Chapter 11 Generative LSTM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from random import random"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def random_rectangle():\n",
    "    width, height = random(), random()\n",
    "    points = list()\n",
    "    points.append([0.0,0.0])\n",
    "    points.append([width,0.0])\n",
    "    points.append([width,height])\n",
    "    points.append([0.0,height])\n",
    "    return points"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.0, 0.0], [0.1702915948268291, 0.0], [0.1702915948268291, 0.6249677004111098], [0.0, 0.6249677004111098]]\n"
     ]
    }
   ],
   "source": [
    "rect = random_rectangle()\n",
    "print(rect)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "rect.append(rect[0])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "axis = pyplot.gca()\n",
    "patch = PathPatch(path)\n",
    "axis.add_path(patch)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_rectangle(rect):\n",
    "    rect.append(rect[0])\n",
    "    codes = [Path.MOVETO, Path.LINETO,Path.LINETO,Path.LINETO,Path.CLOSEPOLY]\n",
    "    path = Path(rect,codes)\n",
    "    axis = pyplot.gca()\n",
    "    patch = PathPatch(path)\n",
    "    axis.add_patch(patch)\n",
    "    axis.set_xlim(-0.1,1.1)\n",
    "    axis.set_ylim(-0.1,1.1)\n",
    "    pyplot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from matplotlib import pyplot\n",
    "from matplotlib.patches import PathPatch\n",
    "from matplotlib.path import Path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_rectangle(rect)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_samples():\n",
    "    rect = random_rectangle()\n",
    "    X,y = list(), list()\n",
    "    for i in range(1,len(rect)):\n",
    "        X.append(rect[i-1])\n",
    "        y.append(rect[i])\n",
    "    X,y = array(X), array(y)\n",
    "    X = X.reshape((X.shape[0],1,2))\n",
    "    return X, y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from numpy import array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 0.] => [0.06479887 0.        ]\n",
      "[0.06479887 0.        ] => [0.06479887 0.70148552]\n",
      "[0.06479887 0.70148552] => [0.         0.70148552]\n"
     ]
    }
   ],
   "source": [
    "X,y = get_samples()\n",
    "for i in range(X.shape[0]):\n",
    "    print(X[i][0], '=>', y[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import LSTM\n",
    "from keras.layers import Dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0814 21:07:37.638391 139849587177280 module_wrapper.py:136] From /home/arvind/.local/lib/python3.6/site-packages/tensorflow_core/python/util/module_wrapper.py:163: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
      "\n",
      "W0814 21:07:37.873475 139849587177280 module_wrapper.py:136] From /home/arvind/.local/lib/python3.6/site-packages/tensorflow_core/python/util/module_wrapper.py:163: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "lstm_1 (LSTM)                (None, 10)                520       \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 2)                 22        \n",
      "=================================================================\n",
      "Total params: 542\n",
      "Trainable params: 542\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "model = Sequential()\n",
    "model.add(LSTM(10, input_shape=(1,2)))\n",
    "model.add(Dense(2,activation='linear'))\n",
    "model.compile(loss='mae',optimizer = 'adam')\n",
    "print(model.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "W0814 21:11:25.785642 139849587177280 deprecation.py:323] From /home/arvind/.local/lib/python3.6/site-packages/tensorflow_core/python/ops/math_grad.py:1423: where (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1\n",
      " - 1s - loss: 0.2461\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0303\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4077\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2062\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2513\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3338\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3058\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4747\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3219\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3105\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0507\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4903\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2347\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2218\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4675\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2046\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.5302\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3187\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4718\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3176\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3870\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1244\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2733\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4014\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1201\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1129\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3773\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2051\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1963\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2629\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0561\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3225\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2585\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4197\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3371\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3233\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4170\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3680\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4578\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2861\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1557\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4064\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2259\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4202\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1386\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3673\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0554\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4357\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2480\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3218\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2598\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1970\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3360\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3472\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1451\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0741\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3429\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.5324\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4263\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2952\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2804\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4247\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4785\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2716\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1709\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3881\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2896\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4526\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0991\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1671\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1775\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1812\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1767\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4063\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0786\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4594\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2883\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3230\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1332\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1993\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2610\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1599\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1009\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3186\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3629\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1345\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1147\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2976\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2374\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2099\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4389\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3916\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3927\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3181\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3754\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1670\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4001\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2250\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2766\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1481\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2875\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3245\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3671\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1074\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1908\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2531\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4013\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0841\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2738\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4175\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2672\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2049\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3132\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1365\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2989\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2399\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4915\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3115\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1867\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1436\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3068\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0706\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4106\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1983\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2614\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2446\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3370\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1902\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1985\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0875\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1814\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1146\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2532\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2867\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1065\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1350\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3626\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3750\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0527\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2818\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2815\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4130\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3413\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2149\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2064\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.5343\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4258\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2481\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2831\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2591\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1205\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4577\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.5242\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2872\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3259\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1963\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3072\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3092\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2277\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2653\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2904\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3616\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2359\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3922\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2750\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1027\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3409\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0636\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0752\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1034\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3431\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3666\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3728\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2547\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1316\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2751\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3610\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3611\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2797\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1372\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2892\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3401\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2927\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3854\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3675\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2077\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1547\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1654\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3978\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2231\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3701\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3181\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1425\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1614\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3049\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2077\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2216\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1421\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2799\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2651\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2651\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3504\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2563\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2993\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0915\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2133\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4574\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2595\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3090\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3228\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3683\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1568\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2963\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1848\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2238\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3272\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3164\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3140\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3690\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2274\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2379\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3029\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3755\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2785\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3175\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2729\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2447\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3349\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2868\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2963\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1456\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2132\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1555\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2509\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2702\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1367\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3201\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1457\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3458\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3131\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3069\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2911\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2311\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3244\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2833\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2222\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2666\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3087\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1748\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1187\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2645\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3190\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2655\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3474\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2728\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2265\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3064\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3133\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3018\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1775\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2439\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2183\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2760\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4070\n",
