train.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import os\n",
    "\n",
    "os.environ['CUDA_VISIBLE_DEVICES'] = '1'\n",
    "os.environ['TF_FORCE_GPU_ALLOW_GROWTH'] = 'true'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8564, 45, 100)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "actions = [\n",
    "    'rewind',\n",
    "    'advance',\n",
    "    'stop',\n",
    "    'OK'\n",
    "]\n",
    "\n",
    "data = np.loadtxt('dataset/data_1648573431.csv', delimiter=',', dtype=np.float32)\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(8564, 45, 99)\n",
      "(8564,)\n"
     ]
    }
   ],
   "source": [
    "x_data = data[:, :-1]\n",
    "labels = data[:,  -1]\n",
    "print(x_data.shape)\n",
    "print(labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8564, 3)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from tensorflow.keras.utils import to_categorical\n",
    "\n",
    "y_data = to_categorical(labels, num_classes=len(actions))\n",
    "y_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7536, 45, 99) (7536, 3)\n",
      "(1028, 45, 99) (1028, 3)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x_data = x_data.astype(np.float32)\n",
    "y_data = y_data.astype(np.float32)\n",
    "x_train, x_val, y_train, y_val = train_test_split(x_data, y_data, test_size=0.12, random_state=2022)\n",
    "print(x_train.shape, y_train.shape)\n",
    "print(x_val.shape, y_val.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "lstm (LSTM)                  (None, 64)                41984     \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 32)                2080      \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 3)                 99        \n",
      "=================================================================\n",
      "Total params: 44,163\n",
      "Trainable params: 44,163\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import LSTM, Dense\n",
    "\n",
    "model = Sequential([\n",
    "    Dense(64, activation='relu', input_shape = x_train.shape[1:2]),\n",
    "    Dense(len(actions), activation='softmax')\n",
    "])\n",
    "\n",
    "model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['acc'])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1000\n",
      "236/236 [==============================] - 12s 43ms/step - loss: 24.0910 - acc: 0.4074 - val_loss: 7.5797 - val_acc: 0.4358\n",
      "\n",
      "Epoch 00001: val_acc improved from -inf to 0.43580, saving model to models\\gesture.h5\n",
      "Epoch 2/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 6.9324 - acc: 0.4820 - val_loss: 6.4111 - val_acc: 0.5068\n",
      "\n",
      "Epoch 00002: val_acc improved from 0.43580 to 0.50681, saving model to models\\gesture.h5\n",
      "Epoch 3/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 36.1899 - acc: 0.4624 - val_loss: 95.0077 - val_acc: 0.3901\n",
      "\n",
      "Epoch 00003: val_acc did not improve from 0.50681\n",
      "Epoch 4/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 97.9491 - acc: 0.4943 - val_loss: 10.7075 - val_acc: 0.5233\n",
      "\n",
      "Epoch 00004: val_acc improved from 0.50681 to 0.52335, saving model to models\\gesture.h5\n",
      "Epoch 5/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 670.9191 - acc: 0.3877 - val_loss: 58.9680 - val_acc: 0.3375\n",
      "\n",
      "Epoch 00005: val_acc did not improve from 0.52335\n",
      "Epoch 6/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 151.3815 - acc: 0.3584 - val_loss: 60.7012 - val_acc: 0.3492\n",
      "\n",
      "Epoch 00006: val_acc did not improve from 0.52335\n",
      "Epoch 7/1000\n",
      "236/236 [==============================] - 9s 38ms/step - loss: 90.9178 - acc: 0.3990 - val_loss: 36.2912 - val_acc: 0.5730\n",
      "\n",
      "Epoch 00007: val_acc improved from 0.52335 to 0.57296, saving model to models\\gesture.h5\n",
      "Epoch 8/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 53.5970 - acc: 0.4662 - val_loss: 24.5083 - val_acc: 0.5846\n",
      "\n",
      "Epoch 00008: val_acc improved from 0.57296 to 0.58463, saving model to models\\gesture.h5\n",
      "Epoch 9/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 48.2743 - acc: 0.4778 - val_loss: 30.2822 - val_acc: 0.4562\n",
      "\n",
      "Epoch 00009: val_acc did not improve from 0.58463\n",
      "Epoch 10/1000\n",
      "236/236 [==============================] - 9s 38ms/step - loss: 36.1554 - acc: 0.5013 - val_loss: 98.8187 - val_acc: 0.3444\n",
      "\n",
      "Epoch 00010: val_acc did not improve from 0.58463\n",
      "Epoch 11/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 36.3670 - acc: 0.5019 - val_loss: 85.3607 - val_acc: 0.5126\n",
      "\n",
      "Epoch 00011: val_acc did not improve from 0.58463\n",
      "Epoch 12/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 40.7314 - acc: 0.5031 - val_loss: 38.5374 - val_acc: 0.5175\n",
      "\n",
      "Epoch 00012: val_acc did not improve from 0.58463\n",
      "Epoch 13/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 33.8976 - acc: 0.5289 - val_loss: 38.4448 - val_acc: 0.4650\n",
      "\n",
      "Epoch 00013: val_acc did not improve from 0.58463\n",
      "Epoch 14/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 36.7876 - acc: 0.5086 - val_loss: 97.0113 - val_acc: 0.6109\n",
      "\n",
      "Epoch 00014: val_acc improved from 0.58463 to 0.61089, saving model to models\\gesture.h5\n",
      "Epoch 15/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 37.8425 - acc: 0.5192 - val_loss: 21.6090 - val_acc: 0.4640\n",
      "\n",
      "Epoch 00015: val_acc did not improve from 0.61089\n",
      "Epoch 16/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 28.0924 - acc: 0.5361 - val_loss: 87.5094 - val_acc: 0.3560\n",
      "\n",
      "Epoch 00016: val_acc did not improve from 0.61089\n",
      "Epoch 17/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 39.2103 - acc: 0.5242 - val_loss: 13.4364 - val_acc: 0.4309\n",
      "\n",
      "Epoch 00017: val_acc did not improve from 0.61089\n",
      "Epoch 18/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 30.4809 - acc: 0.5321 - val_loss: 10.2383 - val_acc: 0.4922\n",
      "\n",
      "Epoch 00018: val_acc did not improve from 0.61089\n",
      "Epoch 19/1000\n",
      "236/236 [==============================] - 9s 39ms/step - loss: 32.5762 - acc: 0.5300 - val_loss: 88.4834 - val_acc: 0.3570\n",
      "\n",
      "Epoch 00019: val_acc did not improve from 0.61089\n",
      "Epoch 20/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 36.9753 - acc: 0.5176 - val_loss: 34.0330 - val_acc: 0.4650\n",
      "\n",
      "Epoch 00020: val_acc did not improve from 0.61089\n",
      "Epoch 21/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 29.9355 - acc: 0.5426 - val_loss: 57.4173 - val_acc: 0.6274\n",
      "\n",
      "Epoch 00021: val_acc improved from 0.61089 to 0.62743, saving model to models\\gesture.h5\n",
      "Epoch 22/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 33.5646 - acc: 0.5333 - val_loss: 29.6681 - val_acc: 0.5233\n",
      "\n",
      "Epoch 00022: val_acc did not improve from 0.62743\n",
      "Epoch 23/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 29.0969 - acc: 0.5431 - val_loss: 44.2674 - val_acc: 0.3570\n",
      "\n",
      "Epoch 00023: val_acc did not improve from 0.62743\n",
      "Epoch 24/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 26.9205 - acc: 0.5344 - val_loss: 19.0424 - val_acc: 0.6294\n",
      "\n",
      "Epoch 00024: val_acc improved from 0.62743 to 0.62938, saving model to models\\gesture.h5\n",
      "Epoch 25/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 37.2468 - acc: 0.5341 - val_loss: 32.9077 - val_acc: 0.3560\n",
      "\n",
      "Epoch 00025: val_acc did not improve from 0.62938\n",
      "Epoch 26/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 23.8202 - acc: 0.5460 - val_loss: 13.8402 - val_acc: 0.5642\n",
      "\n",
      "Epoch 00026: val_acc did not improve from 0.62938\n",
      "Epoch 27/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 30.3161 - acc: 0.5563 - val_loss: 24.8220 - val_acc: 0.5311\n",
      "\n",
      "Epoch 00027: val_acc did not improve from 0.62938\n",
      "Epoch 28/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 21.7207 - acc: 0.5480 - val_loss: 28.3548 - val_acc: 0.4864\n",
      "\n",
      "Epoch 00028: val_acc did not improve from 0.62938\n",
      "Epoch 29/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 28.1356 - acc: 0.5382 - val_loss: 15.5908 - val_acc: 0.5389\n",
      "\n",
      "Epoch 00029: val_acc did not improve from 0.62938\n",
      "Epoch 30/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 24.6653 - acc: 0.5579 - val_loss: 18.1473 - val_acc: 0.5914\n",
      "\n",
      "Epoch 00030: val_acc did not improve from 0.62938\n",
      "Epoch 31/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 30.3829 - acc: 0.5352 - val_loss: 28.9196 - val_acc: 0.3570\n",
      "\n",
      "Epoch 00031: val_acc did not improve from 0.62938\n",
      "Epoch 32/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 32.9898 - acc: 0.5330 - val_loss: 21.8238 - val_acc: 0.5399\n",
      "\n",
      "Epoch 00032: val_acc did not improve from 0.62938\n",
      "Epoch 33/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 25.6295 - acc: 0.5636 - val_loss: 22.5435 - val_acc: 0.4883\n",
      "\n",
      "Epoch 00033: val_acc did not improve from 0.62938\n",
      "Epoch 34/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 24.5134 - acc: 0.5496 - val_loss: 19.5945 - val_acc: 0.5467\n",
      "\n",
      "Epoch 00034: val_acc did not improve from 0.62938\n",
      "Epoch 35/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 32.4760 - acc: 0.5403 - val_loss: 20.9523 - val_acc: 0.6245\n",
      "\n",
      "Epoch 00035: val_acc did not improve from 0.62938\n",
      "Epoch 36/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 18.0997 - acc: 0.5604 - val_loss: 13.2267 - val_acc: 0.6352\n",
      "\n",
      "Epoch 00036: val_acc improved from 0.62938 to 0.63521, saving model to models\\gesture.h5\n",
      "Epoch 37/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 24.5919 - acc: 0.5309 - val_loss: 33.6205 - val_acc: 0.6119\n",
      "\n",
      "Epoch 00037: val_acc did not improve from 0.63521\n",
      "Epoch 38/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 20.9714 - acc: 0.5380 - val_loss: 25.1448 - val_acc: 0.5389\n",
      "\n",
      "Epoch 00038: val_acc did not improve from 0.63521\n",
      "Epoch 39/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 17.7886 - acc: 0.5576 - val_loss: 14.2066 - val_acc: 0.4912\n",
      "\n",
      "Epoch 00039: val_acc did not improve from 0.63521\n",
      "Epoch 40/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 25.7509 - acc: 0.5301 - val_loss: 13.3829 - val_acc: 0.6479\n",
      "\n",
      "Epoch 00040: val_acc improved from 0.63521 to 0.64786, saving model to models\\gesture.h5\n",
      "Epoch 41/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 17.0753 - acc: 0.5427 - val_loss: 6.7606 - val_acc: 0.5175\n",
      "\n",
      "Epoch 00041: val_acc did not improve from 0.64786\n",
      "Epoch 42/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 5s 22ms/step - loss: 16.3735 - acc: 0.5426 - val_loss: 36.9240 - val_acc: 0.4329\n",
      "\n",
      "Epoch 00042: val_acc did not improve from 0.64786\n",
      "Epoch 43/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 15.0380 - acc: 0.5362 - val_loss: 36.0296 - val_acc: 0.3580\n",
      "\n",
      "Epoch 00043: val_acc did not improve from 0.64786\n",
      "Epoch 44/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 17.9008 - acc: 0.5395 - val_loss: 23.5212 - val_acc: 0.4543\n",
      "\n",
      "Epoch 00044: val_acc did not improve from 0.64786\n",
      "Epoch 45/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 16.3971 - acc: 0.5482 - val_loss: 14.5431 - val_acc: 0.5165\n",
      "\n",
      "Epoch 00045: val_acc did not improve from 0.64786\n",
      "Epoch 46/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 14.2439 - acc: 0.5638 - val_loss: 8.0887 - val_acc: 0.5593\n",
      "\n",
      "Epoch 00046: val_acc did not improve from 0.64786\n",
      "Epoch 47/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 13.9155 - acc: 0.5673 - val_loss: 3.5038 - val_acc: 0.6644\n",
      "\n",
      "Epoch 00047: val_acc improved from 0.64786 to 0.66440, saving model to models\\gesture.h5\n",
      "Epoch 48/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 12.7537 - acc: 0.5695 - val_loss: 12.4868 - val_acc: 0.5039\n",
      "\n",
      "Epoch 00048: val_acc did not improve from 0.66440\n",
      "Epoch 49/1000\n",
      "236/236 [==============================] - 9s 39ms/step - loss: 14.6442 - acc: 0.5593 - val_loss: 14.9649 - val_acc: 0.6109\n",
      "\n",
      "Epoch 00049: val_acc did not improve from 0.66440\n",
      "Epoch 50/1000\n",
      "236/236 [==============================] - 9s 40ms/step - loss: 16.4251 - acc: 0.5569 - val_loss: 13.8246 - val_acc: 0.4669\n",
      "\n",
      "Epoch 00050: val_acc did not improve from 0.66440\n",
      "Epoch 51/1000\n",
      "236/236 [==============================] - 9s 36ms/step - loss: 14.0989 - acc: 0.5565 - val_loss: 5.8413 - val_acc: 0.6274\n",
      "\n",
      "Epoch 00051: val_acc did not improve from 0.66440\n",
      "Epoch 52/1000\n",
      "236/236 [==============================] - 10s 42ms/step - loss: 10.3916 - acc: 0.5873 - val_loss: 6.0939 - val_acc: 0.6226\n",
      "\n",
      "Epoch 00052: val_acc did not improve from 0.66440\n",
      "Epoch 53/1000\n",
      "236/236 [==============================] - 9s 39ms/step - loss: 12.1053 - acc: 0.5633 - val_loss: 9.7089 - val_acc: 0.4942\n",
      "\n",
      "Epoch 00053: val_acc did not improve from 0.66440\n",
      "Epoch 54/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 12.9534 - acc: 0.5606 - val_loss: 2.7555 - val_acc: 0.5593\n",
      "\n",
      "Epoch 00054: val_acc did not improve from 0.66440\n",
      "Epoch 55/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 10.4805 - acc: 0.5870 - val_loss: 10.2878 - val_acc: 0.6206\n",
      "\n",
      "Epoch 00055: val_acc did not improve from 0.66440\n",
      "Epoch 56/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 8.3739 - acc: 0.5943 - val_loss: 8.8104 - val_acc: 0.5321\n",
      "\n",
      "Epoch 00056: val_acc did not improve from 0.66440\n",
      "Epoch 57/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 9.5595 - acc: 0.6001 - val_loss: 7.8541 - val_acc: 0.6518\n",
      "\n",
      "Epoch 00057: val_acc did not improve from 0.66440\n",
      "Epoch 58/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 7.4961 - acc: 0.5950 - val_loss: 10.7890 - val_acc: 0.6051\n",
      "\n",
      "Epoch 00058: val_acc did not improve from 0.66440\n",
      "Epoch 59/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 8.8250 - acc: 0.5817 - val_loss: 4.6869 - val_acc: 0.6226\n",
      "\n",
      "Epoch 00059: val_acc did not improve from 0.66440\n",
      "Epoch 60/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 9.0647 - acc: 0.5709 - val_loss: 2.5944 - val_acc: 0.5350\n",
      "\n",
      "Epoch 00060: val_acc did not improve from 0.66440\n",
      "Epoch 61/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 6.2491 - acc: 0.5780 - val_loss: 2.2464 - val_acc: 0.6440\n",
      "\n",
      "Epoch 00061: val_acc did not improve from 0.66440\n",
      "Epoch 62/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 5.4682 - acc: 0.5856 - val_loss: 3.3129 - val_acc: 0.5739\n",
      "\n",
      "Epoch 00062: val_acc did not improve from 0.66440\n",
      "Epoch 63/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 5.4492 - acc: 0.5981 - val_loss: 2.9983 - val_acc: 0.5545\n",
      "\n",
      "Epoch 00063: val_acc did not improve from 0.66440\n",
      "Epoch 64/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 5.4689 - acc: 0.5902 - val_loss: 10.1373 - val_acc: 0.4747\n",
      "\n",
      "Epoch 00064: val_acc did not improve from 0.66440\n",
      "Epoch 65/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 3.4782 - acc: 0.6060 - val_loss: 4.0302 - val_acc: 0.5233\n",
      "\n",
      "Epoch 00065: val_acc did not improve from 0.66440\n",
      "Epoch 66/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 6.4970 - acc: 0.5784 - val_loss: 2.9794 - val_acc: 0.6673\n",
      "\n",
      "Epoch 00066: val_acc improved from 0.66440 to 0.66732, saving model to models\\gesture.h5\n",
      "Epoch 67/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 5.4981 - acc: 0.5933 - val_loss: 9.3125 - val_acc: 0.6274\n",
      "\n",
      "Epoch 00067: val_acc did not improve from 0.66732\n",
      "Epoch 68/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 6.8958 - acc: 0.5913 - val_loss: 3.1285 - val_acc: 0.5214\n",
      "\n",
      "Epoch 00068: val_acc did not improve from 0.66732\n",
      "Epoch 69/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 3.9089 - acc: 0.6136 - val_loss: 9.1636 - val_acc: 0.5389\n",
      "\n",
      "Epoch 00069: val_acc did not improve from 0.66732\n",
      "Epoch 70/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 5.0355 - acc: 0.5962 - val_loss: 14.0574 - val_acc: 0.4660\n",
      "\n",
      "Epoch 00070: val_acc did not improve from 0.66732\n",
      "Epoch 71/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 6.3982 - acc: 0.4772 - val_loss: 8.6114 - val_acc: 0.3881\n",
      "\n",
      "Epoch 00071: val_acc did not improve from 0.66732\n",
      "Epoch 72/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.2350 - acc: 0.5005 - val_loss: 1.9917 - val_acc: 0.5233\n",
      "\n",
      "Epoch 00072: val_acc did not improve from 0.66732\n",
      "Epoch 73/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 2.8225 - acc: 0.5686 - val_loss: 2.2965 - val_acc: 0.4815\n",
      "\n",
      "Epoch 00073: val_acc did not improve from 0.66732\n",
      "Epoch 74/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 2.5670 - acc: 0.5738 - val_loss: 2.1378 - val_acc: 0.6877\n",
      "\n",
      "Epoch 00074: val_acc improved from 0.66732 to 0.68774, saving model to models\\gesture.h5\n",
      "Epoch 75/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 3.0399 - acc: 0.5791 - val_loss: 1.9265 - val_acc: 0.5340\n",
      "\n",
      "Epoch 00075: val_acc did not improve from 0.68774\n",
      "Epoch 76/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 2.9680 - acc: 0.5637 - val_loss: 2.1609 - val_acc: 0.6411\n",
      "\n",
      "Epoch 00076: val_acc did not improve from 0.68774\n",
      "Epoch 77/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 2.1709 - acc: 0.6223 - val_loss: 1.6866 - val_acc: 0.6868\n",
      "\n",
      "Epoch 00077: val_acc did not improve from 0.68774\n",
      "Epoch 78/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 2.1258 - acc: 0.6105 - val_loss: 1.2222 - val_acc: 0.6245\n",
      "\n",
      "Epoch 00078: val_acc did not improve from 0.68774\n",
      "Epoch 79/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 430.3382 - acc: 0.5267 - val_loss: 528.5187 - val_acc: 0.3220\n",
      "\n",
      "Epoch 00079: val_acc did not improve from 0.68774\n",
      "Epoch 80/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 66.4607 - acc: 0.4115 - val_loss: 16.4867 - val_acc: 0.4455\n",
      "\n",
      "Epoch 00080: val_acc did not improve from 0.68774\n",
      "Epoch 81/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 9.3332 - acc: 0.4885 - val_loss: 3.7625 - val_acc: 0.5535\n",
      "\n",
      "Epoch 00081: val_acc did not improve from 0.68774\n",
      "Epoch 82/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 3.2699 - acc: 0.5540 - val_loss: 2.9437 - val_acc: 0.5603\n",
      "\n",
      "Epoch 00082: val_acc did not improve from 0.68774\n",
      "Epoch 83/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 2.4786 - acc: 0.5893 - val_loss: 6.1499 - val_acc: 0.4776\n",
      "\n",
      "Epoch 00083: val_acc did not improve from 0.68774\n",
      "Epoch 84/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 2.7212 - acc: 0.5839 - val_loss: 4.9934 - val_acc: 0.5370\n",
      "\n",
      "Epoch 00084: val_acc did not improve from 0.68774\n",
      "Epoch 85/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 9s 36ms/step - loss: 3.3072 - acc: 0.5669 - val_loss: 3.3065 - val_acc: 0.5681\n",
      "\n",
      "Epoch 00085: val_acc did not improve from 0.68774\n",
      "Epoch 86/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 2.1925 - acc: 0.6287 - val_loss: 1.9487 - val_acc: 0.5837\n",
      "\n",
      "Epoch 00086: val_acc did not improve from 0.68774\n",
      "Epoch 87/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 1.6414 - acc: 0.6482 - val_loss: 1.3474 - val_acc: 0.6498\n",
      "\n",
      "Epoch 00087: val_acc did not improve from 0.68774\n",
      "Epoch 88/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 1.4518 - acc: 0.6456 - val_loss: 1.2922 - val_acc: 0.6984\n",
      "\n",
      "Epoch 00088: val_acc improved from 0.68774 to 0.69844, saving model to models\\gesture.h5\n",
      "Epoch 89/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.2916 - acc: 0.6602 - val_loss: 1.8258 - val_acc: 0.5496\n",
      "\n",
      "Epoch 00089: val_acc did not improve from 0.69844\n",
      "Epoch 90/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.5137 - acc: 0.5515 - val_loss: 2.9328 - val_acc: 0.4835\n",
      "\n",
      "Epoch 00090: val_acc did not improve from 0.69844\n",
      "Epoch 91/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 2.4384 - acc: 0.5387 - val_loss: 1.5638 - val_acc: 0.5924\n",
      "\n",
      "Epoch 00091: val_acc did not improve from 0.69844\n",
      "Epoch 92/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 1.7598 - acc: 0.5908 - val_loss: 1.4220 - val_acc: 0.6323\n",
      "\n",
      "Epoch 00092: val_acc did not improve from 0.69844\n",
      "Epoch 93/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.4802 - acc: 0.6104 - val_loss: 1.0381 - val_acc: 0.6342\n",
      "\n",
      "Epoch 00093: val_acc did not improve from 0.69844\n",
      "Epoch 94/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 1.2769 - acc: 0.6286 - val_loss: 0.9348 - val_acc: 0.6411\n",
      "\n",
      "Epoch 00094: val_acc did not improve from 0.69844\n",
      "Epoch 95/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 1.1269 - acc: 0.6367 - val_loss: 0.8610 - val_acc: 0.6868\n",
      "\n",
      "Epoch 00095: val_acc did not improve from 0.69844\n",
      "Epoch 96/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.9848 - acc: 0.6519 - val_loss: 0.7802 - val_acc: 0.6780\n",
      "\n",
      "Epoch 00096: val_acc did not improve from 0.69844\n",
      "Epoch 97/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.9433 - acc: 0.6673 - val_loss: 0.9927 - val_acc: 0.6907\n",
      "\n",
      "Epoch 00097: val_acc did not improve from 0.69844\n",
      "Epoch 98/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.8051 - acc: 0.6675 - val_loss: 0.8304 - val_acc: 0.6313\n",
      "\n",
      "Epoch 00098: val_acc did not improve from 0.69844\n",
      "Epoch 99/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.8311 - acc: 0.6848 - val_loss: 0.8092 - val_acc: 0.6916\n",
      "\n",
      "Epoch 00099: val_acc did not improve from 0.69844\n",
      "Epoch 100/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 1.0643 - acc: 0.6774 - val_loss: 0.8200 - val_acc: 0.6469\n",
      "\n",
      "Epoch 00100: val_acc did not improve from 0.69844\n",
      "Epoch 101/1000\n",
      "236/236 [==============================] - 6s 24ms/step - loss: 0.8849 - acc: 0.6943 - val_loss: 0.6325 - val_acc: 0.7189\n",
      "\n",
      "Epoch 00101: val_acc improved from 0.69844 to 0.71887, saving model to models\\gesture.h5\n",
      "Epoch 102/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.7672 - acc: 0.6961 - val_loss: 0.6349 - val_acc: 0.7286\n",
      "\n",
      "Epoch 00102: val_acc improved from 0.71887 to 0.72860, saving model to models\\gesture.h5\n",
      "Epoch 103/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.8022 - acc: 0.7021 - val_loss: 0.5712 - val_acc: 0.7296\n",
      "\n",
      "Epoch 00103: val_acc improved from 0.72860 to 0.72957, saving model to models\\gesture.h5\n",
      "Epoch 104/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.7060 - acc: 0.7098 - val_loss: 0.5695 - val_acc: 0.7043\n",
      "\n",
      "Epoch 00104: val_acc did not improve from 0.72957\n",
      "Epoch 105/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.6692 - acc: 0.7130 - val_loss: 0.7552 - val_acc: 0.7004\n",
      "\n",
      "Epoch 00105: val_acc did not improve from 0.72957\n",
      "Epoch 106/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.7396 - acc: 0.7118 - val_loss: 0.5584 - val_acc: 0.7082\n",
      "\n",
      "Epoch 00106: val_acc did not improve from 0.72957\n",
      "Epoch 107/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.6243 - acc: 0.7217 - val_loss: 0.5315 - val_acc: 0.7140\n",
      "\n",
      "Epoch 00107: val_acc did not improve from 0.72957\n",
      "Epoch 108/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.6792 - acc: 0.7107 - val_loss: 0.5529 - val_acc: 0.6848\n",
      "\n",
      "Epoch 00108: val_acc did not improve from 0.72957\n",
      "Epoch 109/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.6375 - acc: 0.7158 - val_loss: 0.4961 - val_acc: 0.7471\n",
      "\n",
      "Epoch 00109: val_acc improved from 0.72957 to 0.74708, saving model to models\\gesture.h5\n",
      "Epoch 110/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.6665 - acc: 0.7156 - val_loss: 0.5523 - val_acc: 0.7062\n",
      "\n",
      "Epoch 00110: val_acc did not improve from 0.74708\n",
      "Epoch 111/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.6450 - acc: 0.7236 - val_loss: 0.5831 - val_acc: 0.6615\n",
      "\n",
      "Epoch 00111: val_acc did not improve from 0.74708\n",
      "Epoch 112/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.5734 - acc: 0.7227 - val_loss: 0.5793 - val_acc: 0.7111\n",
      "\n",
      "Epoch 00112: val_acc did not improve from 0.74708\n",
      "Epoch 113/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.5910 - acc: 0.7244 - val_loss: 0.4772 - val_acc: 0.7578\n",
      "\n",
      "Epoch 00113: val_acc improved from 0.74708 to 0.75778, saving model to models\\gesture.h5\n",
      "Epoch 114/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.5726 - acc: 0.7103 - val_loss: 0.5898 - val_acc: 0.7208\n",
      "\n",
      "Epoch 00114: val_acc did not improve from 0.75778\n",
      "Epoch 115/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 0.6601 - acc: 0.7313 - val_loss: 0.4910 - val_acc: 0.7432\n",
      "\n",
      "Epoch 00115: val_acc did not improve from 0.75778\n",
