2장. 4절. 케라스를 이용한 와인데이터 분류용 DNN 구현.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5ee650ad",
   "metadata": {},
   "source": [
    "## DNN을 이용한 와인 등급 분류"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1f19942c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-11T12:07:15.192754Z",
     "start_time": "2021-06-11T12:06:41.665375Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.2394 - loss: 2.1112\n",
      "Epoch 2/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.4332 - loss: 1.4291 \n",
      "Epoch 3/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4524 - loss: 1.2939 \n",
      "Epoch 4/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.4589 - loss: 1.3092 \n",
      "Epoch 5/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.4598 - loss: 1.2527 \n",
      "Epoch 6/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4750 - loss: 1.1922 \n",
      "Epoch 7/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.4811 - loss: 1.1924 \n",
      "Epoch 8/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4660 - loss: 1.2213 \n",
      "Epoch 9/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4730 - loss: 1.2062 \n",
      "Epoch 10/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.4608 - loss: 1.1988 \n",
      "Epoch 11/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.4985 - loss: 1.1496 \n",
      "Epoch 12/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4791 - loss: 1.1738 \n",
      "Epoch 13/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.4942 - loss: 1.1819 \n",
      "Epoch 14/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4779 - loss: 1.1834 \n",
      "Epoch 15/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4923 - loss: 1.1797 \n",
      "Epoch 16/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4737 - loss: 1.2024 \n",
      "Epoch 17/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4716 - loss: 1.1816 \n",
      "Epoch 18/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.4821 - loss: 1.1649 \n",
      "Epoch 19/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5003 - loss: 1.1686 \n",
      "Epoch 20/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4856 - loss: 1.1671 \n",
      "Epoch 21/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5015 - loss: 1.1725 \n",
      "Epoch 22/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.4973 - loss: 1.1746 \n",
      "Epoch 23/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.4792 - loss: 1.1373 \n",
      "Epoch 24/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5113 - loss: 1.1137 \n",
      "Epoch 25/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5145 - loss: 1.1295 \n",
      "Epoch 26/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.4889 - loss: 1.1542 \n",
      "Epoch 27/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.4995 - loss: 1.1705 \n",
      "Epoch 28/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.4835 - loss: 1.1418 \n",
      "Epoch 29/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5118 - loss: 1.1372 \n",
      "Epoch 30/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5362 - loss: 1.1297 \n",
      "Epoch 31/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5251 - loss: 1.1391 \n",
      "Epoch 32/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5033 - loss: 1.1047 \n",
      "Epoch 33/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5190 - loss: 1.1255 \n",
      "Epoch 34/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5096 - loss: 1.1476 \n",
      "Epoch 35/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5041 - loss: 1.1069 \n",
      "Epoch 36/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5086 - loss: 1.1139 \n",
      "Epoch 37/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5185 - loss: 1.1300 \n",
      "Epoch 38/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5118 - loss: 1.1031 \n",
      "Epoch 39/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.4996 - loss: 1.1262 \n",
      "Epoch 40/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.5172 - loss: 1.0969 \n",
      "Epoch 41/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5158 - loss: 1.0673 \n",
      "Epoch 42/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5238 - loss: 1.1065 \n",
      "Epoch 43/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5365 - loss: 1.1220 \n",
      "Epoch 44/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5384 - loss: 1.0594 \n",
      "Epoch 45/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5138 - loss: 1.1079 \n",
      "Epoch 46/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5170 - loss: 1.0966 \n",
      "Epoch 47/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5344 - loss: 1.0629 \n",
      "Epoch 48/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5071 - loss: 1.1026 \n",
      "Epoch 49/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.4966 - loss: 1.1075 \n",
      "Epoch 50/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5257 - loss: 1.0438 \n",
      "Epoch 51/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5387 - loss: 1.0574 \n",
      "Epoch 52/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5332 - loss: 1.0989 \n",
      "Epoch 53/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5688 - loss: 1.0287 \n",
      "Epoch 54/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5265 - loss: 1.0529 \n",
      "Epoch 55/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5323 - loss: 1.0515 \n",
      "Epoch 56/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5598 - loss: 1.0516 \n",
      "Epoch 57/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5371 - loss: 1.0621 \n",
      "Epoch 58/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5685 - loss: 1.0515 \n",
      "Epoch 59/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5647 - loss: 1.0391 \n",
      "Epoch 60/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5424 - loss: 1.0539 \n",
      "Epoch 61/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5397 - loss: 1.0504 \n",
      "Epoch 62/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5354 - loss: 1.0723 \n",
      "Epoch 63/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5478 - loss: 1.0472 \n",
