train_transf_net.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: pillow==4.0.0 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (4.0.0)\n",
      "Requirement already satisfied: olefile in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from pillow==4.0.0) (0.45.1)\n",
      "\u001b[33mYou are using pip version 10.0.1, however version 19.3.1 is available.\n",
      "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n",
      "Requirement already satisfied: keras in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (2.3.1)\n",
      "Requirement already satisfied: numpy>=1.9.1 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from keras) (1.14.3)\n",
      "Requirement already satisfied: pyyaml in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from keras) (3.12)\n",
      "Requirement already satisfied: keras-applications>=1.0.6 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from keras) (1.0.8)\n",
      "Requirement already satisfied: six>=1.9.0 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from keras) (1.11.0)\n",
      "Requirement already satisfied: scipy>=0.14 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from keras) (1.1.0)\n",
      "Requirement already satisfied: keras-preprocessing>=1.0.5 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from keras) (1.1.0)\n",
      "Requirement already satisfied: h5py in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from keras) (2.8.0)\n",
      "\u001b[33mYou are using pip version 10.0.1, however version 19.3.1 is available.\n",
      "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n",
      "Requirement already satisfied: seaborn==0.9.0 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (0.9.0)\n",
      "Requirement already satisfied: numpy>=1.9.3 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from seaborn==0.9.0) (1.14.3)\n",
      "Requirement already satisfied: scipy>=0.14.0 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from seaborn==0.9.0) (1.1.0)\n",
      "Requirement already satisfied: pandas>=0.15.2 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from seaborn==0.9.0) (0.24.2)\n",
      "Requirement already satisfied: matplotlib>=1.4.3 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from seaborn==0.9.0) (3.0.3)\n",
      "Requirement already satisfied: pytz>=2011k in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from pandas>=0.15.2->seaborn==0.9.0) (2018.4)\n",
      "Requirement already satisfied: python-dateutil>=2.5.0 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from pandas>=0.15.2->seaborn==0.9.0) (2.7.3)\n",
      "Requirement already satisfied: cycler>=0.10 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (0.10.0)\n",
      "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (2.2.0)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (1.0.1)\n",
      "Requirement already satisfied: six>=1.5 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from python-dateutil>=2.5.0->pandas>=0.15.2->seaborn==0.9.0) (1.11.0)\n",
      "Requirement already satisfied: setuptools in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from kiwisolver>=1.0.1->matplotlib>=1.4.3->seaborn==0.9.0) (39.1.0)\n",
      "\u001b[33mYou are using pip version 10.0.1, however version 19.3.1 is available.\n",
      "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n",