      "Epoch 116/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.6448 - acc: 0.7276 - val_loss: 1.0415 - val_acc: 0.6741\n",
      "\n",
      "Epoch 00116: val_acc did not improve from 0.75778\n",
      "Epoch 117/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.5329 - acc: 0.7209 - val_loss: 0.9846 - val_acc: 0.7383\n",
      "\n",
      "Epoch 00117: val_acc did not improve from 0.75778\n",
      "Epoch 118/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.5662 - acc: 0.7139 - val_loss: 0.9245 - val_acc: 0.7432\n",
      "\n",
      "Epoch 00118: val_acc did not improve from 0.75778\n",
      "Epoch 119/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.5447 - acc: 0.7174 - val_loss: 1.1723 - val_acc: 0.7140\n",
      "\n",
      "Epoch 00119: val_acc did not improve from 0.75778\n",
      "Epoch 120/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.5128 - acc: 0.7357 - val_loss: 0.9481 - val_acc: 0.7228\n",
      "\n",
      "Epoch 00120: val_acc did not improve from 0.75778\n",
      "Epoch 121/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 8.2327 - acc: 0.6935 - val_loss: 62.8405 - val_acc: 0.3589\n",
      "\n",
      "Epoch 00121: val_acc did not improve from 0.75778\n",
      "Epoch 122/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 201.6194 - acc: 0.5068 - val_loss: 6.5533 - val_acc: 0.4134\n",
      "\n",
      "Epoch 00122: val_acc did not improve from 0.75778\n",
      "Epoch 123/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 2.7287 - acc: 0.6209 - val_loss: 2.1009 - val_acc: 0.6206\n",
      "\n",
      "Epoch 00123: val_acc did not improve from 0.75778\n",
      "Epoch 124/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.8650 - acc: 0.6392 - val_loss: 1.5995 - val_acc: 0.6362\n",
      "\n",
      "Epoch 00124: val_acc did not improve from 0.75778\n",
      "Epoch 125/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.7946 - acc: 0.6275 - val_loss: 1.2931 - val_acc: 0.6576\n",
      "\n",
      "Epoch 00125: val_acc did not improve from 0.75778\n",
      "Epoch 126/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.4633 - acc: 0.6038 - val_loss: 2.1104 - val_acc: 0.6761\n",
      "\n",
      "Epoch 00126: val_acc did not improve from 0.75778\n",
      "Epoch 127/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 29ms/step - loss: 1.1915 - acc: 0.6241 - val_loss: 1.5158 - val_acc: 0.7043\n",
      "\n",
      "Epoch 00127: val_acc did not improve from 0.75778\n",
      "Epoch 128/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.5626 - acc: 0.6123 - val_loss: 3.0674 - val_acc: 0.5185\n",
      "\n",
      "Epoch 00128: val_acc did not improve from 0.75778\n",
      "Epoch 129/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 1.2073 - acc: 0.6417 - val_loss: 1.2776 - val_acc: 0.6187\n",
      "\n",
      "Epoch 00129: val_acc did not improve from 0.75778\n",
      "Epoch 130/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.9124 - acc: 0.6604 - val_loss: 2.4118 - val_acc: 0.4815\n",
      "\n",
      "Epoch 00130: val_acc did not improve from 0.75778\n",
      "Epoch 131/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 1.3757 - acc: 0.6397 - val_loss: 0.7255 - val_acc: 0.6420\n",
      "\n",
      "Epoch 00131: val_acc did not improve from 0.75778\n",
      "Epoch 132/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.8024 - acc: 0.6782 - val_loss: 2.2697 - val_acc: 0.5700\n",
      "\n",
      "Epoch 00132: val_acc did not improve from 0.75778\n",
      "Epoch 133/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 1.1256 - acc: 0.6529 - val_loss: 1.2812 - val_acc: 0.5263\n",
      "\n",
      "Epoch 00133: val_acc did not improve from 0.75778\n",
      "Epoch 134/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 1.5901 - acc: 0.6334 - val_loss: 2.5781 - val_acc: 0.6800\n",
      "\n",
      "Epoch 00134: val_acc did not improve from 0.75778\n",
      "Epoch 135/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 1.0693 - acc: 0.6564 - val_loss: 1.3912 - val_acc: 0.7053\n",
      "\n",
      "Epoch 00135: val_acc did not improve from 0.75778\n",
      "Epoch 136/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.0951 - acc: 0.6709 - val_loss: 0.9424 - val_acc: 0.6235\n",
      "\n",
      "Epoch 00136: val_acc did not improve from 0.75778\n",
      "Epoch 137/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 1.0349 - acc: 0.6691 - val_loss: 0.6726 - val_acc: 0.6741\n",
      "\n",
      "Epoch 00137: val_acc did not improve from 0.75778\n",
      "Epoch 138/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 1.0225 - acc: 0.6675 - val_loss: 0.5678 - val_acc: 0.7500\n",
      "\n",
      "Epoch 00138: val_acc did not improve from 0.75778\n",
      "Epoch 139/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.2277 - acc: 0.6732 - val_loss: 1.6206 - val_acc: 0.7121\n",
      "\n",
      "Epoch 00139: val_acc did not improve from 0.75778\n",
      "Epoch 140/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.9489 - acc: 0.6818 - val_loss: 1.0814 - val_acc: 0.7053\n",
      "\n",
      "Epoch 00140: val_acc did not improve from 0.75778\n",
      "Epoch 141/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 1.0887 - acc: 0.6718 - val_loss: 0.9727 - val_acc: 0.7218\n",
      "\n",
      "Epoch 00141: val_acc did not improve from 0.75778\n",
      "Epoch 142/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 1.5937 - acc: 0.6482 - val_loss: 0.9846 - val_acc: 0.7296\n",
      "\n",
      "Epoch 00142: val_acc did not improve from 0.75778\n",
      "Epoch 143/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.8515 - acc: 0.6879 - val_loss: 0.8875 - val_acc: 0.6809\n",
      "\n",
      "Epoch 00143: val_acc did not improve from 0.75778\n",
      "Epoch 144/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.2043 - acc: 0.6619 - val_loss: 0.6098 - val_acc: 0.6868\n",
      "\n",
      "Epoch 00144: val_acc did not improve from 0.75778\n",
      "Epoch 145/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.3838 - acc: 0.6660 - val_loss: 0.6864 - val_acc: 0.6673\n",
      "\n",
      "Epoch 00145: val_acc did not improve from 0.75778\n",
      "Epoch 146/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.0960 - acc: 0.6700 - val_loss: 1.5592 - val_acc: 0.6848\n",
      "\n",
      "Epoch 00146: val_acc did not improve from 0.75778\n",
      "Epoch 147/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 1.1510 - acc: 0.6761 - val_loss: 1.2881 - val_acc: 0.6663\n",
      "\n",
      "Epoch 00147: val_acc did not improve from 0.75778\n",
      "Epoch 148/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.4185 - acc: 0.6640 - val_loss: 0.7939 - val_acc: 0.7014\n",
      "\n",
      "Epoch 00148: val_acc did not improve from 0.75778\n",
      "Epoch 149/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.9131 - acc: 0.6870 - val_loss: 2.9243 - val_acc: 0.6897\n",
      "\n",
      "Epoch 00149: val_acc did not improve from 0.75778\n",
      "Epoch 150/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.2304 - acc: 0.6718 - val_loss: 1.2252 - val_acc: 0.6051\n",
      "\n",
      "Epoch 00150: val_acc did not improve from 0.75778\n",
      "Epoch 151/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 1.0956 - acc: 0.6846 - val_loss: 2.0871 - val_acc: 0.6984\n",
      "\n",
      "Epoch 00151: val_acc did not improve from 0.75778\n",
      "Epoch 152/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.3688 - acc: 0.6685 - val_loss: 0.5759 - val_acc: 0.7374\n",
      "\n",
      "Epoch 00152: val_acc did not improve from 0.75778\n",
      "Epoch 153/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.8881 - acc: 0.7021 - val_loss: 1.2499 - val_acc: 0.6926\n",
      "\n",
      "Epoch 00153: val_acc did not improve from 0.75778\n",
      "Epoch 154/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 1.0456 - acc: 0.6829 - val_loss: 2.8808 - val_acc: 0.7091\n",
      "\n",
      "Epoch 00154: val_acc did not improve from 0.75778\n",
      "Epoch 155/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 1.1530 - acc: 0.6830 - val_loss: 1.1634 - val_acc: 0.7014\n",
      "\n",
      "Epoch 00155: val_acc did not improve from 0.75778\n",
      "Epoch 156/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 1.4276 - acc: 0.6657 - val_loss: 2.2290 - val_acc: 0.7179\n",
      "\n",
      "Epoch 00156: val_acc did not improve from 0.75778\n",
      "Epoch 157/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.8793 - acc: 0.7101 - val_loss: 0.5328 - val_acc: 0.7451\n",
      "\n",
      "Epoch 00157: val_acc did not improve from 0.75778\n",
      "Epoch 158/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.8527 - acc: 0.7183 - val_loss: 0.9596 - val_acc: 0.6829\n",
      "\n",
      "Epoch 00158: val_acc did not improve from 0.75778\n",
      "Epoch 159/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.9966 - acc: 0.7042 - val_loss: 0.5211 - val_acc: 0.7636\n",
      "\n",
      "Epoch 00159: val_acc improved from 0.75778 to 0.76362, saving model to models\\gesture.h5\n",
      "Epoch 160/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.9271 - acc: 0.6993 - val_loss: 0.6270 - val_acc: 0.7714\n",
      "\n",
      "Epoch 00160: val_acc improved from 0.76362 to 0.77140, saving model to models\\gesture.h5\n",
      "Epoch 161/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 3.6365 - acc: 0.6876 - val_loss: 12.3289 - val_acc: 0.3696\n",
      "\n",
      "Epoch 00161: val_acc did not improve from 0.77140\n",
      "Epoch 162/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 2.9595 - acc: 0.6488 - val_loss: 1.1547 - val_acc: 0.6897\n",
      "\n",
      "Epoch 00162: val_acc did not improve from 0.77140\n",
      "Epoch 163/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 1.2630 - acc: 0.6939 - val_loss: 1.0567 - val_acc: 0.6819\n",
      "\n",
      "Epoch 00163: val_acc did not improve from 0.77140\n",
      "Epoch 164/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 1.0589 - acc: 0.6924 - val_loss: 1.3108 - val_acc: 0.6994\n",
      "\n",
      "Epoch 00164: val_acc did not improve from 0.77140\n",
      "Epoch 165/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.0998 - acc: 0.6907 - val_loss: 0.7489 - val_acc: 0.7315\n",
      "\n",
      "Epoch 00165: val_acc did not improve from 0.77140\n",
      "Epoch 166/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 0.9783 - acc: 0.6959 - val_loss: 0.6460 - val_acc: 0.7831\n",
      "\n",
      "Epoch 00166: val_acc improved from 0.77140 to 0.78307, saving model to models\\gesture.h5\n",
      "Epoch 167/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.7443 - acc: 0.7237 - val_loss: 1.0623 - val_acc: 0.7451\n",
      "\n",
      "Epoch 00167: val_acc did not improve from 0.78307\n",
      "Epoch 168/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.7463 - acc: 0.7239 - val_loss: 1.0569 - val_acc: 0.7383\n",
      "\n",
      "Epoch 00168: val_acc did not improve from 0.78307\n",
      "Epoch 169/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.6144 - acc: 0.7382 - val_loss: 0.5042 - val_acc: 0.7597\n",
      "\n",
      "Epoch 00169: val_acc did not improve from 0.78307\n",
      "Epoch 170/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 31ms/step - loss: 0.7438 - acc: 0.7262 - val_loss: 0.5650 - val_acc: 0.7364\n",
      "\n",
      "Epoch 00170: val_acc did not improve from 0.78307\n",
      "Epoch 171/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.5854 - acc: 0.7580 - val_loss: 0.6852 - val_acc: 0.6381\n",
      "\n",
      "Epoch 00171: val_acc did not improve from 0.78307\n",
      "Epoch 172/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.5517 - acc: 0.7545 - val_loss: 0.5963 - val_acc: 0.7471\n",
      "\n",
      "Epoch 00172: val_acc did not improve from 0.78307\n",
      "Epoch 173/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.6177 - acc: 0.7614 - val_loss: 0.7026 - val_acc: 0.7607\n",
      "\n",
      "Epoch 00173: val_acc did not improve from 0.78307\n",
      "Epoch 174/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.5371 - acc: 0.7695 - val_loss: 0.6690 - val_acc: 0.6459\n",
      "\n",
      "Epoch 00174: val_acc did not improve from 0.78307\n",
      "Epoch 175/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.5878 - acc: 0.7691 - val_loss: 0.6041 - val_acc: 0.7490\n",
      "\n",
      "Epoch 00175: val_acc did not improve from 0.78307\n",
      "Epoch 176/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.5939 - acc: 0.7768 - val_loss: 1.0332 - val_acc: 0.7656\n",
      "\n",
      "Epoch 00176: val_acc did not improve from 0.78307\n",
      "Epoch 177/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.4574 - acc: 0.8194 - val_loss: 0.4185 - val_acc: 0.8366\n",
      "\n",
      "Epoch 00177: val_acc improved from 0.78307 to 0.83658, saving model to models\\gesture.h5\n",
      "Epoch 178/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.4470 - acc: 0.8274 - val_loss: 0.5279 - val_acc: 0.8239\n",
      "\n",
      "Epoch 00178: val_acc did not improve from 0.83658\n",
      "Epoch 179/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.3789 - acc: 0.8447 - val_loss: 0.5790 - val_acc: 0.7821\n",
      "\n",
      "Epoch 00179: val_acc did not improve from 0.83658\n",
      "Epoch 180/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.3590 - acc: 0.8556 - val_loss: 0.2829 - val_acc: 0.8755: 0.3603 - a\n",
      "\n",
      "Epoch 00180: val_acc improved from 0.83658 to 0.87549, saving model to models\\gesture.h5\n",
      "Epoch 181/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.4329 - acc: 0.8356 - val_loss: 0.2981 - val_acc: 0.8735\n",
      "\n",
      "Epoch 00181: val_acc did not improve from 0.87549\n",
      "Epoch 182/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.3807 - acc: 0.8520 - val_loss: 0.2940 - val_acc: 0.8716\n",
      "\n",
      "Epoch 00182: val_acc did not improve from 0.87549\n",
      "Epoch 183/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.3361 - acc: 0.8641 - val_loss: 0.2750 - val_acc: 0.8716\n",
      "\n",
      "Epoch 00183: val_acc did not improve from 0.87549\n",
      "Epoch 184/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.3538 - acc: 0.8646 - val_loss: 0.3388 - val_acc: 0.8677\n",
      "\n",
      "Epoch 00184: val_acc did not improve from 0.87549\n",
      "Epoch 185/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.3020 - acc: 0.8777 - val_loss: 0.2599 - val_acc: 0.8881\n",
      "\n",
      "Epoch 00185: val_acc improved from 0.87549 to 0.88813, saving model to models\\gesture.h5\n",
      "Epoch 186/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.2781 - acc: 0.8871 - val_loss: 0.2385 - val_acc: 0.9047\n",
      "\n",
      "Epoch 00186: val_acc improved from 0.88813 to 0.90467, saving model to models\\gesture.h5\n",
      "Epoch 187/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2850 - acc: 0.8879 - val_loss: 0.3301 - val_acc: 0.8638\n",
      "\n",
      "Epoch 00187: val_acc did not improve from 0.90467\n",
      "Epoch 188/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.2636 - acc: 0.9029 - val_loss: 0.1967 - val_acc: 0.9270\n",
      "\n",
      "Epoch 00188: val_acc improved from 0.90467 to 0.92704, saving model to models\\gesture.h5\n",
      "Epoch 189/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.2232 - acc: 0.9171 - val_loss: 0.2067 - val_acc: 0.9222\n",
      "\n",
      "Epoch 00189: val_acc did not improve from 0.92704\n",
      "Epoch 190/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.2459 - acc: 0.9099 - val_loss: 0.2699 - val_acc: 0.9037\n",
      "\n",
      "Epoch 00190: val_acc did not improve from 0.92704\n",
      "Epoch 191/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.2153 - acc: 0.9221 - val_loss: 0.4111 - val_acc: 0.8035\n",
      "\n",
      "Epoch 00191: val_acc did not improve from 0.92704\n",
      "Epoch 192/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.1918 - acc: 0.9291 - val_loss: 0.1869 - val_acc: 0.9348\n",
      "\n",
      "Epoch 00192: val_acc improved from 0.92704 to 0.93482, saving model to models\\gesture.h5\n",
      "Epoch 193/1000\n",
      "236/236 [==============================] - 6s 23ms/step - loss: 0.2353 - acc: 0.9147 - val_loss: 0.2502 - val_acc: 0.9125\n",
      "\n",
      "Epoch 00193: val_acc did not improve from 0.93482\n",
      "Epoch 194/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 0.1884 - acc: 0.9310 - val_loss: 0.3622 - val_acc: 0.8940\n",
      "\n",
      "Epoch 00194: val_acc did not improve from 0.93482\n",
      "Epoch 195/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.2046 - acc: 0.9302 - val_loss: 0.2370 - val_acc: 0.9251\n",
      "\n",
      "Epoch 00195: val_acc did not improve from 0.93482\n",
      "Epoch 196/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.1823 - acc: 0.9364 - val_loss: 0.2699 - val_acc: 0.8784\n",
      "\n",
      "Epoch 00196: val_acc did not improve from 0.93482\n",
      "Epoch 197/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.1571 - acc: 0.9443 - val_loss: 0.1291 - val_acc: 0.9504\n",
      "\n",
      "Epoch 00197: val_acc improved from 0.93482 to 0.95039, saving model to models\\gesture.h5\n",
      "Epoch 198/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.1279 - acc: 0.9536 - val_loss: 0.1319 - val_acc: 0.9553\n",
      "\n",
      "Epoch 00198: val_acc improved from 0.95039 to 0.95525, saving model to models\\gesture.h5\n",
      "Epoch 199/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.1161 - acc: 0.9563 - val_loss: 0.1244 - val_acc: 0.9523\n",
      "\n",
      "Epoch 00199: val_acc did not improve from 0.95525\n",
      "Epoch 200/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.1066 - acc: 0.9607 - val_loss: 0.1309 - val_acc: 0.9543\n",
      "\n",
      "Epoch 00200: val_acc did not improve from 0.95525\n",
      "Epoch 201/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0998 - acc: 0.9654 - val_loss: 0.0947 - val_acc: 0.9601\n",
      "\n",
      "Epoch 00201: val_acc improved from 0.95525 to 0.96012, saving model to models\\gesture.h5\n",
      "Epoch 202/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.1042 - acc: 0.9583 - val_loss: 0.1387 - val_acc: 0.9436\n",
      "\n",
      "Epoch 00202: val_acc did not improve from 0.96012\n",
      "Epoch 203/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0853 - acc: 0.9696 - val_loss: 0.0910 - val_acc: 0.9621\n",
      "\n",
      "Epoch 00203: val_acc improved from 0.96012 to 0.96206, saving model to models\\gesture.h5\n",
      "Epoch 204/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.1001 - acc: 0.9627 - val_loss: 0.1026 - val_acc: 0.9591\n",
      "\n",
      "Epoch 00204: val_acc did not improve from 0.96206\n",
      "Epoch 205/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0734 - acc: 0.9721 - val_loss: 0.0789 - val_acc: 0.9698\n",
      "\n",
      "Epoch 00205: val_acc improved from 0.96206 to 0.96984, saving model to models\\gesture.h5\n",
      "Epoch 206/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0775 - acc: 0.9752 - val_loss: 0.0758 - val_acc: 0.9747\n",
      "\n",
      "Epoch 00206: val_acc improved from 0.96984 to 0.97471, saving model to models\\gesture.h5\n",
      "Epoch 207/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.0635 - acc: 0.9780 - val_loss: 0.0734 - val_acc: 0.9737\n",
      "\n",
      "Epoch 00207: val_acc did not improve from 0.97471\n",
      "Epoch 208/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0645 - acc: 0.9768 - val_loss: 0.0464 - val_acc: 0.9835\n",
      "\n",
      "Epoch 00208: val_acc improved from 0.97471 to 0.98346, saving model to models\\gesture.h5\n",
      "Epoch 209/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0664 - acc: 0.9760 - val_loss: 0.0722 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00209: val_acc did not improve from 0.98346\n",
      "Epoch 210/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0528 - acc: 0.9792 - val_loss: 0.0416 - val_acc: 0.9835\n",
      "\n",
      "Epoch 00210: val_acc did not improve from 0.98346\n",
      "Epoch 211/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0406 - acc: 0.9862 - val_loss: 0.2170 - val_acc: 0.9484\n",
      "\n",
      "Epoch 00211: val_acc did not improve from 0.98346\n",
      "Epoch 212/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0840 - acc: 0.9700 - val_loss: 0.0408 - val_acc: 0.9815\n",
      "\n",
      "Epoch 00212: val_acc did not improve from 0.98346\n",
      "Epoch 213/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0421 - acc: 0.9846 - val_loss: 0.0854 - val_acc: 0.9708\n",
      "\n",
      "Epoch 00213: val_acc did not improve from 0.98346\n",
      "Epoch 214/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0446 - acc: 0.9829 - val_loss: 0.0429 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00214: val_acc improved from 0.98346 to 0.98444, saving model to models\\gesture.h5\n",
      "Epoch 215/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0563 - acc: 0.9790 - val_loss: 0.0380 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00215: val_acc improved from 0.98444 to 0.98638, saving model to models\\gesture.h5\n",
      "Epoch 216/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0329 - acc: 0.9878 - val_loss: 0.0740 - val_acc: 0.9689\n",
      "\n",
      "Epoch 00216: val_acc did not improve from 0.98638\n",
      "Epoch 217/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0416 - acc: 0.9839 - val_loss: 0.0431 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00217: val_acc did not improve from 0.98638\n",
      "Epoch 218/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0421 - acc: 0.9837 - val_loss: 0.0949 - val_acc: 0.9650\n",
      "\n",
      "Epoch 00218: val_acc did not improve from 0.98638\n",
      "Epoch 219/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0320 - acc: 0.9898 - val_loss: 0.0360 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00219: val_acc improved from 0.98638 to 0.99027, saving model to models\\gesture.h5\n",
      "Epoch 220/1000\n",
      "236/236 [==============================] - 9s 36ms/step - loss: 0.0171 - acc: 0.9943 - val_loss: 0.2198 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00220: val_acc did not improve from 0.99027\n",
      "Epoch 221/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 0.2028 - acc: 0.9561 - val_loss: 0.0610 - val_acc: 0.9767\n",
      "\n",
      "Epoch 00221: val_acc did not improve from 0.99027\n",
      "Epoch 222/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0519 - acc: 0.9813 - val_loss: 0.0447 - val_acc: 0.9835\n",
      "\n",
      "Epoch 00222: val_acc did not improve from 0.99027\n",
      "Epoch 223/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0405 - acc: 0.9846 - val_loss: 0.0643 - val_acc: 0.9737\n",
      "\n",
      "Epoch 00223: val_acc did not improve from 0.99027\n",
      "Epoch 224/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0378 - acc: 0.9857 - val_loss: 0.0700 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00224: val_acc did not improve from 0.99027\n",
      "Epoch 225/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.0625 - acc: 0.9804 - val_loss: 0.0445 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00225: val_acc did not improve from 0.99027\n",
      "Epoch 226/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0235 - acc: 0.9907 - val_loss: 0.0392 - val_acc: 0.9912\n",
      "\n",
      "Epoch 00226: val_acc improved from 0.99027 to 0.99125, saving model to models\\gesture.h5\n",
      "Epoch 227/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0379 - acc: 0.9863 - val_loss: 0.0333 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00227: val_acc did not improve from 0.99125\n",
      "Epoch 228/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0144 - acc: 0.9956 - val_loss: 0.0357 - val_acc: 0.9912\n",
      "\n",
      "Epoch 00228: val_acc did not improve from 0.99125\n",
      "Epoch 229/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.0149 - acc: 0.9958 - val_loss: 0.0261 - val_acc: 0.9912\n",
      "\n",
      "Epoch 00229: val_acc did not improve from 0.99125\n",
      "Epoch 230/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0131 - acc: 0.9956 - val_loss: 0.0292 - val_acc: 0.9922\n",
      "\n",
      "Epoch 00230: val_acc improved from 0.99125 to 0.99222, saving model to models\\gesture.h5\n",
      "Epoch 231/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0182 - acc: 0.9940 - val_loss: 0.0253 - val_acc: 0.9951\n",
      "\n",
      "Epoch 00231: val_acc improved from 0.99222 to 0.99514, saving model to models\\gesture.h5\n",
      "Epoch 232/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0108 - acc: 0.9962 - val_loss: 0.0323 - val_acc: 0.9912\n",
      "\n",
      "Epoch 00232: val_acc did not improve from 0.99514\n",
      "Epoch 233/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0092 - acc: 0.9971 - val_loss: 0.0330 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00233: val_acc did not improve from 0.99514\n",
      "Epoch 234/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0361 - acc: 0.9882 - val_loss: 0.0233 - val_acc: 0.9942\n",
      "\n",
      "Epoch 00234: val_acc did not improve from 0.99514\n",
      "Epoch 235/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0096 - acc: 0.9968 - val_loss: 0.0410 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00235: val_acc did not improve from 0.99514\n",
      "Epoch 236/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 27.7371 - acc: 0.7131 - val_loss: 8.8457 - val_acc: 0.5700\n",
      "\n",
      "Epoch 00236: val_acc did not improve from 0.99514\n",
      "Epoch 237/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 8.9783 - acc: 0.6505 - val_loss: 0.9389 - val_acc: 0.7121\n",
      "\n",
      "Epoch 00237: val_acc did not improve from 0.99514\n",
      "Epoch 238/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.8409 - acc: 0.7373 - val_loss: 0.6830 - val_acc: 0.7558\n",
      "\n",
      "Epoch 00238: val_acc did not improve from 0.99514\n",
      "Epoch 239/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.7159 - acc: 0.7515 - val_loss: 0.6360 - val_acc: 0.7724\n",
      "\n",
      "Epoch 00239: val_acc did not improve from 0.99514\n",
      "Epoch 240/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.6765 - acc: 0.7622 - val_loss: 0.5830 - val_acc: 0.7597\n",
      "\n",
      "Epoch 00240: val_acc did not improve from 0.99514\n",
      "Epoch 241/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.1867 - acc: 0.7166 - val_loss: 1.5219 - val_acc: 0.7267\n",
      "\n",
      "Epoch 00241: val_acc did not improve from 0.99514\n",
      "Epoch 242/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.4748 - acc: 0.6994 - val_loss: 1.4015 - val_acc: 0.6673\n",
      "\n",
      "Epoch 00242: val_acc did not improve from 0.99514\n",
      "Epoch 243/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 1.3573 - acc: 0.7024 - val_loss: 1.2566 - val_acc: 0.7053\n",