      "Epoch 64/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5162 - loss: 1.0503 \n",
      "Epoch 65/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5267 - loss: 1.0586 \n",
      "Epoch 66/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5510 - loss: 1.0356 \n",
      "Epoch 67/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5643 - loss: 1.0301 \n",
      "Epoch 68/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5584 - loss: 1.0358 \n",
      "Epoch 69/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5468 - loss: 1.0355 \n",
      "Epoch 70/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5418 - loss: 1.0333 \n",
      "Epoch 71/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5664 - loss: 1.0203 \n",
      "Epoch 72/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5679 - loss: 1.0372 \n",
      "Epoch 73/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5135 - loss: 1.0473 \n",
      "Epoch 74/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5257 - loss: 1.0634 \n",
      "Epoch 75/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5341 - loss: 1.0495 \n",
      "Epoch 76/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5647 - loss: 1.0098 \n",
      "Epoch 77/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5692 - loss: 0.9971 \n",
      "Epoch 78/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5365 - loss: 1.0847 \n",
      "Epoch 79/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5561 - loss: 1.0381 \n",
      "Epoch 80/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5781 - loss: 1.0055 \n",
      "Epoch 81/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5707 - loss: 1.0335 \n",
      "Epoch 82/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5347 - loss: 0.9961 \n",
      "Epoch 83/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5721 - loss: 1.0284 \n",
      "Epoch 84/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5660 - loss: 0.9951 \n",
      "Epoch 85/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5608 - loss: 1.0335 \n",
      "Epoch 86/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5395 - loss: 1.0129 \n",
      "Epoch 87/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5523 - loss: 1.0195 \n",
      "Epoch 88/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5553 - loss: 1.0330 \n",
      "Epoch 89/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5466 - loss: 1.0316 \n",
      "Epoch 90/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5461 - loss: 1.0007 \n",
      "Epoch 91/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5493 - loss: 1.0373 \n",
      "Epoch 92/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5446 - loss: 1.0241 \n",
      "Epoch 93/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5442 - loss: 1.0107 \n",
      "Epoch 94/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5423 - loss: 1.0066 \n",
      "Epoch 95/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5438 - loss: 1.0239 \n",
      "Epoch 96/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5588 - loss: 1.0199 \n",
      "Epoch 97/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5586 - loss: 0.9984 \n",
      "Epoch 98/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5614 - loss: 1.0306 \n",
      "Epoch 99/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5607 - loss: 0.9888 \n",
      "Epoch 100/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5798 - loss: 0.9850 \n",
      "Epoch 101/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.5510 - loss: 0.9780 \n",
      "Epoch 102/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5642 - loss: 0.9989 \n",
      "Epoch 103/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5679 - loss: 0.9947 \n",
      "Epoch 104/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5586 - loss: 1.0111 \n",
      "Epoch 105/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5717 - loss: 0.9839 \n",
      "Epoch 106/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5769 - loss: 0.9899 \n",
      "Epoch 107/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5707 - loss: 1.0171 \n",
      "Epoch 108/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5718 - loss: 0.9771 \n",
      "Epoch 109/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5594 - loss: 0.9772 \n",
      "Epoch 110/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5662 - loss: 1.0114 \n",
      "Epoch 111/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5451 - loss: 0.9841 \n",
      "Epoch 112/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5709 - loss: 0.9978 \n",
      "Epoch 113/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5526 - loss: 0.9674 \n",
      "Epoch 114/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5744 - loss: 1.0108 \n",
      "Epoch 115/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5740 - loss: 0.9829 \n",
      "Epoch 116/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5613 - loss: 0.9822 \n",
      "Epoch 117/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5763 - loss: 0.9834 \n",
      "Epoch 118/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5673 - loss: 0.9990 \n",
      "Epoch 119/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5574 - loss: 0.9911 \n",
      "Epoch 120/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5573 - loss: 1.0017 \n",
      "Epoch 121/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5740 - loss: 0.9970 \n",
      "Epoch 122/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5851 - loss: 0.9863 \n",
      "Epoch 123/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5724 - loss: 0.9602 \n",
      "Epoch 124/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5950 - loss: 0.9849 \n",
      "Epoch 125/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5890 - loss: 0.9623 \n",
      "Epoch 126/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5860 - loss: 0.9423 \n",
      "Epoch 127/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5711 - loss: 0.9736 \n",
      "Epoch 128/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5678 - loss: 0.9822 \n",
      "Epoch 129/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5876 - loss: 0.9694 \n",
