
      "Epoch 782/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0698 - val_loss: 0.2473\n",
      "Epoch 783/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0676 - val_loss: 0.2279\n",
      "Epoch 784/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2471\n",
      "Epoch 785/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0680 - val_loss: 0.2408\n",
      "Epoch 786/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0711 - val_loss: 0.2260\n",
      "Epoch 787/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2416\n",
      "Epoch 788/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0689 - val_loss: 0.2665\n",
      "Epoch 789/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0706 - val_loss: 0.2505\n",
      "Epoch 790/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0694 - val_loss: 0.2374\n",
      "Epoch 791/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0689 - val_loss: 0.2280\n",
      "Epoch 792/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0702 - val_loss: 0.2057\n",
      "Epoch 793/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0707 - val_loss: 0.2288\n",
      "Epoch 794/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0679 - val_loss: 0.2415\n",
      "Epoch 795/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0671 - val_loss: 0.2554\n",
      "Epoch 796/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0688 - val_loss: 0.2402\n",
      "Epoch 797/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2448\n",
      "Epoch 798/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0682 - val_loss: 0.2183\n",
      "Epoch 799/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0672 - val_loss: 0.2431\n",
      "Epoch 800/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0690 - val_loss: 0.2052\n",
      "Epoch 801/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0688 - val_loss: 0.2455\n",
      "Epoch 802/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2433\n",
      "Epoch 803/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2495\n",
      "Epoch 804/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2273\n",
      "Epoch 805/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0767 - val_loss: 0.2324\n",
      "Epoch 806/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0710 - val_loss: 0.2340\n",
      "Epoch 807/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0716 - val_loss: 0.2695\n",
      "Epoch 808/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0712 - val_loss: 0.2108\n",
      "Epoch 809/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2087\n",
      "Epoch 810/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0712 - val_loss: 0.2369\n",
      "Epoch 811/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0699 - val_loss: 0.2159\n",
      "Epoch 812/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0682 - val_loss: 0.2023\n",
      "Epoch 813/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0760 - val_loss: 0.1950\n",
      "Epoch 814/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0715 - val_loss: 0.2046\n",
      "Epoch 815/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0676 - val_loss: 0.2164\n",
      "Epoch 816/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0670 - val_loss: 0.2386\n",
      "Epoch 817/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0674 - val_loss: 0.2186\n",
      "Epoch 818/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0677 - val_loss: 0.2355\n",
      "Epoch 819/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0679 - val_loss: 0.2393\n",
      "Epoch 820/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0674 - val_loss: 0.2055\n",
      "Epoch 821/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0677 - val_loss: 0.2342\n",
      "Epoch 822/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0705 - val_loss: 0.2206\n",
      "Epoch 823/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0674 - val_loss: 0.2180\n",
      "Epoch 824/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0744 - val_loss: 0.2372\n",
      "Epoch 825/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0726 - val_loss: 0.2604\n",
      "Epoch 826/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0711 - val_loss: 0.2522\n",
      "Epoch 827/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0710 - val_loss: 0.2179\n",
      "Epoch 828/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0690 - val_loss: 0.2117\n",
      "Epoch 829/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0686 - val_loss: 0.2073\n",
      "Epoch 830/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2386\n",
      "Epoch 831/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0701 - val_loss: 0.2515\n",
      "Epoch 832/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0673 - val_loss: 0.2480\n",
      "Epoch 833/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0732 - val_loss: 0.2194\n",
      "Epoch 834/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0711 - val_loss: 0.2201\n",
      "Epoch 835/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0684 - val_loss: 0.2007\n",
      "Epoch 836/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0696 - val_loss: 0.2482\n",
      "Epoch 837/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0674 - val_loss: 0.2228\n",
      "Epoch 838/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0677 - val_loss: 0.2322\n",
      "Epoch 839/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0671 - val_loss: 0.2244\n",
      "Epoch 840/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0670 - val_loss: 0.2379\n",
      "Epoch 841/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2391\n",
      "Epoch 842/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0721 - val_loss: 0.2097\n",
      "Epoch 843/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0712 - val_loss: 0.2365\n",
      "Epoch 844/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0711 - val_loss: 0.2167\n",
      "Epoch 845/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0725 - val_loss: 0.2041\n",
      "Epoch 846/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0701 - val_loss: 0.1995\n",
      "Epoch 847/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0737 - val_loss: 0.2220\n",
      "Epoch 848/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0687 - val_loss: 0.2259\n",
      "Epoch 849/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2470\n",
      "Epoch 850/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0697 - val_loss: 0.2177\n",
      "Epoch 851/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0686 - val_loss: 0.2302\n",
      "Epoch 852/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0697 - val_loss: 0.2505\n",
      "Epoch 853/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0681 - val_loss: 0.2457\n",
      "Epoch 854/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0679 - val_loss: 0.2555\n",
      "Epoch 855/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0679 - val_loss: 0.2541\n",
      "Epoch 856/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0681 - val_loss: 0.2632\n",
      "Epoch 857/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0720 - val_loss: 0.2068\n",
      "Epoch 858/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0704 - val_loss: 0.2490\n",
      "Epoch 859/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2540\n",
      "Epoch 860/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0682 - val_loss: 0.2393\n",
      "Epoch 861/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0669 - val_loss: 0.2293\n",
      "Epoch 862/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0669 - val_loss: 0.2379\n",
      "Epoch 863/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0705 - val_loss: 0.2649\n",
      "Epoch 864/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0724 - val_loss: 0.2045\n",
      "Epoch 865/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0674 - val_loss: 0.2380\n",
      "Epoch 866/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0662 - val_loss: 0.2503\n",
      "Epoch 867/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0697 - val_loss: 0.1972\n",
      "Epoch 868/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0696 - val_loss: 0.2156\n",
      "Epoch 869/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0668 - val_loss: 0.2463\n",
      "Epoch 870/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0699 - val_loss: 0.2898\n",
      "Epoch 871/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0694 - val_loss: 0.2409\n",
      "Epoch 872/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0673 - val_loss: 0.2473\n",
      "Epoch 873/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0679 - val_loss: 0.2382\n",
      "Epoch 874/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0707 - val_loss: 0.2308\n",
      "Epoch 875/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0683 - val_loss: 0.2425\n",
      "Epoch 876/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0680 - val_loss: 0.2352\n",
      "Epoch 877/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0671 - val_loss: 0.2253\n",
      "Epoch 878/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0677 - val_loss: 0.2376\n",
      "Epoch 879/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0686 - val_loss: 0.2501\n",
      "Epoch 880/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0688 - val_loss: 0.2178\n",
      "Epoch 881/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0673 - val_loss: 0.2376\n",
      "Epoch 882/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0660 - val_loss: 0.2417\n",
      "Epoch 883/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0687 - val_loss: 0.2557\n",
      "Epoch 884/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0680 - val_loss: 0.2287\n",
      "Epoch 885/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0671 - val_loss: 0.2480\n",
      "Epoch 886/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0689 - val_loss: 0.2373\n",
      "Epoch 887/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0669 - val_loss: 0.2479\n",
      "Epoch 888/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0674 - val_loss: 0.2314\n",
      "Epoch 889/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0686 - val_loss: 0.2823\n",
      "Epoch 890/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0691 - val_loss: 0.2301\n",
      "Epoch 891/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0712 - val_loss: 0.2242\n",
      "Epoch 892/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0700 - val_loss: 0.2397\n",
      "Epoch 893/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0675 - val_loss: 0.2516\n",
      "Epoch 894/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0677 - val_loss: 0.2350\n",
      "Epoch 895/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0666 - val_loss: 0.2418\n",
      "Epoch 896/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0698 - val_loss: 0.2544\n",
      "Epoch 897/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0679 - val_loss: 0.2414\n",
      "Epoch 898/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0662 - val_loss: 0.2397\n",
      "Epoch 899/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0664 - val_loss: 0.2475\n",
      "Epoch 900/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0659 - val_loss: 0.2428\n",
      "Epoch 901/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0680 - val_loss: 0.2150\n",
      "Epoch 902/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0670 - val_loss: 0.2416\n",
      "Epoch 903/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0658 - val_loss: 0.2332\n",
      "Epoch 904/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0680 - val_loss: 0.2379\n",
      "Epoch 905/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0688 - val_loss: 0.2533\n",
      "Epoch 906/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0689 - val_loss: 0.2269\n",
      "Epoch 907/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0676 - val_loss: 0.2407\n",
      "Epoch 908/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0682 - val_loss: 0.2218\n",
      "Epoch 909/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0683 - val_loss: 0.2166\n",
      "Epoch 910/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0665 - val_loss: 0.2298\n",
      "Epoch 911/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0656 - val_loss: 0.2305\n",
      "Epoch 912/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0720 - val_loss: 0.2523\n",
      "Epoch 913/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0710 - val_loss: 0.2430\n",
      "Epoch 914/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0737 - val_loss: 0.2596\n",
      "Epoch 915/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0725 - val_loss: 0.2199\n",
      "Epoch 916/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0687 - val_loss: 0.2382\n",
      "Epoch 917/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2584\n",
      "Epoch 918/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0690 - val_loss: 0.2275\n",
      "Epoch 919/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0715 - val_loss: 0.2082\n",
      "Epoch 920/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0710 - val_loss: 0.2531\n",
      "Epoch 921/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0698 - val_loss: 0.2256\n",
      "Epoch 922/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2321\n",
      "Epoch 923/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0675 - val_loss: 0.2413\n",
      "Epoch 924/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0743 - val_loss: 0.2784\n",
      "Epoch 925/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0709 - val_loss: 0.2200\n",
      "Epoch 926/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0686 - val_loss: 0.2540\n",
      "Epoch 927/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0678 - val_loss: 0.2319\n",
      "Epoch 928/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0655 - val_loss: 0.2269\n",
      "Epoch 929/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0669 - val_loss: 0.2371\n",
      "Epoch 930/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0654 - val_loss: 0.2310\n",
      "Epoch 931/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0678 - val_loss: 0.2381\n",