      "\n",
      "Epoch 00243: val_acc did not improve from 0.99514\n",
      "Epoch 244/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 1.2517 - acc: 0.7008 - val_loss: 1.2256 - val_acc: 0.6556\n",
      "\n",
      "Epoch 00244: val_acc did not improve from 0.99514\n",
      "Epoch 245/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 1.5796 - acc: 0.7102 - val_loss: 1.1001 - val_acc: 0.6965\n",
      "\n",
      "Epoch 00245: val_acc did not improve from 0.99514\n",
      "Epoch 246/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 6.6037 - acc: 0.6482 - val_loss: 0.8334 - val_acc: 0.7150\n",
      "\n",
      "Epoch 00246: val_acc did not improve from 0.99514\n",
      "Epoch 247/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.6901 - acc: 0.7111 - val_loss: 0.7624 - val_acc: 0.7695\n",
      "\n",
      "Epoch 00247: val_acc did not improve from 0.99514\n",
      "Epoch 248/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.6710 - acc: 0.7091 - val_loss: 0.6328 - val_acc: 0.7101\n",
      "\n",
      "Epoch 00248: val_acc did not improve from 0.99514\n",
      "Epoch 249/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.5646 - acc: 0.7277 - val_loss: 0.5526 - val_acc: 0.7237\n",
      "\n",
      "Epoch 00249: val_acc did not improve from 0.99514\n",
      "Epoch 250/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.5217 - acc: 0.7415 - val_loss: 0.4938 - val_acc: 0.7354\n",
      "\n",
      "Epoch 00250: val_acc did not improve from 0.99514\n",
      "Epoch 251/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.4692 - acc: 0.7680 - val_loss: 0.4878 - val_acc: 0.7938\n",
      "\n",
      "Epoch 00251: val_acc did not improve from 0.99514\n",
      "Epoch 252/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.4716 - acc: 0.7687 - val_loss: 0.5071 - val_acc: 0.7082\n",
      "\n",
      "Epoch 00252: val_acc did not improve from 0.99514\n",
      "Epoch 253/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 6s 24ms/step - loss: 0.4503 - acc: 0.7760 - val_loss: 0.4253 - val_acc: 0.7685\n",
      "\n",
      "Epoch 00253: val_acc did not improve from 0.99514\n",
      "Epoch 254/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.4462 - acc: 0.7761 - val_loss: 0.4761 - val_acc: 0.7782\n",
      "\n",
      "Epoch 00254: val_acc did not improve from 0.99514\n",
      "Epoch 255/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 0.6135 - acc: 0.7245 - val_loss: 0.6551 - val_acc: 0.6654\n",
      "\n",
      "Epoch 00255: val_acc did not improve from 0.99514\n",
      "Epoch 256/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.5959 - acc: 0.7197 - val_loss: 0.5275 - val_acc: 0.7111\n",
      "\n",
      "Epoch 00256: val_acc did not improve from 0.99514\n",
      "Epoch 257/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.5213 - acc: 0.7326 - val_loss: 0.4915 - val_acc: 0.7276\n",
      "\n",
      "Epoch 00257: val_acc did not improve from 0.99514\n",
      "Epoch 258/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.5151 - acc: 0.7304 - val_loss: 0.4761 - val_acc: 0.7500\n",
      "\n",
      "Epoch 00258: val_acc did not improve from 0.99514\n",
      "Epoch 259/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.5370 - acc: 0.7131 - val_loss: 0.5101 - val_acc: 0.7442\n",
      "\n",
      "Epoch 00259: val_acc did not improve from 0.99514\n",
      "Epoch 260/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.4931 - acc: 0.7341 - val_loss: 0.4592 - val_acc: 0.7529\n",
      "\n",
      "Epoch 00260: val_acc did not improve from 0.99514\n",
      "Epoch 261/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.4982 - acc: 0.7333 - val_loss: 0.4712 - val_acc: 0.7636\n",
      "\n",
      "Epoch 00261: val_acc did not improve from 0.99514\n",
      "Epoch 262/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.5029 - acc: 0.7371 - val_loss: 0.5959 - val_acc: 0.6342\n",
      "\n",
      "Epoch 00262: val_acc did not improve from 0.99514\n",
      "Epoch 263/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.4868 - acc: 0.7428 - val_loss: 0.4415 - val_acc: 0.7831\n",
      "\n",
      "Epoch 00263: val_acc did not improve from 0.99514\n",
      "Epoch 264/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.4961 - acc: 0.7487 - val_loss: 0.4556 - val_acc: 0.7831\n",
      "\n",
      "Epoch 00264: val_acc did not improve from 0.99514\n",
      "Epoch 265/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 0.4680 - acc: 0.7536 - val_loss: 0.4736 - val_acc: 0.7072\n",
      "\n",
      "Epoch 00265: val_acc did not improve from 0.99514\n",
      "Epoch 266/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 0.4492 - acc: 0.7655 - val_loss: 0.4121 - val_acc: 0.8035\n",
      "\n",
      "Epoch 00266: val_acc did not improve from 0.99514\n",
      "Epoch 267/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.4334 - acc: 0.7781 - val_loss: 0.4499 - val_acc: 0.7558\n",
      "\n",
      "Epoch 00267: val_acc did not improve from 0.99514\n",
      "Epoch 268/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.4316 - acc: 0.7771 - val_loss: 0.3988 - val_acc: 0.8025\n",
      "\n",
      "Epoch 00268: val_acc did not improve from 0.99514\n",
      "Epoch 269/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.4195 - acc: 0.7873 - val_loss: 0.4010 - val_acc: 0.8142\n",
      "\n",
      "Epoch 00269: val_acc did not improve from 0.99514\n",
      "Epoch 270/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.4146 - acc: 0.7915 - val_loss: 0.3977 - val_acc: 0.7977\n",
      "\n",
      "Epoch 00270: val_acc did not improve from 0.99514\n",
      "Epoch 271/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.3756 - acc: 0.8247 - val_loss: 0.4160 - val_acc: 0.7733\n",
      "\n",
      "Epoch 00271: val_acc did not improve from 0.99514\n",
      "Epoch 272/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.4384 - acc: 0.7706 - val_loss: 0.4113 - val_acc: 0.7967\n",
      "\n",
      "Epoch 00272: val_acc did not improve from 0.99514\n",
      "Epoch 273/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.4330 - acc: 0.7796 - val_loss: 0.4354 - val_acc: 0.7840\n",
      "\n",
      "Epoch 00273: val_acc did not improve from 0.99514\n",
      "Epoch 274/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.4018 - acc: 0.7998 - val_loss: 0.3505 - val_acc: 0.8288\n",
      "\n",
      "Epoch 00274: val_acc did not improve from 0.99514\n",
      "Epoch 275/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.3691 - acc: 0.8205 - val_loss: 0.3615 - val_acc: 0.8288\n",
      "\n",
      "Epoch 00275: val_acc did not improve from 0.99514\n",
      "Epoch 276/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.3609 - acc: 0.8279 - val_loss: 0.3566 - val_acc: 0.8200\n",
      "\n",
      "Epoch 00276: val_acc did not improve from 0.99514\n",
      "Epoch 277/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.3398 - acc: 0.8451 - val_loss: 0.3528 - val_acc: 0.8366\n",
      "\n",
      "Epoch 00277: val_acc did not improve from 0.99514\n",
      "Epoch 278/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.3263 - acc: 0.8506 - val_loss: 0.3330 - val_acc: 0.8492\n",
      "\n",
      "Epoch 00278: val_acc did not improve from 0.99514\n",
      "Epoch 279/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.3209 - acc: 0.8575 - val_loss: 0.4129 - val_acc: 0.8317\n",
      "\n",
      "Epoch 00279: val_acc did not improve from 0.99514\n",
      "Epoch 280/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.3095 - acc: 0.8661 - val_loss: 0.2866 - val_acc: 0.8784\n",
      "\n",
      "Epoch 00280: val_acc did not improve from 0.99514\n",
      "Epoch 281/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.2701 - acc: 0.8850 - val_loss: 0.2597 - val_acc: 0.8979\n",
      "\n",
      "Epoch 00281: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00281: ReduceLROnPlateau reducing learning rate to 0.0005000000237487257.\n",
      "Epoch 282/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2168 - acc: 0.9112 - val_loss: 0.2267 - val_acc: 0.9173\n",
      "\n",
      "Epoch 00282: val_acc did not improve from 0.99514\n",
      "Epoch 283/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.1970 - acc: 0.9202 - val_loss: 0.2142 - val_acc: 0.9154\n",
      "\n",
      "Epoch 00283: val_acc did not improve from 0.99514\n",
      "Epoch 284/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.1750 - acc: 0.9305 - val_loss: 0.2198 - val_acc: 0.9193\n",
      "\n",
      "Epoch 00284: val_acc did not improve from 0.99514\n",
      "Epoch 285/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.1684 - acc: 0.9350 - val_loss: 0.1886 - val_acc: 0.9290\n",
      "\n",
      "Epoch 00285: val_acc did not improve from 0.99514\n",
      "Epoch 286/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.1525 - acc: 0.9421 - val_loss: 0.1708 - val_acc: 0.9358\n",
      "\n",
      "Epoch 00286: val_acc did not improve from 0.99514\n",
      "Epoch 287/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.1455 - acc: 0.9452 - val_loss: 0.1483 - val_acc: 0.9465\n",
      "\n",
      "Epoch 00287: val_acc did not improve from 0.99514\n",
      "Epoch 288/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.1446 - acc: 0.9479 - val_loss: 0.1567 - val_acc: 0.9407\n",
      "\n",
      "Epoch 00288: val_acc did not improve from 0.99514\n",
      "Epoch 289/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.1726 - acc: 0.9407 - val_loss: 0.1508 - val_acc: 0.9377\n",
      "\n",
      "Epoch 00289: val_acc did not improve from 0.99514\n",
      "Epoch 290/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.9191 - acc: 0.9068 - val_loss: 4.4184 - val_acc: 0.4951\n",
      "\n",
      "Epoch 00290: val_acc did not improve from 0.99514\n",
      "Epoch 291/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 1.8750 - acc: 0.7261 - val_loss: 0.9928 - val_acc: 0.7792\n",
      "\n",
      "Epoch 00291: val_acc did not improve from 0.99514\n",
      "Epoch 292/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.8400 - acc: 0.7858 - val_loss: 0.8410 - val_acc: 0.8191\n",
      "\n",
      "Epoch 00292: val_acc did not improve from 0.99514\n",
      "Epoch 293/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.5511 - acc: 0.8239 - val_loss: 0.5173 - val_acc: 0.8161\n",
      "\n",
      "Epoch 00293: val_acc did not improve from 0.99514\n",
      "Epoch 294/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.4536 - acc: 0.8295 - val_loss: 0.4342 - val_acc: 0.8084\n",
      "\n",
      "Epoch 00294: val_acc did not improve from 0.99514\n",
      "Epoch 295/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.4002 - acc: 0.8364 - val_loss: 0.4211 - val_acc: 0.8414\n",
      "\n",
      "Epoch 00295: val_acc did not improve from 0.99514\n",
      "Epoch 296/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 31ms/step - loss: 3.1524 - acc: 0.8373 - val_loss: 0.3730 - val_acc: 0.8531\n",
      "\n",
      "Epoch 00296: val_acc did not improve from 0.99514\n",
      "Epoch 297/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.3591 - acc: 0.8544 - val_loss: 0.3581 - val_acc: 0.8521\n",
      "\n",
      "Epoch 00297: val_acc did not improve from 0.99514\n",
      "Epoch 298/1000\n",
      "236/236 [==============================] - 10s 40ms/step - loss: 0.3391 - acc: 0.8609 - val_loss: 0.3513 - val_acc: 0.8512\n",
      "\n",
      "Epoch 00298: val_acc did not improve from 0.99514\n",
      "Epoch 299/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 0.3231 - acc: 0.8665 - val_loss: 0.3041 - val_acc: 0.8638\n",
      "\n",
      "Epoch 00299: val_acc did not improve from 0.99514\n",
      "Epoch 300/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.3087 - acc: 0.8746 - val_loss: 0.2903 - val_acc: 0.8755\n",
      "\n",
      "Epoch 00300: val_acc did not improve from 0.99514\n",
      "Epoch 301/1000\n",
      "236/236 [==============================] - 11s 45ms/step - loss: 0.3102 - acc: 0.8731 - val_loss: 0.2757 - val_acc: 0.8774\n",
      "\n",
      "Epoch 00301: val_acc did not improve from 0.99514\n",
      "Epoch 302/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2889 - acc: 0.8787 - val_loss: 0.2730 - val_acc: 0.8872\n",
      "\n",
      "Epoch 00302: val_acc did not improve from 0.99514\n",
      "Epoch 303/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.2800 - acc: 0.8840 - val_loss: 0.2758 - val_acc: 0.8833\n",
      "\n",
      "Epoch 00303: val_acc did not improve from 0.99514\n",
      "Epoch 304/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.2836 - acc: 0.8850 - val_loss: 0.2546 - val_acc: 0.8969\n",
      "\n",
      "Epoch 00304: val_acc did not improve from 0.99514\n",
      "Epoch 305/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2911 - acc: 0.8869 - val_loss: 0.2605 - val_acc: 0.8979\n",
      "\n",
      "Epoch 00305: val_acc did not improve from 0.99514\n",
      "Epoch 306/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.3071 - acc: 0.8847 - val_loss: 0.2809 - val_acc: 0.8823\n",
      "\n",
      "Epoch 00306: val_acc did not improve from 0.99514\n",
      "Epoch 307/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.2614 - acc: 0.8960 - val_loss: 0.2425 - val_acc: 0.8949\n",
      "\n",
      "Epoch 00307: val_acc did not improve from 0.99514\n",
      "Epoch 308/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2463 - acc: 0.9002 - val_loss: 0.2304 - val_acc: 0.9056\n",
      "\n",
      "Epoch 00308: val_acc did not improve from 0.99514\n",
      "Epoch 309/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2387 - acc: 0.9057 - val_loss: 0.2274 - val_acc: 0.9173\n",
      "\n",
      "Epoch 00309: val_acc did not improve from 0.99514\n",
      "Epoch 310/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.2305 - acc: 0.9072 - val_loss: 0.2186 - val_acc: 0.9134\n",
      "\n",
      "Epoch 00310: val_acc did not improve from 0.99514\n",
      "Epoch 311/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2253 - acc: 0.9087 - val_loss: 0.2081 - val_acc: 0.9202\n",
      "\n",
      "Epoch 00311: val_acc did not improve from 0.99514\n",
      "Epoch 312/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2252 - acc: 0.9091 - val_loss: 0.2063 - val_acc: 0.9125\n",
      "\n",
      "Epoch 00312: val_acc did not improve from 0.99514\n",
      "Epoch 313/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.2148 - acc: 0.9172 - val_loss: 0.2057 - val_acc: 0.9134\n",
      "\n",
      "Epoch 00313: val_acc did not improve from 0.99514\n",
      "Epoch 314/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.2058 - acc: 0.9161 - val_loss: 0.1939 - val_acc: 0.9163\n",
      "\n",
      "Epoch 00314: val_acc did not improve from 0.99514\n",
      "Epoch 315/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.2004 - acc: 0.9200 - val_loss: 0.1837 - val_acc: 0.9290\n",
      "\n",
      "Epoch 00315: val_acc did not improve from 0.99514\n",
      "Epoch 316/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2024 - acc: 0.9206 - val_loss: 0.1823 - val_acc: 0.9261\n",
      "\n",
      "Epoch 00316: val_acc did not improve from 0.99514\n",
      "Epoch 317/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1869 - acc: 0.9262 - val_loss: 0.1778 - val_acc: 0.9290\n",
      "\n",
      "Epoch 00317: val_acc did not improve from 0.99514\n",
      "Epoch 318/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.1796 - acc: 0.9285 - val_loss: 0.1736 - val_acc: 0.9348\n",
      "\n",
      "Epoch 00318: val_acc did not improve from 0.99514\n",
      "Epoch 319/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.1719 - acc: 0.9335 - val_loss: 0.1774 - val_acc: 0.9290\n",
      "\n",
      "Epoch 00319: val_acc did not improve from 0.99514\n",
      "Epoch 320/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.1642 - acc: 0.9346 - val_loss: 0.1522 - val_acc: 0.9387\n",
      "\n",
      "Epoch 00320: val_acc did not improve from 0.99514\n",
      "Epoch 321/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1599 - acc: 0.9396 - val_loss: 0.1462 - val_acc: 0.9426\n",
      "\n",
      "Epoch 00321: val_acc did not improve from 0.99514\n",
      "Epoch 322/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1716 - acc: 0.9391 - val_loss: 0.1697 - val_acc: 0.9280\n",
      "\n",
      "Epoch 00322: val_acc did not improve from 0.99514\n",
      "Epoch 323/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.1633 - acc: 0.9351 - val_loss: 0.1354 - val_acc: 0.9446\n",
      "\n",
      "Epoch 00323: val_acc did not improve from 0.99514\n",
      "Epoch 324/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1534 - acc: 0.9412 - val_loss: 0.1504 - val_acc: 0.9387\n",
      "\n",
      "Epoch 00324: val_acc did not improve from 0.99514\n",
      "Epoch 325/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.5295 - acc: 0.8617 - val_loss: 0.5723 - val_acc: 0.8259\n",
      "\n",
      "Epoch 00325: val_acc did not improve from 0.99514\n",
      "Epoch 326/1000\n",
      "236/236 [==============================] - 9s 37ms/step - loss: 0.4768 - acc: 0.8311 - val_loss: 0.3883 - val_acc: 0.8434\n",
      "\n",
      "Epoch 00326: val_acc did not improve from 0.99514\n",
      "Epoch 327/1000\n",
      "236/236 [==============================] - 11s 47ms/step - loss: 0.3707 - acc: 0.8645 - val_loss: 0.3460 - val_acc: 0.8813\n",
      "\n",
      "Epoch 00327: val_acc did not improve from 0.99514\n",
      "Epoch 328/1000\n",
      "236/236 [==============================] - 11s 45ms/step - loss: 0.3257 - acc: 0.8765 - val_loss: 0.2738 - val_acc: 0.8901\n",
      "\n",
      "Epoch 00328: val_acc did not improve from 0.99514\n",
      "Epoch 329/1000\n",
      "236/236 [==============================] - 10s 41ms/step - loss: 0.2701 - acc: 0.8949 - val_loss: 0.2416 - val_acc: 0.9086\n",
      "\n",
      "Epoch 00329: val_acc did not improve from 0.99514\n",
      "Epoch 330/1000\n",
      "236/236 [==============================] - 10s 42ms/step - loss: 0.2366 - acc: 0.9112 - val_loss: 0.2203 - val_acc: 0.9202\n",
      "\n",
      "Epoch 00330: val_acc did not improve from 0.99514\n",
      "Epoch 331/1000\n",
      "236/236 [==============================] - 10s 42ms/step - loss: 0.2211 - acc: 0.9152 - val_loss: 0.2033 - val_acc: 0.9348\n",
      "\n",
      "Epoch 00331: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00331: ReduceLROnPlateau reducing learning rate to 0.0002500000118743628.\n",
      "Epoch 332/1000\n",
      "236/236 [==============================] - 9s 40ms/step - loss: 0.1988 - acc: 0.9245 - val_loss: 0.2104 - val_acc: 0.9270\n",
      "\n",
      "Epoch 00332: val_acc did not improve from 0.99514\n",
      "Epoch 333/1000\n",
      "236/236 [==============================] - 10s 41ms/step - loss: 0.1895 - acc: 0.9270 - val_loss: 0.2153 - val_acc: 0.9193\n",
      "\n",
      "Epoch 00333: val_acc did not improve from 0.99514\n",
      "Epoch 334/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.1815 - acc: 0.9302 - val_loss: 0.1970 - val_acc: 0.9261\n",
      "\n",
      "Epoch 00334: val_acc did not improve from 0.99514\n",
      "Epoch 335/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.1806 - acc: 0.9317 - val_loss: 0.2185 - val_acc: 0.9222\n",
      "\n",
      "Epoch 00335: val_acc did not improve from 0.99514\n",
      "Epoch 336/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1836 - acc: 0.9297 - val_loss: 0.1952 - val_acc: 0.9280\n",
      "\n",
      "Epoch 00336: val_acc did not improve from 0.99514\n",
      "Epoch 337/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.1821 - acc: 0.9285 - val_loss: 0.1911 - val_acc: 0.9290\n",
      "\n",
      "Epoch 00337: val_acc did not improve from 0.99514\n",
      "Epoch 338/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.1681 - acc: 0.9372 - val_loss: 0.1836 - val_acc: 0.9368\n",
      "\n",
      "Epoch 00338: val_acc did not improve from 0.99514\n",
      "Epoch 339/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 32ms/step - loss: 0.1623 - acc: 0.9400 - val_loss: 0.1907 - val_acc: 0.9368\n",
      "\n",
      "Epoch 00339: val_acc did not improve from 0.99514\n",
      "Epoch 340/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.1598 - acc: 0.9384 - val_loss: 0.1659 - val_acc: 0.9407\n",
      "\n",
      "Epoch 00340: val_acc did not improve from 0.99514\n",
      "Epoch 341/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.1528 - acc: 0.9404 - val_loss: 0.1608 - val_acc: 0.9455\n",
      "\n",
      "Epoch 00341: val_acc did not improve from 0.99514\n",
      "Epoch 342/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.1398 - acc: 0.9455 - val_loss: 0.1595 - val_acc: 0.9446\n",
      "\n",
      "Epoch 00342: val_acc did not improve from 0.99514\n",
      "Epoch 343/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1329 - acc: 0.9473 - val_loss: 0.1593 - val_acc: 0.9407\n",
      "\n",
      "Epoch 00343: val_acc did not improve from 0.99514\n",
      "Epoch 344/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1273 - acc: 0.9510 - val_loss: 0.1475 - val_acc: 0.9465\n",
      "\n",
      "Epoch 00344: val_acc did not improve from 0.99514\n",
      "Epoch 345/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1277 - acc: 0.9516 - val_loss: 0.1531 - val_acc: 0.9426\n",
      "\n",
      "Epoch 00345: val_acc did not improve from 0.99514\n",
      "Epoch 346/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1221 - acc: 0.9513 - val_loss: 0.1301 - val_acc: 0.9494\n",
      "\n",
      "Epoch 00346: val_acc did not improve from 0.99514\n",
      "Epoch 347/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1147 - acc: 0.9567 - val_loss: 0.1576 - val_acc: 0.9387\n",
      "\n",
      "Epoch 00347: val_acc did not improve from 0.99514\n",
      "Epoch 348/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.1077 - acc: 0.9589 - val_loss: 0.1189 - val_acc: 0.9562\n",
      "\n",
      "Epoch 00348: val_acc did not improve from 0.99514\n",
      "Epoch 349/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.1091 - acc: 0.9583 - val_loss: 0.1233 - val_acc: 0.9543\n",
      "\n",
      "Epoch 00349: val_acc did not improve from 0.99514\n",
      "Epoch 350/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 0.1051 - acc: 0.9591 - val_loss: 0.1367 - val_acc: 0.9543\n",
      "\n",
      "Epoch 00350: val_acc did not improve from 0.99514\n",
      "Epoch 351/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 0.1041 - acc: 0.9605 - val_loss: 0.1215 - val_acc: 0.9582\n",
      "\n",
      "Epoch 00351: val_acc did not improve from 0.99514\n",
      "Epoch 352/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0909 - acc: 0.9664 - val_loss: 0.1547 - val_acc: 0.9446\n",
      "\n",
      "Epoch 00352: val_acc did not improve from 0.99514\n",
      "Epoch 353/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 0.0980 - acc: 0.9614 - val_loss: 0.1171 - val_acc: 0.9621\n",
      "\n",
      "Epoch 00353: val_acc did not improve from 0.99514\n",
      "Epoch 354/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0820 - acc: 0.9662 - val_loss: 0.1616 - val_acc: 0.9504\n",
      "\n",
      "Epoch 00354: val_acc did not improve from 0.99514\n",
      "Epoch 355/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0868 - acc: 0.9689 - val_loss: 0.1045 - val_acc: 0.9660\n",
      "\n",
      "Epoch 00355: val_acc did not improve from 0.99514\n",
      "Epoch 356/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0791 - acc: 0.9679 - val_loss: 0.0984 - val_acc: 0.9630\n",
      "\n",
      "Epoch 00356: val_acc did not improve from 0.99514\n",
      "Epoch 357/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0735 - acc: 0.9744 - val_loss: 0.1182 - val_acc: 0.9572\n",
      "\n",
      "Epoch 00357: val_acc did not improve from 0.99514\n",
      "Epoch 358/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.0725 - acc: 0.9755 - val_loss: 0.0901 - val_acc: 0.9660\n",
      "\n",
      "Epoch 00358: val_acc did not improve from 0.99514\n",
      "Epoch 359/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.0784 - acc: 0.9719 - val_loss: 0.0868 - val_acc: 0.9708\n",
      "\n",
      "Epoch 00359: val_acc did not improve from 0.99514\n",
      "Epoch 360/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 0.0636 - acc: 0.9764 - val_loss: 0.0821 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00360: val_acc did not improve from 0.99514\n",
      "Epoch 361/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 374.7471 - acc: 0.9080 - val_loss: 0.3562 - val_acc: 0.8726\n",
      "\n",
      "Epoch 00361: val_acc did not improve from 0.99514\n",
      "Epoch 362/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.3195 - acc: 0.8843 - val_loss: 0.3047 - val_acc: 0.8911\n",
      "\n",
      "Epoch 00362: val_acc did not improve from 0.99514\n",
      "Epoch 363/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.2690 - acc: 0.8977 - val_loss: 0.3205 - val_acc: 0.8881\n",
      "\n",
      "Epoch 00363: val_acc did not improve from 0.99514\n",
      "Epoch 364/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.2237 - acc: 0.9144 - val_loss: 0.2355 - val_acc: 0.9125\n",
      "\n",
      "Epoch 00364: val_acc did not improve from 0.99514\n",
      "Epoch 365/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.2106 - acc: 0.9225 - val_loss: 0.2490 - val_acc: 0.9105\n",
      "\n",
      "Epoch 00365: val_acc did not improve from 0.99514\n",
      "Epoch 366/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.2052 - acc: 0.9218 - val_loss: 0.2183 - val_acc: 0.9232\n",
      "\n",
      "Epoch 00366: val_acc did not improve from 0.99514\n",
      "Epoch 367/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1991 - acc: 0.9245 - val_loss: 0.2304 - val_acc: 0.9212\n",
      "\n",
      "Epoch 00367: val_acc did not improve from 0.99514\n",
      "Epoch 368/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1976 - acc: 0.9257 - val_loss: 0.2410 - val_acc: 0.9183\n",
      "\n",
      "Epoch 00368: val_acc did not improve from 0.99514\n",
      "Epoch 369/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1902 - acc: 0.9310 - val_loss: 0.2064 - val_acc: 0.9241\n",
      "\n",
      "Epoch 00369: val_acc did not improve from 0.99514\n",
      "Epoch 370/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 0.1893 - acc: 0.9286 - val_loss: 0.2413 - val_acc: 0.9222\n",
      "\n",
      "Epoch 00370: val_acc did not improve from 0.99514\n",
      "Epoch 371/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1800 - acc: 0.9329 - val_loss: 0.2083 - val_acc: 0.9202\n",
      "\n",
      "Epoch 00371: val_acc did not improve from 0.99514\n",
      "Epoch 372/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1772 - acc: 0.9348 - val_loss: 0.2090 - val_acc: 0.9202\n",
      "\n",
      "Epoch 00372: val_acc did not improve from 0.99514\n",