      "Requirement already satisfied: seaborn==0.9.0 in /usr/local/lib/python3.6/site-packages (0.9.0)\n",
      "Requirement already satisfied: numpy>=1.9.3 in /usr/lib64/python3.6/dist-packages (from seaborn==0.9.0) (1.17.3)\n",
      "Requirement already satisfied: matplotlib>=1.4.3 in /usr/lib64/python3.6/dist-packages (from seaborn==0.9.0) (3.1.1)\n",
      "Requirement already satisfied: pandas>=0.15.2 in /usr/lib64/python3.6/dist-packages (from seaborn==0.9.0) (0.24.2)\n",
      "Requirement already satisfied: scipy>=0.14.0 in /usr/lib64/python3.6/dist-packages (from seaborn==0.9.0) (1.3.1)\n",
      "Requirement already satisfied: cycler>=0.10 in /usr/lib/python3.6/dist-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (0.10.0)\n",
      "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/lib64/python3.6/dist-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (1.1.0)\n",
      "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/lib/python3.6/dist-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (2.4.2)\n",
      "Requirement already satisfied: python-dateutil>=2.1 in /usr/lib/python3.6/dist-packages (from matplotlib>=1.4.3->seaborn==0.9.0) (2.8.0)\n",
      "Requirement already satisfied: pytz>=2011k in /usr/lib/python3.6/dist-packages (from pandas>=0.15.2->seaborn==0.9.0) (2019.3)\n",
      "Requirement already satisfied: six in /usr/lib/python3.6/dist-packages (from cycler>=0.10->matplotlib>=1.4.3->seaborn==0.9.0) (1.12.0)\n",
      "Requirement already satisfied: setuptools in /usr/lib/python3.6/dist-packages (from kiwisolver>=1.0.1->matplotlib>=1.4.3->seaborn==0.9.0) (41.5.0)\n",
      "\u001b[33mYou are using pip version 19.0.2, however version 19.3.1 is available.\n",
      "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "# Downgrade pillow to avoid `UserWarning: Possibly corrupt EXIF data.`\n",
    "!pip install pillow==4.0.0\n",
    "!pip install keras\n",
    "!pip install seaborn==0.9.0\n",
    "!sudo -H pip install seaborn==0.9.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: pydot in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (1.4.1)\n",
      "Requirement already satisfied: pyparsing>=2.1.4 in /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages (from pydot) (2.2.0)\n",
      "\u001b[33mYou are using pip version 10.0.1, however version 19.3.1 is available.\n",
      "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n"
     ]
    }
   ],
   "source": [
    "!pip install pydot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x864 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# !pip install seaborn==0.9.0\n",
    "# !sudo -H pip install seaborn==0.9.0\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "n_vector = np.linspace(1, 100, 100)\n",
    "info_loss = [\n",
    "1.0342,\n",
    "1.0646,\n",
    "1.1141,\n",
    "1.0843,\n",
    "0.9863,\n",
    "1.0026,\n",
    "0.9200,\n",
    "0.8996,\n",
    "0.9565,\n",
    "0.9806,\n",
    "0.8831,\n",
    "0.9379,\n",
    "0.9211,\n",
    "0.8634,\n",
    "0.8797,\n",
    "0.8683,\n",
    "0.8752,\n",
    "0.9039,\n",
    "0.8899,\n",
    "0.8137,\n",
    "0.8621,\n",
    "0.8749,\n",
    "0.7928,\n",
    "0.8580,\n",
    "0.8328,\n",
    "0.8495,\n",
    "0.7947,\n",
    "0.7722,\n",
    "0.8168,\n",
    "0.8167,\n",
    "0.8524,\n",
    "0.8018,\n",