      "Epoch 932/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0672 - val_loss: 0.2512\n",
      "Epoch 933/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0719 - val_loss: 0.2141\n",
      "Epoch 934/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0677 - val_loss: 0.2509\n",
      "Epoch 935/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0668 - val_loss: 0.2484\n",
      "Epoch 936/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2705\n",
      "Epoch 937/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0688 - val_loss: 0.2493\n",
      "Epoch 938/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0681 - val_loss: 0.2400\n",
      "Epoch 939/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0685 - val_loss: 0.2391\n",
      "Epoch 940/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0663 - val_loss: 0.2758\n",
      "Epoch 941/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0667 - val_loss: 0.2240\n",
      "Epoch 942/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0662 - val_loss: 0.2389\n",
      "Epoch 943/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0669 - val_loss: 0.2472\n",
      "Epoch 944/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2516\n",
      "Epoch 945/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0676 - val_loss: 0.2235\n",
      "Epoch 946/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0683 - val_loss: 0.2283\n",
      "Epoch 947/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0667 - val_loss: 0.2341\n",
      "Epoch 948/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0670 - val_loss: 0.2511\n",
      "Epoch 949/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0656 - val_loss: 0.2396\n",
      "Epoch 950/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0650 - val_loss: 0.2236\n",
      "Epoch 951/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0675 - val_loss: 0.2395\n",
      "Epoch 952/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0718 - val_loss: 0.2627\n",
      "Epoch 953/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2227\n",
      "Epoch 954/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0676 - val_loss: 0.2235\n",
      "Epoch 955/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0711 - val_loss: 0.2297\n",
      "Epoch 956/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0681 - val_loss: 0.2271\n",
      "Epoch 957/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0666 - val_loss: 0.2284\n",
      "Epoch 958/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0669 - val_loss: 0.2377\n",
      "Epoch 959/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0674 - val_loss: 0.2661\n",
      "Epoch 960/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0664 - val_loss: 0.2834\n",
      "Epoch 961/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0689 - val_loss: 0.1934\n",
      "Epoch 962/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0677 - val_loss: 0.2324\n",
      "Epoch 963/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0701 - val_loss: 0.2147\n",
      "Epoch 964/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0722 - val_loss: 0.2147\n",
      "Epoch 965/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0691 - val_loss: 0.2406\n",
      "Epoch 966/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0680 - val_loss: 0.2293\n",
      "Epoch 967/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0694 - val_loss: 0.2538\n",
      "Epoch 968/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0678 - val_loss: 0.2292\n",
      "Epoch 969/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0692 - val_loss: 0.2758\n",
      "Epoch 970/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0688 - val_loss: 0.2383\n",
      "Epoch 971/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0669 - val_loss: 0.2408\n",
      "Epoch 972/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0666 - val_loss: 0.2541\n",
      "Epoch 973/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0675 - val_loss: 0.2136\n",
      "Epoch 974/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0685 - val_loss: 0.2107\n",
      "Epoch 975/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0669 - val_loss: 0.2437\n",
      "Epoch 976/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0677 - val_loss: 0.2464\n",
      "Epoch 977/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0666 - val_loss: 0.2605\n",
      "Epoch 978/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0664 - val_loss: 0.2520\n",
      "Epoch 979/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0652 - val_loss: 0.2512\n",
      "Epoch 980/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0657 - val_loss: 0.2285\n",
      "Epoch 981/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0645 - val_loss: 0.2709\n",
      "Epoch 982/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0675 - val_loss: 0.2452\n",
      "Epoch 983/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0662 - val_loss: 0.2497\n",
      "Epoch 984/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0674 - val_loss: 0.2400\n",
      "Epoch 985/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0668 - val_loss: 0.2591\n",
      "Epoch 986/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0664 - val_loss: 0.2069\n",
      "Epoch 987/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0670 - val_loss: 0.2463\n",
      "Epoch 988/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0673 - val_loss: 0.2396\n",
      "Epoch 989/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0662 - val_loss: 0.2277\n",
      "Epoch 990/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0651 - val_loss: 0.2198\n",
      "Epoch 991/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0655 - val_loss: 0.2324\n",
      "Epoch 992/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0691 - val_loss: 0.2090\n",
      "Epoch 993/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0690 - val_loss: 0.2451\n",
      "Epoch 994/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0736 - val_loss: 0.2057\n",
      "Epoch 995/1000\n",
      "180/180 [==============================] - 1s 4ms/step - loss: 0.0737 - val_loss: 0.2469\n",
      "Epoch 996/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0734 - val_loss: 0.2148\n",
      "Epoch 997/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0713 - val_loss: 0.2363\n",
      "Epoch 998/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0693 - val_loss: 0.2652\n",
      "Epoch 999/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0678 - val_loss: 0.2444\n",
      "Epoch 1000/1000\n",
      "180/180 [==============================] - 1s 3ms/step - loss: 0.0662 - val_loss: 0.2396\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.callbacks.History at 0xb75575e90>"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "autoencoder = Sequential()\n",
    "train_shape = train_list.shape\n",
    "# Encoder Layers\n",
    "autoencoder.add(Conv2D(2, (2, 2), activation='relu', padding='same', input_shape=[ 200, 200, 1]))\n",
    "autoencoder.add(Conv2D(3, (2, 2), activation='relu', padding='same'))\n",
    "autoencoder.add(BatchNormalization())\n",
    "autoencoder.add(MaxPooling2D((2, 2), padding='same'))\n",
    "autoencoder.add(Conv2D(3, (2, 2), activation='relu', padding='same'))\n",
    "autoencoder.add(Conv2D(3, (2, 2), activation='relu', padding='same'))\n",
    "autoencoder.add(BatchNormalization())\n",
    "\n",
    "autoencoder.add(Flatten())\n",
    "autoencoder.add(Dense(3, name='bneck'))\n",
    "autoencoder.add(Dense(30000,  activation='relu'))\n",
    "print (autoencoder.summary())\n",
    "autoencoder.add(Reshape( (100, 100, 3) ))\n",
    "\n",
    "print (autoencoder.summary())\n",
    "\n",
    "autoencoder.add(BatchNormalization())\n",
    "\n",
    "autoencoder.add(Conv2D(5, (3, 3), activation='relu', padding='same'))\n",
    "autoencoder.add(Conv2D(5, (3, 3), activation='relu', padding='same'))\n",
    "autoencoder.add(UpSampling2D((2, 2)))\n",
    "autoencoder.add(BatchNormalization())\n",
    "autoencoder.add(Conv2D(5, (3, 3), activation='relu', padding='same'))\n",
    "autoencoder.add(Conv2D(1, (2, 2), activation='sigmoid', padding='same'))\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "encoder = Model(inputs=autoencoder.input, outputs=autoencoder.get_layer('bneck').output)\n",
    "\n",
    "\n",
    "\n",
    "autoencoder.compile(optimizer=sgd, loss='binary_crossentropy', )\n",
    "\n",
    "print (autoencoder.summary())\n",
    "autoencoder.fit(train_list[:180], train_list[:180],\n",
    "                epochs=1000,\n",
    "                batch_size=30,\n",
    "                validation_data=(train_list[180:], train_list[180:]))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Notes:\n",
    "    The number of botle neck nodes are limitating to the compression\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoded_imgs = encoder.predict(train_list)\n",
    "decoded_imgs = autoencoder.predict(train_list)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_plot = pd.DataFrame(\n",
    "    {'x':[x [2] for x in encoded_imgs.tolist()],\n",
    "                        'y':[x [0] for x in encoded_imgs.tolist()],\n",
    "                         'img':['http://0.0.0.0:8889/{}.png'.format(str(i)) for i in range(0, len(encoded_imgs)  )] }\n",
    "                        )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQUAAAD8CAYAAAB+fLH0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzsvXmQXdl52Pc759zt7d2vVzS2HmAwg9kxw+Fwl7iIpEhRluTIKtHWRsuWkpLKVpWTWKWK46SUuOxEscslJ4roRIloy2RYiSRqobmIIsWhORrOwlkxAAZ7Aw309np56333nnPyx73voTHYGuhuoBtzf1UodN9+795z7z3nO9/3ne98n7DWkpGRkdFD3ukGZGRkbC0yoZCRkXEZmVDIyMi4jEwoZGRkXEYmFDIyMi4jEwoZGRmXsWlCQQjxw0KIo0KI40KIX9+s62RkZGwsYjPiFIQQCjgGfBQ4BzwHfNpae3jDL5aRkbGhbJam8BRw3Fp70lrbBb4A/NgmXSsjI2MDcTbpvDuBqVW/nwPeda0PD1eVndztblJTMjIyLJYXX+nOW2tHbvTZzRIK4irHLrNThBC/BPwSwJ6dDt/76u5NakpGRgaA2nH8zFo+t1nmwzlg9SjfBUyv/oC19rPW2iettU+ODKlNakZGRsbNsllC4TnggBDiHiGEB/w08CebdK2MjIwNZFPMB2ttLIT4VeCrgAJ+z1r7+mZcKyMjY2PZLJ8C1tovA1/erPNnZGxltDUAKLH94gO3X4szMjI2lUwoZGRsAkrIdWkJPU3jTpAJhYyMLcidNDsyoZCRkXEZmVDIyMi4jEwoZGRkXEYmFDIyMi4jEwoZGRmXkQmFjIyMy8iEQkZGxmVkQiEjI+MyMqGwzbiTkW4Z62O7vLtMKGRkZFxGJhTuAKGNbmnW0NbQtl0iqzehVRlXY6Nm99672w5s2tbpjGvjoG4ptl0JSQ5vW27H3a5s1LNWQuKyPTKMZb3rDpAN6s1jK9vtzjYRCpmmcBuJrEYiMqGwCURWYzBI5JYderfy3lcLudvVb7LeeZvQ1mAw636xmUC5ktBG/O9L+/hOJ6BluxuiLWwVjeOCbnFBt5jVLZZN+7a065Z7mBBitxDim0KIN4QQrwsh/mF6/L8TQpwXQryU/vvkxjV3e6M3oRrX1a+zNTr07UJby9mwSmSdtxw3/X+R1Tf1XLaK8K1Kj6r0qEiPQNwexX49V4mBf2StfVEIUQJeEEJ8Pf3bv7LW/tb6m7f90Pbq2kCMpmUj8nh3tB13Iw0b4QrNiKoTCOeK+w5tzLLpMqNdHvIcXLFVDYwrycukv9zOnI+3LBSstReAC+nPdSHEGySVod629Oxa7JUvzxcujrw9nfHtIgx6VKTHfz70XXaq/GX33vs5Lzzy0qMkO0RWbyuh0ON2vtMNuZIQYhJ4HHg2PfSrQohXhBC/J4QY3IhrbGW0NTRMh9YNYgi2ymDtqdPX+tt2wxcue5ziDZ+vKxQRmtBGt6ll25N191IhRBH4/4Bfs9auAL8D7AcOkWgS/8s1vvdLQojnhRDPzy1s72AcJSS+cAmEgy/cLTP4r4USEnmVyn6JM/T2+D3uFOqqFQ0zVrOu3iuEcEkEwh9Ya/8QwFo7Y63V1loD/FuSCtRXcLeVjXOFwkFddbBdj+vN2pvJVhdcm4FE9t/TRnGrjsytzHpWHwTwfwJvWGv/5arjO1Z97CeA1269eRkZGbeb9aw+vA/4WeBVIcRL6bHfAD4thDhEUmX6NPDL62rhNuJmZ9+tNrMosXUDfzaCxMG4sXfYe+exjQitpiiCDT3/nWA9qw/f4eol57ddqbjVtvTt8kxrawhtjBICmcWQ3RXM6BjokBO3vj9lK5Sby8KcuTMzpBKSvLg9MQsZa2M9sR2JPwk0FoO95f5ksDftl9poMqGwTdiu6+vbifXOzhGC0Bpc1vauIqtZNh3ywkUJQcfGhNZQFC4+VwZh3S4yobBNuNOzR8b1UUKyS7npz2JNWkdkNdNasVvF+DgEwiEQ3HFzMjNm34K2hnndvOVEKDc6d+/fzdLrYFvNOZlxibz08IWz5iXP0MbUdB6NxRUKX7j4wsUVt5ZvY6O464XCzcQARFYnzj8S599WXMu/24OLtho3K4R71aZv1He0NUghyMuQQKgtZRpuvV6/wTRMuOYXu2w6RGjy0t3Ul7QeYSMRmbZwGzHYyzS8ZNUoomE6GxJ0VjcB0RZ7n3e9TyEv1xZ2nESlWdxNtOk2QvPYitrL3cxqX07v2SvkLafUW32uisjxwSBCify627mR3PU9bC32XWQ1DRsyrT2Wjc4So2b0uZY5sJG5G7caW69FG8xaHrrBYKxlQHapyK1l32Vk3G7ueqGQkZFxc9z1PoW14KDIS5Aipij8LanSbVfuROLRjPWRCQV6Yc7rdx5lXKInDHphu1vlua7e57KV2rWVyJ7IKrZDB+nt29/Ky5LaGho2JGbrpbQ32H6GrJuJ+djKz3ujyTSFbcbqJbKtsKPuavSW27YirlAU8W9649FWe8abydvmTu+WvHyrl8diNDGaWd18W81k62X187vRc9vqWtlm8LYRCh0b31XxBwZLy0S0TMTSbeqzd8vg6PWDG4Wyhzbigm697Yr6vm2EwnYp7rkWtDW0bJc5Y5kzlo7d/Hvr+Qm2u2DoPTuDuW48iraGGR3yB8uPcya2byuh8LbyKRgM2m4tx9etoIQkj8eIDNMjm1/4RQlJke2/XLtWf4cSkj1OkV+rHkbiva0C2t42QuGC7lKS4o4nsNgoXKEoSh9IhN3t4HY9s63kQPWFu2Hn2i5Vu9YtFIQQp4E6oIHYWvukEKIK/D/AJEny1p+y1i6u91rrYUw5/fTe2+HFrIVLs9fdM4tFVtOyXfLCu4vuKiFGp0nbtjYb1cIPWWsPWWufTH//deAb1toDwDfS3+8oRRlsWKGWt6NH+nagrWHRdDgeqQ2rHr2VmNMhoY22/ErYZomtHwN+P/3594Ef36Tr3BHWkkQj4+ZRQlJJC6rWtKZtu3e4RRvLiPJx2NhiNJvBRvRsC3xNCPGCEOKX0mNjaQHaXiHa0bd+6W4qG5eRcTexEY7G91lrp4UQo8DXhRBH1vIla+1ngc8CPPlYkOUYywASx96AbFC3Dq3Ycq8bbaiz706yXe5j3ZqCtXY6/X8W+COS2pEzvfJx6f+z671OxvakZW7eNyCBERmzy9k+A+luYr0FZgtCiFLvZ+BjJLUj/wT4+fRjPw98aT3Xydh8NsOpF1nN812PBdO+6eCfQEgC8bZZMd9SrPepjwF/lNSaxQH+g7X2K0KI54AvCiF+ETgL/K11XmfN9LbGvp2CTdZL4vVvMyhzG+pAlQge89rkRXBT72PCSeIvtrpD7m5lXULBWnsSeOwqxxeAj6zn3LfKed0iL8SGd/C7lVnd5M0ox5we4xP5xQ1fR7+VpDU3azJspWCnu4G7Tj8blh5KZNWU1kpeKO5329zvtvFFYUPPfbsGaSYMNpa7TijkZVa09WbICY/cJsjQ7RLSm3Eld51QuB6Zv+FKNmvg3sp5M0GyNbjrhUKSdiuxObW1hDbGFZJcWgY+64QZGZdz1wuFRCtINQMBeTLzYquSCeitQfYWMjIyLiMTChkZGZeRCYWMLcvdtnV6u5AJhYwtS2jjTDDcATKhkLFlyYLQ7gx3/erDtchCY7c+2Q7JO8PbVihkwiAj4+pkIyMjI+MyMqGQsWFkCW3vDt625sN2RltDaGP8NAnJVjCFeqnZgbuiaMzbmUwobDN6AqFmulTTcbcVitu4Qm16CHnmHL49ZEJhm7FiOrwZuxibo56WjRuQXUrSISe8OzpgNnv3aUya0s1mgmEzyYTCNqFX4PUPG/t4tbWLUbfOE/nTACi3RmRiSsJQlsFdO2Bk6gK7W+9vq3DLQkEIcT9Jabge+4D/FhgA/j4wlx7/DWvtl2+5hRkZGbeVWxYK1tqjwCEAIYQCzpOkeP8M8K+stb+1IS18m9PzIZyKNb918WN85+R+jBaMDNXZtz/JnO+6CwBMaUkUx1Rll3vc4p1s9obSS47TywidJcnZXDbKfPgIcMJae0ZkoakbQm8gLJsOL3fLPNM8wBu1MeJlj9xICyUNSzoPwJzxaRkfKQzPNA/gy4i/XX6dUbWxORfvFN9o+wypZv/3B9zuXZ92L7IaibgjptJGCYWfBj6/6vdfFUL8HPA88I+uVnE6LTH3SwB7dmaujdWENuKNruHzS+9iPizy7ZP3EnccRFOhuoLJoRrLYcBvv/FBAHYPLnH01A6kp8m9niMcsPzBw0/xmwe/xIdyjW0fLrxk8jzTPMB0WAHgH4z+5V1VOepqLJsOhiSxLiQFkm8Xwtr1VWwTQnjANPCQtXZGCDEGzJPUmPxNYIe19u9e7xxPPhbY73119w2v1ZOeV2M7O596KeM6NgZgOrZ8YfmdfO7Fd5M74ZO/YMktGrQnCMuC+JNLjJYaXPyPyTMrThtysxHeQgfrSjpjOVb2Ojg/PM8fPPJ/cY8T3LFZZyM4FTX4s8ZDvFRP7vfvjn6b+92Nr1OxlQhthF41NjdCM1I7jr+wqjL8NdmIKfoTwIvW2hmA3v8AQoh/C/zZei/Qs6vPxDGeMFSlxBcOSggkEom4IpJuu3QWbU0a+BNxLk5ex/+7/CR/MX0/YsWlctIw+L0ZhDaY+Rrs28WRJ4vsr85zfF8iRHLzCm+5S2dHHmEgf7ZO8fU2c40d/NnkI3y8+DqRlbgieUa7nFurx3CniBBc6FZo60QzOB2NUBDTBG6Xorh9M+jt5k7tEt0IofBpVpkOQogdvYrTwE+QlJG7ZbQ1tG2XltUsm4AR1UYK0RcIBkO0SqImgT0GV8AOlbtjTqm1ZiZWQuKiCDCUZATAUpRn5nSV8imJUZZ4pIxqhogVl7jgMTyxzHuqJ3nPD5wE4IXH9vLskX3QFZTedBhr+LjT81ROdPjt5z7EX0we5NDAOQ4VzgAwoqYpbiPXT00HTAbzPJqfAuCAN0NJRrSswLURDmrbCLi1cidNo3UJBSFEHvgo8MurDv9PQohDJObD6bf8bc0smzYANa2