      "Epoch 373/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1748 - acc: 0.9355 - val_loss: 0.1994 - val_acc: 0.9251\n",
      "\n",
      "Epoch 00373: val_acc did not improve from 0.99514\n",
      "Epoch 374/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1688 - acc: 0.9363 - val_loss: 0.2108 - val_acc: 0.9280\n",
      "\n",
      "Epoch 00374: val_acc did not improve from 0.99514\n",
      "Epoch 375/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1611 - acc: 0.9402 - val_loss: 0.2043 - val_acc: 0.9280\n",
      "\n",
      "Epoch 00375: val_acc did not improve from 0.99514\n",
      "Epoch 376/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1547 - acc: 0.9429 - val_loss: 0.1996 - val_acc: 0.9251\n",
      "\n",
      "Epoch 00376: val_acc did not improve from 0.99514\n",
      "Epoch 377/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1508 - acc: 0.9437 - val_loss: 0.1945 - val_acc: 0.9261\n",
      "\n",
      "Epoch 00377: val_acc did not improve from 0.99514\n",
      "Epoch 378/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1543 - acc: 0.9398 - val_loss: 0.2047 - val_acc: 0.9290\n",
      "\n",
      "Epoch 00378: val_acc did not improve from 0.99514\n",
      "Epoch 379/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1467 - acc: 0.9436 - val_loss: 0.2121 - val_acc: 0.9183\n",
      "\n",
      "Epoch 00379: val_acc did not improve from 0.99514\n",
      "Epoch 380/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 0.1496 - acc: 0.9455 - val_loss: 0.1817 - val_acc: 0.9339\n",
      "\n",
      "Epoch 00380: val_acc did not improve from 0.99514\n",
      "Epoch 381/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1426 - acc: 0.9477 - val_loss: 0.1797 - val_acc: 0.9358\n",
      "\n",
      "Epoch 00381: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00381: ReduceLROnPlateau reducing learning rate to 0.0001250000059371814.\n",
      "Epoch 382/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 29ms/step - loss: 0.1350 - acc: 0.9494 - val_loss: 0.1725 - val_acc: 0.9339\n",
      "\n",
      "Epoch 00382: val_acc did not improve from 0.99514\n",
      "Epoch 383/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1310 - acc: 0.9493 - val_loss: 0.1830 - val_acc: 0.9339\n",
      "\n",
      "Epoch 00383: val_acc did not improve from 0.99514\n",
      "Epoch 384/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.1257 - acc: 0.9512 - val_loss: 0.1669 - val_acc: 0.9339\n",
      "\n",
      "Epoch 00384: val_acc did not improve from 0.99514\n",
      "Epoch 385/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.1266 - acc: 0.9512 - val_loss: 0.1599 - val_acc: 0.9377\n",
      "\n",
      "Epoch 00385: val_acc did not improve from 0.99514\n",
      "Epoch 386/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.1242 - acc: 0.9504 - val_loss: 0.1830 - val_acc: 0.9329\n",
      "\n",
      "Epoch 00386: val_acc did not improve from 0.99514\n",
      "Epoch 387/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.1169 - acc: 0.9540 - val_loss: 0.1734 - val_acc: 0.9368\n",
      "\n",
      "Epoch 00387: val_acc did not improve from 0.99514\n",
      "Epoch 388/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.1159 - acc: 0.9574 - val_loss: 0.1526 - val_acc: 0.9397\n",
      "\n",
      "Epoch 00388: val_acc did not improve from 0.99514\n",
      "Epoch 389/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1082 - acc: 0.9569 - val_loss: 0.1507 - val_acc: 0.9446\n",
      "\n",
      "Epoch 00389: val_acc did not improve from 0.99514\n",
      "Epoch 390/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1034 - acc: 0.9601 - val_loss: 0.1421 - val_acc: 0.9436\n",
      "\n",
      "Epoch 00390: val_acc did not improve from 0.99514\n",
      "Epoch 391/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.1037 - acc: 0.9614 - val_loss: 0.1461 - val_acc: 0.9446\n",
      "\n",
      "Epoch 00391: val_acc did not improve from 0.99514\n",
      "Epoch 392/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 0.0997 - acc: 0.9606 - val_loss: 0.1367 - val_acc: 0.9426\n",
      "\n",
      "Epoch 00392: val_acc did not improve from 0.99514\n",
      "Epoch 393/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0992 - acc: 0.9602 - val_loss: 0.1362 - val_acc: 0.9494\n",
      "\n",
      "Epoch 00393: val_acc did not improve from 0.99514\n",
      "Epoch 394/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.1006 - acc: 0.9610 - val_loss: 0.1228 - val_acc: 0.9553\n",
      "\n",
      "Epoch 00394: val_acc did not improve from 0.99514\n",
      "Epoch 395/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0998 - acc: 0.9613 - val_loss: 0.1291 - val_acc: 0.9484\n",
      "\n",
      "Epoch 00395: val_acc did not improve from 0.99514\n",
      "Epoch 396/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0970 - acc: 0.9619 - val_loss: 0.1531 - val_acc: 0.9475\n",
      "\n",
      "Epoch 00396: val_acc did not improve from 0.99514\n",
      "Epoch 397/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0902 - acc: 0.9655 - val_loss: 0.1227 - val_acc: 0.9504\n",
      "\n",
      "Epoch 00397: val_acc did not improve from 0.99514\n",
      "Epoch 398/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0944 - acc: 0.9627 - val_loss: 0.1212 - val_acc: 0.9533\n",
      "\n",
      "Epoch 00398: val_acc did not improve from 0.99514\n",
      "Epoch 399/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0913 - acc: 0.9650 - val_loss: 0.1326 - val_acc: 0.9494\n",
      "\n",
      "Epoch 00399: val_acc did not improve from 0.99514\n",
      "Epoch 400/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0974 - acc: 0.9628 - val_loss: 0.1382 - val_acc: 0.9523\n",
      "\n",
      "Epoch 00400: val_acc did not improve from 0.99514\n",
      "Epoch 401/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0911 - acc: 0.9640 - val_loss: 0.1241 - val_acc: 0.9562\n",
      "\n",
      "Epoch 00401: val_acc did not improve from 0.99514\n",
      "Epoch 402/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0932 - acc: 0.9636 - val_loss: 0.1257 - val_acc: 0.9601\n",
      "\n",
      "Epoch 00402: val_acc did not improve from 0.99514\n",
      "Epoch 403/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0848 - acc: 0.9670 - val_loss: 0.1257 - val_acc: 0.9553\n",
      "\n",
      "Epoch 00403: val_acc did not improve from 0.99514\n",
      "Epoch 404/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0894 - acc: 0.9667 - val_loss: 0.1167 - val_acc: 0.9562\n",
      "\n",
      "Epoch 00404: val_acc did not improve from 0.99514\n",
      "Epoch 405/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0857 - acc: 0.9678 - val_loss: 0.1103 - val_acc: 0.9621\n",
      "\n",
      "Epoch 00405: val_acc did not improve from 0.99514\n",
      "Epoch 406/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0884 - acc: 0.9668 - val_loss: 0.1106 - val_acc: 0.9601\n",
      "\n",
      "Epoch 00406: val_acc did not improve from 0.99514\n",
      "Epoch 407/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0906 - acc: 0.9666 - val_loss: 0.0971 - val_acc: 0.9650\n",
      "\n",
      "Epoch 00407: val_acc did not improve from 0.99514\n",
      "Epoch 408/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0757 - acc: 0.9709 - val_loss: 0.1092 - val_acc: 0.9640\n",
      "\n",
      "Epoch 00408: val_acc did not improve from 0.99514\n",
      "Epoch 409/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0781 - acc: 0.9716 - val_loss: 0.1165 - val_acc: 0.9582\n",
      "\n",
      "Epoch 00409: val_acc did not improve from 0.99514\n",
      "Epoch 410/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0725 - acc: 0.9743 - val_loss: 0.0925 - val_acc: 0.9698\n",
      "\n",
      "Epoch 00410: val_acc did not improve from 0.99514\n",
      "Epoch 411/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0740 - acc: 0.9729 - val_loss: 0.1077 - val_acc: 0.9640\n",
      "\n",
      "Epoch 00411: val_acc did not improve from 0.99514\n",
      "Epoch 412/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0735 - acc: 0.9716 - val_loss: 0.1127 - val_acc: 0.9669\n",
      "\n",
      "Epoch 00412: val_acc did not improve from 0.99514\n",
      "Epoch 413/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0757 - acc: 0.9717 - val_loss: 0.1064 - val_acc: 0.9611\n",
      "\n",
      "Epoch 00413: val_acc did not improve from 0.99514\n",
      "Epoch 414/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0798 - acc: 0.9707 - val_loss: 0.0999 - val_acc: 0.9650\n",
      "\n",
      "Epoch 00414: val_acc did not improve from 0.99514\n",
      "Epoch 415/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0733 - acc: 0.9733 - val_loss: 0.0963 - val_acc: 0.9660\n",
      "\n",
      "Epoch 00415: val_acc did not improve from 0.99514\n",
      "Epoch 416/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0727 - acc: 0.9737 - val_loss: 0.1007 - val_acc: 0.9708\n",
      "\n",
      "Epoch 00416: val_acc did not improve from 0.99514\n",
      "Epoch 417/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 0.0693 - acc: 0.9744 - val_loss: 0.1200 - val_acc: 0.9582\n",
      "\n",
      "Epoch 00417: val_acc did not improve from 0.99514\n",
      "Epoch 418/1000\n",
      "236/236 [==============================] - 6s 28ms/step - loss: 0.0715 - acc: 0.9721 - val_loss: 0.1108 - val_acc: 0.9669\n",
      "\n",
      "Epoch 00418: val_acc did not improve from 0.99514\n",
      "Epoch 419/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0707 - acc: 0.9745 - val_loss: 0.1173 - val_acc: 0.9640\n",
      "\n",
      "Epoch 00419: val_acc did not improve from 0.99514\n",
      "Epoch 420/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0678 - acc: 0.9745 - val_loss: 0.1062 - val_acc: 0.9679\n",
      "\n",
      "Epoch 00420: val_acc did not improve from 0.99514\n",
      "Epoch 421/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.0637 - acc: 0.9756 - val_loss: 0.1108 - val_acc: 0.9660\n",
      "\n",
      "Epoch 00421: val_acc did not improve from 0.99514\n",
      "Epoch 422/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0657 - acc: 0.9765 - val_loss: 0.0925 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00422: val_acc did not improve from 0.99514\n",
      "Epoch 423/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0606 - acc: 0.9786 - val_loss: 0.1072 - val_acc: 0.9708\n",
      "\n",
      "Epoch 00423: val_acc did not improve from 0.99514\n",
      "Epoch 424/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0598 - acc: 0.9793 - val_loss: 0.1141 - val_acc: 0.9591\n",
      "\n",
      "Epoch 00424: val_acc did not improve from 0.99514\n",
      "Epoch 425/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0588 - acc: 0.9793 - val_loss: 0.0931 - val_acc: 0.9718\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00425: val_acc did not improve from 0.99514\n",
      "Epoch 426/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0561 - acc: 0.9789 - val_loss: 0.0948 - val_acc: 0.9689\n",
      "\n",
      "Epoch 00426: val_acc did not improve from 0.99514\n",
      "Epoch 427/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0548 - acc: 0.9800 - val_loss: 0.0996 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00427: val_acc did not improve from 0.99514\n",
      "Epoch 428/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0536 - acc: 0.9785 - val_loss: 0.1004 - val_acc: 0.9689\n",
      "\n",
      "Epoch 00428: val_acc did not improve from 0.99514\n",
      "Epoch 429/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0627 - acc: 0.9780 - val_loss: 0.0949 - val_acc: 0.9747\n",
      "\n",
      "Epoch 00429: val_acc did not improve from 0.99514\n",
      "Epoch 430/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0529 - acc: 0.9813 - val_loss: 0.0873 - val_acc: 0.9767\n",
      "\n",
      "Epoch 00430: val_acc did not improve from 0.99514\n",
      "Epoch 431/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0567 - acc: 0.9809 - val_loss: 0.0902 - val_acc: 0.9689\n",
      "\n",
      "Epoch 00431: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00431: ReduceLROnPlateau reducing learning rate to 6.25000029685907e-05.\n",
      "Epoch 432/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0406 - acc: 0.9867 - val_loss: 0.0869 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00432: val_acc did not improve from 0.99514\n",
      "Epoch 433/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0431 - acc: 0.9855 - val_loss: 0.0833 - val_acc: 0.9757\n",
      "\n",
      "Epoch 00433: val_acc did not improve from 0.99514\n",
      "Epoch 434/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0396 - acc: 0.9870 - val_loss: 0.0822 - val_acc: 0.9767\n",
      "\n",
      "Epoch 00434: val_acc did not improve from 0.99514\n",
      "Epoch 435/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0418 - acc: 0.9862 - val_loss: 0.0869 - val_acc: 0.9708\n",
      "\n",
      "Epoch 00435: val_acc did not improve from 0.99514\n",
      "Epoch 436/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0388 - acc: 0.9863 - val_loss: 0.0927 - val_acc: 0.9718\n",
      "\n",
      "Epoch 00436: val_acc did not improve from 0.99514\n",
      "Epoch 437/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0350 - acc: 0.9886 - val_loss: 0.0892 - val_acc: 0.9708\n",
      "\n",
      "Epoch 00437: val_acc did not improve from 0.99514\n",
      "Epoch 438/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0404 - acc: 0.9861 - val_loss: 0.0910 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00438: val_acc did not improve from 0.99514\n",
      "Epoch 439/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0330 - acc: 0.9895 - val_loss: 0.0898 - val_acc: 0.9737\n",
      "\n",
      "Epoch 00439: val_acc did not improve from 0.99514\n",
      "Epoch 440/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0338 - acc: 0.9891 - val_loss: 0.0911 - val_acc: 0.9737\n",
      "\n",
      "Epoch 00440: val_acc did not improve from 0.99514\n",
      "Epoch 441/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0367 - acc: 0.9877 - val_loss: 0.0964 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00441: val_acc did not improve from 0.99514\n",
      "Epoch 442/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0323 - acc: 0.9906 - val_loss: 0.1064 - val_acc: 0.9708\n",
      "\n",
      "Epoch 00442: val_acc did not improve from 0.99514\n",
      "Epoch 443/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0323 - acc: 0.9887 - val_loss: 0.1133 - val_acc: 0.9698\n",
      "\n",
      "Epoch 00443: val_acc did not improve from 0.99514\n",
      "Epoch 444/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0339 - acc: 0.9885 - val_loss: 0.0936 - val_acc: 0.9698\n",
      "\n",
      "Epoch 00444: val_acc did not improve from 0.99514\n",
      "Epoch 445/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0280 - acc: 0.9907 - val_loss: 0.1005 - val_acc: 0.9718\n",
      "\n",
      "Epoch 00445: val_acc did not improve from 0.99514\n",
      "Epoch 446/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0365 - acc: 0.9886 - val_loss: 0.0990 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00446: val_acc did not improve from 0.99514\n",
      "Epoch 447/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0315 - acc: 0.9895 - val_loss: 0.0883 - val_acc: 0.9737\n",
      "\n",
      "Epoch 00447: val_acc did not improve from 0.99514\n",
      "Epoch 448/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0263 - acc: 0.9916 - val_loss: 0.0842 - val_acc: 0.9767\n",
      "\n",
      "Epoch 00448: val_acc did not improve from 0.99514\n",
      "Epoch 449/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0270 - acc: 0.9912 - val_loss: 0.0922 - val_acc: 0.9728\n",
      "\n",
      "Epoch 00449: val_acc did not improve from 0.99514\n",
      "Epoch 450/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0266 - acc: 0.9911 - val_loss: 0.0943 - val_acc: 0.9767\n",
      "\n",
      "Epoch 00450: val_acc did not improve from 0.99514\n",
      "Epoch 451/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0281 - acc: 0.9915 - val_loss: 0.0923 - val_acc: 0.9757\n",
      "\n",
      "Epoch 00451: val_acc did not improve from 0.99514\n",
      "Epoch 452/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0230 - acc: 0.9924 - val_loss: 0.0803 - val_acc: 0.9767\n",
      "\n",
      "Epoch 00452: val_acc did not improve from 0.99514\n",
      "Epoch 453/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0278 - acc: 0.9907 - val_loss: 0.0906 - val_acc: 0.9767\n",
      "\n",
      "Epoch 00453: val_acc did not improve from 0.99514\n",
      "Epoch 454/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0270 - acc: 0.9911 - val_loss: 0.0818 - val_acc: 0.9776\n",
      "\n",
      "Epoch 00454: val_acc did not improve from 0.99514\n",
      "Epoch 455/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0244 - acc: 0.9931 - val_loss: 0.0944 - val_acc: 0.9757\n",
      "\n",
      "Epoch 00455: val_acc did not improve from 0.99514\n",
      "Epoch 456/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 0.0247 - acc: 0.9914 - val_loss: 0.0833 - val_acc: 0.9776\n",
      "\n",
      "Epoch 00456: val_acc did not improve from 0.99514\n",
      "Epoch 457/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0216 - acc: 0.9928 - val_loss: 0.0892 - val_acc: 0.9776\n",
      "\n",
      "Epoch 00457: val_acc did not improve from 0.99514\n",
      "Epoch 458/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0205 - acc: 0.9935 - val_loss: 0.0835 - val_acc: 0.9786\n",
      "\n",
      "Epoch 00458: val_acc did not improve from 0.99514\n",
      "Epoch 459/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0204 - acc: 0.9940 - val_loss: 0.0875 - val_acc: 0.9747\n",
      "\n",
      "Epoch 00459: val_acc did not improve from 0.99514\n",
      "Epoch 460/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0238 - acc: 0.9910 - val_loss: 0.0856 - val_acc: 0.9786\n",
      "\n",
      "Epoch 00460: val_acc did not improve from 0.99514\n",
      "Epoch 461/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0221 - acc: 0.9920 - val_loss: 0.0865 - val_acc: 0.9767\n",
      "\n",
      "Epoch 00461: val_acc did not improve from 0.99514\n",
      "Epoch 462/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0236 - acc: 0.9924 - val_loss: 0.0772 - val_acc: 0.9786\n",
      "\n",
      "Epoch 00462: val_acc did not improve from 0.99514\n",
      "Epoch 463/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.0179 - acc: 0.9943 - val_loss: 0.0796 - val_acc: 0.9796\n",
      "\n",
      "Epoch 00463: val_acc did not improve from 0.99514\n",
      "Epoch 464/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0207 - acc: 0.9939 - val_loss: 0.0868 - val_acc: 0.9805\n",
      "\n",
      "Epoch 00464: val_acc did not improve from 0.99514\n",
      "Epoch 465/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0175 - acc: 0.9956 - val_loss: 0.0718 - val_acc: 0.9805\n",
      "\n",
      "Epoch 00465: val_acc did not improve from 0.99514\n",
      "Epoch 466/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0198 - acc: 0.9942 - val_loss: 0.0745 - val_acc: 0.9825\n",
      "\n",
      "Epoch 00466: val_acc did not improve from 0.99514\n",
      "Epoch 467/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0156 - acc: 0.9960 - val_loss: 0.0673 - val_acc: 0.9835\n",
      "\n",
      "Epoch 00467: val_acc did not improve from 0.99514\n",
      "Epoch 468/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0189 - acc: 0.9958 - val_loss: 0.0905 - val_acc: 0.9757\n",
      "\n",
      "Epoch 00468: val_acc did not improve from 0.99514\n",
      "Epoch 469/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0315 - acc: 0.9893 - val_loss: 0.0809 - val_acc: 0.9796\n",
      "\n",
      "Epoch 00469: val_acc did not improve from 0.99514\n",
      "Epoch 470/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0206 - acc: 0.9932 - val_loss: 0.0820 - val_acc: 0.9796\n",
      "\n",
      "Epoch 00470: val_acc did not improve from 0.99514\n",
      "Epoch 471/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0157 - acc: 0.9956 - val_loss: 0.0871 - val_acc: 0.9796\n",
      "\n",
      "Epoch 00471: val_acc did not improve from 0.99514\n",
      "Epoch 472/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0245 - acc: 0.9924 - val_loss: 0.0604 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00472: val_acc did not improve from 0.99514\n",
      "Epoch 473/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0234 - acc: 0.9934 - val_loss: 0.0641 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00473: val_acc did not improve from 0.99514\n",
      "Epoch 474/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0159 - acc: 0.9951 - val_loss: 0.0628 - val_acc: 0.9835\n",
      "\n",
      "Epoch 00474: val_acc did not improve from 0.99514\n",
      "Epoch 475/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0160 - acc: 0.9955 - val_loss: 0.0711 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00475: val_acc did not improve from 0.99514\n",
      "Epoch 476/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0141 - acc: 0.9959 - val_loss: 0.0598 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00476: val_acc did not improve from 0.99514\n",
      "Epoch 477/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0124 - acc: 0.9963 - val_loss: 0.0617 - val_acc: 0.9835\n",
      "\n",
      "Epoch 00477: val_acc did not improve from 0.99514\n",
      "Epoch 478/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0159 - acc: 0.9951 - val_loss: 0.0626 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00478: val_acc did not improve from 0.99514\n",
      "Epoch 479/1000\n",
      "236/236 [==============================] - 9s 38ms/step - loss: 0.0141 - acc: 0.9958 - val_loss: 0.0588 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00479: val_acc did not improve from 0.99514\n",
      "Epoch 480/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0125 - acc: 0.9963 - val_loss: 0.0682 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00480: val_acc did not improve from 0.99514\n",
      "Epoch 481/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0114 - acc: 0.9968 - val_loss: 0.0725 - val_acc: 0.9835\n",
      "\n",
      "Epoch 00481: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00481: ReduceLROnPlateau reducing learning rate to 3.125000148429535e-05.\n",
      "Epoch 482/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0104 - acc: 0.9979 - val_loss: 0.0647 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00482: val_acc did not improve from 0.99514\n",
      "Epoch 483/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0098 - acc: 0.9975 - val_loss: 0.0659 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00483: val_acc did not improve from 0.99514\n",
      "Epoch 484/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0099 - acc: 0.9971 - val_loss: 0.0638 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00484: val_acc did not improve from 0.99514\n",
      "Epoch 485/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0094 - acc: 0.9977 - val_loss: 0.0728 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00485: val_acc did not improve from 0.99514\n",
      "Epoch 486/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0088 - acc: 0.9979 - val_loss: 0.0668 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00486: val_acc did not improve from 0.99514\n",
      "Epoch 487/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0094 - acc: 0.9976 - val_loss: 0.0637 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00487: val_acc did not improve from 0.99514\n",
      "Epoch 488/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0113 - acc: 0.9963 - val_loss: 0.0691 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00488: val_acc did not improve from 0.99514\n",
      "Epoch 489/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0098 - acc: 0.9969 - val_loss: 0.0663 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00489: val_acc did not improve from 0.99514\n",
      "Epoch 490/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0090 - acc: 0.9977 - val_loss: 0.0672 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00490: val_acc did not improve from 0.99514\n",
      "Epoch 491/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0083 - acc: 0.9981 - val_loss: 0.0659 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00491: val_acc did not improve from 0.99514\n",
      "Epoch 492/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0084 - acc: 0.9979 - val_loss: 0.0689 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00492: val_acc did not improve from 0.99514\n",
      "Epoch 493/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0086 - acc: 0.9976 - val_loss: 0.0663 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00493: val_acc did not improve from 0.99514\n",
      "Epoch 494/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0104 - acc: 0.9976 - val_loss: 0.1031 - val_acc: 0.9757\n",
      "\n",
      "Epoch 00494: val_acc did not improve from 0.99514\n",
      "Epoch 495/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0096 - acc: 0.9972 - val_loss: 0.0718 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00495: val_acc did not improve from 0.99514\n",
      "Epoch 496/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0138 - acc: 0.9964 - val_loss: 0.0768 - val_acc: 0.9825\n",
      "\n",
      "Epoch 00496: val_acc did not improve from 0.99514\n",
      "Epoch 497/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0093 - acc: 0.9972 - val_loss: 0.0639 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00497: val_acc did not improve from 0.99514\n",
      "Epoch 498/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 0.0077 - acc: 0.9980 - val_loss: 0.0659 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00498: val_acc did not improve from 0.99514\n",
      "Epoch 499/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0077 - acc: 0.9984 - val_loss: 0.0662 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00499: val_acc did not improve from 0.99514\n",
      "Epoch 500/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0075 - acc: 0.9984 - val_loss: 0.0691 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00500: val_acc did not improve from 0.99514\n",
      "Epoch 501/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0133 - acc: 0.9960 - val_loss: 0.0794 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00501: val_acc did not improve from 0.99514\n",
      "Epoch 502/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0093 - acc: 0.9971 - val_loss: 0.0770 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00502: val_acc did not improve from 0.99514\n",