      "Epoch 130/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5710 - loss: 0.9750 \n",
      "Epoch 131/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5429 - loss: 0.9898 \n",
      "Epoch 132/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5587 - loss: 0.9770 \n",
      "Epoch 133/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5576 - loss: 0.9506 \n",
      "Epoch 134/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5713 - loss: 0.9666 \n",
      "Epoch 135/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5814 - loss: 0.9700 \n",
      "Epoch 136/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5798 - loss: 0.9568 \n",
      "Epoch 137/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5829 - loss: 0.9595 \n",
      "Epoch 138/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5433 - loss: 0.9939 \n",
      "Epoch 139/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5678 - loss: 0.9825 \n",
      "Epoch 140/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5705 - loss: 0.9805 \n",
      "Epoch 141/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5874 - loss: 0.9524 \n",
      "Epoch 142/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5741 - loss: 0.9646 \n",
      "Epoch 143/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 13ms/step - accuracy: 0.5755 - loss: 0.9768\n",
      "Epoch 144/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5918 - loss: 0.9624 \n",
      "Epoch 145/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5787 - loss: 0.9654 \n",
      "Epoch 146/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5683 - loss: 0.9560 \n",
      "Epoch 147/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5708 - loss: 0.9729 \n",
      "Epoch 148/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5739 - loss: 0.9760 \n",
      "Epoch 149/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5664 - loss: 0.9589 \n",
      "Epoch 150/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5585 - loss: 1.0079 \n",
      "Epoch 151/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5704 - loss: 0.9688 \n",
      "Epoch 152/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5824 - loss: 0.9815 \n",
      "Epoch 153/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5662 - loss: 0.9688 \n",
      "Epoch 154/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5694 - loss: 0.9754 \n",
      "Epoch 155/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.5699 - loss: 0.9639 \n",
      "Epoch 156/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5725 - loss: 0.9601 \n",
      "Epoch 157/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5854 - loss: 0.9467 \n",
      "Epoch 158/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5735 - loss: 0.9656 \n",
      "Epoch 159/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.5779 - loss: 0.9532 \n",
      "Epoch 160/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5796 - loss: 0.9567 \n",
      "Epoch 161/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5736 - loss: 0.9654 \n",
      "Epoch 162/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5657 - loss: 0.9802 \n",
      "Epoch 163/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5809 - loss: 0.9670 \n",
      "Epoch 164/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5728 - loss: 0.9647 \n",
      "Epoch 165/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5751 - loss: 0.9615 \n",
      "Epoch 166/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5744 - loss: 0.9714 \n",
      "Epoch 167/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6063 - loss: 0.9330 \n",
      "Epoch 168/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5910 - loss: 0.9610 \n",
      "Epoch 169/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5929 - loss: 0.9526 \n",
      "Epoch 170/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5480 - loss: 0.9716 \n",
      "Epoch 171/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5932 - loss: 0.9433 \n",
      "Epoch 172/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5931 - loss: 0.9283 \n",
      "Epoch 173/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5847 - loss: 0.9567 \n",
      "Epoch 174/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5988 - loss: 0.9446 \n",
      "Epoch 175/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6086 - loss: 0.9270 \n",
      "Epoch 176/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5787 - loss: 0.9733 \n",
      "Epoch 177/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5832 - loss: 0.9701 \n",
      "Epoch 178/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5710 - loss: 0.9621 \n",
      "Epoch 179/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5778 - loss: 0.9450 \n",
      "Epoch 180/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5650 - loss: 0.9589 \n",
      "Epoch 181/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5615 - loss: 0.9771 \n",
      "Epoch 182/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5983 - loss: 0.9295 \n",
      "Epoch 183/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5760 - loss: 0.9432 \n",
      "Epoch 184/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5624 - loss: 0.9665 \n",
      "Epoch 185/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5812 - loss: 0.9720 \n",
      "Epoch 186/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5992 - loss: 0.9564 \n",
      "Epoch 187/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5971 - loss: 0.9377 \n",
      "Epoch 188/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5870 - loss: 0.9527 \n",
      "Epoch 189/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5769 - loss: 0.9459 \n",
      "Epoch 190/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5648 - loss: 0.9455 \n",
      "Epoch 191/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5698 - loss: 0.9508 \n",
      "Epoch 192/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5653 - loss: 0.9587 \n",
      "Epoch 193/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5566 - loss: 0.9718 \n",
      "Epoch 194/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5821 - loss: 0.9490 \n",