Z0TmatsCCLgI18qINEhSCuolpWVgwSamx7zQf5qszDzK9UuafPPTnfDg3zfBtdLg1TIe6iVFCMCRzwI21ll54cO8evnbifnZ8WzLw6gLheJHaw3m8eo6B2FB7MM/Hdn6fJ3KnKYjke0/kTvNwaZrnF/fycrALbyXHaG0QZ7HN3i+WmBvfyx//SBX2J9f7cG56k+5+c5hw2uTlSQoi0YwqUuAKRctopuOQghT4Qt6y9hPau9s/cbOst2xcCxh6y7GfXVeLSGzsZzoDAAQiIi9DBkSbQEQsmAJL3TyBiOhYl4txhbrOcaGbOKH+8sJ9zEwPQCx5Y99OHvIuMihvXz2BuomJgLV2sQtxgz9sPMCIs8J/uPBuAArfLjLw/AVs3qe+22P5Q22GBhpUB+b5ldE/4UPBSrrvIZnuYzoMyFc40x5ictc8Z35giDg3THkqxgqBVaC1pGXStPbbzL+wxymy0xrUW0yFvNCcijvMaYnEcq9rb6n6ZCYQLmfLuf21NXwvFPzPpz8OwM/v+i7vDM4yIA3Q5eVujq8vP8TryztYaCUagLGwuFACwLnoUawJ4hw8W5tkl1djQp1kUOVveO23OjJj9FVV056PwxXqMgdeZDWBkGhrqBtLPp3Z8uJKJ1FkNWfjNp858rNMnRkGacmfSD5XvaAxAwWaewosPGr5xIE3eLQ4xY8X32RQBrhvOZ9Ccr/bYSJYoh77sAPOvWeQ1kSOyjHoVgRCWM42BwGoW0P5OubNVtxjcLW29EzD17vjKCx7nfkshmED2HJCIUazoCs0u0nHr5scBoEGAiEoyTZDbpOd+WWksLQij1ArbJR0GqcpkBGYkuVivcR3cgcIZMSHc2eoKv+qg7zn/Y+s7neqjo0JrWFQBiguDRSDZdF0ONwtUVUtqjKmKj1cofrBNdPa59XObg4FZwHY57SQQhBaw1Tscj4e4GR3lH938imWjwwxdBRkDLmF1HF4sU1U8Zl9XDJ0/zxDXoOybJMX6pqdvigDfmP4JWqDIYejCl+uPsbXcgepBWWEtpi2y4naMADfGp3kk4Uz1zWrQhtfVZhtNfY7OXapuXTX6N3rdLydbDmhIJEMqQbLzcQef725k4P+NJFtUZIRTeNTcVrsydUAuEAZIVxyg4kPotNSYCRxyRJpxXPTe/jrqUn+mTS8d9dp3l0+waQ3x4BMPn86GmYqqvKgf54D7iJKxCwZh5PRCBPOItLpUJEBoU2cgFOx4XcXfpAvv/kguSBif3Wen9vxDE/6F1FAy8Kfrxzimxfv4+WhZAntieIZanGRby8c4Nzn9jHyhVcwzSbjkx2GdrRY2Z+j+tw8NpcMQvvGCYIdY3R/cpyJ4gqN2Kdj3X7s/7XwhcsOx6WqOgSDz/Pa0gS1AzFLK3ls06W+kjzTP5k7xLnuEJ8ZeIFRlb9CSIY2ZkZ3mUi93xutXm9keTglJHnhZUV+NpCtox9mZGRsCbacpgCgsPRWGV9emCAnH2MyWGDEWWHFJLNdXoXsydU4VJpizF1CkXxh7sES31q8n+lGBW0F9Y6PlAYlLAUnpCBDzkeDnLBjAPzZ7KO8dmYC6Rge3XWe91dPUNcBC1GByWCBk+4ik+48p6MdAHxz+SCvLEwQLfuY83leWsoz3agwWmgQG0mj6zO9UEEKy3NmDwBFFRJZRTP28OsWG8c4OyeY/uQumrsshXOCcFcFdznZCi3uv4eZdw7yiSe/z0cqhynIkBFVX/O6tUQSiJhG5GEsCGEhFlibmB71bsBinOe5cJR3+rN9sypGM6dDasahbvIMyA4AjtzYJb+t5KvIuJItJxRCG/FSZz+Ok9jwc4sl/nTpYe4dm+eB8kVGvTp52aUkO1RVk0PBWSZUl6Df0S5yKDjLXzQe4nuLk3RjByEsShq6xqFpfGq6wJFGMsiPfWsf+77VwZtuULt3kt/+G5OM7V5ECMuLYjfzKwW6M3m8WjKgopLB5AzegqI4BXHg45+oEk256KLPhR/NkasJumXLvJs4N/9j1+WR0WlGggbff1DiLz2CCjX1fQYzHOEfdpl+v8/gkURNb05I9v74SX5t9BuMKYlJJaTD2mzmyGqOdHdQ7/h02h5R00VEApvKlOUw4EyryonG+/hfI5/lMKAbK3w3xlea+wdmeLgwzYA8AoAruuS4dq6GreCYvJU2ZFWur86WEAoW2/fo/0V7mD++eIjWSjoAll1wLOGwgyMNDwTnmXRrVGWMJwQuAn+VQ8xgqMgQV2gCFeEojbUCbWTf+w6wECZONrcJbq0NWqN9ifBjlDRpu6Db8nDqEpVMmuicwD/nMPimoXSijnUVarHFiZ8bIXhoCTeOCY+XKJ5JnIcAS84A8+UVqn6L+L4WZ4sBsutifAPLLvOPCcr3L9B4POmgP3HPK/zcwLOMKQdfuIREaNYWjh5Zzfe7Dt9aOogjDdYKMCIxFP3EERppyXSjQmwkkZbEOtGzRKz6z+acO8hpP3leeTHPgDT99G9webRdL6r0aqx14N3qoA5tjMFQM70YBkUgnH5gW++ZSORlEbAGi8GgMgv6CraEUIixHIs6/LvFd/OVqQdYnCnjLiRNc5cFOmephz6xkexxFtnngC9y/Y4Yo/tx4pE1hFbRMS6e1ChhCbVCG8Fss8hKGKCtoBMl52/u0sw9NcDgG21aw5J8uYObCoVO7OD4MdGgQnWSzwsNlVOGyqsLmDdPIYREv+Mg4kCDJ8en+Kv/9DB2Z4eolqNwPjmPtyi5sFJm2Q/I50O8gy18J0ZbgRSWwIl5ojrFgVwSIf6R/DF2OX7fsaixhNYQCH3drEOR1RyLuvzuzMeZbZfIeRGqbGl6Hu2VAJUKBSUt9Y5PlAoBKRNzzVqBtYJW7FHrFpiLywBMyxZHoxyHOzv52uwDTK+U2TdY48HyBf7O4LPc46jLBEZvsDZsRNdaRpR/XWdlZDWhjVgy8Q0/2zv/2bjFtM6j8YisYioaoq5zjDgrVFUDgO+1kmit7y1O8mjlPD9QPMJuZ5kRZTGQti2LU3grW0IoWGupmcSOj7QCLbBpv+9WLPGA5gPjJ/ibg8/zgCeveIkK2YvjwbcO97ghD+fOMdct4UhDLA2xdoi1JJIyvWbyBW+8xULBZ3l/jnhAM+xFGCuIjKQZJhqIOxAS5nT6PXD/WiE6XYTnoR87wJt/32V3ZYGX5ib6bepWLH4tbZSFVsdDSoM2EtD4TkwnduhEDpFWGCsopOnVloxHxXRRCDrW0LFgAEmXQKhrqvISwYgy3FeYpRV71Ls+nhOj8galDE4q7CpBB5UzhDq5dqvr0g49jEn8D1JYpDBoe+kaXatoGQ9tJcZIap08x9UISxWfiBAXBelsH9qYNyI4He3gYlzhIf8897srVywJ97SDlu0SWkPLikS4X8N10lsBmtMhf1R/lDF3mf3uLAjIy5Dv1ffx1eaDDHhtym6H12qJiegqzXMLeznZHEYKy/7CHMNOA1fEHArO8oDXppJGn2ZsFaEANI3PQljAGEFhtEmnnAx8E0uGqw0+OfAyD7odfHH9ICQlJBWR45A/zZncMEdXRomNJAS0kXRjhRSXhILjGLyhJoWJLoETE2lFJ3ZotH3Ctou1giDfZWSwDiT9df7hcZz2CH4hx+kfzfOLT34TV2ieXZzk1L2S5akK3rLAykR7kRpQhh2lOmcWB9FG4KvkWsZIWh2XqfZgX1N4szvGVBwhMShhGFcrDMguHWsxaFyhr6r2JstzivuCC5xojXBWq/59lnMdAidRsUeCBiW3Q1u7dLTL+UaFZtvHGoHrQmwkzdjnQpRElZZUm1pcZLZbSoQ20I5cpuoDzOkykZ0hQvdNimOR5V+c/yRvLoywtFBEOIbBwQb/5OCXedK/yEgqHMwqk6hpLCAwmCtMDm0Ni6bN/zD7gwAcWR6j1s7zQxNHGS8v0bEuLePz1acPMXBUcN4RtCYsJg3p2PH4RU6eHyV30iNYsLzqPcjKAzH5kSaDhTbDuSb/494/5j7X29Tgp16ODEjM3NupodxM6v0tIRQMgghFK/bQWjI+UKc6nCTVqEcBShgUFnkTu8aGlWLCXcRYQTP0WLpQJnfOoZOz6HvaFPKpp1/Y/sTUG0AA3a6DaTuIQFMIulT8xKlgrKD2+AqnR4vIaIDhB+eYDgeIjKIZ+TTbHqXjiqHDXdpDPRMFCn6XQEUUgxBtJNpIck7E8GADRxo8qXlu5R4A5sIixgreXT3FHm+eAdlCE6GwuIhrxiskuy01kXW40Cozf6qKakrMRIdH9iSOTgBXagoqRAlL1yTaQtRxsB2FjiU1NyYyifYC0NIeM2GZZ85NwktlhIbFR1qMVVdY0QEtazFE9Mb46909nFoaYvn0APmLkrBqqXsxSzpPmH5GCdkPSS7ik3csoY2QyCRBq+31DUvLdnm5W+bl2k4gEe6BE3O2XeVZdS8T3iKBiPB2N2k2ipgDTd49eYqPV18HklD5f9b9YbpvDOO0QUaW4KJDS+RxlMFVmte744yo6TXvV7kVYnT/3d0o5mSjuZn72RJCIbKKqe4QUlisFRgr2FNI8rIoDGdaVc5GVe53V8gLvSZp7qLIy5BIK5amywy86jD8SovuoMfZqodTagEQa9W/bmzSxKDS4PsR7UgilWUg16bqJ59fiQLGKnW6xRae0gz6LV5ZmGCl4xPHimgpoNyweIshrdHk8XarhnG/y3I31/dXCGGp+i0OFGfZ4S5xpL2Dp6f3AbB4sooKBSuPBXxk/ChKWDp2iRHVpCpj/Ldk5enZ8POmy0XtcyYc5ujpHex4WmAcy8ygw/7iHOP+MgDnw0FcoXGEphH51FsBtuUkAiSSrPhBYmqlWkFHO7x0ejeV7wWMf6uGLvkcnwhQQ5aO9WgaSROo22Tm+6PZx5m/WCZ/UVI6axBW0hpxOdYeZ583S1W2Llvm7AmIngMwspoOIaE1XNSKrzYe5oun38HCQpJibrDaoByEzLRL1MI8PzL2Kg/mzvOH7/xdzDsFFakZUZd8Mi3b5eL+Z/iXJ36E3JzAbSamnfYc2hWXpuvxreUHAPhAcB6A6hp8GzfLaofsVl712BJCIS81P1Y8ygH/Il/MP8WFdoWLncTJ1dWKC80yX9TvZHmkwHvzbzKh2leNxFvNsulypjuCqzSqKSnMamRXE5YlotAlcBNVOpK2ryHERmJSZ5vvxthC8nMYO6xEQf8zPdv8wmKZM+0hpGsYGawjA0t7JaA15jD9gRKNycQPEYw3CZwIKSy5IMIRmiG/xYDbYtBpUlIdYqNopz4MhEX7sNwOeHl5J282Ryk4IeP+Cju9RcadJfY4i/38CBpBZCVLpsJUNMSJ1gjOvEvhfJvWRIBTjBj3lxlxEhMoNC4d4xLbRCACYMFKsI7FGEG9kaPRTO65W1HIGZ/SVIxcWEI2PIZe3Mk5McrZsSGW/Sk61k13scLJxSFypz2qRzS5ix2cMEB7Pl8fPkhoHOqVwzzhzVNJE4f09pD0iNDUtGZKFznc2cmfTz/CQq2I1cn79hyNr2Ieqlzg/vxFPpA7zoQjyAv/qhNGHo+PFo5w9P3jPH3PfhZWclgjUK7BczVzMxW+tlxgZTLg0M5kB2nlOr6NW2W7JLrZEkJBIdjhFCnJJs3Bl/mKeJRXFhKn3XI7oFkPuNga4vUju/nXpQ+Ty4d87R2fZYdz7eSkU9rnaGscX8XIXS2W7i0iTJ76HslQtXHp2tIghe37G6wVyN5SHiCloRF6fVU650aJOaMMnhcT1nJYFXFwcJYd/jIvuHs41pkACTsn5/vfSUyEmCG/RU51KTsdKk6ifTSNjyM1fiqo2r4BN4mtOFEbpt1xMVohpQFhiZYC/BkHbyW5h7BqmXhqmvcMn6KhfSKbmAzzj+VRXUuQ6zLi1CmlwUjDTp2ZKAnuKjohA8UWM20XQ2Iu6UhhlzxEnNzzbKSwlZile13yZ4dQy00G3myjgxyHHx3n0XyVWlzkcCt9Z8t5Bmcs5ZcuYuZrlJo7Mc4As+Uqf7z4GN8evJeHhy/wieqb5mzGAAAgAElEQVQrABzyp6lKyAs3Taaj+MvmQ3xveZKlMMe5+QFYcZGVZKt42QvJO12eKJ7hCX+K3Y68Yc2LAQk/WX2OTw68woIu8lzjHubDIi+c303xsIewHt+N9zE1lkxGY6pxzXPdKoZb28W5EWw7n4JIRXJOeIyqOp6M+y6osOPCokewIDGepesZbE7QstdeA9fW8FJnL6/UJnCV5n2Tp7gwWib6IcWnR4/x/sIxApF4sl8Jd/Ps8j7eXB6h1XUJ3JiCl3S+duRSzbW4tzTHhytvAHDAnUNiMQg6VhE9qghETEVGBAL00HcJ703a4a6aaZL9FMO8GY4x2y3T1i4NPUhbuyhhuTc3y3958OsADD3UIC9DyiIkL2PmdI7Xw118Ze4h3pgZJw4lbhNSRQGTvkWNZDJYoFps8IPvOMaJh0Z5obaHkVyDJZ2nmeZrCE3qxLWSrlHk3YiBgSatwMNxDK26j+xI0pwvRC0ncf6WHGbaFYYO+4RVl8YecKTh9fYuXl2eYGoliWswHYfOkCAeKePEmm4lR7ckMIEBK1huBLwuxtGpoD1eGGfMXWa3u8DFuMJXFh6hFuap+i3eP3yC9wyd4s3mKK04abcUltgoXBETCIPuxR2s2uWq0t2qkDj1isLlMa9NaJt07QIPedOc1xWO1kbxzhbIT3fQXp5vPPIQAAMDz/OotzZTda1cK5Zjq7H1dZmMjIzbypbQFHooIcnLDopLqk7cVXjLgvyMJc4LuiOJen86rrBDta+6vTe0Mb9/5t3MLFTYO7bAwcJF3lU+ySPBFPucFhV5aelpr3OMhg441xxgmYBK0KHkdhJbO4BHytM8nDvH+4NkubAig77E7y0vSVxWp1W5mvayw2p2q1kCEfGV8BGWoxwXWmU6sUtkJB/Zd5j3Bqk922+fg8TlHsfysHeCAdXkv5/7FLYY0x5xcZtJO3TOkHMiFIZd3gL3ubNUZMSH88f5WPk1XmxPcqE7QJBO/dpK2tqlGXssd3NoIyn6XYp+N/WpQKujEEuXuoerNKWBDhff6RCVAmQMzoEVKm6Hv56/h2OnxhFhOqvmNK2dmrl3FPGXC8S+oLFL4I60sUagI0Vtqch3lxLz7+nOfSAtuVKyItReCShVm9y7c54n86cYUXUuFsq8GY4D8HpjJzOdErW4SN1zUCYmEh1a1iKBklS4VvWXSB0UjkhqbgZCE1lNSRqqap79g/Oc04M4L59gLLiPr773IABz3RI/PfzXPOk31hzDcKPozTvpT+iX3FsDW0ooAOx1LB8oH+OVpWT5yWqJ0II4D90SCN8ggK8sP4qqvMzjfpOi8Pvfj0nCfKdPD4NnGPRb7PJqDKkGu53WFQE0JelQVB1yToTnaHJORMXrINMBP+zWGVBN8qknerU6uTrPAlz/pbtCMajy/EBQ52HvW9SMw6vhTqajAULj8q5gmtE0Ecxbz6OAisjxDv88ShlylQ662CUIEjPnkaE5Plw9wqQ3z7haoaqivoiSwtAxLrNhiWacCNDYSLrGoR27dGIHk0ZWChJzYLDQxnM0jVLiaAzcxN9hrWCg2qTxmEY5mkM7pqnHPtMrZURL0dtc4eYiyEUsviNthRaoUkS12CaKVd93o1PHoXFSIWsEShmkm0SiGgQLukjHunxh9ilen02EQs6LyLsRyzpPPc0mtQQsmQCFYUK1KElDPn0KziqtXSLx0+cbKMN/NfEVfvKT+9i3eB86p1hcTATVs3oPC2GevzP+LB/Pz64pxfpGDvo7uS9jTUJBCPF7wKeAWWvtw+mxKknZuEmSXIw/Za1dFEk1mH8NfBJoAb9grX1xrQ2qyBzv8i/yOTfdoRfEREUXqyAqWZwgwgJfPvUgX4oeYffwEh8aPcY9flKlTmH4wsWnkC2J7QoCFTOgmgzIFiUhr0iy4gpFILoU3bC/58GTMb6Mie2N7cnVwShrKXyalx6+cBhTlgl1lml9njldoCpvXBy2KiWu0jjS4CjDD00cBeCp4kn2ufOURIwrkmQ0kbUsGclcXGYhKjDTLnF2MbH5raUf4ZhEMCbn95xk34erdKIZBMnMndjwknaUbC4bra6wt7TIZH6B060hdCq4rUqehXIMrhtTKrYJ3BhPJasFoXboSAdHJdeNU6EQOpqo6xCFDmEswQrCyGG6VeHzzaeYaZSIvjpM9VTiiD33EQn3LSBFb2+DpGk9joQThMZlyZ9ht7PEiEyEZkVy2b4H6L0nxaOe5t/84L/nHy79AkMvC5zz6ca34wGv5Qb4x3t3MvSuz/Gk37qtBXwNljiN4ATQ1hKhCYSzpkS1bxUqzk24ONeqKfzfwL8BPrfq2K8D37DW/nMhxK+nv/9jkpTvB9J/7wJ+J/1/zZTkpWYVCx1WxgVhpFBBTC6I8N2YpU6OqOExHxSYrgz0P98xLsvdAOtZZCiohZciIJMs0Jc/TF+4fLo0w98snmNpV0zdSGomSAKFhGZAdskLMOmjWv2wk/iAKF1Sk2vuML11+ZL0GLAh07Gzpm3RZRnwTw/+GSe7oyzHeR7LJ5mddjs1RmRMSTpE1tCxhhntcqQ7zlRUTVYkzKVOYa3AGIkVFkcZ8m7UXzLtCQAjBJ6TdErfiRnwWhSdLqF2aGuXmXaJ1+bGaZwv4yxLbNFAORm0QliMkRiRJLoR0F/hSeJQAJukiAMIvAhHGTqhS7fjQCQIXZf5doHpEyNUDivGXmhgVapZBB5SWEacOuOqlQhLYTjovknXWkrppijnGolXVr8nieAJb56n3n2U58OD/aAptwFWCQywYgI6to4vbt/qgSsU2opLUZ/CJBrdDfrZtVYZNjx4yVr7bSHE5FsO/xjwwfTn3we+RSIUfgz4nE2qzPy1EGLgLWnfb4grFCaNuw+8CFNK4gZUmhfBkYYop1iJFO2Oy8V2qe/JPtOocnpqBG9RElVMfymxIkPca9yum6Y5K8pLKeWB9AUkQTC9VGuhjVFpVGFkNZE1/e/fLJc2PF07SnE1Skh+KLfEjDfLmbjczx6VlzGuEP02NK3hSHecV1pJ5idfxskyaqoJGaMwRoBKlmQH/DZDfhODINQOS90cjjAU3ERTyDsRDxUvsNebp2l8znaHePbUJJXvBOw7EhIOupz/oCBX6vTbam1ipgidCBqb1rToCYJu7PRjJPJ+EjeipEFria5f8s/YnMZtStRKh9rjVQCC4SbGClyhyQvISxcHRU5cek43gxKCfYV5nt3dwTR611aIVHid6Y6w310gL64s5baeeiNJ/9FpDZMr84DC6hULSU7cWEN4676Sa7XzeqzHpzDWG+jW2gtCiNH0+E5gatXnzqXH1iwUAHbll4Akmg6g4ncY8Nvsy8+z01+koQNOtoc5tjxKK/ZYjhJnkBSWXLlD19f8jYOv8gvV/8QDnsRhbSXalZDkVs0wve+sHvSXtvkKfJxbDll1hWKPU2SP04E1zkG+cBiQMVPCUEkdhwNpcZweoU1iH/KqS2QV0lgCFeM5iWDrDUbXSWx3KQyO1MRG0TWKyChiIzlUSkyydxROsd+dY0B2mYrLXIgGMA2X0nmNd24Rt+azy5RZujfJqF1/oEtxqNU3T4wViDQoDMAYidayvzekGysarQBzukD5VHIPi09aCm6X+/Ze5NyPVph/YhBbSDSR8WIr0RTUCnmp1lX3QQnJoAz4leozPP6uMzxdvw+AZ2f3srBUxHM1F7oV5oICY6qe9INUW+xFkkJiPkrkmvNavrUC1I242eXMtz6PO+1ovFrrr7j7G9WS3JdLOuRSlCMyitFcnT25Rd5fPMZuZ4mmdbjXH0IJy9HlURY7l8yEH93/GvcFF/lo4fgtF4RZixfZWEtElCSWtbcnWk0JiRSCuglwnSTAJi/dS/aygDntMhNViNIt5PUowCD68RexSdyoguTFdLSLsTKJNuwGtCIXYwUPpPUh3uGfT1V0B0gipnLDLZb2l8mfDZC1OoXjZykeThyBx4fH8MZiXGX6JkOPSMt+OHlvTISRi3qlyJ7vtHFePI4cGmTpgQlKXoe80+VAeQ6zRzCf5sBYDnN0jWLcqZPfADvfFYodTpFP5Oc56CWrTEUVcrgwzpnlQULj0LQeHWsJejUirCFGs2ySZxoBJSGvmyuypxnAJSHiXmP2Xx1n0eN2BT+tRyjM9MwCIcQOYDY9fg5YXRhyF0mtycuw1n4W+CwktSTf+vdhJ+l8T5TP4lZ0Gt7cpSI9fOETo9nnXGC3u8Cb5fH+MmbHuuzzZtmpGgzLm9v1djMqVg/Jlc7LzURbg0rt4GGVaEeX3aNNNgBFVnGqNURsFCtRQDt2Lw3G9KMPVS/wQOFC+mxDIgs14/FSZw/fq9/DgGql5080FF847FCSn6q8wK6Ha3x+8Cmm2cOOpyWy1cJUEsEcF5LnWPJDCk6Xstem6HRZiQJqYZ75VqGfywGSTWb58xZ3ehnTCbE5H+sZ6t0gDSkP+7knACZLC+wOFtnr2A0NLnKFopQ6J02aV2J5pcDrhR0cyM0wouoYwnT/jSS0hvoqgVeR4rqrBpJLJl5kLyWqudZ3VguGXn6KoevUz7zetW9mH8d6hMKfAD8P/PP0/y+tOv6rQogvkDgYl2/GnwDJQOt1yFGnzoSzzL2OxBeX9jsoJI5UPOzGjKszfQW+N6xLQm56DYBb9SWsh5gkym5YuVe9thKSYRWRVyHHFkf7Nvzq2bp37PHiWT6QP869roMvimhr2IVlQh2/7JxNIxlOH7AvHO5xFKX8SdgJ/9tHc1ywI+wQgsUH09obo03CyMVXdfYX59gd1BhQLWaiCsfVKJFRNLpev03dWLG8H/JzQ+SBpceGCIabLHZyKGmoeq3EtEk3aN1bmeWduVMbnoLeYGimgnMpznNuaQCmA462d/BlpakNFhl26wSiy053kbrJsaQTQZiXIe8NzpNXl8/mq0sD9MoIQJI4J7KJtpmX7g1Xrlo24nC3xPuC6JrawkZNTGtdkvw8iVNxWAhxDvinJMLgi0KIXwTOAn8r/fiXSZYjj5MsSX7mVhrWe9j7vVnGlca/ipqYePEFEij0bX+JxqbHb94O6+1534olRXod7HqaSVKwNvl5IGjz/uETPJw7hytilnSifr/U3MPR+hgl1UmyOaXdQAkJ1pCXipJq989Zty4RMX76uWTGE2gr+eV9T6P+C0P+V0L2u4nJt6ALfH3lYdraY8htMuIksR5N4+NKjbGiH6sAidMx2hsy7fgEB8dpTRgKbow2gpUw4KIqkXci8ulqyLvyJ3jMa+OKjUuMElnNuTjkcDfZvxFqh9ZKQH5eYGser9rdzO4sstIKaC8HlIea7K/O00mL3o7l6rQGfJ4IzjKR5u4z0PcbuELQsbb/blTaNTtWMEJEWV69x12Kp/HY564Am58MZq2rD5++xp8+cpXPWuBX1tMoiWA+TQV2nzeDe4PdZXUjCdI1ch+JK9bmzX8rq9OIJZWYbq9o6M0ivdqSq2sjrk7QcS0uVecuczGssLdY42OlVzngREghMDaplbHTWSQ076BjXJrWoWPj/jq2SVO/eUJTkpcEg7EWI+yq7c2Wugn4YP4oex1NUfr9ylWhbTHuPMNz7Uk61qUkk5J/rtDodNdpp+v2hYLWyRZ1s6dDuE8TuDpZGtWKlhXMUGJncZnxIDEpJ50GxRsk27lZQhvxQriTc92kCmJORXj5Lsb18RchmnNYnhpl9MWY/OkVmvvLvPSxMvkdiV9nammAl2Z24rsxgROzvzzPmL/CZDDPTrfGgGyhsAT9epiJgOtYRcfGFG8wETkoSuLquTA3OtAp2/uQkZFxGVsuzBkSlekzlaSCfbIGfe0Q06TGQZLHEMAQ4VqBK5LttGulN8vWTJdlo8iL6Lbvf189C7SMJq9sf/mrF/B0vVWOXln7B7061aHvMqcL3O/GFMXly7EfCGKeHH86veblHnAFDMqAH8rNE9levoZkrWK15zwvFT9Tfp1S6vhdjS9cHnAlE+oENQNjKlmqe39whr9dOsnMWMyb0VB/L8NzK3t5Y2Gc5XoOa5Nw55VmDjUVUJgWLJWg816H8R2JphBcJQhtvUTW8Gx9P3/0+qHkwKIHg13sPV06YwpKMd4Rn2C2jQi7RAWJKIcU06jPHiU/ZNBvUfWaSGE53Rkmsord7gJl2cEVl5YGXQEFkqrpDRNSltdeNo/R1IwBOhRl8rxbtsuMNnSsYkJpFIJyGo69nuezJYUCsKaCsJAMpKpS1E3ysJcMaAtKaCbUtesVrHYAre7skU2Cie4EvXbm8Vimfdmxt37mWrhCMSRzVKRmQq2QF1d2NCUkxevUXew5UFc/o9BG1EyXQAjywiUnHJS8wV4PmSMvYvKyJ5wVvnApStjlNHjYS9Kl7fYW+KJ5J883dieRkMYSHM6x86+aqFdPIsZHeGN/FTmR2ucb/H5CG/F0Z5g//fq7uP/zSXyMaC7y5t8bJ7h/GTlk8RzN/D7J3FKR6lGX1pikOtjsp+mLbZKpaihoMh6s0NQ+890ip+tVfLWbHxl7jcdyZ4jS0PkIAdZSNy5LJseCDNnntBlctbqwOg7inI440h3FE5pARHyz8QD//pWnKLyUwziw6+Nn+PHxl/i5NNBjPU7YLSsU1ooSkiI+pJmQa3Evg7KhY5Mkp2/17K6OWnSFuiyirBfV2Dv3nUAiqEh1047SHolWIfHV+tKJ9e8/jdosCJP6Dtbma+nVebwavnAZS8OWH/Rm2FeY543cGLGRSGFRixZnvoHpdiHnITxDO3XqRWusgbEWtDV8P5T819//z5h4OkbOJH6XeHIMPd6l4Hept320kZRHGyy/L2DpPYKDe07zxOAUYZrM4kRjmMUwT05FFFXIS/URjs8Oo08XETH8zn2DPLpjP++onAFgt1ujaxVnu8O0tEdRhTycm+KQP0sgEud5y1qOdJP9Kq91djMTlftL7189dxD3ZA7VgWgYyl6HQEZENxGkdC22vVCApPMF6a0sGZcB2aEgTTrbWSK6RCbZE9DbYqLTfrXLcW96Nr4dBFcJfb3T5OXVl0Fvld65qhL2+AsMF5uJEzJ2WLjP4tVHqDqK5fsrDAws9aNWu9ZuiHMttBF/0S7xm8c+RXcmT3NMkhtPHI0X3lvkvj1n6MQuSzpHHDmUim32jNf46NgRnsifRmKopSnomtqnGSVqvRKGQEWEDZ/x71tycxGdl4scHT7Is48kGXjK43UcpZEiCTUXwvJicTdfCRpoK2jGPucaAyy3E43OdTQFr0ukFe3IoTZfwvUtcUEQlQ2OMNR1rq8xF8WtP5+7Qihoa6jpRFP4duMRPlQ8zIiM+zvKAHIC8lyKKOtxu1cY1kJvpt9KXK8K1HoZVgV+qTLNZ8pTtGyXczH85eRBPnvP+6g9NIQOLKNuTCfNvDSjPUbU1dPc33Ib9tVYHCpQ+3AysD984GXuK1zk1fpOpmYHsQs+TaWp5DosRAVebu/hmdo+lsJEUNVaORrLOY41J3CrHSaqKwwO12nsHsIJnST/pQLS0G5tZCIMZLJb1ZGGQMWUnCT1flc6BE5E5PeS21o8qSl5IU5es7O8wsrOgOmFCo4yLHQKPLeylwk3SXj83mD6snSF2y4d23oJbcxXW0km5K9MP8iuyRoj6hRVLp/1Vdq1b5bIbmxarrcjazHJXKGoiBxF1zBSPszMPWWer+xhtlFECks9nYmfa99DSR5jr3PtSL21ahKBiNBG4jsxD+y8yKdG07yRwRnqJsBYydPRAYpTkoYf0Cx2+NLRRzHzPju/ST+oX40rql1L6VzMyp4CUz+gGB1aYfGBkDjnk5uxhAOg0v0bxiR7QXrb1HvJgBUGYyX1yEcKS8m75MiMjGLCb7InV8MVmpb2OFusMtcustjJcbFeopXmzDhameIfVF/u54Ho+SbWwl0hFAB2uwsA/L3J7/CB3EmGlbrmHvKkPkJiSAxeJ2w0I+HSysfNh4G/9Txr/VxRuP3M2tYK2l23X+rvW4v3E1mHTxVfZ+IagiFGE6da4bWCvVom4ny0I9nKDThCM+4kjsZxFTIkQzq5qTQjVJK3shW6iKkcgycgd7FFlBYtkrGkeD7Gn2+jxkrYWBJGDrlSSGefJRxysYHpJ+eFZA9KJ3LwlKbgdMmpCI3EkJQ50OZSm989fIrH82c44M5RkZp57fJKuJOlOM+FVpmVZkB8tsCrryVpBF7cs4ef+eALFNNT3Iyv4a4QCq5QPJamNn7QXb4iu9JqkmKkXY5GyY6+x7wVhlXhuueXiOvOdDrdHHO31yRcj/C8le82tN/fbGa0Io6Tc7xyYYITi8Pk94d8IH+CXcr2Vzh672JOh3jp3oK8UPi4l9WVaNmI17olvrn0AFIkOSPyTsRQmsW5JCQIeNCd5+HxC7w4VsLkNELAwEMLLO7OY1WeqJBco7nbYKWDCn2MC0IZIq1QylAod6CcRjmay3eKdoxLOQgpe22qXpNARtQJkGnmqd7W//cUjvOkX6MkPRx8BmSXpp3liJNEYMZnC+z5WkxwKnGUrjw6wkvvGWWHqiMRzOi3mfmQ7AVIBvaN1EaJxBOCuklswTldJy+6q5bNrs/Vzm+w162BmHHzuELxqYGXON38CPXFPLlSiOsms50QFiEsr7Z2oZHsdGu801/ARaCxtKxlOs5RTcONXWnQJLU553TMybjCgh7hG0sPcrYxSKxlv37mapIMWfDf7Ppz/uqTx5mJyqzEAZPBAnkZ8p3JA1xoJZG3YexwYajCygEfndMUCl08J+5ntQIw9tIeFJ3uVF29J8VYQSP2eWVhglojT+BFfc1CI1CI/mRXFAHv9g33Dz/Ds6Vj/OqZn8G/2EA0Wv3z9TKWQxI5uVbuCqGwmhvNSBJBXijyIl3CNAGBbrHnBt7+XohxZDUul5YLYzQdG2OsxbVq2xT82Oq4QvGIt8iu/BKvtvdAiX6gkE33TvzpG4/wpcYTuDVFNKh532PHeKR0HlckS9EH/IsAaGeRCMHJaJj/4/wHOLM42M8O1Suo28tb2UmrXGnaKJEMwnudmInyYea04Gg0StP4jKgVJkfn+wVwnlm5lwMDc9T3+iyGecI0qUzPVwCXtqwDffMg1kkJwVbscSIaph4FnD89jKoruhMd/DQP59HOBI948/z/7d1rbBzXdcDx/5nZF3eXokhKop6WJVl2I9W26qiJC9euk7qNk6JxXfThfEmaBk0CxEC/xmiBpP3SFkUQoI+kcFDXcdDmgRpJ/cGobaRo7bZxYxu2FTu2bMmRLFoSJYoUxeWSu7Mzpx9mdrUrcbUvcne5PD9A4nI4y713d+bwzp1779lQ9R6Fk98yfCh1iT/6xef4/uEPM/ZaWP7FcYcdsUs4hEPlM7IO+xSaVV5E5eZE2Es75cfxomblNWefRZcPy3U4piRGrIlVcUxrUuKQcQtkti5w3ehs5QRb8BIEmkSmkoyccBh+z6eYdfnf0k28unM7u0dn2ZhYZDoTnrDvJS4wXRrm/2av58JiOhwH4QSV5ehirs/h8Xe5PXucA/Ewtd6wczntnCtCCpeMUwqT/kaJf8edhUra+9SIx8niJo4tbmHJjxN3wlZNOlYkHfPYlznPjsRsJbO4S8A7hS28uTDBVH4D88UUZ+aHyV1ME7vkUsr6jA7nKUT9KEfzE5zNJJlY5q5LUmLcv+Fl/umeX2LmlrDFnNiRiya7lfdp/n23o9gYU2PdtRQgSlcf9SHMBD7n/SEm3AIj15iKWz03oEYTKziXeVdkMDKh5dY59DUgKTHuHD7Kma0j7M+cY3IpHN13RjdQKMVwduRZ8DKMvVEic6pIIpdmfudGXr81zfD4Au8OhftvSC4xuzRUuXsB4DpKIuaRiG4H3p49zgdTp69auKZctiX1WQgc5oMhhp1FMlJks1skHXVmbnTO4CNMeRtIx8L8GaXAZXMqx9bkJe7IvMX18YsMy+Wp1CcTU4zE8jwX7GdyfmN4azLl4W1wyU7k2Dkyx4loBe4wyXAcX4tX9V254jDsBOzfdo7UzrAfYV92upI5Cw2Y9ruzyMqaVu5YvDHuM+XnKaiQD4rLLqJ5La3sWz2fYBC1Ozy8PLcirz7JaEkzByEpMT6anuee657Gw2faD5vDPylu5eX8btI7iiy9L85jOz9I5uVhEhcVPwFuyicZq70F54oyFC8Rd31yhWRlEeCRxBKjyTy3JU8zsczSfdVDxkedgBvi56p+OhQu7Q+4eOyLn2ffxvMMj3lsdKjkl3ArZ3HVOp8oaSkRqMPu9Ay7hmbZFM/hqcvx/GYSTonzS1lK0QpVGTdMIxh2atdOo89reDdtZjHNUDwMCkFGiFftc6Rw5dKp9a2poLAaCTLi4jLhDjEXLPGWp+yMLTLSwloKrZZpEFoI9U7+8uKguaBAoBoFWGlqubq8+rxQGGd//ALbXa+SY8Hl8l/tbHR3YMw5x0YnT1xKxMVn5/tn+NaW2zn7/Db8hDI6skAmUWQstQDA1qF53vAmcFQYTy1UVnAC2JBYZEfqImNu41W0yuWpfS+i254OTLgFNjoxkrL88VO+XQqQD4oUNMaSxtiXOseW2CXG3ByBOuxJnmemlOU1tnMmE3YtXvTSXNIkM0GOYQk4WXL5n8UbcAkYdpc4tjSBHzgsRR2cjigZ5/JCQ1uijOPNWFNBoZUTqvoDaHRQhmvzJXi3NMIlzbPLzbE9lmxq3MEgnOStKL+vDlefIJcHi5V4pxSr5MzY0mDGa1xcshJnszsf/YW9ej4KXH6vkxIjLiW2unlGHGFv9ji/9nPH+MneTVwoZUk5HjOlLNcnwqzfu2OzbN0azt0oJ1WZD3zmg7DZnRGHEae9RVuqF/ENuPYSfWGHdfh41EkxnhBuTrwdPj9arm0uKJISj/OlYbYmL7FhIrytmnaLnChu4qKfJiE+bxa28er8Lk4vjIStn2KS6cmNxObC137qpgR/vOm5yuseTs40XTUv+PMAAA39SURBVKc1FRSaUZ5uOul7zPgpFjTBzYlLDVdSiuFyyhvnPy4dIOmU+OToj9gT06bHL6wnk6UCY45z1fz/8uMUMV5Z2sWO+Cx74zNsqfeLqrgiuIS5IJuZY7GkcdJCNJjHZZRwSbQlncXTgPlAGY4GCYRrPkQBPvrVIys8G9ZBiEPDYfT1+qbc6P8RB7bHFomLz6ZYjpHkQmWfGT/L8/M3MFNMc2Epw9R8ltypDbgLDok5YeJUQCIX/iGcnh9l5lCMbdELxVuoY8OgUCdl3F8DvwkUgePAp1X1YpQw5g3gaPT051X1802XZgUEKK4I212XCbeIi1e3OVethI9LwJw3xLyX5MTwKCPOtAWFK5TweTJ3kF/JHOUGKS07NTqGy/7kWVwCUqIN17z0NcBXZcItMuwk656k+Wg59emgCAwx5ta25lycyveb3MYn/Uq28lxx2OKmO/6d+cDjYhAjHyRIO0US0W3FooYJkmLik3BKFEox8vkkKPjpAPe8S/pcidkbws9j051n2B3TSnlOl5qfbt5MS+FRrk4Z9wzwkKqWROSvgIcIs0MBHFfVQ02XYIVUZ9SJS7ylIce+BuQDD09jFPwYJ+fG+M/M+0iNHGGb6627S4RryQceT5y9BWdbQCrzJnvqNJe3uguc9TNcDGKMOY1XsYqLy2a3/iWbpz5/O3sQgNHYArckTzXMj9joc2s2OXCzOv0dvgYc9ZKc8DYxV0qTl2RlUZZCEGfayzJfCic4ZRMFEokSi0NxYhmP+WEX8ZPMHwrHQXxp7zM1iZe9FkYfNAwKy6WMU9Wnq759Hvidpl9xFXT64brikHbi7E1OcSY9QjFw2ZGcJSNFbOxyLR9lcyrHXCnNWT/DhLu47K3cnW6cMWeBAHC4doC+3IG3/Enua8CUv8iPZsKZsHeNv91ZJapet5+44nBLosjPJ85wT3qSKd/hhaXdALxR3M7J/FjYgegW+eXx43x84lXi4pNxChTV5a1f2FZJtHxz4ixwuZ9kbwsdBSvRp/CHhNmny/aIyMuEqYT+VFWfW4HXqMvXgNlgsZIqPkl7i5MkJc6B+DQX0xmSTombU6fY7C5S/caud74GLKmyJTVP2i0wH6Qo6EJl/EXlejlacSmpMQK07aHfnvrMBkv8ILefZ2dv5OxCmFfiRGacvclzLOr0NZeVW4vKl6tZYMTx8XkXgClvhLeDzWTjBYZcj1vTJ9kfv1C59VnQgDuGTlTaA8NO7R+zbiWDQUT+BCgB/xxtOgNcp6oXROT9wA9E5KCqXlrmuddMG9esEj7zgZJyO7//vyeeZXvsLPPpd0mJS/yKBUnXuwBlzEnwZxPPVZLsupKo2zG43C28VuSCAidKCV7JXceQ63HzeJhT6MMjb3Br4uzAfz5xcTkYD9/bm8Z+ytLoEQpRYB6Lbn2WA66vAeNXxN3qQFyoSmvfSNtno4h8irAD8lejXA+oagEoRI9fEpHjwI3Ai1c+v1HauGaEiU/8yq0g6LxJmJT4Ndc27PX6jb0S3tUJ8yhe2blYWfg2+rpSk8JmgoD/XriJs4vD7MtOsy8VDhy6LXmWiWv0PwyS6oxo5fouNzam0fvttTBorq1PTkTuJexY/Liq5qu2bxYJz04R2QvsB95p5zWMMb3RzC3J5VLGPQQkgWckHPtdvvV4F/DnIlICfODzqtr8qIkWleeVJ6V7y6WVr5HXm0bLwl+570qYcGN8JPs6B1LvcSBxgTEnPFyHpPNbf/2qmU7zduqedlawT6FOyrh/rLPv48DjTb/6Cun2+omDekC2a7WCZFLi7I55TLgzjDhXz0vopdUYcg+XRzYCK7owbaPbt7X7mpb004HZL1YrSJaHDGcb79p1q1lnX4US/qoFnkYsKBjTh/JBmJDX0ShrWIcja1sJLtYONqbPBCgnSy4zQYmCllpann0lrJugMMjrGJjB85a3hVOlNEsaZjbrpnVz+bDWOgd7dT1pei8uLvdnZvDUx+3BAK11ExTWgvIswLmgyMnSEBNujjHXJSv1Zw6awXStdRlWmx1pfSSnHjn1OO0nOFLYxSk/y3zg1yy9Zcxqs6DQR+IIcYQFTXB8aQszfpaCxQPThPIw9JVglw99JOuE148H4gtsHH2evbHBHr1nVlY+8K45b6dZdrQZY2pYS6GPlIcLDzsJ9opfSSNuTDP8Fep7sqDQR5abJmtMM1xxGHXqJzNqZZyOXT4YMyCu1fdkw5yNMW2zoGCMqWFBwRhTw4KCMaaGBQVjTI2GQUFEHhGRcyLyWtW2L4vIeyLySvTvY1U/e0hEjonIURH5yGoV3DRm08VNO5ppKTwK3LvM9q+q6qHo35MAInIAeAA4GD3na+XVnU13eeqT00Kvi2HWoIZBQVWfBZpdkfk+4DuqWlDVnwHHgA90UD7TJk99JkvWWjCt66RP4UERORJdXoxG23YAp6r2mYy2mS5LOwluiLeXQs+sb+0eMV8H9gGHCFPFfSXavtxa38sOyBaRz4rIiyLy4vkLfpvFMMastLaCgqpOqaqvqgHwDS5fIkwCu6p23QmcrvM7HlbVw6p6ePO4dTusBps/YdrRbtq4bVXf3g+U70w8ATwgIkkR2UOYNu7HnRXRGNNN7aaNu1tEDhFeGpwAPgegqq+LyPeAnxJmo/6Cqtq1gTFriEQJo3vq8K0p/fFTuxrvaIxpm7vt2EuqerjRftY1bYypYUHBGFPDgoIxpoYFBWNMDQsKxpgaFhSMMTUsKBhjalhQMMbUsKBgjKlhyWBWiXfF6G4HsWnMZk2wo3QVeOoz7S/iqV8JDhYQzFphR+oqKKjH26UsHj4efiVHpDFrgV0+rIKsk+KOZIAr9XP7GdOvrKWwSuxywaxVduQaY2pYUDDG1LA+hRVWUA8AB4e4pbwwa5C1FFaIrwHHvRwPTt7NU/kRzviL+BpY3gWz5lhLYQXtjCX5i+1PM+KkiEu218Uxpi3t5pL8blUeyRMi8kq0/XoRWaz62T+sZuGNMSuvmZbCo8DfAY+VN6jq75cfi8hXgLmq/Y+r6qGVKuBa4YqDi0PS7b9cC+VLGLtNaprRMCio6rMicv1yPxMRAX4P+PDKFmt98TVY1RPWgoFpRadHy53AlKq+XbVtj4i8LCL/JSJ31nuipY0LdaMj0jo7TSs67Wj8BPDtqu/PANep6gUReT/wAxE5qKqXrnyiqj4MPAxh3ocOy7FmdeOvuLUUTCvaPlpEJAb8NvDd8rYoBf2F6PFLwHHgxk4LabrnuJdjLljsdTFMD3XyJ+Qe4E1VnSxvEJHNIuGIHRHZS5hL8p3Oimi6xVOfBY2RlkSvi2J6qJlbkt8GfgTcJCKTIvKZ6EcPUHvpAHAXcEREXgX+Ffi8qs6sZIHN6iiox7S/yFbXt5GY61wzdx8+UWf7Hyyz7XHg8c6LNVjKC6308+pLDg5b3HTfls90j41oXEXlXv9cUMAn7EsddhLEcPvu5LPWgSmzoLCKyif+Bie17HZj+pEFhS7oJAjYaETTbXakGWNqWFDoknanUV8KlghYt2O7TA9YUOiiVk5uXwPmgkWeW9p0VQ6JVtkwZ9MK61PoknAWZWv7j8gQv5HO4XY4mMj6I0wr7Gjpc3ZCm26zI84YU8OCgjGmhgUFY0wNCwp9zu4cmG6zoNDnbIyC6TYLCn3M14Apf5F8UOx1Ucw6YuMU+pgrDhPukM1gNF1lLQVjTA0LCn2uk1aCpa0z7bDLhwFmoyFNO+yoMcbUsKBgjKlhQWGNsL4B04mCek3vK6q9HxwjIueBBWC612VZBZsYzHrB4NZtUOu1W1U3N9qpL4ICgIi8qKqHe12OlTao9YLBrdug1qtZdvlgjKlhQcEYU6OfgsLDvS7AKhnUesHg1m1Q69WUvulTMMb0h35qKRhj+kDPg4KI3CsiR0XkmIh8sdfl6ZSInBCRn4jIKyLyYrRtTESeEZG3o6+jvS5nIyLyiIicE5HXqrYtWw8J/U30GR4Rkdt6V/LG6tTtyyLyXvS5vSIiH6v62UNR3Y6KyEd6U+ru6WlQEBEX+Hvgo8AB4BMicqCXZVohH1LVQ1W3tb4I/FBV9wM/jL7vd48C916xrV49Pgrsj/59Fvh6l8rYrke5um4AX40+t0Oq+iRAdDw+AByMnvO16LgdWL1uKXwAOKaq76hqEfgOcF+Py7Qa7gO+GT3+JvBbPSxLU1T1WWDmis316nEf8JiGngc2isi27pS0dXXqVs99wHdUtaCqPwOOER63A6vXQWEHcKrq+8lo21qmwNMi8pKIfDbaNqGqZwCir1t6VrrO1KvHoHyOD0aXP49UXeINSt2a1uugIMtsW+u3Q+5Q1dsIm9RfEJG7el2gLhiEz/HrwD7gEHAG+Eq0fRDq1pJeB4VJYFfV9zuB0z0qy4pQ1dPR13PA9wmbmlPl5nT09VzvStiRevVY85+jqk6pqq+qAfANLl8irPm6tarXQeEFYL+I7BGRBGGHzhM9LlPbRCQjIsPlx8CvA68R1ulT0W6fAv6tNyXsWL16PAF8MroLcTswV77MWCuu6AO5n/Bzg7BuD4hIUkT2EHam/rjb5eumnq68pKolEXkQeApwgUdU9fVelqlDE8D3RQTC9/ZfVPXfReQF4Hsi8hngXeB3e1jGpojIt4G7gU0iMgl8CfhLlq/Hk8DHCDvh8sCnu17gFtSp290icojw0uAE8DkAVX1dRL4H/BQoAV9Q7TANeJ+zEY3GmBq9vnwwxvQZCwrGmBoWFIwxNSwoGGNqWFAwxtSwoGCMqWFBwRhTw4KCMabG/wPfLdipzcEoVgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQUAAAD8CAYAAAB+fLH0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzsvXd4HNXZuH2fme1a9V4s23K3sTEuFGPA2NghgCmhBBJa4sDrkJAQQn4vL18SUkgPpBFIIBACmF5NANNNN+6496Jqq1iWtdo+c74/RitUVtJK2pV25b196ZI1OztzZubMc57znKcIKSVJkiRJEkIZ6gYkSZIkvkgKhSRJknQgKRSSJEnSgaRQSJIkSQeSQiFJkiQdSAqFJEmSdCBmQkEIca4QYqcQYo8Q4vZYnSdJkiTRRcTCT0EIoQK7gIVAJbAGuEpKuS3qJ0uSJElUiZWmcDKwR0q5T0rpB54CLorRuZIkSRJFTDE6bjFQ0e7vSuCU7nbOyVLlqBHmGDUlSZIkAOs2+eqllLm97RcroSDCbOswTxFC3AjcCFBabGL1GyNi1JQkSZIAqIV7DkayX6ymD5VA+7e8BKhuv4OU8gEp5Swp5azcbDVGzUgcNKmz3e9mu9/N3oALt+4Pu19AarzqtrGkfC61Wgu1WssgtzQ+CEit23uUZGDESlNYA4wTQowGqoArga/F6FwJz96Aix2BHL7/8k0AjHhLo36amc233NdhvyXlc/n8X1M5cqaPzI+tXF49GYCf/vkhzrT5MYuuwtWt+7EKE6oYXqvPOjpWEavuG300qSfMM4hJK6WUQeC7wBvAduAZKeXWWJwr0QiNcJrUAaOzjDI5MKMx4Y8HmPDHA1jf28SIB7fzverZbd/b6POx/vFpNJ4g+dq0NXhyBUfHmDg6xkSG4ulwDk3q1GstPONKxyeDCdMZ+4KCMiyvKx6IyZJkX5l1ok0Od5vCSo/Cb8aeiGK1Iux2ms4Zj/vqJl476V+kKibOXH8duRfu7PI9z8Un485RyX+3Bq2iGs+506lYJBjznB8RNJ6d6a5a/jD6ecaaTZhQWdacx8O3XoL1tTU8VfEJmapjsC93SInHUTke2qQW7lknpZzV237xdeeSJEky5CTOpCyB0aTOt9dfQ8EiK+XnmsjYLkg5rHGs2c5vas9mde1IrE9mojiMEV2kOkGXkJFK1ZkKi89aQ/rNHp589UzKfrqGSRvykceaYUQhAFUvjeKKM7/FlPxDVLvS8byQT8HWSoLAGl86ixyBIbz6wWcwR+TQNLC3cw61ltAXkkJhEPhJ7XRyl9k5dJqJmxe9xmNjT0b+O5P0NDdmoXG4IpMJu5oRxQUAHLysAN0CQodHL76X021Gh/rONau59s6zCVZWAVB+2QkA/OWaB5loaWSbP5vbHlxC8QOfEGw9981PfYst37g3rBGyVmshXbFgFfHtIxIPqndnyoMuNAm21sX3QpNzaBsURZJCIcqs8/mZbjGs/QGpAfDyvqmM3NaAtTGN+yediRASu4CFI3byo5yP+b/zPiT9AlvYF7f9DC9PTeHAsgnoe51k7IBLLv4IoFUTcFKoupl9yWaqf9f6BSEY+aqbGVOu4fkZD1KiGi+/SwZY0TKSX/x3Kfdc9CgXprhjeEcGToPuIU9NGepmdCBfteLWAxzV9aFuStRJCoUost3v5iezLmD7b8bw2qK/8OKxkwDIetyJtmcH6kETpfpkjo22YXIHSVW9ZCr2Po2Cb556P7bTBA2aoFmGRnjjt45Oqf0Ih6afiNLUQu28IppHg/1twbce+QH2l1e3HUdxOBifspffrr2W7F/fx0iTm5JuRruhHqnTFcuQnbs7rMKMVTVjFdH3lRjq+51cfYgil+09B9fZR1FGl1K5uIDURYcAuHLEOna6C3h3+UxKX2tCbWxh510ZvDv3XkqjoHYGpNZFywhIjfKgh1KTHR2dnQGNJT//AQDZT6xHmz2JqrMdaDbJ6NPK+dXoF5lpjb+XL0n0iHT1IakpRJH9j48jJ/gp1DdS/I6Vi771GQBLM6og8yDu//mQey6fxhO7ZvHW7HspjtJSYbhph1mojDGHBI7KNIuZhd/7GIC1u2ZQfYaDMxZvQJcKQalgE1pU2pJk6Ef6gZIUClFiu99N2sEgpoJ8ZFY6ylEXU22hmDCjgzgUCz/O2cGPc3YAg2+YOjd9EwDvF89BCcKB5mzMqkauzUWuOvzmxkNFIgsESAqFqNCoubnw6R+Skympv3oMWTuCpKw5yqFgRusex4a0fSE+ck0AIKXKS0ql5FhVCXUzwDXtEI6SZPxJEoOkUOgn+wMuikxWqoM+vrX7azgmHGXBl7ZwTupWnjlyMh88N4Ns9e2hbmYH7sgxPCYvWraRK//+QxyHJdYGwd/HP4lTsQ9x65LEC0mh0E++O+/r1J5dREuhIOiU/P6yxzjLVotVmGjJ3MiCJduYZ49PlXyKxc6rN/+ed91l/H7LIiaarUPdJCDx5+LDhaRQ6AdNuofg/oPkutzYzijjjJ98yvmOJszCMBye7/AC3qFtZC+Umpxcn1bL9XMeJ1683XUkyUnMFwyVkIyP3pAkSZK4ISkU+sFtVecAoNXVYav1MS91ezfeiB1x6d4BJ0UJ+doPdJ945LDm6X2nCIn1PQhIrc1jNRL60x6X9EXlOH0lKRT6yEqPwt6fTGr723zEzRtNUyP6rlOxDdhdN1xH6UyizssLoxjiHet7YBZqRANBiP60Jz2M8Xcwnm1i9p4h5Paf3oh9U6v/gaIiD1bxxsFJPX6nSff0aVTpiXAdZbgwFMLsin0LqI/jlHYu3YtLH1z7VNLQGCEhtd+fJvCcUEL1t8cgJ7qwrHUyOa9rcpT2NOsa6abwo8pQW9zDuUgPZL9E476RyzHHsWZlF4Pvep4UChHwQFMR92xeAECqFw6damXZtX9mpCnAz8rm88uC94COqm+T7sGtG9pBVjcBPfEw9+/uRd/k93L5ZzcCUPbrAFJRUKrrmLqilt/lbxzMJsaUnDiLvuzMUAwY/RYKQogRwKNAAaADD0gp/yKE+BlwA1DXuusdUsrXBtrQoeSPL15E8QdGhgJvpuSpb/2JaRYbYOHG3PfDpjtLV+w4WuMJlLAZ7/v+wAdTqziq27BajeQsO5ekIc2S7LVpvFedDsNIKCTpykA0hSDwQynleiFEKrBOCPFW62d/klL+ceDNG3oaNTdj7vq87e/DP5zOBPMXo+tIU/dRpqFROCC1Aa+/u3Qvbqm1JUVpr2VEW1BoUuephjlYTYZQu+Ssjyg0H+WT6WMotB2jUXMfd3kfjyf6LRSklDVATev/m4UQ2zEqQw0r5q1bQoF3V9vfOZuD/OfYSK5LM+pqOCKY85mFik8ao25/sxw5FVuHECpVKDGbfuhIsswtuNbkAPDo4dM4beJeTIpGS9BKi9SxDtPU8UmitPoghBgFnAR81rrpu0KITUKIh4UQmdE4x1Dw6/oJFP1Ugq61/Tj+u577/3Yx/24aRXXQF7HxTWn9NxR0JzwCUuvwWa3WQpPuQUFwY9anFH3ko+gjH5N/Wcv+eyew4YUTSDH5KFQdWIWJIMM33NonA2hSR5N6m0CPJp3vfTwx4F4qhHACzwO3SCmPAfcDY4DpGJrE3d1870YhxFohxNq6hvjsXI+8Nh9RXYdpRAlqbi5qbi6KM4WmSRplllpy1cgVrb6ua0eCjkSn9yQ53Y3mTbqXGu2LVGx5agrprZmgSk1O5t6zirn3rCJQmEnWR5WMeLGGSfZqVGHUXIj33I4DIVRXIlaakFmocatlDWj1QQhhxhAIy6SULwBIKQ+3+/xB4L/hviulfAB4AIzMSwNpR6w4b+EaVm2ZhVQELUWGsdB6ej0vTf0LY00K1gEuF7U3HNZrLSw9YBTmfm5MZNGVISHTXwNkumKDHtbAf55r1O/Z+9RnnPfojxj1k0/Z78sFavt8rkSjvQAP2XCG8iU2NBeJYxBS0/X7KoUQAngI2C6lvKfd9sJ2u10CbOl/85IkSTLY9DtHoxBiLvAhsBljSRLgDuAqjKmDBA4A/9NqlOyWeM3RWBl0cdmW67l+1Kd8NXU3AGmKLWYjRsjrUUEM2qjk1v0dRh+X7sWp2Lrs97X9Z9NweiP7f3Ma7379D5SYnEM+eg5H6rUW0pXuMnsPjJjnaJRSfkT4kvMJ65PwVHMm8x2V5KkpaFLnyWMnUlufxpzJe0lrfVFi+RIMhcdg53OG86ALSI1Pt49hgnkTI1d4WaD/iDPP2cSfit/BKboKkCT9J0dNwa2HLxY8WCTFfDuuTG0kW7Gz0qNw0pqv89YNc5n4y6Nc/P5N1GjuDka5eLUc95XOna+90AtZ3193p5K5zkz99TOpmG8ndR+sevZEAsPkHsQbDsVCedA1ZOdPujl3QhUKP7/pmxS+sRYADUjdMIfAPONzY4kqyGHNT5HJ2qMFPtKSYvHOBvcofBkCZjVhVXXMO9IJpAlapE649ebOU5LkNKPvlJqcQxZvknxSYbB/uqvD3558SYaikKGElqkENgGVwZ7DmGO5pDUYhNr/vay1ePN07K+nwXuZpO13Y2uQVAY7RmyGcgz0pH3EE/Gu7Q3VFCI+n9YQox3rmH258GONCk2hQjO8CE2opCsWFIzU7n3pXG49+hWFBoMRb2lk/+tTih7ejLJxF+l73KzxlIW9nr525kjvX2g6Ey12Bbwd2h8LJ6VEJCkUOuHW/SgnTsJUkI86pTUl+t5GljWeyrLGUwFj5HMoFopMVgpUw4ko0oxIVpF4M7ZM1UHzt5tQc7KNDRNGc2SKgxp/egcB0F8HrUgcsCD6mpcqZIf2mpIZIoGkTaENTeroSHwyyOmPbWDZc/NRAlCYNwOfXWWSvRroqAqH6gmCsZRnx9Jjpx1sNTqac/l1M5+h7I9LyP7AQsNZPr550ntcmrYehYGvPgyFmqxJnc2+QvLVwzixogqlx8Sx3S3VDkeOW6HQ2QioCgUVY1T8v+xtBC5VafA7abnIwsrPJzLKXN/j8ZyKzeg4cbBEFxIG0RZCL5x1H5cEv8PSGR9ya9YOFGLnsxFryoNu9vnHYXZ88Vy7C3GH/geyJSLHrVBoU1nDjKaqUPDqZmal7idLdVF8ylHKzCE7Q8dyb+1H44+86Zxpax4UV9TuCBn7Ione7CtVWjpFb6jUz3AmfBamUpODq9M+x6F88Tx7EnD9WQnY5Pe25t1ILBJTzEeB0Pw3XEcISI0X3jqNX3x2AVs8I3i9YhI2IbCJriNJKITZpXvJUNw0DbEh0SzUmAml/33wm6S+tIGX3j41boxy/c1fqAqFHLVrvsvujJk1mr9HI3G47xSp8Rno1xvHrVBIkiRJeIaFUIj2evMzrjzSd4Opysqr1VPQX8+hTlOo08LfLlUomIXKqTaVHNXebXtW+wJxM8L2lfKgi+LffYIM+Mn/TOew1nuq+VijSR0twpWLcDTq3i6agU8G8clgl2d4VLeweMdlPNUcPj1IZ41Tk3qv+R/j1U9iWAiFaBu7fvbSFTTM8VN2ajk1e3Jxn+VipXs8K93ju03VHjJE9RQnf7LVnLAGq9dcE0Ax5tRpO46yeP0NQ+pzoUkdj/TjFP2vg5mnpqAjWek1t01DHIoFh9J1FWmm1cI7k5dzQUqPsX1t9NYnA1KjUY9e8ZtoMiyEQjTZ7ncz5qljXDTtcxYXbMJRqZKf0YwuFXSpcGftSXEr4WPFRp+PuzcuBEBYrYgjTRT9TPDr+pmD2o4m3dP245I+AlFYcjULlQnmpohTqUdrWfKw5sHdLkK5p+xY0aoZEinH7epDdzx85HSC6Vb+u2Mq00szMLmheXkh+d8/CsCS9N2oQ5CLfyjZFchDAqbSYo6eXMShxT6+MmUjZzp39Ot43flPNOmeHovdxKoQTonJ2ftOMThnaOrS0/LxUKzyDGuh0JPzTkj6Tnp/CRvP/AdOxUZAaky01/D89TOw7rJRsWIcdk0nfZeL6VbDecmhxHedgL4SiYNTs2Zn1shy9s2bQOrVVWyb9PyApkHhnISadA9qD34Cw5FYJt8dCMN6+tBTZw8tSd42/S22BdS2bUvSD7Fv0UM8/Y178GYLUg96Uavq0aRAk7HttCFVMV6SevqkYRh161ampVZxrAxKnY0DtouEcxJyCutx4zHYnnh0/oq/Fg0yS9LLOap1rWEwzWJj0uU7ODLZTrDmEDfsuJobdlwds3ZoUqdJ99Kke9EZvFDjns7zvsfB+x4H7zWM5/26cThq4OP3ThiwwAp3zmhcb+e5d7QDqI4XjnuhYBYqc23hC4zeXPg2jZMk3sUn434pH/dL+Swpn0uj5o665d1wpkkhR02JixUKTers9hew21/AEW8K++uyyP+4kbF37+KRY0VD1qaeXvIPvJYOgiGI1msa+p6OebwKmeNeKADdRi4+XHsmWZsFFZcF0ewCzS6oua6AGa99nyMJGgIdKUE0Htg1lwd2zaVqYyHqNieKy4uwWnmvcWJMLeIPNRWEFbq9xXOcYm2hqZ2HYyS1NkLHc+lGGHWt1kKt1sL+gIv9wY4Vn0NBU27dP2Dh8OPaqQP6fiwZsKFRCHEAaMZIUhSUUs4SQmQBTwOjMJK3XiGlbBzouWJFuI6mSZ0N/5lKzk43R6bZ0VsH771X53Lvgn8PicV6MAlIDc/2DADy1+q0FKhomSmgpJJvPRTTc8+x78Oh9L0snVOxcTjgIqfVihmp5V4VSlsgm64bL7tDNXxPOh8jWvU77srbPOBjxIpoaQpnSymnt8sUezvwjpRyHPBO698JhSoUnBcfQgR18leB6jF+gk7JxF4iJiF+vdUixanYuOLLH3HFlz8i4FDI2BtArW7Al21jakplVF6MkCGz872aZOl/nUqvDN+uSJ+HU7HhVGwoKFG5xvbnDWkY8d43YjV9uAj4T+v//wNcHKPzhCVaN/3+CU/gy7XhrPRicktMbknWZsEPDlzW63fj0arcV1Showodi6v1flotHJ5t5jJneVSOr0mJJiUuGT2X6Xw1/LPv6/OIVjIcVShtgu/J5tKESNEXjdZJ4E0hxDohxI2t2/JDtR5af+d1/lIilI1LkuR4JBri8HQpZbUQIg94SwgRkZtbrMrGhQKO1CjIuwbdjuLT8WeYMfmMJlqP6mzeXULNaBeFw9iu4JMBHlt5BgBljX4aJtvw5BTxtcvfjdgluDdC6nm0cj9oUsctZVQyTkVzNA+tJl2