      "Epoch 503/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0079 - acc: 0.9980 - val_loss: 0.0781 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00503: val_acc did not improve from 0.99514\n",
      "Epoch 504/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0075 - acc: 0.9988 - val_loss: 0.0752 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00504: val_acc did not improve from 0.99514\n",
      "Epoch 505/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0069 - acc: 0.9984 - val_loss: 0.0732 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00505: val_acc did not improve from 0.99514\n",
      "Epoch 506/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0067 - acc: 0.9980 - val_loss: 0.0752 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00506: val_acc did not improve from 0.99514\n",
      "Epoch 507/1000\n",
      "236/236 [==============================] - 9s 36ms/step - loss: 0.0070 - acc: 0.9988 - val_loss: 0.0735 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00507: val_acc did not improve from 0.99514\n",
      "Epoch 508/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0066 - acc: 0.9985 - val_loss: 0.0746 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00508: val_acc did not improve from 0.99514\n",
      "Epoch 509/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0071 - acc: 0.9987 - val_loss: 0.0836 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00509: val_acc did not improve from 0.99514\n",
      "Epoch 510/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0102 - acc: 0.9975 - val_loss: 0.0767 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00510: val_acc did not improve from 0.99514\n",
      "Epoch 511/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0072 - acc: 0.9980 - val_loss: 0.0737 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00511: val_acc did not improve from 0.99514\n",
      "Epoch 512/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0063 - acc: 0.9988 - val_loss: 0.0707 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00512: val_acc did not improve from 0.99514\n",
      "Epoch 513/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0064 - acc: 0.9988 - val_loss: 0.0735 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00513: val_acc did not improve from 0.99514\n",
      "Epoch 514/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0056 - acc: 0.9991 - val_loss: 0.0771 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00514: val_acc did not improve from 0.99514\n",
      "Epoch 515/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0058 - acc: 0.9991 - val_loss: 0.0798 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00515: val_acc did not improve from 0.99514\n",
      "Epoch 516/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0057 - acc: 0.9988 - val_loss: 0.0778 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00516: val_acc did not improve from 0.99514\n",
      "Epoch 517/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0109 - acc: 0.9968 - val_loss: 0.0828 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00517: val_acc did not improve from 0.99514\n",
      "Epoch 518/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0060 - acc: 0.9988 - val_loss: 0.0860 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00518: val_acc did not improve from 0.99514\n",
      "Epoch 519/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0056 - acc: 0.9991 - val_loss: 0.0845 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00519: val_acc did not improve from 0.99514\n",
      "Epoch 520/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0054 - acc: 0.9989 - val_loss: 0.0780 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00520: val_acc did not improve from 0.99514\n",
      "Epoch 521/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0052 - acc: 0.9989 - val_loss: 0.0822 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00521: val_acc did not improve from 0.99514\n",
      "Epoch 522/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0054 - acc: 0.9989 - val_loss: 0.0850 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00522: val_acc did not improve from 0.99514\n",
      "Epoch 523/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0058 - acc: 0.9991 - val_loss: 0.0743 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00523: val_acc did not improve from 0.99514\n",
      "Epoch 524/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0075 - acc: 0.9981 - val_loss: 0.0713 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00524: val_acc did not improve from 0.99514\n",
      "Epoch 525/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0078 - acc: 0.9976 - val_loss: 0.0779 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00525: val_acc did not improve from 0.99514\n",
      "Epoch 526/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 0.0051 - acc: 0.9993 - val_loss: 0.0741 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00526: val_acc did not improve from 0.99514\n",
      "Epoch 527/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0052 - acc: 0.9991 - val_loss: 0.0765 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00527: val_acc did not improve from 0.99514\n",
      "Epoch 528/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0046 - acc: 0.9992 - val_loss: 0.0770 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00528: val_acc did not improve from 0.99514\n",
      "Epoch 529/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0046 - acc: 0.9989 - val_loss: 0.0770 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00529: val_acc did not improve from 0.99514\n",
      "Epoch 530/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0069 - acc: 0.9984 - val_loss: 0.0870 - val_acc: 0.9825\n",
      "\n",
      "Epoch 00530: val_acc did not improve from 0.99514\n",
      "Epoch 531/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0048 - acc: 0.9989 - val_loss: 0.0858 - val_acc: 0.9835\n",
      "\n",
      "Epoch 00531: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00531: ReduceLROnPlateau reducing learning rate to 1.5625000742147677e-05.\n",
      "Epoch 532/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.0043 - acc: 0.9991 - val_loss: 0.0827 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00532: val_acc did not improve from 0.99514\n",
      "Epoch 533/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0041 - acc: 0.9995 - val_loss: 0.0795 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00533: val_acc did not improve from 0.99514\n",
      "Epoch 534/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.0042 - acc: 0.9992 - val_loss: 0.0839 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00534: val_acc did not improve from 0.99514\n",
      "Epoch 535/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0041 - acc: 0.9995 - val_loss: 0.0764 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00535: val_acc did not improve from 0.99514\n",
      "Epoch 536/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 0.0040 - acc: 0.9992 - val_loss: 0.0776 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00536: val_acc did not improve from 0.99514\n",
      "Epoch 537/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0040 - acc: 0.9991 - val_loss: 0.0798 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00537: val_acc did not improve from 0.99514\n",
      "Epoch 538/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0041 - acc: 0.9993 - val_loss: 0.0811 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00538: val_acc did not improve from 0.99514\n",
      "Epoch 539/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 0.0040 - acc: 0.9991 - val_loss: 0.0786 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00539: val_acc did not improve from 0.99514\n",
      "Epoch 540/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0041 - acc: 0.9995 - val_loss: 0.0798 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00540: val_acc did not improve from 0.99514\n",
      "Epoch 541/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0038 - acc: 0.9993 - val_loss: 0.0775 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00541: val_acc did not improve from 0.99514\n",
      "Epoch 542/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0036 - acc: 0.9996 - val_loss: 0.0865 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00542: val_acc did not improve from 0.99514\n",
      "Epoch 543/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0038 - acc: 0.9995 - val_loss: 0.0820 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00543: val_acc did not improve from 0.99514\n",
      "Epoch 544/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0043 - acc: 0.9993 - val_loss: 0.0842 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00544: val_acc did not improve from 0.99514\n",
      "Epoch 545/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0038 - acc: 0.9995 - val_loss: 0.0834 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00545: val_acc did not improve from 0.99514\n",
      "Epoch 546/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0036 - acc: 0.9997 - val_loss: 0.0821 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00546: val_acc did not improve from 0.99514\n",
      "Epoch 547/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0033 - acc: 0.9996 - val_loss: 0.0877 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00547: val_acc did not improve from 0.99514\n",
      "Epoch 548/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 0.0033 - acc: 0.9996 - val_loss: 0.0864 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00548: val_acc did not improve from 0.99514\n",
      "Epoch 549/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0035 - acc: 0.9993 - val_loss: 0.0854 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00549: val_acc did not improve from 0.99514\n",
      "Epoch 550/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0056 - acc: 0.9983 - val_loss: 0.0951 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00550: val_acc did not improve from 0.99514\n",
      "Epoch 551/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0034 - acc: 0.9996 - val_loss: 0.0905 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00551: val_acc did not improve from 0.99514\n",
      "Epoch 552/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0032 - acc: 0.9995 - val_loss: 0.0845 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00552: val_acc did not improve from 0.99514\n",
      "Epoch 553/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0032 - acc: 0.9996 - val_loss: 0.0800 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00553: val_acc did not improve from 0.99514\n",
      "Epoch 554/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 0.0033 - acc: 0.9995 - val_loss: 0.0833 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00554: val_acc did not improve from 0.99514\n",
      "Epoch 555/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0036 - acc: 0.9992 - val_loss: 0.0934 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00555: val_acc did not improve from 0.99514\n",
      "Epoch 556/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0033 - acc: 0.9995 - val_loss: 0.0829 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00556: val_acc did not improve from 0.99514\n",
      "Epoch 557/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0041 - acc: 0.9996 - val_loss: 0.0831 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00557: val_acc did not improve from 0.99514\n",
      "Epoch 558/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0031 - acc: 0.9996 - val_loss: 0.0829 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00558: val_acc did not improve from 0.99514\n",
      "Epoch 559/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0029 - acc: 0.9996 - val_loss: 0.0836 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00559: val_acc did not improve from 0.99514\n",
      "Epoch 560/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0027 - acc: 0.9999 - val_loss: 0.0841 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00560: val_acc did not improve from 0.99514\n",
      "Epoch 561/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0029 - acc: 0.9997 - val_loss: 0.0852 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00561: val_acc did not improve from 0.99514\n",
      "Epoch 562/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0029 - acc: 0.9996 - val_loss: 0.0847 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00562: val_acc did not improve from 0.99514\n",
      "Epoch 563/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0027 - acc: 0.9997 - val_loss: 0.0860 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00563: val_acc did not improve from 0.99514\n",
      "Epoch 564/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0026 - acc: 0.9997 - val_loss: 0.0866 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00564: val_acc did not improve from 0.99514\n",
      "Epoch 565/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0028 - acc: 0.9999 - val_loss: 0.0863 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00565: val_acc did not improve from 0.99514\n",
      "Epoch 566/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0054 - acc: 0.9988 - val_loss: 0.0857 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00566: val_acc did not improve from 0.99514\n",
      "Epoch 567/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0027 - acc: 0.9996 - val_loss: 0.0866 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00567: val_acc did not improve from 0.99514\n",
      "Epoch 568/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0026 - acc: 0.9999 - val_loss: 0.0880 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00568: val_acc did not improve from 0.99514\n",
      "Epoch 569/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0026 - acc: 0.9997 - val_loss: 0.0895 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00569: val_acc did not improve from 0.99514\n",
      "Epoch 570/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 0.0051 - acc: 0.9989 - val_loss: 0.0866 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00570: val_acc did not improve from 0.99514\n",
      "Epoch 571/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0025 - acc: 0.9997 - val_loss: 0.0907 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00571: val_acc did not improve from 0.99514\n",
      "Epoch 572/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0024 - acc: 0.9999 - val_loss: 0.0893 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00572: val_acc did not improve from 0.99514\n",
      "Epoch 573/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0023 - acc: 0.9997 - val_loss: 0.0885 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00573: val_acc did not improve from 0.99514\n",
      "Epoch 574/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0022 - acc: 0.9999 - val_loss: 0.0893 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00574: val_acc did not improve from 0.99514\n",
      "Epoch 575/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0022 - acc: 0.9997 - val_loss: 0.0893 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00575: val_acc did not improve from 0.99514\n",
      "Epoch 576/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0021 - acc: 0.9997 - val_loss: 0.0921 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00576: val_acc did not improve from 0.99514\n",
      "Epoch 577/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0021 - acc: 0.9999 - val_loss: 0.0897 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00577: val_acc did not improve from 0.99514\n",
      "Epoch 578/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0022 - acc: 0.9997 - val_loss: 0.0943 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00578: val_acc did not improve from 0.99514\n",
      "Epoch 579/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0025 - acc: 0.9996 - val_loss: 0.0881 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00579: val_acc did not improve from 0.99514\n",
      "Epoch 580/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0024 - acc: 0.9997 - val_loss: 0.0893 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00580: val_acc did not improve from 0.99514\n",
      "Epoch 581/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0023 - acc: 0.9999 - val_loss: 0.0925 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00581: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00581: ReduceLROnPlateau reducing learning rate to 7.812500371073838e-06.\n",
      "Epoch 582/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0021 - acc: 0.9997 - val_loss: 0.0909 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00582: val_acc did not improve from 0.99514\n",
      "Epoch 583/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0021 - acc: 0.9999 - val_loss: 0.0941 - val_acc: 0.9864\n",
      "\n",
      "Epoch 00583: val_acc did not improve from 0.99514\n",
      "Epoch 584/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0020 - acc: 0.9999 - val_loss: 0.0908 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00584: val_acc did not improve from 0.99514\n",
      "Epoch 585/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0020 - acc: 0.9999 - val_loss: 0.0904 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00585: val_acc did not improve from 0.99514\n",
      "Epoch 586/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0020 - acc: 0.9999 - val_loss: 0.0913 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00586: val_acc did not improve from 0.99514\n",
      "Epoch 587/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0020 - acc: 0.9999 - val_loss: 0.0913 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00587: val_acc did not improve from 0.99514\n",
      "Epoch 588/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0019 - acc: 0.9999 - val_loss: 0.0917 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00588: val_acc did not improve from 0.99514\n",
      "Epoch 589/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0019 - acc: 0.9999 - val_loss: 0.0914 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00589: val_acc did not improve from 0.99514\n",
      "Epoch 590/1000\n",
      "236/236 [==============================] - 9s 38ms/step - loss: 0.0019 - acc: 0.9999 - val_loss: 0.0942 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00590: val_acc did not improve from 0.99514\n",
      "Epoch 591/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0024 - acc: 0.9997 - val_loss: 0.0908 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00591: val_acc did not improve from 0.99514\n",
      "Epoch 592/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0020 - acc: 0.9997 - val_loss: 0.0932 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00592: val_acc did not improve from 0.99514\n",
      "Epoch 593/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0018 - acc: 0.9999 - val_loss: 0.0945 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00593: val_acc did not improve from 0.99514\n",
      "Epoch 594/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0018 - acc: 0.9999 - val_loss: 0.0946 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00594: val_acc did not improve from 0.99514\n",
      "Epoch 595/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0018 - acc: 0.9999 - val_loss: 0.0949 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00595: val_acc did not improve from 0.99514\n",
      "Epoch 596/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0018 - acc: 0.9999 - val_loss: 0.0943 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00596: val_acc did not improve from 0.99514\n",
      "Epoch 597/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0018 - acc: 0.9999 - val_loss: 0.0951 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00597: val_acc did not improve from 0.99514\n",
      "Epoch 598/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0017 - acc: 0.9999 - val_loss: 0.0919 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00598: val_acc did not improve from 0.99514\n",
      "Epoch 599/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 0.0017 - acc: 0.9999 - val_loss: 0.0954 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00599: val_acc did not improve from 0.99514\n",
      "Epoch 600/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0017 - acc: 0.9999 - val_loss: 0.0962 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00600: val_acc did not improve from 0.99514\n",
      "Epoch 601/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0017 - acc: 0.9999 - val_loss: 0.0982 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00601: val_acc did not improve from 0.99514\n",
      "Epoch 602/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0016 - acc: 0.9999 - val_loss: 0.1061 - val_acc: 0.9844\n",
      "\n",
      "Epoch 00602: val_acc did not improve from 0.99514\n",
      "Epoch 603/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0022 - acc: 0.9997 - val_loss: 0.0945 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00603: val_acc did not improve from 0.99514\n",
      "Epoch 604/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0017 - acc: 0.9999 - val_loss: 0.0938 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00604: val_acc did not improve from 0.99514\n",
      "Epoch 605/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0016 - acc: 0.9999 - val_loss: 0.0963 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00605: val_acc did not improve from 0.99514\n",
      "Epoch 606/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0016 - acc: 0.9999 - val_loss: 0.0970 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00606: val_acc did not improve from 0.99514\n",
      "Epoch 607/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0015 - acc: 0.9999 - val_loss: 0.0959 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00607: val_acc did not improve from 0.99514\n",
      "Epoch 608/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0015 - acc: 0.9999 - val_loss: 0.0975 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00608: val_acc did not improve from 0.99514\n",
      "Epoch 609/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0014 - acc: 0.9999 - val_loss: 0.0967 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00609: val_acc did not improve from 0.99514\n",
      "Epoch 610/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 0.0027 - acc: 0.9997 - val_loss: 0.0954 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00610: val_acc did not improve from 0.99514\n",
      "Epoch 611/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 0.0015 - acc: 0.9999 - val_loss: 0.0949 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00611: val_acc did not improve from 0.99514\n",
      "Epoch 612/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 0.0015 - acc: 0.9999 - val_loss: 0.0962 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00612: val_acc did not improve from 0.99514\n",
      "Epoch 613/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0014 - acc: 0.9999 - val_loss: 0.0948 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00613: val_acc did not improve from 0.99514\n",
      "Epoch 614/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0015 - acc: 0.9999 - val_loss: 0.0971 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00614: val_acc did not improve from 0.99514\n",
      "Epoch 615/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 0.0014 - acc: 0.9999 - val_loss: 0.0993 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00615: val_acc did not improve from 0.99514\n",
      "Epoch 616/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 0.0014 - acc: 0.9999 - val_loss: 0.0948 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00616: val_acc did not improve from 0.99514\n",
      "Epoch 617/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0013 - acc: 0.9999 - val_loss: 0.1031 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00617: val_acc did not improve from 0.99514\n",
      "Epoch 618/1000\n",
      "236/236 [==============================] - 9s 39ms/step - loss: 0.0013 - acc: 0.9999 - val_loss: 0.0973 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00618: val_acc did not improve from 0.99514\n",
      "Epoch 619/1000\n",
      "236/236 [==============================] - 9s 38ms/step - loss: 0.0013 - acc: 0.9999 - val_loss: 0.0975 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00619: val_acc did not improve from 0.99514\n",
      "Epoch 620/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 0.0013 - acc: 0.9999 - val_loss: 0.0972 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00620: val_acc did not improve from 0.99514\n",
      "Epoch 621/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.0026 - acc: 0.9997 - val_loss: 0.1202 - val_acc: 0.9854\n",
      "\n",
      "Epoch 00621: val_acc did not improve from 0.99514\n",
      "Epoch 622/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0015 - acc: 0.9999 - val_loss: 0.1096 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00622: val_acc did not improve from 0.99514\n",
      "Epoch 623/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 0.0013 - acc: 0.9999 - val_loss: 0.1112 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00623: val_acc did not improve from 0.99514\n",
      "Epoch 624/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.0012 - acc: 0.9999 - val_loss: 0.1084 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00624: val_acc did not improve from 0.99514\n",
      "Epoch 625/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0012 - acc: 1.0000 - val_loss: 0.1079 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00625: val_acc did not improve from 0.99514\n",
      "Epoch 626/1000\n",
      "236/236 [==============================] - 6s 23ms/step - loss: 0.0012 - acc: 1.0000 - val_loss: 0.1097 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00626: val_acc did not improve from 0.99514\n",
      "Epoch 627/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.0011 - acc: 1.0000 - val_loss: 0.1052 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00627: val_acc did not improve from 0.99514\n",
      "Epoch 628/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0012 - acc: 1.0000 - val_loss: 0.1080 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00628: val_acc did not improve from 0.99514\n",
      "Epoch 629/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 0.0011 - acc: 1.0000 - val_loss: 0.1115 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00629: val_acc did not improve from 0.99514\n",
      "Epoch 630/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 0.0011 - acc: 1.0000 - val_loss: 0.1076 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00630: val_acc did not improve from 0.99514\n",