    "0.8416,\n",
    "0.7751,\n",
    "0.7823,\n",
    "0.7790,\n",
    "0.7864,\n",
    "0.7744,\n",
    "0.7602,\n",
    "0.7693,\n",
    "0.7994,\n",
    "0.7580,\n",
    "0.7727,\n",
    "0.7904,\n",
    "0.7633,\n",
    "0.7582,\n",
    "0.7470,\n",
    "0.7359,\n",
    "0.7831,\n",
    "0.7776,\n",
    "0.7603,\n",
    "0.7666,\n",
    "0.7465,\n",
    "0.7626,\n",
    "0.7662,\n",
    "0.7861,\n",
    "0.7354,\n",
    "0.7243,\n",
    "0.7830,\n",
    "0.7481,\n",
    "0.7229,\n",
    "0.7437,\n",
    "0.7135,\n",
    "0.7737,\n",
    "0.7112,\n",
    "0.7247,\n",
    "0.7230,\n",
    "0.7359,\n",
    "0.7464,\n",
    "0.7486,\n",
    "0.7485,\n",
    "0.7233,\n",
    "0.7058,\n",
    "0.7123,\n",
    "0.7273,\n",
    "0.6953,\n",
    "0.7408,\n",
    "0.7128,\n",
    "0.7010,\n",
    "0.7115,\n",
    "0.7003,\n",
    "0.6922,\n",
    "0.7364,\n",
    "0.7081,\n",
    "0.6874,\n",
    "0.6862,\n",
    "0.6872,\n",
    "0.6949,\n",
    "0.7115,\n",
    "0.7215,\n",
    "0.7074,\n",
    "0.7042,\n",
    "0.6933,\n",
    "0.7251,\n",
    "0.6946,\n",
    "0.7111,\n",
    "0.7099,\n",
    "0.6885,\n",
    "0.7037,\n",
    "0.6941]    \n",
    "info_val_loss = [\n",
    "0.6977,\n",
    "0.6994,\n",
    "0.7052,\n",
    "1.2656,\n",
    "1.9046,\n",
    "1.6511,\n",
    "1.5540,\n",
    "1.7777,\n",
    "1.4850,\n",
    "1.7448,\n",
    "1.7236,\n",
    "1.7089,\n",
    "1.6378,\n",
    "1.9452,\n",
    "1.7653,\n",
    "1.6470,\n",
    "1.4801,\n",
    "1.1215,\n",
    "1.6753,\n",
    "1.4486,\n",
    "1.6196,\n",
    "1.4025,\n",
    "1.5974,\n",
    "1.8315,\n",
    "1.5801,\n",
    "1.5183,\n",
    "1.4985,\n",
    "1.5267,\n",
    "1.2313,\n",
    "1.3673,\n",
    "1.5033,\n",
    "1.3914,\n",
    "1.6173,\n",
    "1.4350,\n",
    "1.5576,\n",
    "1.2085,\n",
    "1.4645,\n",
    "1.4834,\n",
    "1.3494,\n",
    "1.7563,\n",
    "1.4095,\n",
    "1.6371,\n",
    "1.3658,\n",
    "1.6366,\n",
    "1.4620,\n",
    "1.7982,\n",
    "1.5273,\n",
    "2.0862,\n",
    "1.6126,\n",
    "1.5763,\n",
    "1.7656,\n",
    "1.3623,\n",
    "1.4690,\n",
    "1.5387,\n",
    "1.2767,\n",
    "1.4798,\n",
    "1.3852,\n",
    "1.2538,\n",
    "1.3447,\n",
    "1.5597,\n",
    "1.8820,\n",
    "1.4845,\n",
    "1.2506,\n",
    "1.3851,\n",
    "1.3919,\n",
    "1.2077,\n",
    "1.4491,\n",
    "1.3574,\n",
    "1.6551,\n",
    "1.4182,\n",
    "1.4200,\n",
    "1.2136,\n",
    "1.3870,\n",
    "1.5277,\n",
    "1.3115,\n",
    "1.2856,\n",
    "1.0861,\n",
    "1.3493,\n",
    "1.7437,\n",
    "1.8764,\n",
    "1.5451,\n",
    "1.3768,\n",
    "1.1351,\n",
    "1.1815,\n",
    "0.9732,\n",
    "1.0815,\n",
    "1.1921,\n",
    "1.3799,\n",
    "1.4467,\n",
    "1.1733,\n",
    "1.2933,\n",
    "1.1475,\n",
    "1.3488,\n",
    "1.4886,\n",
    "1.6898,\n",
    "1.4532,\n",
    "1.6644,\n",
    "1.2321,\n",
    "1.3576,\n",
    "1.2964]\n",
    "info_acc = [\n",
    "0.5170,\n",
    "0.5142,\n",
    "0.4873,\n",
    "0.5255,\n",
    "0.5354,\n",
    "0.5071,\n",
    "0.5212,\n",
    "0.5467,\n",
    "0.5397,\n",
    "0.4703,\n",
    "0.5340,\n",
    "0.4943,\n",
    "0.5142,\n",
    "0.5326,\n",
    "0.5283,\n",
    "0.5482,\n",
    "0.5283,\n",
    "0.5028,\n",
    "0.5113,\n",
    "0.5425,\n",
    "0.5255,\n",
    "0.5028,\n",
    "0.5269,\n",
    "0.5283,\n",
    "0.5340,\n",
    "0.5142,\n",