fVk0i2PYHLBSklNWtv2uFEC8CJwOHhRCFUsqa1jJyg1Z8MPQAotExZli8/O1f95KlaLhbxdbPq8+jdsUJzP3gZn41+yXm2CsoHWbCwScDzFn/dZRcw/L+2KP3kqPaCUgtquXnY+HCaxMCnwxiZWDtjEV69XifNoQYaIHZFECRUja3/n8R8AtgOXAd8NvW3y8PtKF9JYiGS/cNyF/eqdiY0mkQ+3nRa3yl7gRG//UAjxYt4sECJ7//1/3MtA6PeIhpd99EycPbyWncxd17NwG0aUTxnm1JFQp5agoBafgnDERbjPdrjSUD1RTygReNWrOYgCeklCuEEGuAZ4QQS4By4PIBnqfP7AxojIxBZEeqIlCCEu/MMuqmW/HmSC5/5Wa2X3pvXDgdDZSilU3orhaE2UKGEgrt7X948lBgvNDH70s9UAb02kgp9wEnhtneACwYyLEHyiSzOSYJTXQgtTKIyaPhzZE4KwS6WWW7X2d6Yr07XVjtCyDXGeXnEYJVnjIAplurhrBV4akJDm+bzlCSkFGSlUFXr85D0XIy6UyDJjA3B1HWbqdMm4C6s4KDSyeRrwZItBG1PQGp8Y0Hvk9p6mbkxFEA/O2xVADmL/0D483xFR2aFAixIzEsH53YE0gblPOE83d4o2UylvJ6pN8Pq7eiNTaStUNjvT9nUNoUC+q1FhZffD0jH9xJ9Ten8s1ly7nm8dcQEoSEb3/zewmbRi5J30lITeEMW5DBkGfhrMWLnVt4fNGXyV8h8Y/KJeA0oZvgu+9cw765b1JmPUyu2syJlviuFbCsOZsrnLUEpMbs125h0u5dNC2cyLO3fqEVzL/p90BoVI7fa0kSXRJOKPhkYEhftjFmJ3+8/Z/cxv9Q8PUD3D7iTd5rnszTb85l+dL5mHdUIhw2mqcXUjfduL3mkxrZdPKTQ9bmcPzj/y7jse9WEtBVJjzoQfoD1M1QKFG/uLeRquihStyh6Vr7aVu91tJrAtMk8UVCCQWjFuPQj1jz7DorfvrHts4+z/Y50y4u5+FnL0QW5XDgJyonFOylsS4fgL9OfXoom9uFgNTwpSrUHMpF6oLCURaqlpzA7xY8QaUWoIS+2WRUoXTrgJQUCEM/kPWVhBIK8eT80b6zq0JhurUa3aJSd0oqXx69Cpdm5cR8I4XbaTYfg6l+N+keHMLS7Ut9V/00cj+sIWNPBuaqBnyjc/jnwsc42+7FJ1VUIaJqpI1mTctEpDLowyF85Kj2qAVZxfJ+Hr9PKkmSJGFJKE0hHD1lAB7MEWpfIIuWIiu+THizfCKW19NpGm/4RteV/JdCVUVHovPFFKhWayEvBuq1U1hxSR+6LrEKE1ZhQkeyrjUY8dHVc5hka8K0q4pgfT1qXjqr3WNYYN8WkzyJx7OWAIYdKpqoQsGt+2NWHjChhUJAasx47yZ2zX+oS8cLSK3HKsLRpkA9hpCQUiXR6tLJf3YH+TlGNaGz8m9mdtlBDtw3nux39oPdhrSY0R1WVry6LOptUYVCQNdZ8McfUfJCORVXlBI87RjB1gS1zl1m3KPTSXF7oUHBVNXAv9+dx82XbSBdxCaN+kA43qcf4bAKU8zuS0ILBbNQ2bvg34SbBbmlP2Z1AsIx3Wrl6l/8l+du+hLmRg+NX56Aq8hol6lccvjxMrI+2w52O76R2bQUWvCn9V9oaVJvq5wdbp76kTef4mU7IcWBCMLV49dwQdrnAKinSVqkie/vuBLrvTNQqlqw1it86s1god2TfAETgFg+o4QUCpGo3YMhEDb5vTzfNJOf5mxGFQpLM6rI+OeL/GTdRXzjhPf4RsY6AA5rZr6S9j3Gv9OCCASpn2bj2AwftJjCWqbdup9LSk5mzz2nsuOrf+/y0mtSZ9G1N1A70/CgXHbTPW0lzzWpU6W5ufPvN1FQ/wmKz09aeRHnpG5hkvmL0nYAK6c9xXk/uJRDL5TgKQmyw1fISNM2xpqtHc5ZGXSRoZj6PbVo1Nxkqv3PpqQjoZ0j2fEktLQhuO6Eu7ua1Lls29VxkdJqmsXGz3O3dnhYV6Y2snveI9yRs5NCk5NCk5PpVivPXvA3lDGj8J02EXeh4SqYulvlS1svb0tSGkrVtWDzVQA4apQOnoSa1GnU3Pzw0MmY3l1P8TtNFL/TxCUv3cLfGkfysVfnzroTOevNWyh+9ZBR+1HXSVtTSZ2W2mWZ0SrMrJj0ImctWU36NhNv103ipt1X8dW957LSo1ATdFETdFEZ7L+ADV3bQDwizULFJX0sa85jWXPeceVdGcrUNJiCMOE0hSAa8p95PP/bTK5wNg11cyJmglmnelGekQBWSrLft5K7qh79/RRmn3sLgTTJqOVuALIqGwgCBas83H7FWfy04F1Weor46x1XkvLCWtCDAG3BS2PXwYpRJ/JK0TyUgM7kmhqC1YdQ7DaE3Yaek86Ttady7sh3gY4jjlmonJG6i92vFFPJaJxVGu6X1/HbqVdR9sBeAH6Y9zZaP6IONakbBVuxtDo49X+9Pl2xM601MEtJvG6bUCTE3W1vULEKMx/e+88hblHfsQsL865fzYq9k9CqU1CCgpr5OfjPbmLpxFfJNx/ldoehIUz8hx0lJQVTRQOr/jGDcxaX4TpmJ61QRb9iNpkr9xM8dLjD8fddV8Kll3zIXOcu/FLlB69cS8GnkpYCFWeNRsM/J7HqTkMonN5pFrDdW4Q81kzxf6uRqoK0WNCtKqen7Qag1NS/9fX2wkcVyoCzYU23Jm7AWSKREEJhOMwhVaFwc85Kxtjq+G/2VOacvY+znDvIV13kqpLKoAmRbawZ+vNTsWgFyIZG8t4/xGG1ADkziLqonkMHMknfkgmdhMKCC9ZxR866tmWqqZf+kXNyjVWPzzaPZdKPd3HHkaUALL//L202F58M0BhwoB9zIY80ojidoKq4RtjJVl3A8Z1wJJGIVraohBAKw4UxZic3Zx7k5syD7bYaBrhMRUepNoZw58/3cVPxu1QHM1ndXAYuH/ePeplRJj/XplzBvq+WMvOfmdhVY2699d4TuCDz3x3WrUebneyY/y8AGktf4Yy62xh913oAZj1+Ky9deQ82ofGaawrvPXgKuYFPAdCbmwEwt+h81jIGgIX2LcNCMA93oiW8hZRRy5nab2adaJOr3xgB0JoOW0bdMSMRAnMWbl8MwDPjn+5grS8PushRLDgUC3sDLt5qmcAN6RVtL+pGnw+r0Hqsl7C8xcHfx40HwFQ2it03FlJw0iEqDuQw8f4W9I3bOn7h1Gl4fm4IiFemPNHBbdonA5hQk4IiisTSGSmEWrhnXbvaLN0Sd5qCVZiJhc9RvAsEgLcmvdL6v44vd/vEsGPMTsZkVNF+4SiSufaLDTNBGNMBPF6sDYKzC3Zx8+THcZxvpkn3owpBnaaweOV3GfWEoGJvLgCvjypiqrWaKRZjyhGJsbBRc7d6UuphKy0l6UisBUJf6LdQEEJMwCgNF6IM+CmQAdwA1LVuv0NK+Vq/W5gkSZJBpd/6n5RyZ2upuOnATMANvNj68Z9CnyUFwuDT2Ydjo8/H7j9MRrHbUex2pNdH9tYASzM/I0dNwaFYKDQ5yVNTmGKxs33hPzg804LiVVC8Cj994UrurFjcp/NvC9hwyQDVmkZl0NP7l5LEDdGaFC4A9kopD/a6Zy+Ec0yJB0elRKL9XL9Wa+GSN26mqUwl/S0b6W/Z2PX3kSCgWYafp1mFmZGLDmCrV7DVKxSvDODVIvcv8MkgRaobq1DIUKDU1H9vxiSDT7SEwpVA+9RC3xVCbBJCPCyEyAz3he7KxoWbrwbpWlYu5AwTy5LoiU5Aanzizcea5WHO5Rv4TclyflOynHtmP4NuFjxy5LSw3/PJAIddTnI3BMjdEEBTvnPPAAAgAElEQVQqgq07RnQoyx4Ot+7HrfsJoFFispOu2ElXLEmDZB9Y5/P3+TvRHjQHvPoghLAA1cAUKeVhIUQ+UI9RY/KXQKGU8ps9HaP96kOk1ARdaDDsS8IPlFDglI6OqdUr0SP9nHLfraheePcHf2gzwvpkgH81lXH32+cx/jE3UjU0CX+6BdWvs/9CC785/0l0qXCOo7KD8ba7AK1khGPf6OsqRF9WggZz9eHLwHop5WGA0G8AIcSDwH+jcI4u5KmOZGcLQ2jUCN0bw5MQ2hdHsWPBl6Uz4e4DLPLcRvNoGPOsC7WhGd1hY0ymH1+2jaDDOIY7VyH/uZ1MPJjG3Ru+BsAHN2/gb0WfoAqlLUdjuM480EpNxxt9XYWIRZq3aAiFq2g3dQjVkGz98xJgSxTO0YWkQAhP6CXtbZ/Jsw8Q9PrI//cGCkaWgCKQTccQditH/7eFh054lLLW3qGgMO2sGxn7Jw1XieDE87ezJOcDQt1Hp3u/kk+9VubZkzahRGJAb5YQwgEsBF5ot/n3QojNQohNwNnADwZyjp5IGiDDoyPbohO746bi9xCZ6eheL9ru/cj9FWgNRxAtHm4Y8xFTzBacig2nYsOhWNh45j/wFNjxZetY1SABqfaYzyHEZEtzVK+tO3q73uGMJnVqtZZebT6RMiChIKV0SymzpZRN7bZdI6WcKqWcJqW8sJ3WEHVqNHesDp3QmIXaqxq615+HpywbYbWCrqF7vZhKitmzpJjFKbu6aGJOxcbCX31Axg7B9nuncNVbS9nqD/ballikmwtHPDn/DDY6kjw1JWqp9BJaBx8ORsb2I1xAaoO2mnLPivOx1ragFhWgZqQDUP61kTz49fvJ65QQJZTn4easDTSNgfTHVzHqRbjkre8OSluTdI8m9ah7iya0UBgOOBQLTbrHSJ5Scyo/rDl1UM5bOLkW790eKr5SjMgyVo3dhTozLN42LaFJ99Cke/hF/VR2BbxUByWpB1rbvbmKcY/4k1O4ISaZo3GY4pM6LzSP55VNRgHvvxat6ddxXLo3YhVyxQlPYBVmto8LcDW3UvSXSvLWwL5LYJrFGIHOWGusJBf/VPLoTaeDVWPyy/sJAsGaw5iBWs2dLPY6REQrVLozSU0hSZIkHYi70OnjmSbdiBEYzCzUAPsDLpaOOgM1O4vqqyYw59r1jLbX8eIvFwKQ9tIGapbOZMLlO9lel4/9pXTS93nx5FmovijA7gX/Si4RDwF91RQSNnS6P7SPl2jvzJFo3nSDLQxCjDY7QUq0+gYKH/6cA0+lsr8ZUr2rAMM1taVU576Ry0kfbWPDdJ2vrvgO9iqVrA8Uzi9ZzIqJr7Ydr7+5K1Z5NU61JUOsh5qEeGMCUuvRoNWs+7EKcxfvrkQSCEPNwZ/PQZ00Dr2lBa2uDt3rRc3MRM3MpOLHc3jtsrvJUVMwC5WZFpXbznodX7aOsybI/o9L2R9wta2eODoJ5kjJVZPRlH0hVsWO4v6tcelezKJ7325N6uwM2Ad1OW84cu0l79AwKwdh/mK9Xx9TjD6mmL9945+MN3csqDvbvo/5czZTc5qJMY/XMf/1W9nk19jkN55ByKHmnG2XUK+1RNSGaJdXG85Ey1EpHHEvFHqzpqtCYXnTDMqDHsqTcfv9Jt3kRugd7UveXDveXDsjTMfatgWkRpPuoUj1cU3ux2ijvegHKhj/sIfb913K7fsuBYznogAtfgt12uCV7zseCEgNr9RipgkPC5tCuuohXx0WlzJkPLRnDnk7mpGBL5ypdIvxMjfrX0wHfDLAnoBKlqLRrNvIXmFD+nywejP1romAUecQIFOxs/LEZagxCNo5njELNabpBRPyTepsQLwjZyfQVaPQpM6HXlMyICcCrh3zGVnLXEyxVlMRzOLh6rns+8ww+r3RPJUTLFtQUGjWg/yq4hIK7M28uWsS41fXoVutKBnppNqMFPU+GcQhjDwKDnH8uh8nKgkpFB45VsSS9EO97ncg6KbIJID4T9o6lASkxrfSd7RO1SzMtLo4f+yrXM55ADz40VnocwU5pmbWHBvNjjfG0fKOC/sZdpqnWEg9XId0ewg+XAbAm3dlsdhxLGnoTVDiViiElhk7ryjUai3ktpvj9kR/KxsdT7h1Py4Z6BK4ZBYqE9KM1Bi++zJ5ftt8mue4kYdspHjhjAdWc2vWJlQhOKL5OOuTmyi7ei0At0++Fu2qx7kwpTF5/xOQuBXlJlRMqG0p15p0D+VBFw2a4EuOyGpIHs8dMtxKTChdWnu/Dodi6TaScYq9kin2SvStOyl6pYKcV204yxXS92vcmrUJh2LBKswUmpxsnPsvZDCIDAYZeecnPHTJl3myOT/scRs1N8tbHPhk4LgqFjsQQqs5g0HcCoVQpd1QzYB0xU6x6qDM3NUfoa8kchBPpG0PJxDrdT/1rT4dkXCG/QBn2A+gWK1Irw9ntR+pQuX5WpdQZYdiQVitRii2okJ1Lb/e9OWwL71TsTLbWovS+i+Ub7MxGQrfI4MVhh63QiFJkiRDQ0IIhfajo6mbkuiROi/tDbj4S+NY9gdc7A+4otbGcNRqLVHNiAPQqHv6nWWoULVTqEbuSq0K40f3esHjxbJhLya35HdnPBt+/5xs1JxslBQHWmMjo36nc+raa7o8F2NJzY6CQEGgIzms+VBE0p+hO3oy2kZb841bQ2N7Opc070yT7kFtdfmMxI5wSeomcgfBryGz1fEqmraNHDWlLXCqr4TaEWlMyHPHpgHQcukptBSqFL1ahaVZEpBd793egAvpb50qaIYQEFv3kvPHSbz+cCoXpnScGrS/JyodS+OFhN5wyaYU6xicaB87oqO11m+oFUJsabctSwjxlhBid+vvzNbtQgjxVyHEntbaDzOi1dhwF69JnbU+Z1s+wZ5w637yVRMlJntE+zdq7gG5TpuFGhNjZ6SBUy7dG7b93d3HzsxP2c78lO088ae7eeX//Z5Df7VydLwgQ20x6jtIDZfu5dFjOVx154+QbjfS7Ub3GEJL93pRPtzA/y67PmJX5ybdw/6gRqU2fAyQ0XxpB8OVP9Lh8hHgXuDRdttuB96RUv5WCHF769//i5HyfVzrzynA/a2/o06oI59p80M304rOHNGDlJp6T0Ti0r3sDFjDRu31JPnjKTLTqdio11pwCHOvo264Nn9RuNb4vW7mM/ytbCT/rplLc946dnoLeXzryVg2Oygo96FNGwtA7ewUmiYHUVsU8taAyQ0vu8b06FuiSZ0gGgGps8NfyBxbdY/tjVV18nhnMFbUIhIKUsoPhBCjOm2+CJjX+v//ACsxhMJFwKPSSNSwSgiR0Snte9Roq20Q4f4OxcIObwqlpt5HocNakFNt4YVHEI2aoBubEKz1ZeGXKkc0J58eG8OvCt/u4oLq1v1YhanDixeQGgoi5gIk2u6wU20VvPjLhfwtfQyp+1sYs3ELUpeYigqoXVQKwJxr1vOjvLdxS5Vvn/A1Wg5lUmA+2u0xfTKAWw9gFSbSFRs2xU9OL7YPqzDjkrELCjqeGcjEOj/0okspa4QQea3bi4GKdvtVtm6LWVbnvnCW3Q30viTXU8SeVZgpMZkpD7owiyAL7R5U4eb6tFrCeU+GG80adS8HghZOtkZHKAyWhtIiLZjf3oA9PxetJBclIx1ZkEvD74O8P+0vQOh6nWhS5/yiLfzj4Dw+dY3jHPv6LvkuACqDhnt0qcmYbp3v8BKJqO9r9uJ40uLimVjcoXAm5C7pnbqrJRlrollRp9TkZJ4t0O+OtsNXGLW29KcN/bFan2U7iqm0GK2uHuWYB1mUy96rM3lj2mM4FEsHAagKhc+PlWCvMPPy/qk82VzM8640ztt5Hk81Z7LRH2SjP8iHnjIcIvaq8XASCLH0tRmIpnA4NC0QQhQCta3bK4H2udVKMGpNdkBK+QDwABjp2AbQjiGlvx0tT03h2rT6fn030hFPkzpveeytmkzX/fvTdruwgKYjg0G0nXtQJ4zFNvFoWONnQGqsWj+eUZ/6OFaTzq+3X4q9ViA0KLhlBdMtRvebaa1Fk8nK1H0hlgJuIEdeDlzX+v/rgJfbbb+2dRXiVKAplgVhEoloSfdjEfo9bPQHadCcUe1AqlBAVUAIFJsN4fPjKk8Lu0zapHtJ3adi31pF3jNbGfP0UUqe3kf+xeXMs+ttXqttx40iiey1OtREpCkIIZ7EMCrmCCEqgTuB3wLPCCGWAOXA5a27vwacB+wB3MA3otzmNkLBPIYV2lA9hyrPYSREq+M7FWvvOwEzrRZmWhuics72/GblM1z6yVJsGx1YzqjHuVLlpHe+w4YFfweMZ6BJnfc9hViaJDIQQPpbHa5UlT+MeY5woe7RZDhNFQabSFcfrurmowVh9pXAdwbSqEg5Zc31ePam8c1F73GC3bBtzrHVRWRxD/nkKyhxGTilSZ0dAR8TzdYuHTxWufnanzuI1q39ZYrZwlOnPUDG6X4KVQun69cy7jcWFnxwKwCN87x8aeJ2thwpJGgTCLMZzR9ALT+E7vOxxjOKaZbulyd9MhCTasr94Xg0Th5fV5skSZJeSfi6D6FRTWu9DqswoSPjcvSPBgGpoaPHfCTtabTWpM4Kj4M51iOkKTZ+eOhkdp2biXAYU4KWKQU0l5g4eoaXqaXVtAQtHPXYUZ/MImvFLrbfNZb3z7+ng2tz++u7q34apZYGrk2r6uCaHeJ4G7mjxbCu+9BepVOFgorSYSHUo3vxyUDUqvDGE8ZLMhhebd2fwyeD/LtmLrai95hldVHekgmKIFhhLDJZK6qxpziQFxTxq5EvMdIksAozPy6eyQtTTyNnRAN3VF7A3qZsAFZOe6pNAG0PBHjijTMpe87F3YtScZ5Wx9iMetZ+OJGlF7wBwI0Z27p9tqG+kehq/1DWwEg4oRBaZptra+q2Y4Tce+2to0sid46hoqd75lAsPFn2BrsCfpbsX8ymz8YywV6FYjeeh5KbTeVFJXw84x6c7Qy/d+Wt48arPsIvFdb7RnBXlZHu7buV8/hj8dv4pM6H7onkrZX4s2w4T6vjypHr8Opmtk/K5+9vLgJg7OJDnO9w9bjMGq1nHqt6jT2hSZ1pFo3BEP7hSDihoAqFubYmjuhBnD08d4cwc0z3oghBuoivFYlEHcXat9ssVKZY7KxfP5a8jRI9zYHwGEulLZPz+fXND3cR2mahUqhaqNH8vN04GbExFYCtz09lxvnjmT9+F29vmcSkDbWIoMaRgAO3buHl8mnYLAFSJhuuMNmqC4/0E9B1jug6o0yOmN1PHZ3+vpxu3d/n2IyA1Ibc0JrQNoUm3YNTGMtznTtFcg46OMJno8/HDb+8hcYFXrQWY4wxN5hYc809YZeH1/n8XP7hUkY/IrDtb3Xe0nX8pTnsvdKCVCX2ShMj//Q5orgAT1kWliNe6k9ycnSi0VenzDjAj0tfYXnTDJatP4XioiO8N/XZYWFH6vzMustV2h8itSkktFAISVRN6uitntQ+GcDeml48GgyF+tgbjZqbTHVgHoD1Wgvpim3A16ZJnWO6F7Mw0r8DvO8ZwSm2CqNGJXBX/UTmpuxirPkYF25cQsE369HqGxAmQ4goo0vRMlPYc1UKf7rgUTJUNz9b+i3Mb65tO4+am4ssygGgaWI6tbME4x4/iv75dhCC3Y+cxL6FDw/oWuKB9kKhfZh0T88pUuE/rA2NIULSUxUKtGoGa30OTrA0Ry06MN4EAhgrLAMVVjlqSlS8/lShtAmo0HRuhq2Shc/dxhXzP+GwL431j07j5fOm8bfJT1KafpTac8bRUjgB/5xmAHxuMxmrrHzlrFV82dGMWah8595n+N2vv072hqM0TUwnY2k5F+V/AMBrdVPxLRuN3LbXOKGUjP+bn8A58SfA+0r7lzvSKNpoa4MJrSmEY3mLg33+vHZ1DIYnLt0bt9dXGXSxpHRuh23Vt83hZzc+TkCqbHCP5DvZH5GlGGNSnRbkrppz+aRiNBvnPNym/b3qdnLLx1cxuqSOlyc+03a9Phlg2kdLGPOdKrT6Lzw2v7TlGLdm7Ru8C00wItUUht1k+8IUN9elbeP6/RcMdVNiijIIj66/RXs/8nQV8On7Nda1jCJLdbE06yNKTV9kyxptdvLAiA+4f8aytqVQVShcmOJm+8J/8M7k5R0EoFWYeXj2I9QtHo+pIB+lNe/FI3s65vLRpJ7QMRBDVTB52AkFgEzVwT9Gvdz7jglOvdZCvdZCTTB6CWjdup/9ARc+Geh3Ork7Pruky7bmUpVCSxMWoYXNj6kKpS1Iqj3dOlChIC6rZ9svS9Gnjwcg60En/1N5GrsCLTRqbp5y5fKUK5d1Pn+HFyyUUn5/wNXvJLiDwVBNhRLaptATsSzAGQ84FAuqNDy2tChOAR2KhdJOWaL6ytnjd1HZ7m/F4QAJZdbDTLY045Uw0KLzE80tfDz9KfTpOifnXovy1hyOnhjgyFMncVHaSXgLguSuMl4qT65gzuUb+FXh22Qq9jaHt9HmrgInJDyG6oVs0j1tAWVDtWo2LDWFJEmS9J9hqymEoz/OJPFMaP6tiOgai8ONUPVaC2ahtPke9DSSzc/czqPt8uzIQBAlACda6slWHDToHjb5veitms4Ui6lPI7Mm9XbVklSemv4QTdOsTDD7uGHShXiuthM8WIFpRAkA/rJcGi5MMdyQehl9Yx2B2huh+zuUvjXHlVDoTiAMtQdZCK0Pbtkd4j9i2irjXKmd7l24NoZU78eqTgWqOPSDOdx4wytcnrqDbGUVqjAmDZmKDa/0tBWmaS8Q9gdclPbiodj5s5EmE6BhFTZuK36DX4qONo3Ds+wsG/VyRGXXukt/H42XtPMycrwOUgkhFKLhrNMTAdl97oB4ZTBGkvbza7WHmeboN5aweOomtjcVGBvuzEahij9850EWOQJ0TmZrFmrYCEkwKlKVB91tjk+REKp8bRNBbtlxLZl1RqKvYGUVAGkHiknv48tXq7W0CZEgGroceORt+++HWxUJJ3z+3DgKgG9n7AYGJ/9HQgiFWAoE6HtW4FgR8sqMF/ebSDtf2eOwyTYds8vwaDRvPYAGTLY00leTYqnJ2edlRKswk6poXL33ItRHcvDMy8Ra50XdVQ5A+qoKqoO+sIbF7sjstAQabVSh4BBdta/2Wqsmda5O2wqAVyo06xqpioqTrol3oklCCIW+0nn+myiE/NwTzSvP9M66Dh0ptPj3trusNe193+hPh3cqNn418iUuWPg9zp++iZ3H8vD/fhwAljfWcm/9mdxduD7i44WWZAebUBVuHWl4NLbaOJp0jWI1doFfHdvQC92UjPuDEGJHa1m4F4UQGa3bRwkhPEKIja0//4hl47sjU7EnnEAAIiplNxCadE+/61B2x/OutG4/+9ePLwk76sfKKWeKxc5/FjzIbwvf55aRb+HLNOHLNMTVa8tP5RlXesTHGirt0SxUVGFMEXQkXqnjlTqlpugm4O2JSM7yCHBup21vASdIKacBu4D/a/fZXinl9NafpdFpZt84XqMie+PWyoXcWrlw0M7nPOjmY1/XZ+HSfdRGWFuyrxzw53BYC/J03clkrj5E5mojF2TZP/fy78vP48TVVw2Zp2B7XLqXj70dBWZngd2ke8lTHeRFYfrclylZr2+PlPID4EinbW9KKYOtf67CqO0QlzRq7t53Ok64b8R73Dfivage84KUBuTp0xHmL+bH7ktOQbxbzI+eeoJTrF1L9GWqDrIVOy7di0v38ufGUZRHwStTkzrL66bzdNNMPtw2HtwecHtQc7KRgQD65l1on2bysXfojMoh1+vnXKWYhdb2su4NuPhvSwmudun7c9SUDmnwB0JfjhGNIfWbwOvt/h4thNgghHhfCHFGFI7fKz1JwfX+1MFowqDTl5E2dH9MqD2mWesJVze1Jkyo7PmaFbUgr23b4Su8vDJhOQvs3a/qqELhM18Kn/lSeH1KBreVX9RrG3qLZVCFwp9HvsT3sj7nspPWoZUYpe3ITEcW56FmZeDPkOzyF/R6rlhzTeohTraaUYXCq24bix/4f8yxHYyLVbABGRqFEP8fEASWtW6qAUqllA1CiJnAS0KIKVLKY2G+eyNwI0Bpcf+aEZAabulvS7QSjgX2oVcVY4G1j6PHQNfEe5pjj55QQ2BENi2nFgPw+py7MYsvVh1cuhezUDsEO8EXz2ZW5acc0TSadLVbW5ARr2Aop52t9u3JV+2YhcqP8z7hy2OMAKmMN6tQ0lPBmULqARjfQ3r5WODSvWhIqoOSsWajr7Y3Yp7v8HL+d+5j4M7f0aHfmoIQ4jrgAuDrrbUekFL6pJQNrf9fB+wFxof7vpTyASnlLCnlrNzs/o1eTzbn4xtkH/F4CaBp1iMXdqpQYuIk45MBXNLHvr35BB0m6r7ipe4rXkaZOs6BA1KnWfe3Lbl2Jl2xM9rsRO8hhiN0DaHr6E5jaPOtQMWTq+DJVdAaG5FNzSAE3mzBGbZg2/4u3dsnA2R/2BeEp5vHUGQSKIgh95rsjX69TUKIczHKzl8opXS3254rhCEChRBlwDggGeCeJEkC0ave3k3JuP8DrMBbQgiAVa0rDWcCvxBCBDGWq5dKKY+EPXAU+HpqLaoY3GhIq4gP145UZWh9GTSpM/npm9EdOuMf9qBbVEbnGQlPOmtuaYotIm0urQ/LgN0dT8ewO1RqAXwZxraj15xGwAm5G1r4/jUvdfiuU7FxhbMp4vOG6IuXbZYS5BtpFZjjLIFwd/Taw7spGfdQN/s+Dzw/0EZFnHNuCJYe42W5syc7ymCwK+Bl/E82o6SlIv1+TKlOtu0rND6c9MV+fYkb6Ou9/Ub5GZS7Mlkx6UUCUuNNTxbnOzRUoVCimik+2ygleGBtCRNOOYBypeRrqfuIRh3LkEAITVcOax5yVWtYQ2FJNy7d8Up89PBOxMuLF8/09R5FOwORjqDpgqloxTkQCKJVHcJWYcZWYW7zzATajIPRplFzc2JqBRX1GRzWPDgUCxenuNoMeA7FwuVF67i8aB3pe2BS2iH+NOq5qDslhRLRlJicmFC7XaVJJOJDF04yIPYHXNgENEtBiWqO2KjYW3RoyJlGlxJFCJzC2uafv6zxFKQKgTQrVosZeewYOVuMUdOtB7CqxnH7a+AMZYnuTkXXkHwnYy9Xn76VHLXrSOzW/TxfPQMAe73O6wcmMy9tO2PMsXtpVaHg1TWOBl0Raweb/F6KVC2ukgIlhUI3RJoYNR4SqAYQjDalUNjd51ILa/HuLc9juOVBt+7ndXcOz+2cznk/XMvFGev50fbLyLkjF+eKzQC85h7B11Mbunw3RHf3rFZrIbM17Xz7LNHhCL1E3b1MJ7z9bQpfNbp3xsZD2Ooy+cF53+Cc6/4aU1+Avr7c0yzR6TuNmjti201vJKSeHombansVtj+s99t69YYM1ZiIhCbdE7MkomNMPRuwmnRv2OXAymDf4yAcioVUxcPvZ77AnwvXMs+u8/70x2mcmoHudqO73fx0zYXdXqtPBnjBVdLBpXd/wMX3qmdTp0WnO2pSZ8LSrTif/Qzns58R3H8QddUWsjdLDgbjY0k5mrh1P1/fcxkNUYprSUhNIZLotYGOBqdbdXR6N+ZFKpljGaDVWxtCeQnbo0md3YFMRpv7Ljzn2lo4on8RFu1QLByZKsgePRKA0sdUriubz39Gvhs2Eeu1afXsD2j8rm6asU0J8v3c9xjTLodCXxLOdEZHEpw1EfPWg8axGhuRmjGQbPYVkq8ebpsKDQaxzrfok0EeHfts1KYgCakpDAahSLWeGEqXVJ8MRJyCPVyH/NBr4hRbF0fTDnTnVuxQLKh8obH5ZID0XeAZl4tnXC6aTaH+eyWMXXEj2/3usMcYbXYyPeUg01MOstOVT47a8V436B6qNOO7fQ1gqtc8HDrVgSzORxbnGxulJGNbE/+pmcNzzaMjNoBGq2BOLMlUHVG1SSSFwiDRqLmpj2JkoAm1zfLdXcftKVZgnl3vVXsJBeOEO4ZXQnnQQ73WwnUHvkRKTZDGcRbjZ7wJf5aNsY9qfPWvt/FSS0bY6dwVziaucDbx99JXqQ7KDvvkqSlt4cJ9yWvQqLm5p/4M3EU6IhBEBIyXX508nj1fy+BAYyZ3fbQ4YqFwPK6Exd30IZTiuq/EKjV3bwU+I8nvuDfgYp2vmIX2mgG3J2Q0bN9Zu+u4r7jT+LKjscdUapEQEgztz5OhKLRIHZtQmZ+1g4dyJuDLND4zecB6uAVReZjAWRNp0JxYRfdaSabqwKn0veRbyDCpIHiwaQT//PtF6BZome3G7BIEs43RU5s/k8xf7ufzkY+jCsHTzYUc0iAzsXLZDBpxJxT6IxBW+wLMtMTmCff2wkcyhUhXBFMsh3AoA5tuBKTGM6485tsPUhjBktfFKS4gOlOccF6Kaa3bv+LczbKjGun7DCOe4gsi9lfBiEL84zxc7NxN5zyNnelrNmcdiRnB/qCXm/d8Ffe9xaT7A9RNM2PaY0ezwdNP39/WVqP9hlH42rR6fDL2Uz+fDGBCjYm2EcvCx8efbpQkSZIeGRZCQe0m+i5eyFFTcAiN6qCvbZsm9YhtDCGDYr3m4ZX6E3nBNanb/fqyfSC0T/7xuT8Ne40b8fFGxMcbkWu3oHu8CJcb+xY7FcHojsoNugezUMlUHWQpcPDjEdjqfHgzVEa8dYxRr7h49vI/k6k6yOwmr6EJtc2Iud3vxq37+3SfOttIwtldTKgcCMYmyU9ftYSt/siXK4eFUJhptUQtQ02sKDHZyVAUmnQPlUEXQTTS++D0ZBYqhSYnfy19hSpfJpWdMhX1pE4qiDajYyx8JWZZXdTd6cc0eiSm0SOp+PEc9j06iYpLR1D8h8/41t230Ni6klAedA3Yh8TR7jpz1BS0sR4QguxPapBrt+DNsTHF0vPMONRfzEKlzGx4gfb1RTfgOcQAABmOSURBVAtIjcqgi8qgi/Kgu4tQUYXSYZk1VkRyP0ebIr+2uLMpDFdCIxtAeh9lV3tvxDw1heuzPuWwZqFeMzSPrf4iLkipJL2bKLzOwjLaXpjpip3VM57iwEpjVMxXTWhImk7XuOzIj8jYE+DC7/2AIxNVgimS1675A2O6qePY+cUMt8b/mS+FM23+tn0lYP58L7J1WVP19T1FfF+xCjOa1GnWW7UlXzFmewWFPWRc1qROsDXXdTSXsz/3wzhTz1GbfXE3TwqFBEBHQruXY7TJ1uHlmW5tACI30EbqhdkXwo2KTqHz3duf5Xf/uYKiTzw4K204q/z8ftFC/lnyaYd967UWflc3l0JLE99I39LWwRt1D9WayliTwuPNowC4+4WLCI70cuesV9jsHsGoBxU0VwuKzXA2082CyqCnTwVl+nvNkyxGO8ebG9oqYPW0f2glKJoOTSeYJWoUhUxSKCQAW/1BGnRHW/qygWbuiaQzPu9K41Jnz85NkeCVFmwNkooFNoIpEpPPzJvrp+Iuer9DFqWTX7yVtD0qzWU6j5fN5s2THiZTsVOtqXz1kVvJ+VyjucQQhHlVGqmPNPFk1Xh0txuzbRsiOwsAU242B+dYOKzZKTIZanWsVgDa09fjR7p/aLoXpPt8l9HOqjXshIImdTzSH1X1uD+1JtvP3Xt6oJEcY5JFAbyAedBKlF+Q0sBAlzN1JM9Wz8TkAdUrMLUIHNVuRr5i5ptTz+XO4v8C8MP9l1H2YgBPDrhGCnyrspmvf4unpz+EV5rRbJLaGQpzFhoBV3Y1wMqXZiD0AmZeuIWv5KwnVz3GI3Vn8PbGUdizj7G86STysz4BoMhkHbCvxlDxRb3QwWu/kD3kxRssZp1ok6vfGNH7jhES7fXheq2lz26kbt1PU2s+x6xukm+A0db3PE4W2j3ttgX5yJvCbGtTl3liOOeleGZ5i4Of3Hc9aQc1bHV+TBv34D5rEhWLBErA0HjG/HAVAGL2VA7PTiWtIkjKriMcvDQPT4HGuCc8CE2y71bjms8YvZfVz01j0VWrOlR9atTc/LZ+Dq8fnERwXSaeIsNr8fo5H3Fn7rZBvvL4Qy3cs05KOau3/YalUIgH+uKZWRl0kaGYaNaNTuyWhrFuqEOyo8X+gIvz//X/MB8Dfd5RPp39MHZhYYXHEHh/HTsRADU/D1mQjXLUhTzmIjipFF+2FbMriPmIB+H+/9s78/ioynOPf585M5nJZAFCFjbBCAQFBayIonVBb13A61KqFfGqVUutuNXaVm179WPV2ivUulQtCld7WxHvB61eRVCxbqgVkCAERSBhCQmEJQQmk8xyznv/OBOSkAyZSWbP+X4++WTmzJmZ5z3LM+/7vM/7/MzAarA4H8fmWu799G1Od7V3jtVBDxeX38CAe0DZzdeahuTy7txn006OL9ZE6hQiqdE4H7Nqc51S6vjQtvuBHwO7Q7vdq5RaHHrtHuAGzBqNtymllnarBWlONOXSWiLWuaHr22v4CZA5pelLHbnccdU/GOOsDt3EprM7L9vM03gitJ++qw521WEAnDqWylk2BhXupqqihL4bnBTNXwWAbPQTBBoNJ9B+Om6IPZebRn7MoopWHQrnGnh072juLdwQ13ZmCpHEFF4AngL+etj2x5RSs9tuEJHRwJXAGGAQ8J6IlCmVAjpdCSaS7n2D0YShVIeCpYcHjroTR/CpALpScSnt3h1m9qnh8LSYll/u7A9LWLuqlFHP7wPdYOfkIt759ezWIdsJZvLN1fafA1DwjY+s5RX8qmIa5054qcOxqQ92HOq9+vg5/PyBdSkhtpLqdEs27ghcArwc0n+oAjYBE3tgX0ZjKEVjm7nrnnB44kyzCvJeU1+8hp+qQM8l2eLJohFv89602Xz76xwqZxRx8vXlHWI4Y7KyufbmxVx782KOfmgDe6afiHqvgOlV32sX1NWVwby3/q3DdxR/vo9rt5wft0I3mURPZh9uEZFrgJXAz5VS9cBgTG3JFqpD2zKOI2UQRqrG1E9zk68MKgJ+xjjCB0Yj6SUcPk3Zx5bNxTleIIvSFOkthEMTG6WOXDZN/m8CZ4c/rrP6bgbA0W8LH/32Ex68+lp2f1vK6bdfwYIxL+AQeK7+FPIqYcevTiOrwYyXFX7lZc9IN2flf5k2Adpk0t0j9AwwHBiPKRU3J7S9swn0TiOZIjJTRFaKyMrde3vd6MLCImXpVk9BKbWr5bGIPAe8GXpaDbSdRhgC1IT5jLnAXDBnH7pjR6LobCahs1+zzQEPN9z0M7bPCPLl2U9HNPugiY2AsrEt6GVoSG6tN/+aHWmGoO1rk5w69oZm5NM1ZC2Fm/guvikns316kJW/nUO+zUV9qGbhKR/cgn0bzC3/Lnecs96KK3RBd2Xj2hYOvgxYF3r8BnCliDhFpBRTNu6LnpmYfLwR6jYebXcz79nHWHD6XDYFIp/+GpuloQms9QdY6++4uMUMGh55LFwfo6Kd6YS/uH3cofCeKr49Z96hlZGFWg6FWg5rJj9D/ri9jPyTnxM+upF3vI64rBzNFLorG3e2iIzHHBpsAX4CoJSqEJFXgPWYatSz0mnmobMov64MCrTIphdb8v+HE11tv5ZfwDcPjgMgz7aSApsNp9jRRDho+HGLI6za8h69kX/5+jPVnf5CJJHiMXztBqbamFE8OnRep+sPcm0ufnfs6zy26jhKp8McxvDEBwN5s+ztQ/t0J2u1LYnKNE0EMZWNC+3/EPBQT4xKFp2d1LaLWKL9rEhLxAWUzn7DztjsbUBr9810CrZDwirQKpJSowslmul4+thcvcohAOTanOw93kXxB+bzyh8WUKSFv5zNnIZWKioHo49svZGjcQidOQBNbFHpS3ZFbdATUXWteJAZri0Kyn0+vIb/0F88aSms2tX3eJUfl+iMd9Yx3lnHILszbHGQFpGUIs2sj+gKfUdvwyEa7ot2mk9EcO0WanSdPXojnzd37Jw+tXVyu+elL6moCqD4VACfCvC138vmMHoZ7zYNjFo2Lpy2SJ8kzhj1OqdwQDkJoOMUe8IUpLuanuxjy2a4PZsSzUnJEdZJtKW/LZtsyYrLMuh0YfnYV/l1ZTl1/xhF380Bzl9yBzdvvZi7N01jXsMAaoMeaoMelnid7H/dnBnXSorh1LE4Dvj5QfmNEccWnOKgWQV5of40NFSnN/8VuQ3oYaqABZSO1/C3K3SjK4N3mwZ2akMyk84ybpVkVxjKhivULY81keYndEa0w5QDRnNU0u1d0RPbk8mZLvh8wv8wevOtDPgAvt44CsMBixadzqKdZQAETziGQbU1qML+bL1+BN4RfgYt0TDe16g9sYnBRyiM0pY+tmzKXDsZas8Ou3zda+gdiuis8vk5aLjoawuiY65vGWYPYADV/hE4cht6cghiTq9zCmdnG8SqwvHhJPKmitXYtQW3LSttg2VOcTDvkrnM2n0TedsMNL+Cuj1Q3B+Aystc/HLqF3zP/S0BhLW+gdxfcBGGIaz398eWtTdiQdi+mveIw7XO4gAnObMAA9opjpnXyp0FlZE2s0c0RDE71eucQiJIt+XNmcDxWQcJ5ipEAWL+ijceWwTA6z94jDFZ2RySuZMa/jzuJZ6tncxP37+GnEoH6257OqLvWbBrItNy34tHE+KKK4qhsuUUIiCeNfZTiXR2YluDDgZ8ppPVEAClUIEg26aar5kOoZW+NjuTnDo7i76kbkEp9o9X84srTuTRAau7/J7Lilen5fUQzexK+l4FCcJjNLMp4Ot6x8NI5xssHdkeLMDu1XGsrcS+eiNGYyOTxm5k0tiNHfa1YVZxPsO1A/t+HyoY5O1XJnWpMg4wI29vj8vhpTrWlWthYdEOa/jQBdmShSa9L4U43bg0x8P2J9/j+eencuC4AIOX2tgUkrGjtHU/XRkE0NGVwVp/P7Td+wkChWsDfNhcHJLaOzKZ3gvM7NbFAJ8KElDRHSYDa81+Mri5bxU33vgWi897nPqrPAx8PYuBr2cdqifhUwHu3nUSbzYOYZVf5xdP/Zhg9Q4Aspet5Tfzrkmm+SmD1VPoArctizHdmGms173k2swpqHQLSqUrmtiY2XcTTnHz2oS53PC3OwGYU3cucwZ9wtagnzfeOpXPv5xI9XkKxvloXjgWf4OTouV2ila3FqSJt2ZEKmM5hTixU4dX9p0AwN2Fa6zlugmi5TiX2l14i80e3uKK4/lBwQp2BAeQtw2UDe48cylX53+N22ZmKj588iSWP3gK5yz5GQBfTnk85rkg6UKvcwqJmE5yioNhdsVF+eWh5+mXKZiOtF3pGFA62XvMYZx3i5NHhl7I9vq+5Dcqxt69hlv7bQXMm94pDh4sXsXEa44ld7kpKnPyxzfzhwmvcqF7T1pmevaEXhdTSFRX3m3LYnyWnfFdCJ1mAl2tH4h2kVCkHC6saqf13K4LCPnf7Cf/m/0M+MzPgb8cxbB7mhEdfj/wnx0+yyEaN45cTm61QW61wfCrynnosRl81JwXF9sTTTT1I3qdU0gkqa6EHSu6crTx0K/QldFhSNb2WB9j91M1rT9V0/rjrqil379qUNt2sOtUwlbE2ubrjxggoThxycf1PLH93JjbngyicQqZ/zMWByJdPNSyGq43OIZE09UxLdRymH3NfACemnMSRqOpMeEY3Njp/g1GE3W+PIKu1sQkY/1Gdiw5BUbFyOgkEs0QyLpau0GkB7i39BRSlanuZqa6m5n/9VJOWm3g/f4paOV5HRYH1etepm+cxqfLjsdVr+Oq17EPKEEcdoa9vL3XlW6zrliLqKkNpraOxOEMtOfycMlX3PrIy2TthxNfu4MvfAGerB9G2Ys/5aw/3UXwN8WMmFeLq86Hq85H85ghaEWF6DvrupXmns5Yw4cEkq5Lkw8nWWXCesoU9y5+O0hx3Owa7vjkVhyNBiOq9uEZ2YfH//40x2W1n4LcHPAw80e384stBe3qOWY6XV6hIjJfROpEZF2bbQtFpDz0t0VEykPbjxaRpjavPRtP4y0sLGJPt7QklVI/bHksInOAtqVjNiulxsfKwExBVwYGimTkNmZKD6WnVAZh4GdBglu3k7d1OwAyeBC6o2+HXgLAcEcuO09xYuwu5Hz9In53zGtMdGZ+ElqPtCRFRIArgAUxtivj0MSWtHRnyyGYjvE/t15C9o72sw+N4waz/8rwMZLrZizF9c88tKv8zPjsx71Ci7KnV8sZwC6lVNtF66UislpEPhSRM8K90ZKNs0gkmth4pnQR316X3267t9jOzcd9FPZ90/PX0FwINDVT+mR0FaDTlZ46hem07yXUAkOVUicCdwIviUh+Z29USs1VSk1QSk0o6m8tGEomveHXD8wA6aVnfIH96KEEzzkJbXQZhh2OdXaqbAjAl75isusU+oED2FZUcOmTv+yQSdlCvDI3E023nYKI2IHvAwtbtoUk6PeGHq8CNgNlPTUyXfAa/rSb094W9LDW37Us3ZHwqUDatLs46yCOF3088PxzHPXCdvz5gq5s+FSA6qCHZU0ay5o0AsrUkLh92dUUrzKHFyoYZPDjKzn197d3cAA+FWDSiusp96X/9KUo1bW2q4gcDbyplDq+zbYLgHuUUme12VYE7FNK6SJyDPAxcIJSqtOYRAsTxrnUF0uPOtIuaUGLelM6ra5b4nUyzF5PmcOVlrGHaEvTtw26BpRO2Vs3IU0aT059gY2+Afxl4RRzv9EeAk0Ohr6q0ThAw3bpHvZ9059hbwUI5Gs4b63l72UvH/rc0z6exYg/Btl4dS7fXP7nlFwurw3ctEopNaGr/bqlJamUmgdcSccA45nAAyISBHTgpq4cQibRot6UTgyz1zPMHh8djEQQ7QrGtu10iIaWE6Tsns3cXjid44fUULjO7PG4H6yg8uGJ/OHxpxhmb6JAc9I8LsjCKcPx6C4K7B5mVn6fyjeGA3DM517UynUck30ir1xYzIy8vbFrZILprpYkSqnrOtm2CFjUc7PSh54KkyabEQ5nSv6qJQrNrqP8AfI+zWbN2KEc7Q2tVxlxNG9Nn02ZI4eW0vBOcTCzT2v84bLcf/Dw1ZMAWOo/jZLl4NjZwH2r/p0rz5qfto7WymjsAaa+Q2QnPhm5ApE4rN7sEHwqgK7bMA4eZMAn9eiufmg+M1aw7+Qi8uTIQ2uX2Fmy9TgA+n9txhL0jZWUzeoH6470ztTGcgo9IJobqlb3RqxCFCtSqQcTix5VSzDUQMXEme3TfZxWWsmOc0/CtamO4i+z2T3WXOb9wxuWUdzFUNCGDffCPgC4avbhuXQiSoS89XvTtpcAllNIGIl2CKmGPYa5nLHq3RRrbh4fsoQJF41m1PNu9g93ctZ/rADg3sINdDU5Z0NwXV8LwNebSzhxdBXnFa7nmW/PjIl9ySJ93ZmFhUVciGhKMt5kypRkphCPOpapvP6i3Ofj8v+9gzmXvcjFOd3LWGwwmviwqT/nZO8LaYWkXlsjnZJMPcstkobHaGaP3ogNiXkykk8FY/p5saTMIRhZCq/h7HrnMPSxZbPcU4aN9C+sk97WW8SUXJuLQi0nLou3Urki8kp/FkUr4Jm7Lqcq4Dmk/RAt7z85ifvqTomxdYnHCjSmAKnctc50vvI3c93i2zl25R70DZu4OvsuAN794xNRObJtQQ8F8z/jq3Un4Hvti5Sa+YkW60pMASyHkBzq9EauXXMduVUanmMLsA8eRP6GBvI3NLAuEJ2y9P95zHwF29dbOL/i8niYmzCsq9Gi1+JAGFVYx8ERQbZPVTRMOgp0Bbpitx6d3sMf35kKgHHwIK6pNXwb6LxqdDpgOYVeQrqsYkwk/TQ3Y/N2YG/QICjsHm+jeXAuzYNzmVcTthQIEFKvVjq31ZzMbTUnUzZ/f+uLysDdRTZkKmPFFHoBlkPonIDSee6zMxm18ADeYTnsHiu03Ms1c4dz1U+DbDvYD5lbxB2PLOB8dx0vHhjJS/dNwd6sUBo495q1FWxfrcbmdqN0HeXz4VXRDT9SCcsppAD1upd+mvvQzVurNzE0hhmQXuUnV7o/3ZapOERj9uSF/NfyGbh3BcAGVZebN/Ok0RvYWF+E428F5NV4+dwznMH2emZ/dj6jV9Ri9Mkh+8k9zBn2GgAlWhYO0TAw0JXCbctJZtN6hOUUkoyujEOS9TbMC3Kg1rmsWXcJKAND4lc0NlErRVtEXA6XfdOVgUf5wsrBHYlpuQc466HZrA/k8IlnFKflmJUFJ7l8HBzqZ1LFXfRd7WWzp5D37aMZ+K4dI9/NNzfnUjViAS0rKFvRIH07CYDlFJKOJrZDN2vLLESsb948W9YhhxMPEjX9Fu6m18RGH+m+Iy3UcjhTg0nO9YeOkyYOnJqD2y5azOIHhrD+n+No/K6TxgE2XHtyeOyczK1VbAUaLSws2mE5hV5AvFJvtwU9GVP0VVcGNqSD/udI505shQX02aTYWF1Mc6HCW+JgkL0+idbGF8sp9ALiVUhlqD03IxKvWjJKO2vLBW4fFyz+CrtP4drowj8wwNyH/pTRojBWTMGi19OVY7u131a+88jT6NgYafekrZZmpFhOwcIiAk532UI9isx2CGANHzol1uNkXRlsDngO5SF4DX+35Nx1ZbC82eDRfcN5+WA/fCoQVpgk3Ptb/iIhoPQuE58i+axUTJ4KKJ3qoIc9emNE9nkNP2v9AbZ147ylGylRZEVEdgONwJ5k2xIHCsnMdkHmti1T2zVMKVXU1U4p4RQARGRlJFVh0o1MbRdkbtsytV2RYg0fLCws2mE5BQsLi3akklOYm2wD4kSmtgsyt22Z2q6ISJmYgoWFRWqQSj0FCwuLFCDpTkFELhCRDSKySUTuTrY9PUVEtojIWhEpF5GVoW0FIvKuiGwM/e+XbDu7QkTmi0idiKxrs63TdojJE6Fz+JWIfCd5lndNmLbdLyI7QuetXESmtHntnlDbNojI+cmxOnEk1SmIiAb8GbgQGA1MF5HRybQpRkxWSo1vM611N7BMKTUSWBZ6nuq8AFxw2LZw7bgQGBn6mwk8kyAbu8sLdGwbwGOh8zZeKbUYIHQ9XgmMCb3n6dB1m7Eku6cwEdiklKpUSvmBl4FLkmxTPLgEeDH0+EXg0iTaEhFKqY+AfYdtDteOS4C/KpPPgb4iMjAxlkZPmLaF4xLgZaWUTylVBWzCvG4zlmQ7hcHA9jbPq0Pb0hkFvCMiq0RkZmhbiVKqFiD0vzhp1vWMcO3IlPN4S2j4M7/NEC9T2hYxyXYKnZUDSvfpkNOVUt/B7FLPEpH0liCOjEw4j88Aw4HxQC0wJ7Q9E9oWFcl2CtVAW2XZIUBNkmyJCUqpmtD/OuA1zK7mrpbudOh/XfIs7BHh2pH251EptUsppSulDOA5WocIad+2aEm2U1gBjBSRUhHJwgzovJFkm7qNiOSISF7LY+A8YB1mm64N7XYt8HpyLOwx4drxBnBNaBbiVKChZZiRLhwWA7kM87yB2bYrRcQpIqWYwdQvEm1fIklqPQWlVFBEbgGWYtYrna+UqkimTT2kBHhNRMA8ti8ppZaIyArgFRG5AdgGpLyumIgsAM4GCkWkGrgPeITO27EYmIIZhPMCP0q4wVEQpm1ni8h4zKHBFuAnAEqpChF5BVgPBIFZSqXgWvAYYmU0WlhYtCPZwwcLC4sUw3IKFhYW7bCcgoWFRTssp2BhYdEOyylYWFi0w3IKFhYW7bCcgoWFRTssp2BhYdGO/wdFFATz/Em+cwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "count = 0\n",
    "for e, im in zip(decoded_imgs[:10], train_list[:10]) :\n",
    "    \n",
    "    plt.imshow(e.reshape(200,200))\n",
    "    plt.show()\n",
    "\n",
    "    plt.imshow(im.reshape(200,200))\n",
    "    plt.savefig(str(count)+'.png')\n",
    "    count +=1\n",
    "    plt.show()\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "    <div class=\"bk-root\">\n",
       "        <a href=\"https://bokeh.pydata.org\" target=\"_blank\" class=\"bk-logo bk-logo-small bk-logo-notebook\"></a>\n",
       "        <span id=\"b1e54ee4-a589-4322-85e6-cbf694b3b102\">Loading BokehJS ...</span>\n",
       "    </div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/javascript": [
       "\n",
       "(function(root) {\n",
       "  function now() {\n",
       "    return new Date();\n",
       "  }\n",
       "\n",
       "  var force = true;\n",
       "\n",
       "  if (typeof (root._bokeh_onload_callbacks) === \"undefined\" || force === true) {\n",
       "    root._bokeh_onload_callbacks = [];\n",
       "    root._bokeh_is_loading = undefined;\n",
       "  }\n",
       "\n",
       "\n",
       "  \n",
       "  if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n",
       "    root._bokeh_timeout = Date.now() + 5000;\n",
       "    root._bokeh_failed_load = false;\n",
       "  }\n",
       "\n",
       "  var NB_LOAD_WARNING = {'data': {'text/html':\n",
       "     \"<div style='background-color: #fdd'>\\n\"+\n",
       "     \"<p>\\n\"+\n",
       "     \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n",
       "     \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n",
       "     \"</p>\\n\"+\n",
       "     \"<ul>\\n\"+\n",
       "     \"<li>re-rerun `output_notebook()` to attempt to load from CDN again, or</li>\\n\"+\n",
       "     \"<li>use INLINE resources instead, as so:</li>\\n\"+\n",
       "     \"</ul>\\n\"+\n",
       "     \"<code>\\n\"+\n",
       "     \"from bokeh.resources import INLINE\\n\"+\n",
       "     \"output_notebook(resources=INLINE)\\n\"+\n",
       "     \"</code>\\n\"+\n",
       "     \"</div>\"}};\n",
       "\n",
       "  function display_loaded() {\n",
       "    if (root.Bokeh !== undefined) {\n",
       "      var el = document.getElementById(\"b1e54ee4-a589-4322-85e6-cbf694b3b102\");\n",
       "      if (el != null) {\n",
       "        el.textContent = \"BokehJS \" + Bokeh.version + \" successfully loaded.\";\n",
       "      }\n",
       "    } else if (Date.now() < root._bokeh_timeout) {\n",
       "      setTimeout(display_loaded, 100)\n",
       "    }\n",
       "  }\n",
       "\n",
       "\n",
       "  function run_callbacks() {\n",
       "    try {\n",
       "      root._bokeh_onload_callbacks.forEach(function(callback) { callback() });\n",
       "    }\n",
       "    finally {\n",
       "      delete root._bokeh_onload_callbacks\n",
       "    }\n",
       "    console.info(\"Bokeh: all callbacks have finished\");\n",
       "  }\n",
       "\n",
       "  function load_libs(js_urls, callback) {\n",
       "    root._bokeh_onload_callbacks.push(callback);\n",
       "    if (root._bokeh_is_loading > 0) {\n",
       "      console.log(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n",
       "      return null;\n",
       "    }\n",
       "    if (js_urls == null || js_urls.length === 0) {\n",
       "      run_callbacks();\n",
       "      return null;\n",
       "    }\n",
       "    console.log(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n",
       "    root._bokeh_is_loading = js_urls.length;\n",
       "    for (var i = 0; i < js_urls.length; i++) {\n",
       "      var url = js_urls[i];\n",
       "      var s = document.createElement('script');\n",
       "      s.src = url;\n",
       "      s.async = false;\n",
       "      s.onreadystatechange = s.onload = function() {\n",
       "        root._bokeh_is_loading--;\n",
       "        if (root._bokeh_is_loading === 0) {\n",
       "          console.log(\"Bokeh: all BokehJS libraries loaded\");\n",
       "          run_callbacks()\n",
       "        }\n",
       "      };\n",
       "      s.onerror = function() {\n",
       "        console.warn(\"failed to load library \" + url);\n",
       "      };\n",
       "      console.log(\"Bokeh: injecting script tag for BokehJS library: \", url);\n",
       "      document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
       "    }\n",
       "  };var element = document.getElementById(\"b1e54ee4-a589-4322-85e6-cbf694b3b102\");\n",
       "  if (element == null) {\n",
       "    console.log(\"Bokeh: ERROR: autoload.js configured with elementid 'b1e54ee4-a589-4322-85e6-cbf694b3b102' but no matching script tag was found. \")\n",
       "    return false;\n",
       "  }\n",
       "\n",
       "  var js_urls = [\"https://cdn.pydata.org/bokeh/release/bokeh-0.12.7.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-widgets-0.12.7.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-tables-0.12.7.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-gl-0.12.7.min.js\"];\n",
       "\n",
       "  var inline_js = [\n",
       "    function(Bokeh) {\n",
       "      Bokeh.set_log_level(\"info\");\n",
       "    },\n",
       "    \n",
       "    function(Bokeh) {\n",
       "      \n",