      "Epoch 1/1\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - 0s - loss: 0.0876\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2303\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1583\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2727\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0988\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1531\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0897\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2801\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2310\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4270\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2931\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2890\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3366\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1497\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0763\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3600\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1997\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1333\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1642\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2620\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2379\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3335\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0770\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3863\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2500\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1431\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0777\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2368\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2608\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4048\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2591\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2318\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2975\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1688\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3194\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3381\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1857\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2712\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3239\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3626\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2552\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2977\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1809\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0928\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0979\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2415\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3938\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2037\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2327\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1403\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1763\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2117\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4023\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2663\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1256\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3060\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3488\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2173\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2136\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2475\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2025\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3271\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2145\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1919\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3183\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3315\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2354\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2979\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1537\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3263\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3275\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2653\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0959\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2572\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1576\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3372\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2778\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4153\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3083\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2004\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1949\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1931\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2427\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2687\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2647\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2825\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2425\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.4011\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1624\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2378\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1598\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3471\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1385\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1886\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1707\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2300\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2841\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1507\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2200\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1977\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2100\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1917\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3739\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1495\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1219\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2719\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1501\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1797\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2393\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2354\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2630\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1333\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1949\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3474\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2419\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3416\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1819\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2269\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1290\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1603\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1685\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0893\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2047\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2055\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1387\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2105\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2609\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2152\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2445\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2002\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1281\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2292\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1948\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2510\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1972\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3468\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2342\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2157\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1613\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2410\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1199\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3073\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3171\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2849\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3538\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2771\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2875\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1923\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2544\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2571\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1622\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1508\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2557\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1670\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2905\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1431\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1480\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2731\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3491\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2816\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2723\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2223\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1392\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1279\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1162\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2367\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0934\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1331\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2596\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2766\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1771\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2699\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1525\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1874\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1843\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1990\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1559\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1412\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1452\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2095\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1467\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2410\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0743\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1884\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1556\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1046\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2151\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2493\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2521\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1695\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0841\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1691\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2006\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3224\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2094\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2088\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0723\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1335\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1794\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1996\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1564\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2779\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1891\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1301\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3195\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2675\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1051\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1690\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1152\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1742\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3588\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0781\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1373\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1838\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2482\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1864\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2712\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2595\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3826\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1320\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1164\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1650\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1639\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3489\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1543\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1994\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3227\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1696\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1694\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1997\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3035\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1483\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1834\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2783\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0973\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2097\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2491\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3594\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2902\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2844\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1993\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2095\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2437\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0684\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1821\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3083\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1067\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2948\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1963\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1675\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3286\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1175\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0963\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2556\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2251\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1618\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2776\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1353\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1876\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2713\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2213\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1790\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2528\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1196\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2423\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1\n",
      " - 0s - loss: 0.2069\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1371\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2250\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2716\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1513\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1623\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2035\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3335\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2017\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2179\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2212\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0811\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1249\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1943\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2364\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1362\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2089\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2487\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1650\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0703\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1664\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1882\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3075\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1146\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2300\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1568\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2590\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2021\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2288\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3646\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1750\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0958\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2715\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1052\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3405\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1642\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1615\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2229\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2175\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2811\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2844\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1669\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2357\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1139\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1664\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2040\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1245\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1389\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0957\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1374\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1891\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3886\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1750\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2343\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2128\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2811\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1548\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1167\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3336\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2575\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0778\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0748\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2655\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0728\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2018\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1134\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2086\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1288\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3108\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1057\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2520\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1159\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2381\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1449\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3265\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1418\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1646\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2105\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2801\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1543\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1562\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1604\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1238\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2823\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2689\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1491\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1527\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2067\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1799\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2531\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2738\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0773\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0764\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1309\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0764\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1142\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2740\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1504\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2795\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0870\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1342\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1718\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1765\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1364\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2681\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1026\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1948\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3079\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1121\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2309\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2567\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2111\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3002\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2431\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1267\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2492\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2432\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1405\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2041\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1169\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2896\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1032\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2560\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2911\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