      "Epoch 631/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 0.0010 - acc: 1.0000 - val_loss: 0.1075 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00631: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00631: ReduceLROnPlateau reducing learning rate to 3.906250185536919e-06.\n",
      "Epoch 632/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 0.0010 - acc: 1.0000 - val_loss: 0.1075 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00632: val_acc did not improve from 0.99514\n",
      "Epoch 633/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 9.8238e-04 - acc: 1.0000 - val_loss: 0.1070 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00633: val_acc did not improve from 0.99514\n",
      "Epoch 634/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 9.9561e-04 - acc: 1.0000 - val_loss: 0.1074 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00634: val_acc did not improve from 0.99514\n",
      "Epoch 635/1000\n",
      "236/236 [==============================] - 9s 37ms/step - loss: 9.7999e-04 - acc: 1.0000 - val_loss: 0.1074 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00635: val_acc did not improve from 0.99514\n",
      "Epoch 636/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 9.6291e-04 - acc: 1.0000 - val_loss: 0.1070 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00636: val_acc did not improve from 0.99514\n",
      "Epoch 637/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 9.7837e-04 - acc: 1.0000 - val_loss: 0.1065 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00637: val_acc did not improve from 0.99514\n",
      "Epoch 638/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 9.3536e-04 - acc: 1.0000 - val_loss: 0.1067 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00638: val_acc did not improve from 0.99514\n",
      "Epoch 639/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 9.3531e-04 - acc: 1.0000 - val_loss: 0.1063 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00639: val_acc did not improve from 0.99514\n",
      "Epoch 640/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 9.3159e-04 - acc: 1.0000 - val_loss: 0.1055 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00640: val_acc did not improve from 0.99514\n",
      "Epoch 641/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 31ms/step - loss: 9.0343e-04 - acc: 1.0000 - val_loss: 0.1061 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00641: val_acc did not improve from 0.99514\n",
      "Epoch 642/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 9.1981e-04 - acc: 1.0000 - val_loss: 0.1064 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00642: val_acc did not improve from 0.99514\n",
      "Epoch 643/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 9.0807e-04 - acc: 1.0000 - val_loss: 0.1063 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00643: val_acc did not improve from 0.99514\n",
      "Epoch 644/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 9.0477e-04 - acc: 1.0000 - val_loss: 0.1064 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00644: val_acc did not improve from 0.99514\n",
      "Epoch 645/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 8.9504e-04 - acc: 1.0000 - val_loss: 0.1067 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00645: val_acc did not improve from 0.99514\n",
      "Epoch 646/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 8.8198e-04 - acc: 1.0000 - val_loss: 0.1070 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00646: val_acc did not improve from 0.99514\n",
      "Epoch 647/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 8.7268e-04 - acc: 1.0000 - val_loss: 0.1067 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00647: val_acc did not improve from 0.99514\n",
      "Epoch 648/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 8.4082e-04 - acc: 1.0000 - val_loss: 0.1080 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00648: val_acc did not improve from 0.99514\n",
      "Epoch 649/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 8.5648e-04 - acc: 1.0000 - val_loss: 0.1073 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00649: val_acc did not improve from 0.99514\n",
      "Epoch 650/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 8.5334e-04 - acc: 1.0000 - val_loss: 0.1074 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00650: val_acc did not improve from 0.99514\n",
      "Epoch 651/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 8.2302e-04 - acc: 1.0000 - val_loss: 0.1078 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00651: val_acc did not improve from 0.99514\n",
      "Epoch 652/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 8.4523e-04 - acc: 1.0000 - val_loss: 0.1074 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00652: val_acc did not improve from 0.99514\n",
      "Epoch 653/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 8.3150e-04 - acc: 1.0000 - val_loss: 0.1079 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00653: val_acc did not improve from 0.99514\n",
      "Epoch 654/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 8.1268e-04 - acc: 1.0000 - val_loss: 0.1083 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00654: val_acc did not improve from 0.99514\n",
      "Epoch 655/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 8.3104e-04 - acc: 1.0000 - val_loss: 0.1084 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00655: val_acc did not improve from 0.99514\n",
      "Epoch 656/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 8.0054e-04 - acc: 1.0000 - val_loss: 0.1089 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00656: val_acc did not improve from 0.99514\n",
      "Epoch 657/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 7.7912e-04 - acc: 1.0000 - val_loss: 0.1089 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00657: val_acc did not improve from 0.99514\n",
      "Epoch 658/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 7.9478e-04 - acc: 1.0000 - val_loss: 0.1082 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00658: val_acc did not improve from 0.99514\n",
      "Epoch 659/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 7.7192e-04 - acc: 1.0000 - val_loss: 0.1094 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00659: val_acc did not improve from 0.99514\n",
      "Epoch 660/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 8.2271e-04 - acc: 1.0000 - val_loss: 0.1085 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00660: val_acc did not improve from 0.99514\n",
      "Epoch 661/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 7.8208e-04 - acc: 1.0000 - val_loss: 0.1097 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00661: val_acc did not improve from 0.99514\n",
      "Epoch 662/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 7.5373e-04 - acc: 1.0000 - val_loss: 0.1101 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00662: val_acc did not improve from 0.99514\n",
      "Epoch 663/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 7.5045e-04 - acc: 1.0000 - val_loss: 0.1098 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00663: val_acc did not improve from 0.99514\n",
      "Epoch 664/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 7.4956e-04 - acc: 1.0000 - val_loss: 0.1097 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00664: val_acc did not improve from 0.99514\n",
      "Epoch 665/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 7.4374e-04 - acc: 1.0000 - val_loss: 0.1095 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00665: val_acc did not improve from 0.99514\n",
      "Epoch 666/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 7.4452e-04 - acc: 1.0000 - val_loss: 0.1104 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00666: val_acc did not improve from 0.99514\n",
      "Epoch 667/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 0.0012 - acc: 0.9999 - val_loss: 0.1036 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00667: val_acc did not improve from 0.99514\n",
      "Epoch 668/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 7.3750e-04 - acc: 1.0000 - val_loss: 0.1038 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00668: val_acc did not improve from 0.99514\n",
      "Epoch 669/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 7.2906e-04 - acc: 1.0000 - val_loss: 0.1039 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00669: val_acc did not improve from 0.99514\n",
      "Epoch 670/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 7.1867e-04 - acc: 1.0000 - val_loss: 0.1045 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00670: val_acc did not improve from 0.99514\n",
      "Epoch 671/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 7.1206e-04 - acc: 1.0000 - val_loss: 0.1047 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00671: val_acc did not improve from 0.99514\n",
      "Epoch 672/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 7.0453e-04 - acc: 1.0000 - val_loss: 0.1048 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00672: val_acc did not improve from 0.99514\n",
      "Epoch 673/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 7.2921e-04 - acc: 1.0000 - val_loss: 0.1045 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00673: val_acc did not improve from 0.99514\n",
      "Epoch 674/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 6.9613e-04 - acc: 1.0000 - val_loss: 0.1050 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00674: val_acc did not improve from 0.99514\n",
      "Epoch 675/1000\n",
      "236/236 [==============================] - 9s 36ms/step - loss: 6.8601e-04 - acc: 1.0000 - val_loss: 0.1052 - val_acc: 0.9903 1s - loss: 7.0391\n",
      "\n",
      "Epoch 00675: val_acc did not improve from 0.99514\n",
      "Epoch 676/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 6.9597e-04 - acc: 1.0000 - val_loss: 0.1054 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00676: val_acc did not improve from 0.99514\n",
      "Epoch 677/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 6.8696e-04 - acc: 1.0000 - val_loss: 0.1056 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00677: val_acc did not improve from 0.99514\n",
      "Epoch 678/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 6.8858e-04 - acc: 1.0000 - val_loss: 0.1052 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00678: val_acc did not improve from 0.99514\n",
      "Epoch 679/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 6.8436e-04 - acc: 1.0000 - val_loss: 0.1062 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00679: val_acc did not improve from 0.99514\n",
      "Epoch 680/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 6.6709e-04 - acc: 1.0000 - val_loss: 0.1062 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00680: val_acc did not improve from 0.99514\n",
      "Epoch 681/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 6.8971e-04 - acc: 1.0000 - val_loss: 0.1048 - val_acc: 0.9912\n",
      "\n",
      "Epoch 00681: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00681: ReduceLROnPlateau reducing learning rate to 1.9531250927684596e-06.\n",
      "Epoch 682/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 6.6920e-04 - acc: 1.0000 - val_loss: 0.1058 - val_acc: 0.9912\n",
      "\n",
      "Epoch 00682: val_acc did not improve from 0.99514\n",
      "Epoch 683/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 8s 33ms/step - loss: 6.5435e-04 - acc: 1.0000 - val_loss: 0.1061 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00683: val_acc did not improve from 0.99514\n",
      "Epoch 684/1000\n",
      "236/236 [==============================] - 9s 36ms/step - loss: 6.4961e-04 - acc: 1.0000 - val_loss: 0.1064 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00684: val_acc did not improve from 0.99514\n",
      "Epoch 685/1000\n",
      "236/236 [==============================] - 11s 46ms/step - loss: 6.4466e-04 - acc: 1.0000 - val_loss: 0.1066 - val_acc: 0.9912\n",
      "\n",
      "Epoch 00685: val_acc did not improve from 0.99514\n",
      "Epoch 686/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 6.4701e-04 - acc: 1.0000 - val_loss: 0.1066 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00686: val_acc did not improve from 0.99514\n",
      "Epoch 687/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 6.3917e-04 - acc: 1.0000 - val_loss: 0.1068 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00687: val_acc did not improve from 0.99514\n",
      "Epoch 688/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 6.3911e-04 - acc: 1.0000 - val_loss: 0.1068 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00688: val_acc did not improve from 0.99514\n",
      "Epoch 689/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 6.3967e-04 - acc: 1.0000 - val_loss: 0.1069 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00689: val_acc did not improve from 0.99514\n",
      "Epoch 690/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 6.3291e-04 - acc: 1.0000 - val_loss: 0.1070 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00690: val_acc did not improve from 0.99514\n",
      "Epoch 691/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 6.3067e-04 - acc: 1.0000 - val_loss: 0.1068 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00691: val_acc did not improve from 0.99514\n",
      "Epoch 692/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 6.3274e-04 - acc: 1.0000 - val_loss: 0.1069 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00692: val_acc did not improve from 0.99514\n",
      "Epoch 693/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 6.2707e-04 - acc: 1.0000 - val_loss: 0.1071 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00693: val_acc did not improve from 0.99514\n",
      "Epoch 694/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 6.2506e-04 - acc: 1.0000 - val_loss: 0.1072 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00694: val_acc did not improve from 0.99514\n",
      "Epoch 695/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 6.2339e-04 - acc: 1.0000 - val_loss: 0.1073 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00695: val_acc did not improve from 0.99514\n",
      "Epoch 696/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 6.1884e-04 - acc: 1.0000 - val_loss: 0.1075 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00696: val_acc did not improve from 0.99514\n",
      "Epoch 697/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 6.1591e-04 - acc: 1.0000 - val_loss: 0.1073 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00697: val_acc did not improve from 0.99514\n",
      "Epoch 698/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 6.1691e-04 - acc: 1.0000 - val_loss: 0.1079 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00698: val_acc did not improve from 0.99514\n",
      "Epoch 699/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 6.0876e-04 - acc: 1.0000 - val_loss: 0.1078 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00699: val_acc did not improve from 0.99514\n",
      "Epoch 700/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 6.1061e-04 - acc: 1.0000 - val_loss: 0.1077 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00700: val_acc did not improve from 0.99514\n",
      "Epoch 701/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 6.0439e-04 - acc: 1.0000 - val_loss: 0.1081 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00701: val_acc did not improve from 0.99514\n",
      "Epoch 702/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 6.1074e-04 - acc: 1.0000 - val_loss: 0.1079 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00702: val_acc did not improve from 0.99514\n",
      "Epoch 703/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 5.9999e-04 - acc: 1.0000 - val_loss: 0.1080 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00703: val_acc did not improve from 0.99514\n",
      "Epoch 704/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 6.0487e-04 - acc: 1.0000 - val_loss: 0.1081 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00704: val_acc did not improve from 0.99514\n",
      "Epoch 705/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 5.9928e-04 - acc: 1.0000 - val_loss: 0.1082 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00705: val_acc did not improve from 0.99514\n",
      "Epoch 706/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.9155e-04 - acc: 1.0000 - val_loss: 0.1092 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00706: val_acc did not improve from 0.99514\n",
      "Epoch 707/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.8758e-04 - acc: 1.0000 - val_loss: 0.1090 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00707: val_acc did not improve from 0.99514\n",
      "Epoch 708/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 5.8715e-04 - acc: 1.0000 - val_loss: 0.1109 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00708: val_acc did not improve from 0.99514\n",
      "Epoch 709/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 5.8654e-04 - acc: 1.0000 - val_loss: 0.1107 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00709: val_acc did not improve from 0.99514\n",
      "Epoch 710/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 5.8152e-04 - acc: 1.0000 - val_loss: 0.1091 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00710: val_acc did not improve from 0.99514\n",
      "Epoch 711/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.7250e-04 - acc: 1.0000 - val_loss: 0.1088 - val_acc: 0.9912\n",
      "\n",
      "Epoch 00711: val_acc did not improve from 0.99514\n",
      "Epoch 712/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.7721e-04 - acc: 1.0000 - val_loss: 0.1120 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00712: val_acc did not improve from 0.99514\n",
      "Epoch 713/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.7290e-04 - acc: 1.0000 - val_loss: 0.1147 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00713: val_acc did not improve from 0.99514\n",
      "Epoch 714/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 5.6760e-04 - acc: 1.0000 - val_loss: 0.1114 - val_acc: 0.9903\n",
      "\n",
      "Epoch 00714: val_acc did not improve from 0.99514\n",
      "Epoch 715/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 5.6845e-04 - acc: 1.0000 - val_loss: 0.1130 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00715: val_acc did not improve from 0.99514\n",
      "Epoch 716/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 5.6939e-04 - acc: 1.0000 - val_loss: 0.1140 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00716: val_acc did not improve from 0.99514\n",
      "Epoch 717/1000\n",
      "236/236 [==============================] - 9s 38ms/step - loss: 5.6584e-04 - acc: 1.0000 - val_loss: 0.1131 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00717: val_acc did not improve from 0.99514\n",
      "Epoch 718/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.6380e-04 - acc: 1.0000 - val_loss: 0.1143 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00718: val_acc did not improve from 0.99514\n",
      "Epoch 719/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 5.6223e-04 - acc: 1.0000 - val_loss: 0.1148 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00719: val_acc did not improve from 0.99514\n",
      "Epoch 720/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 5.5719e-04 - acc: 1.0000 - val_loss: 0.1145 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00720: val_acc did not improve from 0.99514\n",
      "Epoch 721/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 5.5586e-04 - acc: 1.0000 - val_loss: 0.1151 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00721: val_acc did not improve from 0.99514\n",
      "Epoch 722/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 5.5183e-04 - acc: 1.0000 - val_loss: 0.1150 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00722: val_acc did not improve from 0.99514\n",
      "Epoch 723/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 5.5293e-04 - acc: 1.0000 - val_loss: 0.1163 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00723: val_acc did not improve from 0.99514\n",
      "Epoch 724/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 5.5031e-04 - acc: 1.0000 - val_loss: 0.1168 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00724: val_acc did not improve from 0.99514\n",
      "Epoch 725/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 5.4064e-04 - acc: 1.0000 - val_loss: 0.1156 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00725: val_acc did not improve from 0.99514\n",
      "Epoch 726/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 5s 21ms/step - loss: 5.4326e-04 - acc: 1.0000 - val_loss: 0.1163 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00726: val_acc did not improve from 0.99514\n",
      "Epoch 727/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 5.4284e-04 - acc: 1.0000 - val_loss: 0.1168 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00727: val_acc did not improve from 0.99514\n",
      "Epoch 728/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 5.3726e-04 - acc: 1.0000 - val_loss: 0.1171 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00728: val_acc did not improve from 0.99514\n",
      "Epoch 729/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.3561e-04 - acc: 1.0000 - val_loss: 0.1174 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00729: val_acc did not improve from 0.99514\n",
      "Epoch 730/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.3292e-04 - acc: 1.0000 - val_loss: 0.1176 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00730: val_acc did not improve from 0.99514\n",
      "Epoch 731/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 5.3074e-04 - acc: 1.0000 - val_loss: 0.1178 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00731: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00731: ReduceLROnPlateau reducing learning rate to 9.765625463842298e-07.\n",
      "Epoch 732/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.2188e-04 - acc: 1.0000 - val_loss: 0.1179 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00732: val_acc did not improve from 0.99514\n",
      "Epoch 733/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 5.1886e-04 - acc: 1.0000 - val_loss: 0.1180 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00733: val_acc did not improve from 0.99514\n",
      "Epoch 734/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 5.1894e-04 - acc: 1.0000 - val_loss: 0.1180 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00734: val_acc did not improve from 0.99514\n",
      "Epoch 735/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 5.1898e-04 - acc: 1.0000 - val_loss: 0.1180 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00735: val_acc did not improve from 0.99514\n",
      "Epoch 736/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 5.1659e-04 - acc: 1.0000 - val_loss: 0.1180 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00736: val_acc did not improve from 0.99514\n",
      "Epoch 737/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 5.1475e-04 - acc: 1.0000 - val_loss: 0.1181 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00737: val_acc did not improve from 0.99514\n",
      "Epoch 738/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 5.1345e-04 - acc: 1.0000 - val_loss: 0.1183 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00738: val_acc did not improve from 0.99514\n",
      "Epoch 739/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 5.1188e-04 - acc: 1.0000 - val_loss: 0.1184 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00739: val_acc did not improve from 0.99514\n",
      "Epoch 740/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 5.1175e-04 - acc: 1.0000 - val_loss: 0.1183 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00740: val_acc did not improve from 0.99514\n",
      "Epoch 741/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 5.1018e-04 - acc: 1.0000 - val_loss: 0.1184 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00741: val_acc did not improve from 0.99514\n",
      "Epoch 742/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 5.0790e-04 - acc: 1.0000 - val_loss: 0.1185 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00742: val_acc did not improve from 0.99514\n",
      "Epoch 743/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 5.0670e-04 - acc: 1.0000 - val_loss: 0.1186 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00743: val_acc did not improve from 0.99514\n",
      "Epoch 744/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 5.0598e-04 - acc: 1.0000 - val_loss: 0.1187 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00744: val_acc did not improve from 0.99514\n",
      "Epoch 745/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 5.0462e-04 - acc: 1.0000 - val_loss: 0.1186 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00745: val_acc did not improve from 0.99514\n",
      "Epoch 746/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 5.0129e-04 - acc: 1.0000 - val_loss: 0.1188 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00746: val_acc did not improve from 0.99514\n",
      "Epoch 747/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 5.0188e-04 - acc: 1.0000 - val_loss: 0.1189 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00747: val_acc did not improve from 0.99514\n",
      "Epoch 748/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 5.0076e-04 - acc: 1.0000 - val_loss: 0.1190 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00748: val_acc did not improve from 0.99514\n",
      "Epoch 749/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 5.0214e-04 - acc: 1.0000 - val_loss: 0.1190 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00749: val_acc did not improve from 0.99514\n",
      "Epoch 750/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.9860e-04 - acc: 1.0000 - val_loss: 0.1191 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00750: val_acc did not improve from 0.99514\n",
      "Epoch 751/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.9734e-04 - acc: 1.0000 - val_loss: 0.1192 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00751: val_acc did not improve from 0.99514\n",
      "Epoch 752/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.9627e-04 - acc: 1.0000 - val_loss: 0.1192 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00752: val_acc did not improve from 0.99514\n",
      "Epoch 753/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.9606e-04 - acc: 1.0000 - val_loss: 0.1193 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00753: val_acc did not improve from 0.99514\n",