      "Epoch 195/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5895 - loss: 0.9302 \n",
      "Epoch 196/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5714 - loss: 0.9541 \n",
      "Epoch 197/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5938 - loss: 0.9396 \n",
      "Epoch 198/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5848 - loss: 0.9462 \n",
      "Epoch 199/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6012 - loss: 0.9398 \n",
      "Epoch 200/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5914 - loss: 0.9298 \n",
      "Epoch 201/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5811 - loss: 0.9395 \n",
      "Epoch 202/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.6100 - loss: 0.9053 \n",
      "Epoch 203/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5918 - loss: 0.9668 \n",
      "Epoch 204/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5959 - loss: 0.9419 \n",
      "Epoch 205/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5825 - loss: 0.9597 \n",
      "Epoch 206/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5796 - loss: 0.9223 \n",
      "Epoch 207/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5873 - loss: 0.9324 \n",
      "Epoch 208/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.5889 - loss: 0.9275 \n",
      "Epoch 209/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5570 - loss: 0.9633 \n",
      "Epoch 210/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5851 - loss: 0.9542 \n",
      "Epoch 211/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5725 - loss: 0.9505 \n",
      "Epoch 212/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5821 - loss: 0.9425 \n",
      "Epoch 213/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5635 - loss: 0.9783 \n",
      "Epoch 214/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5871 - loss: 0.9558 \n",
      "Epoch 215/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5796 - loss: 0.9453 \n",
      "Epoch 216/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5707 - loss: 0.9429 \n",
      "Epoch 217/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5801 - loss: 0.9360 \n",
      "Epoch 218/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.5934 - loss: 0.9286 \n",
      "Epoch 219/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5877 - loss: 0.9483 \n",
      "Epoch 220/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.5783 - loss: 0.9650 \n",
      "Epoch 221/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5633 - loss: 0.9430 \n",
      "Epoch 222/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5803 - loss: 0.9231 \n",
      "Epoch 223/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5917 - loss: 0.9022 \n",
      "Epoch 224/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5865 - loss: 0.9267 \n",
      "Epoch 225/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5785 - loss: 0.9512 \n",
      "Epoch 226/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5979 - loss: 0.9274 \n",
      "Epoch 227/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5887 - loss: 0.9414 \n",
      "Epoch 228/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5776 - loss: 0.9224 \n",
      "Epoch 229/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5831 - loss: 0.9546 \n",
      "Epoch 230/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5808 - loss: 0.9255 \n",
      "Epoch 231/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5836 - loss: 0.9449 \n",
      "Epoch 232/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5882 - loss: 0.9211 \n",
      "Epoch 233/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5747 - loss: 0.9461 \n",
      "Epoch 234/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5800 - loss: 0.9381 \n",
      "Epoch 235/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5804 - loss: 0.9240 \n",
      "Epoch 236/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5840 - loss: 0.9205 \n",
      "Epoch 237/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5666 - loss: 0.9302 \n",
      "Epoch 238/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5873 - loss: 0.9242 \n",
      "Epoch 239/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6053 - loss: 0.9097 \n",
      "Epoch 240/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5966 - loss: 0.9074 \n",
      "Epoch 241/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5787 - loss: 0.9369 \n",
      "Epoch 242/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6149 - loss: 0.8964 \n",
      "Epoch 243/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5777 - loss: 0.9559 \n",
      "Epoch 244/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6056 - loss: 0.9282 \n",
      "Epoch 245/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5933 - loss: 0.9498 \n",
      "Epoch 246/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5631 - loss: 0.9582 \n",
      "Epoch 247/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5900 - loss: 0.9255 \n",
      "Epoch 248/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5725 - loss: 0.9541 \n",
      "Epoch 249/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5735 - loss: 0.9301 \n",
      "Epoch 250/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5934 - loss: 0.9347 \n",
      "Epoch 251/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5985 - loss: 0.9235 \n",
      "Epoch 252/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5939 - loss: 0.9216 \n",
      "Epoch 253/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5977 - loss: 0.9119 \n",
      "Epoch 254/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5954 - loss: 0.9220 \n",
      "Epoch 255/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5856 - loss: 0.9235 \n",
      "Epoch 256/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5931 - loss: 0.9347 \n",
      "Epoch 257/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5849 - loss: 0.9145 \n",
      "Epoch 258/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5831 - loss: 0.9320 \n",
      "Epoch 259/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6188 - loss: 0.8940 \n",