2035\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2393\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1313\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3228\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2885\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1235\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1592\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2475\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1429\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1612\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1497\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1374\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1104\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0769\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0772\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2037\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0915\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1343\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2035\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1596\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2203\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0905\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1289\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1917\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1718\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2105\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1851\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2092\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2661\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1099\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1484\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1166\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3300\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1889\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2151\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1642\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1993\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3592\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3145\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2761\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1767\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2105\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3245\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1246\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1977\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1373\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1589\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2680\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2059\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2704\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1961\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1515\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1248\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1881\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2488\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2019\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2261\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3099\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1802\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2587\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1917\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2282\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1955\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2357\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1396\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2137\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1077\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2439\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1346\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1846\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1036\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1540\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1149\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2008\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2189\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1534\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2609\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2013\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1418\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1456\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2623\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1462\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1633\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1471\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2613\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1454\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1560\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2033\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1208\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1492\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1123\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1786\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1276\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1534\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1434\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2326\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1978\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1411\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2712\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1914\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2751\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1346\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1960\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2853\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2483\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0842\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1492\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3657\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2118\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2064\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3046\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1635\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0831\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2906\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1347\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2472\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2058\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1637\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1555\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2296\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1002\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1529\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1230\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1941\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0989\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1231\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3187\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3040\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1905\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3239\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1632\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1046\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1777\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1832\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2137\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1694\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2003\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2060\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2557\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1444\n",
      "Epoch 1/1\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " - 0s - loss: 0.1464\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1512\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2086\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2471\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1662\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3154\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0937\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2268\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2279\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2581\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2161\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1698\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1940\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1554\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1095\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1557\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1667\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0786\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2089\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1985\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2629\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2009\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2318\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2096\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2848\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2538\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1810\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2355\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2708\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0758\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3330\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1979\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0893\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1809\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2304\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1671\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1288\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1797\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1558\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2970\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2285\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0926\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1699\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1370\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1122\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3544\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1598\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1706\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2772\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1313\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1168\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1813\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1736\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1194\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1501\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1048\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2538\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1979\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2447\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1918\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1488\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2157\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2671\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1884\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2756\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1239\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2630\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2195\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1948\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2182\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1319\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1705\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2538\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1381\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3790\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0944\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1115\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2870\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1487\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1305\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1640\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1513\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1535\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2151\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2830\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2464\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1353\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1790\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1257\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2651\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2742\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1135\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1539\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1594\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2587\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2524\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1022\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1976\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1095\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1588\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2081\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3363\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1886\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1739\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2794\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1712\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2949\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2394\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1057\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1059\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1013\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2174\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2302\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1737\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2029\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1734\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1319\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2845\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1629\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1410\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3132\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1810\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2251\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1514\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1239\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0892\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2136\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1922\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2467\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2750\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1243\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1883\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2402\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2601\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2209\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2852\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0878\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0985\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2450\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1913\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2673\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1933\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1015\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2388\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2519\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1739\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1909\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1561\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3458\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0909\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1113\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0873\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2118\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2284\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2559\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1872\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2228\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1154\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1684\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1957\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2733\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1961\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2200\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1667\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2037\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2280\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1858\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1603\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1652\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1211\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1879\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1133\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2130\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1282\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1896\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2254\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2609\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1258\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1706\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2620\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1518\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1208\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2249\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2181\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2576\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2810\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2136\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1505\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1207\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2239\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1761\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1158\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1753\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1291\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2186\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1578\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2501\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2335\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1844\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1836\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.3107\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2219\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1576\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2257\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.1948\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.2617\n",
      "Epoch 1/1\n",
      " - 0s - loss: 0.0925\n"
     ]
    }
   ],
   "source": [
    "for i in range(1000):\n",
    "    X,y = get_samples()\n",
    "    model.fit(X,y, epochs = 1, verbose=2, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_rectangle(model):\n",
    "    rect = list()\n",
    "    last = array([0.0,0.0]).reshape((1,1,2))\n",
    "    rect.append([[y for y in x] for x in last[0]][0])\n",
    "    for i in range(3):\n",
    "        yhat = model.predict(last, verbose=0)\n",
    "        last = yhat.reshape((1,1,2))\n",
    "        rect.append([[y for y in x] for x in last[0]][0])\n",
    "    return rect"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "rect = generate_rectangle(model)\n",
    "plot_rectangle(rect)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}