       "    },\n",
       "    \n",
       "    function(Bokeh) {\n",
       "      \n",
       "      document.getElementById(\"b1e54ee4-a589-4322-85e6-cbf694b3b102\").textContent = \"BokehJS is loading...\";\n",
       "\n",
       "    },\n",
       "    function(Bokeh) {\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-0.12.7.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-0.12.7.min.css\");\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-widgets-0.12.7.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-widgets-0.12.7.min.css\");\n",
       "      console.log(\"Bokeh: injecting CSS: https://cdn.pydata.org/bokeh/release/bokeh-tables-0.12.7.min.css\");\n",
       "      Bokeh.embed.inject_css(\"https://cdn.pydata.org/bokeh/release/bokeh-tables-0.12.7.min.css\");\n",
       "    }\n",
       "  ];\n",
       "\n",
       "  function run_inline_js() {\n",
       "    \n",
       "    if ((root.Bokeh !== undefined) || (force === true)) {\n",
       "      for (var i = 0; i < inline_js.length; i++) {\n",
       "        inline_js[i].call(root, root.Bokeh);\n",
       "      }if (force === true) {\n",
       "        display_loaded();\n",
       "      }} else if (Date.now() < root._bokeh_timeout) {\n",
       "      setTimeout(run_inline_js, 100);\n",
       "    } else if (!root._bokeh_failed_load) {\n",
       "      console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n",
       "      root._bokeh_failed_load = true;\n",
       "    } else if (force !== true) {\n",
       "      var cell = $(document.getElementById(\"b1e54ee4-a589-4322-85e6-cbf694b3b102\")).parents('.cell').data().cell;\n",
       "      cell.output_area.append_execute_result(NB_LOAD_WARNING)\n",
       "    }\n",
       "\n",
       "  }\n",
       "\n",
       "  if (root._bokeh_is_loading === 0) {\n",
       "    console.log(\"Bokeh: BokehJS loaded, going straight to plotting\");\n",
       "    run_inline_js();\n",
       "  } else {\n",
       "    load_libs(js_urls, function() {\n",
       "      console.log(\"Bokeh: BokehJS plotting callback run at\", now());\n",
       "      run_inline_js();\n",
       "    });\n",
       "  }\n",
       "}(window));"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/lxf453/miniconda2/lib/python2.7/site-packages/bokeh/util/deprecation.py:34: BokehDeprecationWarning: ResizeTool is removed in Bokeh 0.12.7, adding it is a no-op. In the future, accessing ResizeTool will be an error\n",
      "  warn(message)\n",
      "/Users/lxf453/miniconda2/lib/python2.7/site-packages/bokeh/util/deprecation.py:34: BokehDeprecationWarning: ResizeTool is removed in Bokeh 0.12.7, adding it is a no-op. In the future, accessing ResizeTool will be an error\n",
      "  warn(message)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "\n",
       "    <div class=\"bk-root\">\n",
       "        <div class=\"bk-plotdiv\" id=\"a6d1ec7f-976a-4f68-b591-f3022c75cc91\"></div>\n",
       "    </div>\n",
       "<script type=\"text/javascript\">\n",
       "  \n",
       "  (function(root) {\n",
       "    function now() {\n",
       "      return new Date();\n",
       "    }\n",
       "\n",
       "    var force = false;\n",
       "\n",
       "    if (typeof (root._bokeh_onload_callbacks) === \"undefined\" || force === true) {\n",
       "      root._bokeh_onload_callbacks = [];\n",
       "      root._bokeh_is_loading = undefined;\n",
       "    }\n",
       "\n",
       "\n",
       "    \n",
       "    if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n",
       "      root._bokeh_timeout = Date.now() + 0;\n",
       "      root._bokeh_failed_load = false;\n",
       "    }\n",
       "\n",
       "    var NB_LOAD_WARNING = {'data': {'text/html':\n",
       "       \"<div style='background-color: #fdd'>\\n\"+\n",
       "       \"<p>\\n\"+\n",
       "       \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n",
       "       \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n",
       "       \"</p>\\n\"+\n",
       "       \"<ul>\\n\"+\n",
       "       \"<li>re-rerun `output_notebook()` to attempt to load from CDN again, or</li>\\n\"+\n",
       "       \"<li>use INLINE resources instead, as so:</li>\\n\"+\n",
       "       \"</ul>\\n\"+\n",
       "       \"<code>\\n\"+\n",
       "       \"from bokeh.resources import INLINE\\n\"+\n",
       "       \"output_notebook(resources=INLINE)\\n\"+\n",
       "       \"</code>\\n\"+\n",
       "       \"</div>\"}};\n",
       "\n",
       "    function display_loaded() {\n",
       "      if (root.Bokeh !== undefined) {\n",
       "        var el = document.getElementById(\"a6d1ec7f-976a-4f68-b591-f3022c75cc91\");\n",
       "        if (el != null) {\n",
       "          el.textContent = \"BokehJS \" + Bokeh.version + \" successfully loaded.\";\n",
       "        }\n",
       "      } else if (Date.now() < root._bokeh_timeout) {\n",
       "        setTimeout(display_loaded, 100)\n",
       "      }\n",
       "    }\n",
       "\n",
       "\n",
       "    function run_callbacks() {\n",
       "      try {\n",
       "        root._bokeh_onload_callbacks.forEach(function(callback) { callback() });\n",
       "      }\n",
       "      finally {\n",
       "        delete root._bokeh_onload_callbacks\n",
       "      }\n",
       "      console.info(\"Bokeh: all callbacks have finished\");\n",
       "    }\n",
       "\n",
       "    function load_libs(js_urls, callback) {\n",
       "      root._bokeh_onload_callbacks.push(callback);\n",
       "      if (root._bokeh_is_loading > 0) {\n",
       "        console.log(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n",
       "        return null;\n",
       "      }\n",
       "      if (js_urls == null || js_urls.length === 0) {\n",
       "        run_callbacks();\n",
       "        return null;\n",
       "      }\n",
       "      console.log(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n",
       "      root._bokeh_is_loading = js_urls.length;\n",
       "      for (var i = 0; i < js_urls.length; i++) {\n",
       "        var url = js_urls[i];\n",
       "        var s = document.createElement('script');\n",
       "        s.src = url;\n",
       "        s.async = false;\n",
       "        s.onreadystatechange = s.onload = function() {\n",
       "          root._bokeh_is_loading--;\n",
       "          if (root._bokeh_is_loading === 0) {\n",
       "            console.log(\"Bokeh: all BokehJS libraries loaded\");\n",
       "            run_callbacks()\n",
       "          }\n",
       "        };\n",
       "        s.onerror = function() {\n",
       "          console.warn(\"failed to load library \" + url);\n",
       "        };\n",
       "        console.log(\"Bokeh: injecting script tag for BokehJS library: \", url);\n",
       "        document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
       "      }\n",
       "    };var element = document.getElementById(\"a6d1ec7f-976a-4f68-b591-f3022c75cc91\");\n",
       "    if (element == null) {\n",
       "      console.log(\"Bokeh: ERROR: autoload.js configured with elementid 'a6d1ec7f-976a-4f68-b591-f3022c75cc91' but no matching script tag was found. \")\n",
       "      return false;\n",
       "    }\n",
       "\n",
       "    var js_urls = [];\n",
       "\n",
       "    var inline_js = [\n",
       "      function(Bokeh) {\n",
       "        (function() {\n",
       "          var fn = function() {\n",
       "            var docs_json = {\"e6eabf9a-1352-4e21-a3e1-1ca93778e61e\":{\"roots\":{\"references\":[{\"attributes\":{\"bottom_units\":\"screen\",\"fill_alpha\":{\"value\":0.5},\"fill_color\":{\"value\":\"lightgrey\"},\"left_units\":\"screen\",\"level\":\"overlay\",\"line_alpha\":{\"value\":1.0},\"line_color\":{\"value\":\"black\"},\"line_dash\":[4,4],\"line_width\":{\"value\":2},\"plot\":null,\"render_mode\":\"css\",\"right_units\":\"screen\",\"top_units\":\"screen\"},\"id\":\"fd9a85fc-a251-48a7-b852-f71f809f69b3\",\"type\":\"BoxAnnotation\"},{\"attributes\":{\"dimension\":1,\"plot\":{\"id\":\"392d6bf8-37b9-4be4-a99f-64420b0a2546\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"f1857cc1-da5f-4b93-a8f9-969d4d961e6c\",\"type\":\"BasicTicker\"}},\"id\":\"6d4e3c93-155b-422e-8c29-16e5ef8a847c\",\"type\":\"Grid\"},{\"attributes\":{\"axis_label\":\"y\",\"formatter\":{\"id\":\"0d4e3e68-e310-480f-a528-8fc485d97153\",\"type\":\"BasicTickFormatter\"},\"plot\":{\"id\":\"392d6bf8-37b9-4be4-a99f-64420b0a2546\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"f1857cc1-da5f-4b93-a8f9-969d4d961e6c\",\"type\":\"BasicTicker\"}},\"id\":\"a8894c2f-5765-48e1-b8f0-515c8a897445\",\"type\":\"LinearAxis\"},{\"attributes\":{\"callback\":null,\"column_names\":[\"y\",\"x\",\"img\",\"index\"],\"data\":{\"img\":[\"http://0.0.0.0:8889/0.png\",\"http://0.0.0.0:8889/1.png\",\"http://0.0.0.0:8889/2.png\",\"http://0.0.0.0:8889/3.png\",\"http://0.0.0.0:8889/4.png\",\"http://0.0.0.0:8889/5.png\",\"http://0.0.0.0:8889/6.png\",\"http://0.0.0.0:8889/7.png\",\"http://0.0.0.0:8889/8.png\",\"http://0.0.0.0:8889/9.png\",\"http://0.0.0.0:8889/10.png\",\"http://0.0.0.0:8889/11.png\",\"http://0.0.0.0:8889/12.png\",\"http://0.0.0.0:8889/13.png\",\"http://0.0.0.0:8889/14.png\",\"http://0.0.0.0:8889/15.png\",\"http://0.0.0.0:8889/16.png\",\"http://0.0.0.0:8889/17.png\",\"http://0.0.0.0:8889/18.png\",\"http://0.0.0.0:8889/19.png\",\"http://0.0.0.0:8889/20.png\",\"http://0.0.0.0:8889/21.png\",\"http://0.0.0.0:8889/22.png\",\"http://0.0.0.0:8889/23.png\",\"http://0.0.0.0:8889/24.png\",\"http://0.0.0.0:8889/25.png\",\"http://0.0.0.0:8889/26.png\",\"http://0.0.0.0:8889/27.png\",\"http://0.0.0.0:8889/28.png\",\"http://0.0.0.0:8889/29.png\",\"http://0.0.0.0:8889/30.png\",\"http://0.0.0.0:8889/31.png\",\"http://0.0.0.0:8889/32.png\",\"http://0.0.0.0:8889/33.png\",\"http://0.0.0.0:8889/34.png\",\"http://0.0.0.0:8889/35.png\",\"http://0.0.0.0:8889/36.png\",\"http://0.0.0.0:8889/37.png\",\"http://0.0.0.0:8889/38.png\",\"http://0.0.0.0:8889/39.png\",\"http://0.0.0.0:8889/40.png\",\"http://0.0.0.0:8889/41.png\",\"http://0.0.0.0:8889/42.png\",\"http://0.0.0.0:8889/43.png\",\"http://0.0.0.0:8889/44.png\",\"http://0.0.0.0:8889/45.png\",\"http://0.0.0.0:8889/46.png\",\"http://0.0.0.0:8889/47.png\",\"http://0.0.0.0:8889/48.png\",\"http://0.0.0.0:8889/49.png\",\"http://0.0.0.0:8889/50.png\",\"http://0.0.0.0:8889/51.png\",\"http://0.0.0.0:8889/52.png\",\"http://0.0.0.0:8889/53.png\",\"http://0.0.0.0:8889/54.png\",\"http://0.0.0.0:8889/55.png\",\"http://0.0.0.0:8889/56.png\",\"http://0.0.0.0:8889/57.png\",\"http://0.0.0.0:8889/58.png\",\"http://0.0.0.0:8889/59.png\",\"http://0.0.0.0:8889/60.png\",\"http://0.0.0.0:8889/61.png\",\"http://0.0.0.0:8889/62.png\",\"http://0.0.0.0:8889/63.png\",\"http://0.0.0.0:8889/64.png\",\"http://0.0.0.0:8889/65.png\",\"http://0.0.0.0:8889/66.png\",\"http://0.0.0.0:8889/67.png\",\"http://0.0.0.0:8889/68.png\",\"http://0.0.0.0:8889/69.png\",\"http://0.0.0.0:8889/70.png\",\"http://0.0.0.0:8889/71.png\",\"http://0.0.0.0:8889/72.png\",\"http://0.0.0.0:8889/73.png\",\"http://0.0.0.0:8889/74.png\",\"http://0.0.0.0:8889/75.png\",\"http://0.0.0.0:8889/76.png\",\"http://0.0.0.0:8889/77.png\",\"http://0.0.0.0:8889/78.png\",\"http://0.0.0.0:8889/79.png\",\"http://0.0.0.0:8889/80.png\",\"http://0.0.0.0:8889/81.png\",\"http://0.0.0.0:8889/82.png\",\"http://0.0.0.0:8889/83.png\",\"http://0.0.0.0:8889/84.png\",\"http://0.0.0.0:8889/85.png\",\"http://0.0.0.0:8889/86.png\",\"http://0.0.0.0:8889/87.png\",\"http://0.0.0.0:8889/88.png\",\"http://0.0.0.0:8889/89.png\",\"http://0.0.0.0:8889/90.png\",\"http://0.0.0.0:8889/91.png\",\"http://0.0.0.0:8889/92.png\",\"http://0.0.0.0:8889/93.png\",\"http://0.0.0.0:8889/94.png\",\"http://0.0.0.0:8889/95.png\",\"http://0.0.0.0:8889/96.png\",\"http://0.0.0.0:8889/97.png\",\"http://0.0.0.0:8889/98.png\",\"http://0.0.0.0:8889/99.png\",\"http://0.0.0.0:8889/100.png\",\"http://0.0.0.0:8889/101.png\",\"http://0.0.0.0:8889/102.png\",\"http://0.0.0.0:8889/103.png\",\"http://0.0.0.0:8889/104.png\",\"http://0.0.0.0:8889/105.png\",\"http://0.0.0.0:8889/106.png\",\"http://0.0.0.0:8889/107.png\",\"http://0.0.0.0:8889/108.png\",\"http://0.0.0.0:8889/109.png\",\"http://0.0.0.0:8889/110.png\",\"http://0.0.0.0:8889/111.png\",\"http://0.0.0.0:8889/112.png\",\"http://0.0.0.0:8889/113.png\",\"http://0.0.0.0:8889/114.png\",\"http://0.0.0.0:8889/115.png\",\"http://0.0.0.0:8889/116.png\",\"http://0.0.0.0:8889/117.png\",\"http://0.0.0.0:8889/118.png\",\"http://0.0.0.0:8889/119.png\",\"http://0.0.0.0:8889/120.png\",\"http://0.0.0.0:8889/121.png\",\"http://0.0.0.0:8889/122.png\",\"http://0.0.0.0:8889/123.png\",\"http://0.0.0.0:8889/124.png\",\"http://0.0.0.0:8889/125.png\",\"http://0.0.0.0:8889/126.png\",\"http://0.0.0.0:8889/127.png\",\"http://0.0.0.0:8889/128.png\",\"http://0.0.0.0:8889/129.png\",\"http://0.0.0.0:8889/130.png\",\"http://0.0.0.0:8889/131.png\",\"http://0.0.0.0:8889/132.png\",\"http://0.0.0.0:8889/133.png\",\"http://0.0.0.0:8889/134.png\",\"http://0.0.0.0:8889/135.png\",\"http://0.0.0.0:8889/136.png\",\"http://0.0.0.0:8889/137.png\",\"http://0.0.0.0:8889/138.png\",\"http://0.0.0.0:8889/139.png\",\"http://0.0.0.0:8889/140.png\",\"http://0.0.0.0:8889/141.png\",\"http://0.0.0.0:8889/142.png\",\"http://0.0.0.0:8889/143.png\",\"http://0.0.0.0:8889/144.png\",\"http://0.0.0.0:8889/145.png\",\"http://0.0.0.0:8889/146.png\",\"http://0.0.0.0:8889/147.png\",\"http://0.0.0.0:8889/148.png\",\"http://0.0.0.0:8889/149.png\",\"http://0.0.0.0:8889/150.png\",\"http://0.0.0.0:8889/151.png\",\"http://0.0.0.0:8889/152.png\",\"http://0.0.0.0:8889/153.png\",\"http://0.0.0.0:8889/154.png\",\"http://0.0.0.0:8889/155.png\",\"http://0.0.0.0:8889/156.png\",\"http://0.0.0.0:8889/157.png\",\"http://0.0.0.0:8889/158.png\",\"http://0.0.0.0:8889/159.png\",\"http://0.0.0.0:8889/160.png\",\"http://0.0.0.0:8889/161.png\",\"http://0.0.0.0:8889/162.png\",\"http://0.0.0.0:8889/163.png\",\"http://0.0.0.0:8889/164.png\",\"http://0.0.0.0:8889/165.png\",\"http://0.0.0.0:8889/166.png\",\"http://0.0.0.0:8889/167.png\",\"http://0.0.0.0:8889/168.png\",\"http://0.0.0.0:8889/169.png\",\"http://0.0.0.0:8889/170.png\",\"http://0.0.0.0:8889/171.png\",\"http://0.0.0.0:8889/172.png\",\"http://0.0.0.0:8889/173.png\",\"http://0.0.0.0:8889/174.png\",\"http://0.0.0.0:8889/175.png\",\"http://0.0.0.0:8889/176.png\",\"http://0.0.0.0:8889/177.png\",\"http://0.0.0.0:8889/178.png\",\"http://0.0.0.0:8889/179.png\",\"http