      "Epoch 260/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6052 - loss: 0.9262 \n",
      "Epoch 261/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6002 - loss: 0.9045 \n",
      "Epoch 262/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5950 - loss: 0.9209 \n",
      "Epoch 263/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5910 - loss: 0.9065 \n",
      "Epoch 264/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5932 - loss: 0.9076 \n",
      "Epoch 265/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.6036 - loss: 0.9089 \n",
      "Epoch 266/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5929 - loss: 0.9249 \n",
      "Epoch 267/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5987 - loss: 0.9168 \n",
      "Epoch 268/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5892 - loss: 0.9099 \n",
      "Epoch 269/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6167 - loss: 0.9090 \n",
      "Epoch 270/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5973 - loss: 0.9362 \n",
      "Epoch 271/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step - accuracy: 0.6173 - loss: 0.9090 \n",
      "Epoch 272/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5940 - loss: 0.9022 \n",
      "Epoch 273/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.6177 - loss: 0.9108 \n",
      "Epoch 274/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6075 - loss: 0.8887 \n",
      "Epoch 275/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6136 - loss: 0.9106 \n",
      "Epoch 276/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6078 - loss: 0.9013 \n",
      "Epoch 277/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6065 - loss: 0.8858 \n",
      "Epoch 278/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5874 - loss: 0.9160 \n",
      "Epoch 279/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5979 - loss: 0.9201 \n",
      "Epoch 280/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5936 - loss: 0.9193 \n",
      "Epoch 281/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5977 - loss: 0.9076 \n",
      "Epoch 282/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6016 - loss: 0.9044 \n",
      "Epoch 283/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.6081 - loss: 0.9158 \n",
      "Epoch 284/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 3ms/step - accuracy: 0.5943 - loss: 0.9160 \n",
      "Epoch 285/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5911 - loss: 0.8975 \n",
      "Epoch 286/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6091 - loss: 0.8757 \n",
      "Epoch 287/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6064 - loss: 0.9065 \n",
      "Epoch 288/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 6ms/step - accuracy: 0.6103 - loss: 0.9087 \n",
      "Epoch 289/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6078 - loss: 0.8913 \n",
      "Epoch 290/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6070 - loss: 0.8938 \n",
      "Epoch 291/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.6107 - loss: 0.9061 \n",
      "Epoch 292/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6019 - loss: 0.9046 \n",
      "Epoch 293/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5953 - loss: 0.9108 \n",
      "Epoch 294/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.6014 - loss: 0.9113 \n",
      "Epoch 295/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5949 - loss: 0.9122 \n",
      "Epoch 296/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.6012 - loss: 0.9262 \n",
      "Epoch 297/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5954 - loss: 0.9036 \n",
      "Epoch 298/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.6286 - loss: 0.8743 \n",
      "Epoch 299/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 5ms/step - accuracy: 0.5754 - loss: 0.9442 \n",
      "Epoch 300/300\n",
      "\u001b[1m6/6\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4ms/step - accuracy: 0.5971 - loss: 0.9108 \n"
     ]
    }
   ],
   "source": [
    "from tensorflow.keras import Sequential\n",
    "from tensorflow.keras.layers import Input, Dense, Dropout\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "redwine = pd.read_csv(\"winequality-red.csv\", sep=\";\")\n",
    "X, y = redwine.iloc[:, :-1], redwine.iloc[:, -1]\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "train_X, test_X, train_y, test_y = train_test_split(X, y)\n",
    "\n",
    "model = Sequential()\n",
    "model.add(Input(shape=(11,)))\n",
    "model.add(Dense(100, activation='sigmoid'))\n",
    "model.add(Dropout(0.2))\n",
    "model.add(Dense(200, activation='relu'))\n",
    "model.add(Dropout(0.4))\n",
    "model.add(Dense(50, activation='tanh'))\n",
    "model.add(Dropout(0.1))\n",
    "model.add(Dense(10, activation='softmax'))\n",
    "\n",
    "model.compile(optimizer='adam',\n",
    "              loss='sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])\n",
    "\n",
    "history = model.fit(train_X, train_y,\n",
    "                    batch_size=200, epochs=300)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1a9f00a4",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-11T12:08:24.145036Z",
     "start_time": "2021-06-11T12:08:23.881121Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.977449893951416 0.6000000238418579\n"
     ]
    }
   ],
   "source": [
    "loss, accuracy = model.evaluate(test_X, test_y, verbose=0)\n",
    "print(loss, accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "07261484",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-11T12:08:25.764548Z",
     "start_time": "2021-06-11T12:08:24.967142Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig, ax1 = plt.subplots(figsize=(6,4))\n",
    "ax1.plot(history.history['loss'], 'r-', label=\"train_loss\")\n",
    "ax1.set_ylabel(\"train loss\")\n",
    "ax1.legend()\n",
    "\n",
    "ax2 = ax1.twinx()\n",
    "ax2.plot(history.history['accuracy'],'g:',label=\"train_acc\")\n",
    "ax2.set_ylabel(\"train acc\")\n",
    "ax2.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8be514c2",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.7"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}