      "Epoch 754/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.9387e-04 - acc: 1.0000 - val_loss: 0.1194 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00754: val_acc did not improve from 0.99514\n",
      "Epoch 755/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.9330e-04 - acc: 1.0000 - val_loss: 0.1196 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00755: val_acc did not improve from 0.99514\n",
      "Epoch 756/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 4.9238e-04 - acc: 1.0000 - val_loss: 0.1196 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00756: val_acc did not improve from 0.99514\n",
      "Epoch 757/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.9119e-04 - acc: 1.0000 - val_loss: 0.1196 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00757: val_acc did not improve from 0.99514\n",
      "Epoch 758/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.8947e-04 - acc: 1.0000 - val_loss: 0.1198 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00758: val_acc did not improve from 0.99514\n",
      "Epoch 759/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.8753e-04 - acc: 1.0000 - val_loss: 0.1199 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00759: val_acc did not improve from 0.99514\n",
      "Epoch 760/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.8768e-04 - acc: 1.0000 - val_loss: 0.1200 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00760: val_acc did not improve from 0.99514\n",
      "Epoch 761/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.8489e-04 - acc: 1.0000 - val_loss: 0.1200 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00761: val_acc did not improve from 0.99514\n",
      "Epoch 762/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.8491e-04 - acc: 1.0000 - val_loss: 0.1201 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00762: val_acc did not improve from 0.99514\n",
      "Epoch 763/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.8309e-04 - acc: 1.0000 - val_loss: 0.1203 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00763: val_acc did not improve from 0.99514\n",
      "Epoch 764/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.8166e-04 - acc: 1.0000 - val_loss: 0.1203 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00764: val_acc did not improve from 0.99514\n",
      "Epoch 765/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.8241e-04 - acc: 1.0000 - val_loss: 0.1204 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00765: val_acc did not improve from 0.99514\n",
      "Epoch 766/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.7540e-04 - acc: 1.0000 - val_loss: 0.1203 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00766: val_acc did not improve from 0.99514\n",
      "Epoch 767/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.8036e-04 - acc: 1.0000 - val_loss: 0.1205 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00767: val_acc did not improve from 0.99514\n",
      "Epoch 768/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 30ms/step - loss: 4.7880e-04 - acc: 1.0000 - val_loss: 0.1206 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00768: val_acc did not improve from 0.99514\n",
      "Epoch 769/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.7550e-04 - acc: 1.0000 - val_loss: 0.1207 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00769: val_acc did not improve from 0.99514\n",
      "Epoch 770/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.7415e-04 - acc: 1.0000 - val_loss: 0.1207 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00770: val_acc did not improve from 0.99514\n",
      "Epoch 771/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.7245e-04 - acc: 1.0000 - val_loss: 0.1208 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00771: val_acc did not improve from 0.99514\n",
      "Epoch 772/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.7089e-04 - acc: 1.0000 - val_loss: 0.1209 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00772: val_acc did not improve from 0.99514\n",
      "Epoch 773/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.7122e-04 - acc: 1.0000 - val_loss: 0.1211 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00773: val_acc did not improve from 0.99514\n",
      "Epoch 774/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.6834e-04 - acc: 1.0000 - val_loss: 0.1211 - val_acc: 0.9893\n",
      "\n",
      "Epoch 00774: val_acc did not improve from 0.99514\n",
      "Epoch 775/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.6960e-04 - acc: 1.0000 - val_loss: 0.1213 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00775: val_acc did not improve from 0.99514\n",
      "Epoch 776/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.6812e-04 - acc: 1.0000 - val_loss: 0.1213 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00776: val_acc did not improve from 0.99514\n",
      "Epoch 777/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.6530e-04 - acc: 1.0000 - val_loss: 0.1214 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00777: val_acc did not improve from 0.99514\n",
      "Epoch 778/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.6435e-04 - acc: 1.0000 - val_loss: 0.1214 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00778: val_acc did not improve from 0.99514\n",
      "Epoch 779/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.6352e-04 - acc: 1.0000 - val_loss: 0.1216 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00779: val_acc did not improve from 0.99514\n",
      "Epoch 780/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.6293e-04 - acc: 1.0000 - val_loss: 0.1216 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00780: val_acc did not improve from 0.99514\n",
      "Epoch 781/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.5997e-04 - acc: 1.0000 - val_loss: 0.1216 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00781: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00781: ReduceLROnPlateau reducing learning rate to 4.882812731921149e-07.\n",
      "Epoch 782/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.5719e-04 - acc: 1.0000 - val_loss: 0.1217 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00782: val_acc did not improve from 0.99514\n",
      "Epoch 783/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.5660e-04 - acc: 1.0000 - val_loss: 0.1217 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00783: val_acc did not improve from 0.99514\n",
      "Epoch 784/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.5737e-04 - acc: 1.0000 - val_loss: 0.1219 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00784: val_acc did not improve from 0.99514\n",
      "Epoch 785/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.5508e-04 - acc: 1.0000 - val_loss: 0.1219 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00785: val_acc did not improve from 0.99514\n",
      "Epoch 786/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.5515e-04 - acc: 1.0000 - val_loss: 0.1220 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00786: val_acc did not improve from 0.99514\n",
      "Epoch 787/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.5449e-04 - acc: 1.0000 - val_loss: 0.1220 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00787: val_acc did not improve from 0.99514\n",
      "Epoch 788/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.5364e-04 - acc: 1.0000 - val_loss: 0.1221 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00788: val_acc did not improve from 0.99514\n",
      "Epoch 789/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.5262e-04 - acc: 1.0000 - val_loss: 0.1221 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00789: val_acc did not improve from 0.99514\n",
      "Epoch 790/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.5217e-04 - acc: 1.0000 - val_loss: 0.1222 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00790: val_acc did not improve from 0.99514\n",
      "Epoch 791/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.5102e-04 - acc: 1.0000 - val_loss: 0.1222 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00791: val_acc did not improve from 0.99514\n",
      "Epoch 792/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.5205e-04 - acc: 1.0000 - val_loss: 0.1223 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00792: val_acc did not improve from 0.99514\n",
      "Epoch 793/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.5051e-04 - acc: 1.0000 - val_loss: 0.1224 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00793: val_acc did not improve from 0.99514\n",
      "Epoch 794/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.5044e-04 - acc: 1.0000 - val_loss: 0.1224 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00794: val_acc did not improve from 0.99514\n",
      "Epoch 795/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.4914e-04 - acc: 1.0000 - val_loss: 0.1225 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00795: val_acc did not improve from 0.99514\n",
      "Epoch 796/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.4942e-04 - acc: 1.0000 - val_loss: 0.1225 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00796: val_acc did not improve from 0.99514\n",
      "Epoch 797/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.4770e-04 - acc: 1.0000 - val_loss: 0.1225 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00797: val_acc did not improve from 0.99514\n",
      "Epoch 798/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.4926e-04 - acc: 1.0000 - val_loss: 0.1225 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00798: val_acc did not improve from 0.99514\n",
      "Epoch 799/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.4641e-04 - acc: 1.0000 - val_loss: 0.1226 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00799: val_acc did not improve from 0.99514\n",
      "Epoch 800/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.4685e-04 - acc: 1.0000 - val_loss: 0.1226 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00800: val_acc did not improve from 0.99514\n",
      "Epoch 801/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.4527e-04 - acc: 1.0000 - val_loss: 0.1227 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00801: val_acc did not improve from 0.99514\n",
      "Epoch 802/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 4.4553e-04 - acc: 1.0000 - val_loss: 0.1227 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00802: val_acc did not improve from 0.99514\n",
      "Epoch 803/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 4.4487e-04 - acc: 1.0000 - val_loss: 0.1227 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00803: val_acc did not improve from 0.99514\n",
      "Epoch 804/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 4.4481e-04 - acc: 1.0000 - val_loss: 0.1228 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00804: val_acc did not improve from 0.99514\n",
      "Epoch 805/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 4.4512e-04 - acc: 1.0000 - val_loss: 0.1228 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00805: val_acc did not improve from 0.99514\n",
      "Epoch 806/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 4.4338e-04 - acc: 1.0000 - val_loss: 0.1229 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00806: val_acc did not improve from 0.99514\n",
      "Epoch 807/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.4324e-04 - acc: 1.0000 - val_loss: 0.1229 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00807: val_acc did not improve from 0.99514\n",
      "Epoch 808/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 4.4201e-04 - acc: 1.0000 - val_loss: 0.1230 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00808: val_acc did not improve from 0.99514\n",
      "Epoch 809/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 4.4192e-04 - acc: 1.0000 - val_loss: 0.1230 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00809: val_acc did not improve from 0.99514\n",
      "Epoch 810/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 5s 20ms/step - loss: 4.4104e-04 - acc: 1.0000 - val_loss: 0.1231 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00810: val_acc did not improve from 0.99514\n",
      "Epoch 811/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 4.3963e-04 - acc: 1.0000 - val_loss: 0.1231 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00811: val_acc did not improve from 0.99514\n",
      "Epoch 812/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.3951e-04 - acc: 1.0000 - val_loss: 0.1231 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00812: val_acc did not improve from 0.99514\n",
      "Epoch 813/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.3997e-04 - acc: 1.0000 - val_loss: 0.1232 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00813: val_acc did not improve from 0.99514\n",
      "Epoch 814/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 4.3936e-04 - acc: 1.0000 - val_loss: 0.1232 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00814: val_acc did not improve from 0.99514\n",
      "Epoch 815/1000\n",
      "236/236 [==============================] - 5s 19ms/step - loss: 4.3813e-04 - acc: 1.0000 - val_loss: 0.1233 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00815: val_acc did not improve from 0.99514\n",
      "Epoch 816/1000\n",
      "236/236 [==============================] - 5s 20ms/step - loss: 4.3771e-04 - acc: 1.0000 - val_loss: 0.1233 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00816: val_acc did not improve from 0.99514\n",
      "Epoch 817/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.3706e-04 - acc: 1.0000 - val_loss: 0.1233 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00817: val_acc did not improve from 0.99514\n",
      "Epoch 818/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.3691e-04 - acc: 1.0000 - val_loss: 0.1234 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00818: val_acc did not improve from 0.99514\n",
      "Epoch 819/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.3580e-04 - acc: 1.0000 - val_loss: 0.1234 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00819: val_acc did not improve from 0.99514\n",
      "Epoch 820/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.3530e-04 - acc: 1.0000 - val_loss: 0.1235 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00820: val_acc did not improve from 0.99514\n",
      "Epoch 821/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.3526e-04 - acc: 1.0000 - val_loss: 0.1236 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00821: val_acc did not improve from 0.99514\n",
      "Epoch 822/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.3382e-04 - acc: 1.0000 - val_loss: 0.1236 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00822: val_acc did not improve from 0.99514\n",
      "Epoch 823/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.3441e-04 - acc: 1.0000 - val_loss: 0.1236 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00823: val_acc did not improve from 0.99514\n",
      "Epoch 824/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.3353e-04 - acc: 1.0000 - val_loss: 0.1237 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00824: val_acc did not improve from 0.99514\n",
      "Epoch 825/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.3271e-04 - acc: 1.0000 - val_loss: 0.1237 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00825: val_acc did not improve from 0.99514\n",
      "Epoch 826/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 4.3281e-04 - acc: 1.0000 - val_loss: 0.1238 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00826: val_acc did not improve from 0.99514\n",
      "Epoch 827/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.3200e-04 - acc: 1.0000 - val_loss: 0.1238 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00827: val_acc did not improve from 0.99514\n",
      "Epoch 828/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.3048e-04 - acc: 1.0000 - val_loss: 0.1239 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00828: val_acc did not improve from 0.99514\n",
      "Epoch 829/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.3066e-04 - acc: 1.0000 - val_loss: 0.1239 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00829: val_acc did not improve from 0.99514\n",
      "Epoch 830/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.3052e-04 - acc: 1.0000 - val_loss: 0.1239 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00830: val_acc did not improve from 0.99514\n",
      "Epoch 831/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.3000e-04 - acc: 1.0000 - val_loss: 0.1240 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00831: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00831: ReduceLROnPlateau reducing learning rate to 2.4414063659605745e-07.\n",
      "Epoch 832/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2776e-04 - acc: 1.0000 - val_loss: 0.1240 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00832: val_acc did not improve from 0.99514\n",
      "Epoch 833/1000\n",
      "236/236 [==============================] - 6s 25ms/step - loss: 4.2739e-04 - acc: 1.0000 - val_loss: 0.1241 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00833: val_acc did not improve from 0.99514\n",
      "Epoch 834/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.2726e-04 - acc: 1.0000 - val_loss: 0.1241 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00834: val_acc did not improve from 0.99514\n",
      "Epoch 835/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.2669e-04 - acc: 1.0000 - val_loss: 0.1241 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00835: val_acc did not improve from 0.99514\n",
      "Epoch 836/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2671e-04 - acc: 1.0000 - val_loss: 0.1242 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00836: val_acc did not improve from 0.99514\n",
      "Epoch 837/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2634e-04 - acc: 1.0000 - val_loss: 0.1242 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00837: val_acc did not improve from 0.99514\n",
      "Epoch 838/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2598e-04 - acc: 1.0000 - val_loss: 0.1242 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00838: val_acc did not improve from 0.99514\n",
      "Epoch 839/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.2594e-04 - acc: 1.0000 - val_loss: 0.1243 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00839: val_acc did not improve from 0.99514\n",
      "Epoch 840/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.2619e-04 - acc: 1.0000 - val_loss: 0.1243 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00840: val_acc did not improve from 0.99514\n",
      "Epoch 841/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2520e-04 - acc: 1.0000 - val_loss: 0.1243 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00841: val_acc did not improve from 0.99514\n",
      "Epoch 842/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.2492e-04 - acc: 1.0000 - val_loss: 0.1243 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00842: val_acc did not improve from 0.99514\n",
      "Epoch 843/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.2454e-04 - acc: 1.0000 - val_loss: 0.1243 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00843: val_acc did not improve from 0.99514\n",
      "Epoch 844/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.2468e-04 - acc: 1.0000 - val_loss: 0.1244 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00844: val_acc did not improve from 0.99514\n",
      "Epoch 845/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.2417e-04 - acc: 1.0000 - val_loss: 0.1243 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00845: val_acc did not improve from 0.99514\n",
      "Epoch 846/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.2364e-04 - acc: 1.0000 - val_loss: 0.1244 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00846: val_acc did not improve from 0.99514\n",
      "Epoch 847/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.2369e-04 - acc: 1.0000 - val_loss: 0.1244 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00847: val_acc did not improve from 0.99514\n",
      "Epoch 848/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2376e-04 - acc: 1.0000 - val_loss: 0.1245 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00848: val_acc did not improve from 0.99514\n",
      "Epoch 849/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.2335e-04 - acc: 1.0000 - val_loss: 0.1245 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00849: val_acc did not improve from 0.99514\n",
      "Epoch 850/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.2284e-04 - acc: 1.0000 - val_loss: 0.1245 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00850: val_acc did not improve from 0.99514\n",
      "Epoch 851/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.2312e-04 - acc: 1.0000 - val_loss: 0.1245 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00851: val_acc did not improve from 0.99514\n",
      "Epoch 852/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 31ms/step - loss: 4.2217e-04 - acc: 1.0000 - val_loss: 0.1246 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00852: val_acc did not improve from 0.99514\n",
      "Epoch 853/1000\n",
      "236/236 [==============================] - 6s 27ms/step - loss: 4.2188e-04 - acc: 1.0000 - val_loss: 0.1246 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00853: val_acc did not improve from 0.99514\n",
      "Epoch 854/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2165e-04 - acc: 1.0000 - val_loss: 0.1245 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00854: val_acc did not improve from 0.99514\n",
      "Epoch 855/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.2173e-04 - acc: 1.0000 - val_loss: 0.1247 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00855: val_acc did not improve from 0.99514\n",
      "Epoch 856/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 4.2112e-04 - acc: 1.0000 - val_loss: 0.1247 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00856: val_acc did not improve from 0.99514\n",
      "Epoch 857/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.2097e-04 - acc: 1.0000 - val_loss: 0.1247 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00857: val_acc did not improve from 0.99514\n",
      "Epoch 858/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.2064e-04 - acc: 1.0000 - val_loss: 0.1248 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00858: val_acc did not improve from 0.99514\n",
      "Epoch 859/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2047e-04 - acc: 1.0000 - val_loss: 0.1248 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00859: val_acc did not improve from 0.99514\n",
      "Epoch 860/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.2029e-04 - acc: 1.0000 - val_loss: 0.1248 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00860: val_acc did not improve from 0.99514\n",
      "Epoch 861/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1992e-04 - acc: 1.0000 - val_loss: 0.1249 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00861: val_acc did not improve from 0.99514\n",
      "Epoch 862/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1946e-04 - acc: 1.0000 - val_loss: 0.1249 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00862: val_acc did not improve from 0.99514\n",
      "Epoch 863/1000\n",
      "236/236 [==============================] - 14s 59ms/step - loss: 4.1927e-04 - acc: 1.0000 - val_loss: 0.1250 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00863: val_acc did not improve from 0.99514\n",
      "Epoch 864/1000\n",
      "236/236 [==============================] - 11s 48ms/step - loss: 4.1904e-04 - acc: 1.0000 - val_loss: 0.1250 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00864: val_acc did not improve from 0.99514\n",
      "Epoch 865/1000\n",
      "236/236 [==============================] - 9s 36ms/step - loss: 4.1877e-04 - acc: 1.0000 - val_loss: 0.1250 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00865: val_acc did not improve from 0.99514\n",
      "Epoch 866/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1861e-04 - acc: 1.0000 - val_loss: 0.1251 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00866: val_acc did not improve from 0.99514\n",
      "Epoch 867/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1834e-04 - acc: 1.0000 - val_loss: 0.1251 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00867: val_acc did not improve from 0.99514\n",
      "Epoch 868/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.1813e-04 - acc: 1.0000 - val_loss: 0.1251 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00868: val_acc did not improve from 0.99514\n",
      "Epoch 869/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.1762e-04 - acc: 1.0000 - val_loss: 0.1251 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00869: val_acc did not improve from 0.99514\n",
      "Epoch 870/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.1770e-04 - acc: 1.0000 - val_loss: 0.1251 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00870: val_acc did not improve from 0.99514\n",
      "Epoch 871/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1689e-04 - acc: 1.0000 - val_loss: 0.1251 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00871: val_acc did not improve from 0.99514\n",
      "Epoch 872/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 4.1720e-04 - acc: 1.0000 - val_loss: 0.1252 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00872: val_acc did not improve from 0.99514\n",
      "Epoch 873/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1665e-04 - acc: 1.0000 - val_loss: 0.1252 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00873: val_acc did not improve from 0.99514\n",
      "Epoch 874/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.1640e-04 - acc: 1.0000 - val_loss: 0.1253 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00874: val_acc did not improve from 0.99514\n",
      "Epoch 875/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1586e-04 - acc: 1.0000 - val_loss: 0.1252 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00875: val_acc did not improve from 0.99514\n",
      "Epoch 876/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1624e-04 - acc: 1.0000 - val_loss: 0.1253 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00876: val_acc did not improve from 0.99514\n",
      "Epoch 877/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.1613e-04 - acc: 1.0000 - val_loss: 0.1254 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00877: val_acc did not improve from 0.99514\n",