://0.0.0.0:8889/180.png\",\"http://0.0.0.0:8889/181.png\",\"http://0.0.0.0:8889/182.png\",\"http://0.0.0.0:8889/183.png\",\"http://0.0.0.0:8889/184.png\",\"http://0.0.0.0:8889/185.png\",\"http://0.0.0.0:8889/186.png\",\"http://0.0.0.0:8889/187.png\",\"http://0.0.0.0:8889/188.png\",\"http://0.0.0.0:8889/189.png\",\"http://0.0.0.0:8889/190.png\",\"http://0.0.0.0:8889/191.png\",\"http://0.0.0.0:8889/192.png\",\"http://0.0.0.0:8889/193.png\",\"http://0.0.0.0:8889/194.png\",\"http://0.0.0.0:8889/195.png\",\"http://0.0.0.0:8889/196.png\",\"http://0.0.0.0:8889/197.png\",\"http://0.0.0.0:8889/198.png\",\"http://0.0.0.0:8889/199.png\",\"http://0.0.0.0:8889/200.png\",\"http://0.0.0.0:8889/201.png\"],\"index\":[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199,200,201],\"x\":{\"__ndarray__\":\"AAAAwOJEJkAAAAAgEjkBwAAAAEAAYgRAAAAAIHpg7z8AAABABXEDQAAAAIBGjR1AAAAAgCcy5b8AAACA12gRwAAAAKD+RPW/AAAAoKV3DEAAAABABXEDQAAAAEA5JxTAAAAAgLtPCUAAAADg0EfovwAAAIDXaBHAAAAAIBscBkAAAACggdLrPwAAAGCUuTFAAAAAoC9TI0AAAAAg0GMQwAAAAIDKCQtAAAAAAJ61GEAAAADg5dUVQAAAAMC2ISJAAAAAoBK4BEAAAACgh8/vvwAAAKD/zyfAAAAAID6AMcAAAABgENwJQAAAACCsdw/AAAAAwB7NHcAAAAAgHOIlQAAAAOAlIQvAAAAAAJLr9z8AAACgSIUcQAAAAGAwngDAAAAAQPy9EsAAAACAwOwTQAAAAMDdKjDAAAAAYE0uMcAAAABAXID1vwAAAMD0PSpAAAAAwN0qMMAAAACAnVUQwAAAAKCHz++/AAAAwNsG+j8AAABAbaMxwAAAAACG6RVAAAAAgNdoEcAAAABgjK0UQAAAAIDlVgZAAAAAoOu0FEAAAABAEYf/vwAAAEDz9eQ/AAAAQOynE0AAAACg59IMQAAAAGBSJhdAAAAAQPP15D8AAABgHhYNwAAAAKBIhRxAAAAAoMBDC8AAAAAAKxEMQAAAAABnOey/AAAAQKKByz8AAAAANjckwAAAAICgD7y/AAAAgDeC6r8AAACgm+wuwAAAAEDP7wjAAAAAgJ1VEMAAAADAD5YwwAAAAMAPljDAAAAAgOUCMMAAAADA4kQmQAAAAKARgPo/AAAA4GZXwr8AAABgEJ4JQAAAAEC88yxAAAAAQLzzLEAAAACgfS8BQAAAAKCyQvM/AAAAQG2jMcAAAADgK5n0PwAAAOArmfQ/AAAAoOu0FEAAAABAinwfQAAAAEDHLP8/AAAAgIX5AkAAAADAPhHmPwAAAABQfBlAAAAAgNgOAkAAAACAsRobQAAAACBYeAHAAAAAQGgRLUAAAADAD5YwwAAAAKD7GBxAAAAA4Lj1LsAAAABgSxMTwAAAAEDMQwxAAAAAIAwaIUAAAABA97oTQAAAAMAPljDAAAAAgJM8+z8AAADARB8cQAAAAEBbNCFAAAAAQDBDEUAAAAAAacoRQAAAAIAKuSZAAAAAoCcu9r8AAABgCn0lwAAAAGBSJhdAAAAAYEs8378AAACgwwgPQAAAAKCYHtu/AAAAYNnREUAAAAAAnz0JwAAAAAAQ6tU/AAAAAH5TLMAAAACAt1MQQAAAAIBGjR1AAAAAAPRmHkAAAACg2p0cQAAAAIBJsghAAAAAQJMFEkAAAADg8YEAwAAAAGBVYfg/AAAAQEXRHUAAAABAnYwTQAAAACC0lCXAAAAAYPBeKsAAAADAINkoQAAAAMAg2ShAAAAA4EjAIMAAAACA4IgywAAAAOD3lgzAAAAA4CykDkAAAABgP9EPQAAAAODFaBlAAAAA4MVoGUAAAAAgt1fmPwAAAOBHxyFAAAAAIGCPCkAAAABAyUsbQAAAAMDiRCZAAAAA4NX8J8AAAADg1fwnwAAAACAgvBjAAAAAADY3JMAAAABgUiYXQAAAAEA/sfA/AAAAwB7NHcAAAACA12gRwAAAAMDiRCZAAAAAYKahJUAAAACAFBnfPwAAAGDs0QNAAAAAQIIXKkAAAABAC3gFwAAAAKANSdW/AAAAAABxGEAAAACg59IMQAAAAABSrh1AAAAAICQnB0AAAABgwR4eQAAAACBtJ+A/AAAAAJ89CcAAAADAh0glwAAAACBtJ+A/AAAAYPBeKsAAAADgxT4rQAAAAAC3dwVAAAAAoOfSDEAAAACAEZogwAAAAMAX/C1AAAAAwBf8LUAAAADAQJbaPwAAAKCWiiLAAAAAICQnB0AAAABA1OoBwAAAAIB9RSXAAAAAQATiD0AAAACgSIUcQAAAAICTVgBAAAAAQPP15D8AAACACkATQAAAACDTLPw/AAAAAGs6CkAAAABgUiYXQAAAACBgjwpAAAAAYNFg+j8AAADgSp0AwAAAAAD6JA5AAAAAYDCeAMAAAACA26/1vwAAAGApTwrAAAAAYKahJUAAAABgTS4xwAAAAKDcVgnAAAAAgOVWBkAAAAAg+Eb8PwAAAIDgiDLAAAAAINMs/D8=\",\"dtype\":\"float64\",\"shape\":[202]},\"y\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[202]}}},\"id\":\"f260c07d-23c6-41a3-815f-a6999652c4c4\",\"type\":\"ColumnDataSource\"},{\"attributes\":{},\"id\":\"b1468e98-fe5d-49da-915a-52adc680b7fa\",\"type\":\"ResetTool\"},{\"attributes\":{\"callback\":null},\"id\":\"689ec22d-10e4-46ee-850b-99c43c49fa4c\",\"type\":\"DataRange1d\"},{\"attributes\":{\"active_drag\":\"auto\",\"active_inspect\":\"auto\",\"active_scroll\":\"auto\",\"active_tap\":\"auto\",\"tools\":[{\"id\":\"128705c9-e05d-4121-820d-8754cc6283bd\",\"type\":\"BoxZoomTool\"},{\"id\":\"5f6f161e-4db4-4b63-80d5-6d9a6c1e4b4e\",\"type\":\"BoxSelectTool\"},{\"id\":\"b1468e98-fe5d-49da-915a-52adc680b7fa\",\"type\":\"ResetTool\"},{\"id\":\"3c306978-ed33-4d04-8010-214a8fef333a\",\"type\":\"SaveTool\"},{\"id\":\"cb422d22-d26a-4bfe-89d7-0317d3636293\",\"type\":\"HoverTool\"}]},\"id\":\"39c2ea40-4f3d-4c91-8c5a-28313e9712db\",\"type\":\"Toolbar\"},{\"attributes\":{},\"id\":\"3c306978-ed33-4d04-8010-214a8fef333a\",\"type\":\"SaveTool\"},{\"attributes\":{\"callback\":null,\"tooltips\":\"<img src=@img>\"},\"id\":\"cb422d22-d26a-4bfe-89d7-0317d3636293\",\"type\":\"HoverTool\"},{\"attributes\":{\"data_source\":{\"id\":\"f260c07d-23c6-41a3-815f-a6999652c4c4\",\"type\":\"ColumnDataSource\"},\"glyph\":{\"id\":\"72d6571c-7693-442e-9455-6cea51b5f02b\",\"type\":\"Circle\"},\"hover_glyph\":null,\"muted_glyph\":null,\"nonselection_glyph\":{\"id\":\"cbfa4dbe-127b-45a6-988c-80a6693c9e4a\",\"type\":\"Circle\"},\"selection_glyph\":null,\"view\":{\"id\":\"de6757f5-0ea6-4e71-8c2d-538b40ed7ce4\",\"type\":\"CDSView\"}},\"id\":\"a2859309-8777-4e2c-8cee-dd3670c08707\",\"type\":\"GlyphRenderer\"},{\"attributes\":{\"bottom_units\":\"screen\",\"fill_alpha\":{\"value\":0.5},\"fill_color\":{\"value\":\"lightgrey\"},\"left_units\":\"screen\",\"level\":\"overlay\",\"line_alpha\":{\"value\":1.0},\"line_color\":{\"value\":\"black\"},\"line_dash\":[4,4],\"line_width\":{\"value\":2},\"plot\":null,\"render_mode\":\"css\",\"right_units\":\"screen\",\"top_units\":\"screen\"},\"id\":\"52332999-3b1e-44c9-ba2c-597e8dbefaef\",\"type\":\"BoxAnnotation\"},{\"attributes\":{\"plot\":null,\"text\":\"test\"},\"id\":\"22dcfeb4-560a-4207-993d-a827704ba5ec\",\"type\":\"Title\"},{\"attributes\":{},\"id\":\"8860e1b0-7592-48ae-8cb4-147436810f6d\",\"type\":\"LinearScale\"},{\"attributes\":{\"source\":{\"id\":\"f260c07d-23c6-41a3-815f-a6999652c4c4\",\"type\":\"ColumnDataSource\"}},\"id\":\"de6757f5-0ea6-4e71-8c2d-538b40ed7ce4\",\"type\":\"CDSView\"},{\"attributes\":{\"axis_label\":\"x\",\"formatter\":{\"id\":\"b9758331-d27d-47e0-bf11-c6cb9c0f9426\",\"type\":\"BasicTickFormatter\"},\"plot\":{\"id\":\"392d6bf8-37b9-4be4-a99f-64420b0a2546\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"fc171877-33ec-48ac-b4de-ec8225789e71\",\"type\":\"BasicTicker\"}},\"id\":\"63713382-c415-4ffc-a2fb-c6fa67162c1a\",\"type\":\"LinearAxis\"},{\"attributes\":{\"callback\":null,\"overlay\":{\"id\":\"fd9a85fc-a251-48a7-b852-f71f809f69b3\",\"type\":\"BoxAnnotation\"},\"renderers\":[{\"id\":\"a2859309-8777-4e2c-8cee-dd3670c08707\",\"type\":\"GlyphRenderer\"}]},\"id\":\"5f6f161e-4db4-4b63-80d5-6d9a6c1e4b4e\",\"type\":\"BoxSelectTool\"},{\"attributes\":{},\"id\":\"f1857cc1-da5f-4b93-a8f9-969d4d961e6c\",\"type\":\"BasicTicker\"},{\"attributes\":{\"below\":[{\"id\":\"63713382-c415-4ffc-a2fb-c6fa67162c1a\",\"type\":\"LinearAxis\"}],\"left\":[{\"id\":\"a8894c2f-5765-48e1-b8f0-515c8a897445\",\"type\":\"LinearAxis\"}],\"renderers\":[{\"id\":\"63713382-c415-4ffc-a2fb-c6fa67162c1a\",\"type\":\"LinearAxis\"},{\"id\":\"ef0060a1-1845-4880-ba15-d84317efced3\",\"type\":\"Grid\"},{\"id\":\"a8894c2f-5765-48e1-b8f0-515c8a897445\",\"type\":\"LinearAxis\"},{\"id\":\"6d4e3c93-155b-422e-8c29-16e5ef8a847c\",\"type\":\"Grid\"},{\"id\":\"52332999-3b1e-44c9-ba2c-597e8dbefaef\",\"type\":\"BoxAnnotation\"},{\"id\":\"fd9a85fc-a251-48a7-b852-f71f809f69b3\",\"type\":\"BoxAnnotation\"},{\"id\":\"a2859309-8777-4e2c-8cee-dd3670c08707\",\"type\":\"GlyphRenderer\"}],\"title\":{\"id\":\"22dcfeb4-560a-4207-993d-a827704ba5ec\",\"type\":\"Title\"},\"toolbar\":{\"id\":\"39c2ea40-4f3d-4c91-8c5a-28313e9712db\",\"type\":\"Toolbar\"},\"x_range\":{\"id\":\"689ec22d-10e4-46ee-850b-99c43c49fa4c\",\"type\":\"DataRange1d\"},\"x_scale\":{\"id\":\"4783e4c7-355c-43d5-a709-85d021e20ec1\",\"type\":\"LinearScale\"},\"y_range\":{\"id\":\"cf337d02-bc57-4c69-a9ed-25262ea818a1\",\"type\":\"DataRange1d\"},\"y_scale\":{\"id\":\"8860e1b0-7592-48ae-8cb4-147436810f6d\",\"type\":\"LinearScale\"}},\"id\":\"392d6bf8-37b9-4be4-a99f-64420b0a2546\",\"subtype\":\"Figure\",\"type\":\"Plot\"},{\"attributes\":{\"fill_alpha\":{\"value\":0.2},\"fill_color\":{\"value\":\"#1f77b4\"},\"line_color\":{\"value\":\"#1f77b4\"},\"size\":{\"units\":\"screen\",\"value\":10},\"x\":{\"field\":\"x\"},\"y\":{\"field\":\"y\"}},\"id\":\"72d6571c-7693-442e-9455-6cea51b5f02b\",\"type\":\"Circle\"},{\"attributes\":{\"callback\":null},\"id\":\"cf337d02-bc57-4c69-a9ed-25262ea818a1\",\"type\":\"DataRange1d\"},{\"attributes\":{},\"id\":\"fc171877-33ec-48ac-b4de-ec8225789e71\",\"type\":\"BasicTicker\"},{\"attributes\":{},\"id\":\"4783e4c7-355c-43d5-a709-85d021e20ec1\",\"type\":\"LinearScale\"},{\"attributes\":{\"fill_alpha\":{\"value\":0.1},\"fill_color\":{\"value\":\"#1f77b4\"},\"line_alpha\":{\"value\":0.1},\"line_color\":{\"value\":\"#1f77b4\"},\"size\":{\"units\":\"screen\",\"value\":10},\"x\":{\"field\":\"x\"},\"y\":{\"field\":\"y\"}},\"id\":\"cbfa4dbe-127b-45a6-988c-80a6693c9e4a\",\"type\":\"Circle\"},{\"attributes\":{},\"id\":\"b9758331-d27d-47e0-bf11-c6cb9c0f9426\",\"type\":\"BasicTickFormatter\"},{\"attributes\":{},\"id\":\"0d4e3e68-e310-480f-a528-8fc485d97153\",\"type\":\"BasicTickFormatter\"},{\"attributes\":{\"overlay\":{\"id\":\"52332999-3b1e-44c9-ba2c-597e8dbefaef\",\"type\":\"BoxAnnotation\"}},\"id\":\"128705c9-e05d-4121-820d-8754cc6283bd\",\"type\":\"BoxZoomTool\"},{\"attributes\":{\"plot\":{\"id\":\"392d6bf8-37b9-4be4-a99f-64420b0a2546\",\"subtype\":\"Figure\",\"type\":\"Plot\"},\"ticker\":{\"id\":\"fc171877-33ec-48ac-b4de-ec8225789e71\",\"type\":\"BasicTicker\"}},\"id\":\"ef0060a1-1845-4880-ba15-d84317efced3\",\"type\":\"Grid\"}],\"root_ids\":[\"392d6bf8-37b9-4be4-a99f-64420b0a2546\"]},\"title\":\"Bokeh Application\",\"version\":\"0.12.7\"}};\n",
       "            var render_items = [{\"docid\":\"e6eabf9a-1352-4e21-a3e1-1ca93778e61e\",\"elementid\":\"a6d1ec7f-976a-4f68-b591-f3022c75cc91\",\"modelid\":\"392d6bf8-37b9-4be4-a99f-64420b0a2546\"}];\n",
       "            \n",
       "            Bokeh.embed.embed_items(docs_json, render_items);\n",
       "          };\n",
       "          if (document.readyState != \"loading\") fn();\n",
       "          else document.addEventListener(\"DOMContentLoaded\", fn);\n",
       "        })();\n",
       "\n",
       "      },\n",
       "      function(Bokeh) {\n",
       "      }\n",
       "    ];\n",
       "\n",
       "    function run_inline_js() {\n",
       "      \n",
       "      if ((root.Bokeh !== undefined) || (force === true)) {\n",
       "        for (var i = 0; i < inline_js.length; i++) {\n",
       "          inline_js[i].call(root, root.Bokeh);\n",
       "        }if (force === true) {\n",
       "          display_loaded();\n",
       "        }} else if (Date.now() < root._bokeh_timeout) {\n",
       "        setTimeout(run_inline_js, 100);\n",
       "      } else if (!root._bokeh_failed_load) {\n",
       "        console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n",
       "        root._bokeh_failed_load = true;\n",
       "      } else if (force !== true) {\n",
       "        var cell = $(document.getElementById(\"a6d1ec7f-976a-4f68-b591-f3022c75cc91\")).parents('.cell').data().cell;\n",
       "        cell.output_area.append_execute_result(NB_LOAD_WARNING)\n",
       "      }\n",
       "\n",
       "    }\n",
       "\n",
       "    if (root._bokeh_is_loading === 0) {\n",
       "      console.log(\"Bokeh: BokehJS loaded, going straight to plotting\");\n",
       "      run_inline_js();\n",
       "    } else {\n",
       "      load_libs(js_urls, function() {\n",
       "        console.log(\"Bokeh: BokehJS plotting callback run at\", now());\n",
       "        run_inline_js();\n",
       "      });\n",
       "    }\n",
       "  }(window));\n",
       "</script>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "# Configure default tools for plots\n",
    "TOOLS = 'box_zoom,box_select,resize,reset,save'\n",
    "\n",
    "# Configure for output in the notebook\n",
    "output_notebook()\n",
    "\n",
    "source = ColumnDataSource(df_plot)\n",
    "\n",
    "p = figure(title=\"Molecules\", tools = TOOLS)\n",
    "p.xaxis.axis_label = 'x'\n",
    "p.yaxis.axis_label = 'y'\n",
    "\n",
    "# Make scatterplot\n",
    "p.circle(\"x\", \"y\", fill_alpha=0.2, size=10, source=source)\n",
    "\n",
    "\n",
    "#source.add(df.Molecule.apply(lambda x: oenb.draw_mol_to_html(x,300,200)),name=\"img\")\n",
    "\n",
    "p = figure(title=\"test\", tools=TOOLS)\n",
    "p.xaxis.axis_label = 'x'\n",
    "p.yaxis.axis_label = 'y'\n",
    "\n",
    "# Make scatterplot\n",
    "p.circle(\"x\", \"y\", fill_alpha=0.2, size=10, source=source)\n",
    "\n",
    "# Create tooltips referencing stored images\n",
    "tooltips = \"\"\"<img src=@img>\"\"\"\n",
    "\n",
    "# Connect tooltips to plot\n",
    "p.add_tools(HoverTool(tooltips=tooltips))\n",
    "\n",
    "# Uncomment to create a html file of the plot.\n",
    "# output_file(\"bokeh_plot.html\")\n",
    "\n",
    "show(p)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(90, 200, 200, 1)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_list.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "'module' object is not callable",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-84-4f80df4f9056>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mnp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrandom\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m200\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m200\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mTypeError\u001b[0m: 'module' object is not callable"
     ]
    }
   ],
   "source": [
    "np.random([200,200,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "../../Downloads//files.interaction (1).txt.gz\n",
      "../../Downloads//files.interaction (2).txt.gz\n",
      "../../Downloads//files.interaction (3).txt.gz\n",
      "../../Downloads//files.interaction (4).txt.gz\n",
      "../../Downloads//files.interaction (5).txt.gz\n",
      "../../Downloads//files.interaction (6).txt.gz\n",
      "../../Downloads//files.interaction (7).txt.gz\n",
      "../../Downloads//files.interaction.txt.gz\n",
      "../../Downloads//files.interaction.txt.gz (1).tbi\n",
      "../../Downloads//files.interaction.txt.gz (2).tbi\n",
      "../../Downloads//files.interaction.txt.gz (3).tbi\n",
      "../../Downloads//files.interaction.txt.gz (4).tbi\n",
      "../../Downloads//files.interaction.txt.gz (5).tbi\n",
      "../../Downloads//files.interaction.txt.gz (6).tbi\n",
      "../../Downloads//files.interaction.txt.gz (7).tbi\n",
      "../../Downloads//files.interaction.txt.gz (8).tbi\n",
      "../../Downloads//files.interaction.txt.gz (9).tbi\n",
      "../../Downloads//files.interaction.txt.gz.tbi\n"
     ]
    }
   ],
   "source": [
    "!find ../../Downloads/ | grep interaction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}