      "Epoch 878/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.1528e-04 - acc: 1.0000 - val_loss: 0.1253 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00878: val_acc did not improve from 0.99514\n",
      "Epoch 879/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1543e-04 - acc: 1.0000 - val_loss: 0.1253 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00879: val_acc did not improve from 0.99514\n",
      "Epoch 880/1000\n",
      "236/236 [==============================] - 7s 32ms/step - loss: 4.1514e-04 - acc: 1.0000 - val_loss: 0.1254 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00880: val_acc did not improve from 0.99514\n",
      "Epoch 881/1000\n",
      "236/236 [==============================] - 9s 38ms/step - loss: 4.1494e-04 - acc: 1.0000 - val_loss: 0.1254 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00881: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00881: ReduceLROnPlateau reducing learning rate to 1.2207031829802872e-07.\n",
      "Epoch 882/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.1377e-04 - acc: 1.0000 - val_loss: 0.1254 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00882: val_acc did not improve from 0.99514\n",
      "Epoch 883/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.1362e-04 - acc: 1.0000 - val_loss: 0.1255 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00883: val_acc did not improve from 0.99514\n",
      "Epoch 884/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1365e-04 - acc: 1.0000 - val_loss: 0.1255 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00884: val_acc did not improve from 0.99514\n",
      "Epoch 885/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1334e-04 - acc: 1.0000 - val_loss: 0.1255 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00885: val_acc did not improve from 0.99514\n",
      "Epoch 886/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1322e-04 - acc: 1.0000 - val_loss: 0.1255 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00886: val_acc did not improve from 0.99514\n",
      "Epoch 887/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.1323e-04 - acc: 1.0000 - val_loss: 0.1256 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00887: val_acc did not improve from 0.99514\n",
      "Epoch 888/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.1287e-04 - acc: 1.0000 - val_loss: 0.1256 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00888: val_acc did not improve from 0.99514\n",
      "Epoch 889/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.1284e-04 - acc: 1.0000 - val_loss: 0.1256 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00889: val_acc did not improve from 0.99514\n",
      "Epoch 890/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.1280e-04 - acc: 1.0000 - val_loss: 0.1256 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00890: val_acc did not improve from 0.99514\n",
      "Epoch 891/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1257e-04 - acc: 1.0000 - val_loss: 0.1256 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00891: val_acc did not improve from 0.99514\n",
      "Epoch 892/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.1239e-04 - acc: 1.0000 - val_loss: 0.1256 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00892: val_acc did not improve from 0.99514\n",
      "Epoch 893/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.1222e-04 - acc: 1.0000 - val_loss: 0.1257 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00893: val_acc did not improve from 0.99514\n",
      "Epoch 894/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 8s 34ms/step - loss: 4.1229e-04 - acc: 1.0000 - val_loss: 0.1256 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00894: val_acc did not improve from 0.99514\n",
      "Epoch 895/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.1218e-04 - acc: 1.0000 - val_loss: 0.1257 - val_acc: 0.9883\n",
      "\n",
      "Epoch 00895: val_acc did not improve from 0.99514\n",
      "Epoch 896/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.1198e-04 - acc: 1.0000 - val_loss: 0.1257 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00896: val_acc did not improve from 0.99514\n",
      "Epoch 897/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1171e-04 - acc: 1.0000 - val_loss: 0.1258 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00897: val_acc did not improve from 0.99514\n",
      "Epoch 898/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.1164e-04 - acc: 1.0000 - val_loss: 0.1258 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00898: val_acc did not improve from 0.99514\n",
      "Epoch 899/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1147e-04 - acc: 1.0000 - val_loss: 0.1258 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00899: val_acc did not improve from 0.99514\n",
      "Epoch 900/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.1152e-04 - acc: 1.0000 - val_loss: 0.1258 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00900: val_acc did not improve from 0.99514\n",
      "Epoch 901/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.1140e-04 - acc: 1.0000 - val_loss: 0.1258 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00901: val_acc did not improve from 0.99514\n",
      "Epoch 902/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1128e-04 - acc: 1.0000 - val_loss: 0.1258 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00902: val_acc did not improve from 0.99514\n",
      "Epoch 903/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.1106e-04 - acc: 1.0000 - val_loss: 0.1258 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00903: val_acc did not improve from 0.99514\n",
      "Epoch 904/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 4.1103e-04 - acc: 1.0000 - val_loss: 0.1259 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00904: val_acc did not improve from 0.99514\n",
      "Epoch 905/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.1067e-04 - acc: 1.0000 - val_loss: 0.1259 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00905: val_acc did not improve from 0.99514\n",
      "Epoch 906/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 4.1074e-04 - acc: 1.0000 - val_loss: 0.1259 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00906: val_acc did not improve from 0.99514\n",
      "Epoch 907/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 4.1055e-04 - acc: 1.0000 - val_loss: 0.1259 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00907: val_acc did not improve from 0.99514\n",
      "Epoch 908/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.1030e-04 - acc: 1.0000 - val_loss: 0.1259 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00908: val_acc did not improve from 0.99514\n",
      "Epoch 909/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.1031e-04 - acc: 1.0000 - val_loss: 0.1259 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00909: val_acc did not improve from 0.99514\n",
      "Epoch 910/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.1010e-04 - acc: 1.0000 - val_loss: 0.1260 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00910: val_acc did not improve from 0.99514\n",
      "Epoch 911/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0997e-04 - acc: 1.0000 - val_loss: 0.1260 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00911: val_acc did not improve from 0.99514\n",
      "Epoch 912/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.1011e-04 - acc: 1.0000 - val_loss: 0.1260 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00912: val_acc did not improve from 0.99514\n",
      "Epoch 913/1000\n",
      "236/236 [==============================] - 9s 37ms/step - loss: 4.0985e-04 - acc: 1.0000 - val_loss: 0.1261 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00913: val_acc did not improve from 0.99514\n",
      "Epoch 914/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0970e-04 - acc: 1.0000 - val_loss: 0.1260 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00914: val_acc did not improve from 0.99514\n",
      "Epoch 915/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0957e-04 - acc: 1.0000 - val_loss: 0.1260 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00915: val_acc did not improve from 0.99514\n",
      "Epoch 916/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.0946e-04 - acc: 1.0000 - val_loss: 0.1261 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00916: val_acc did not improve from 0.99514\n",
      "Epoch 917/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.0914e-04 - acc: 1.0000 - val_loss: 0.1261 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00917: val_acc did not improve from 0.99514\n",
      "Epoch 918/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.0900e-04 - acc: 1.0000 - val_loss: 0.1261 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00918: val_acc did not improve from 0.99514\n",
      "Epoch 919/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.0903e-04 - acc: 1.0000 - val_loss: 0.1261 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00919: val_acc did not improve from 0.99514\n",
      "Epoch 920/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0898e-04 - acc: 1.0000 - val_loss: 0.1261 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00920: val_acc did not improve from 0.99514\n",
      "Epoch 921/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.0873e-04 - acc: 1.0000 - val_loss: 0.1262 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00921: val_acc did not improve from 0.99514\n",
      "Epoch 922/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0868e-04 - acc: 1.0000 - val_loss: 0.1262 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00922: val_acc did not improve from 0.99514\n",
      "Epoch 923/1000\n",
      "236/236 [==============================] - 7s 28ms/step - loss: 4.0847e-04 - acc: 1.0000 - val_loss: 0.1262 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00923: val_acc did not improve from 0.99514\n",
      "Epoch 924/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.0837e-04 - acc: 1.0000 - val_loss: 0.1262 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00924: val_acc did not improve from 0.99514\n",
      "Epoch 925/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0853e-04 - acc: 1.0000 - val_loss: 0.1262 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00925: val_acc did not improve from 0.99514\n",
      "Epoch 926/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0809e-04 - acc: 1.0000 - val_loss: 0.1262 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00926: val_acc did not improve from 0.99514\n",
      "Epoch 927/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0811e-04 - acc: 1.0000 - val_loss: 0.1262 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00927: val_acc did not improve from 0.99514\n",
      "Epoch 928/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.0781e-04 - acc: 1.0000 - val_loss: 0.1263 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00928: val_acc did not improve from 0.99514\n",
      "Epoch 929/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0792e-04 - acc: 1.0000 - val_loss: 0.1262 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00929: val_acc did not improve from 0.99514\n",
      "Epoch 930/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0751e-04 - acc: 1.0000 - val_loss: 0.1263 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00930: val_acc did not improve from 0.99514\n",
      "Epoch 931/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.0739e-04 - acc: 1.0000 - val_loss: 0.1263 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00931: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00931: ReduceLROnPlateau reducing learning rate to 6.103515914901436e-08.\n",
      "Epoch 932/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0699e-04 - acc: 1.0000 - val_loss: 0.1263 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00932: val_acc did not improve from 0.99514\n",
      "Epoch 933/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.0683e-04 - acc: 1.0000 - val_loss: 0.1263 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00933: val_acc did not improve from 0.99514\n",
      "Epoch 934/1000\n",
      "236/236 [==============================] - 7s 29ms/step - loss: 4.0685e-04 - acc: 1.0000 - val_loss: 0.1263 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00934: val_acc did not improve from 0.99514\n",
      "Epoch 935/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.0676e-04 - acc: 1.0000 - val_loss: 0.1263 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00935: val_acc did not improve from 0.99514\n",
      "Epoch 936/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0675e-04 - acc: 1.0000 - val_loss: 0.1263 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00936: val_acc did not improve from 0.99514\n",
      "Epoch 937/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0673e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00937: val_acc did not improve from 0.99514\n",
      "Epoch 938/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0667e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00938: val_acc did not improve from 0.99514\n",
      "Epoch 939/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0649e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00939: val_acc did not improve from 0.99514\n",
      "Epoch 940/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0652e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00940: val_acc did not improve from 0.99514\n",
      "Epoch 941/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0658e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00941: val_acc did not improve from 0.99514\n",
      "Epoch 942/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.0636e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00942: val_acc did not improve from 0.99514\n",
      "Epoch 943/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0635e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00943: val_acc did not improve from 0.99514\n",
      "Epoch 944/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0630e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00944: val_acc did not improve from 0.99514\n",
      "Epoch 945/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0617e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00945: val_acc did not improve from 0.99514\n",
      "Epoch 946/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0612e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00946: val_acc did not improve from 0.99514\n",
      "Epoch 947/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0612e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00947: val_acc did not improve from 0.99514\n",
      "Epoch 948/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0600e-04 - acc: 1.0000 - val_loss: 0.1264 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00948: val_acc did not improve from 0.99514\n",
      "Epoch 949/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0603e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00949: val_acc did not improve from 0.99514\n",
      "Epoch 950/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.0585e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00950: val_acc did not improve from 0.99514\n",
      "Epoch 951/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0591e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00951: val_acc did not improve from 0.99514\n",
      "Epoch 952/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0577e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00952: val_acc did not improve from 0.99514\n",
      "Epoch 953/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0569e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00953: val_acc did not improve from 0.99514\n",
      "Epoch 954/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0564e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00954: val_acc did not improve from 0.99514\n",
      "Epoch 955/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0557e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00955: val_acc did not improve from 0.99514\n",
      "Epoch 956/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0552e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00956: val_acc did not improve from 0.99514\n",
      "Epoch 957/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0545e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00957: val_acc did not improve from 0.99514\n",
      "Epoch 958/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0536e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00958: val_acc did not improve from 0.99514\n",
      "Epoch 959/1000\n",
      "236/236 [==============================] - 7s 31ms/step - loss: 4.0534e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00959: val_acc did not improve from 0.99514\n",
      "Epoch 960/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0527e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00960: val_acc did not improve from 0.99514\n",
      "Epoch 961/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0522e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00961: val_acc did not improve from 0.99514\n",
      "Epoch 962/1000\n",
      "236/236 [==============================] - 9s 36ms/step - loss: 4.0511e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00962: val_acc did not improve from 0.99514\n",
      "Epoch 963/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0509e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00963: val_acc did not improve from 0.99514\n",
      "Epoch 964/1000\n",
      "236/236 [==============================] - 9s 37ms/step - loss: 4.0507e-04 - acc: 1.0000 - val_loss: 0.1265 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00964: val_acc did not improve from 0.99514\n",
      "Epoch 965/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0492e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00965: val_acc did not improve from 0.99514\n",
      "Epoch 966/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.0486e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00966: val_acc did not improve from 0.99514\n",
      "Epoch 967/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.0484e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00967: val_acc did not improve from 0.99514\n",
      "Epoch 968/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 4.0469e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874c: 1.\n",
      "\n",
      "Epoch 00968: val_acc did not improve from 0.99514\n",
      "Epoch 969/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0480e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00969: val_acc did not improve from 0.99514\n",
      "Epoch 970/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.0469e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00970: val_acc did not improve from 0.99514\n",
      "Epoch 971/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.0461e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00971: val_acc did not improve from 0.99514\n",
      "Epoch 972/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0456e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00972: val_acc did not improve from 0.99514\n",
      "Epoch 973/1000\n",
      "236/236 [==============================] - 8s 32ms/step - loss: 4.0440e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00973: val_acc did not improve from 0.99514\n",
      "Epoch 974/1000\n",
      "236/236 [==============================] - 9s 37ms/step - loss: 4.0437e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00974: val_acc did not improve from 0.99514\n",
      "Epoch 975/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.0439e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00975: val_acc did not improve from 0.99514\n",
      "Epoch 976/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 4.0433e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00976: val_acc did not improve from 0.99514\n",
      "Epoch 977/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.0419e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00977: val_acc did not improve from 0.99514\n",
      "Epoch 978/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.0420e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Epoch 00978: val_acc did not improve from 0.99514\n",
      "Epoch 979/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 4.0411e-04 - acc: 1.0000 - val_loss: 0.1266 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00979: val_acc did not improve from 0.99514\n",
      "Epoch 980/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.0407e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00980: val_acc did not improve from 0.99514\n",
      "Epoch 981/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 4.0397e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00981: val_acc did not improve from 0.99514\n",
      "\n",
      "Epoch 00981: ReduceLROnPlateau reducing learning rate to 3.051757957450718e-08.\n",
      "Epoch 982/1000\n",
      "236/236 [==============================] - 5s 23ms/step - loss: 4.0373e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00982: val_acc did not improve from 0.99514\n",
      "Epoch 983/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.0366e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00983: val_acc did not improve from 0.99514\n",
      "Epoch 984/1000\n",
      "236/236 [==============================] - 6s 26ms/step - loss: 4.0366e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00984: val_acc did not improve from 0.99514\n",
      "Epoch 985/1000\n",
      "236/236 [==============================] - 5s 22ms/step - loss: 4.0363e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00985: val_acc did not improve from 0.99514\n",
      "Epoch 986/1000\n",
      "236/236 [==============================] - 5s 21ms/step - loss: 4.0358e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00986: val_acc did not improve from 0.99514\n",
      "Epoch 987/1000\n",
      "236/236 [==============================] - 7s 30ms/step - loss: 4.0352e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00987: val_acc did not improve from 0.99514\n",
      "Epoch 988/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0357e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00988: val_acc did not improve from 0.99514\n",
      "Epoch 989/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0350e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00989: val_acc did not improve from 0.99514\n",
      "Epoch 990/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0346e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00990: val_acc did not improve from 0.99514\n",
      "Epoch 991/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0347e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00991: val_acc did not improve from 0.99514\n",
      "Epoch 992/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0339e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00992: val_acc did not improve from 0.99514\n",
      "Epoch 993/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0337e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00993: val_acc did not improve from 0.99514\n",
      "Epoch 994/1000\n",
      "236/236 [==============================] - 9s 36ms/step - loss: 4.0337e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00994: val_acc did not improve from 0.99514\n",
      "Epoch 995/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0329e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00995: val_acc did not improve from 0.99514\n",
      "Epoch 996/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0331e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00996: val_acc did not improve from 0.99514\n",
      "Epoch 997/1000\n",
      "236/236 [==============================] - 8s 34ms/step - loss: 4.0332e-04 - acc: 1.0000 - val_loss: 0.1267 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00997: val_acc did not improve from 0.99514\n",
      "Epoch 998/1000\n",
      "236/236 [==============================] - 8s 35ms/step - loss: 4.0326e-04 - acc: 1.0000 - val_loss: 0.1268 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00998: val_acc did not improve from 0.99514\n",
      "Epoch 999/1000\n",
      "236/236 [==============================] - 8s 33ms/step - loss: 4.0323e-04 - acc: 1.0000 - val_loss: 0.1268 - val_acc: 0.9874\n",
      "\n",
      "Epoch 00999: val_acc did not improve from 0.99514\n",
      "Epoch 1000/1000\n",
      "236/236 [==============================] - 8s 36ms/step - loss: 4.0319e-04 - acc: 1.0000 - val_loss: 0.1268 - val_acc: 0.9874\n",
      "\n",
      "Epoch 01000: val_acc did not improve from 0.99514\n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras.callbacks import ModelCheckpoint, ReduceLROnPlateau\n",
    "import time\n",
    "created_time = int(time.time())\n",
    "history = model.fit(\n",
    "    x_train,\n",
    "    y_train,\n",
    "    validation_data=(x_val, y_val),\n",
    "    epochs=1000,\n",
    "    callbacks=[\n",
    "        ModelCheckpoint('models/model_{created_time}.h5', monitor='val_loss', verbose=1, save_best_only=True, mode='auto'),\n",
    "        ReduceLROnPlateau(monitor='val_acc', factor=0.5, patience=50, verbose=1, mode='auto')\n",
    "    ]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1152x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig, loss_ax = plt.subplots(figsize=(16, 10))\n",
    "acc_ax = loss_ax.twinx()\n",
    "\n",
    "loss_ax.plot(history.history['loss'], 'y', label='train loss')\n",
    "loss_ax.plot(history.history['val_loss'], 'r', label='val loss')\n",
    "loss_ax.set_xlabel('epoch')\n",
    "loss_ax.set_ylabel('loss')\n",
    "loss_ax.legend(loc='upper left')\n",
    "\n",
    "acc_ax.plot(history.history['acc'], 'b', label='train acc')\n",
    "acc_ax.plot(history.history['val_acc'], 'g', label='val acc')\n",
    "acc_ax.set_ylabel('accuracy')\n",
    "acc_ax.legend(loc='upper right')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[742,   0],\n",
       "        [  5, 281]],\n",
       "\n",
       "       [[649,   5],\n",
       "        [  0, 374]],\n",
       "\n",
       "       [[660,   0],\n",
       "        [  0, 368]]], dtype=int64)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import multilabel_confusion_matrix\n",
    "from tensorflow.keras.models import load_model\n",
    "\n",
    "model = load_model('models/model_1648573934.h5')\n",
    "\n",
    "y_pred = model.predict(x_val)\n",
    "\n",
    "multilabel_confusion_matrix(np.argmax(y_val, axis=1), np.argmax(y_pred, axis=1))"
   ]
  },
  {
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}