    "0.5510,\n",
    "0.5567,\n",
    "0.5297,\n",
    "0.5283,\n",
    "0.5042,\n",
    "0.5184,\n",
    "0.5085,\n",
    "0.5496,\n",
    "0.5354,\n",
    "0.5212,\n",
    "0.5156,\n",
    "0.5453,\n",
    "0.5340,\n",
    "0.5581,\n",
    "0.5142,\n",
    "0.5581,\n",
    "0.5567,\n",
    "0.5283,\n",
    "0.5354,\n",
    "0.5567,\n",
    "0.5510,\n",
    "0.5666,\n",
    "0.5411,\n",
    "0.5510,\n",
    "0.5680,\n",
    "0.5482,\n",
    "0.5297,\n",
    "0.5382,\n",
    "0.5496,\n",
    "0.5212,\n",
    "0.5283,\n",
    "0.5779,\n",
    "0.5156,\n",
    "0.5326,\n",
    "0.5779,\n",
    "0.5439,\n",
    "0.5567,\n",
    "0.5510,\n",
    "0.5680,\n",
    "0.5793,\n",
    "0.5892,\n",
    "0.5652,\n",
    "0.5581,\n",
    "0.5467,\n",
    "0.5397,\n",
    "0.5836,\n",
    "0.5708,\n",
    "0.5666,\n",
    "0.5510,\n",
    "0.5949,\n",
    "0.5382,\n",
    "0.5935,\n",
    "0.5751,\n",
    "0.5822,\n",
    "0.5765,\n",
    "0.5977,\n",
    "0.5411,\n",
    "0.5878,\n",
    "0.5907,\n",
    "0.6176,\n",
    "0.5907,\n",
    "0.5836,\n",
    "0.5652,\n",
    "0.5779,\n",
    "0.5751,\n",
    "0.5595,\n",
    "0.6048,\n",
    "0.5765,\n",
    "0.5864,\n",
    "0.5609,\n",
    "0.5722,\n",
    "0.5722,\n",
    "0.5652,\n",
    "0.5892    \n",
    "]\n",
    "info_val_acc = [\n",
    "0.5000,\n",
    "0.5000,\n",
    "0.5071,\n",
    "0.5000,\n",
    "0.4929,\n",
    "0.5071,\n",
    "0.5500,\n",
    "0.4857,\n",
    "0.5429,\n",
    "0.5071,\n",
    "0.5143,\n",
    "0.5143,\n",
    "0.5643,\n",
    "0.5000,\n",
    "0.5143,\n",
    "0.5214,\n",
    "0.5214,\n",
    "0.5571,\n",
    "0.5286,\n",
    "0.5286,\n",
    "0.5214,\n",
    "0.5643,\n",
    "0.5571,\n",
    "0.5214,\n",
    "0.5714,\n",
    "0.5429,\n",
    "0.5643,\n",
    "0.5286,\n",
    "0.5357,\n",
    "0.5429,\n",
    "0.5429,\n",
    "0.5214,\n",
    "0.5357,\n",
    "0.5286,\n",
    "0.5357,\n",
    "0.5714,\n",
    "0.5071,\n",
    "0.5571,\n",
    "0.5571,\n",
    "0.5000,\n",
    "0.5286,\n",
    "0.6000,\n",
    "0.5357,\n",
    "0.5786,\n",
    "0.5429,\n",
    "0.5500,\n",
    "0.5500,\n",
    "0.5857,\n",
    "0.5500,\n",
    "0.5357,\n",
    "0.5071,\n",
    "0.5286,\n",
    "0.6357,\n",
    "0.5857,\n",
    "0.6143,\n",
    "0.5929,\n",
    "0.5786,\n",
    "0.6214,\n",
    "0.5929,\n",
    "0.5786,\n",
    "0.5357,\n",
    "0.6143,\n",
    "0.5929,\n",
    "0.6071,\n",
    "0.6000,\n",
    "0.6214,\n",
    "0.5714,\n",
    "0.6143,\n",
    "0.5929,\n",
    "0.5786,\n",
    "0.6357,\n",
    "0.6000,\n",
    "0.6500,\n",
    "0.6000,\n",
    "0.6214,\n",
    "0.6071,\n",
    "0.6286,\n",
    "0.5786,\n",
    "0.6000,\n",
    "0.5643,\n",
    "0.5929,\n",
    "0.6071,\n",
    "0.6357,\n",
    "0.6143,\n",
    "0.6143,\n",
    "0.6643,\n",
    "0.6000,\n",
    "0.5857,\n",
    "0.5714,\n",
    "0.6500,\n",
    "0.6571,\n",
    "0.6643,\n",
    "0.6429,\n",
    "0.6143,\n",
    "0.6071,\n",
    "0.6429,\n",
    "0.6143,\n",
    "0.6071,\n",
    "0.5857,\n",
    "0.6000\n",
    "]\n",
    "\n",
    "# loss graphs\n",
    "fig1 = plt.figure(figsize=(12, 12))\n",
    "ax = fig1.add_subplot(1, 1, 1)\n",
    "sns.lineplot(\n",
    "    x=n_vector,\n",
    "    y=info_loss,\n",
    "    ax=ax\n",
    "#     data=model_info,\n",
    "#     legend='full'\n",
    ")\n",
    "\n",
    "sns.lineplot(\n",
    "    x=n_vector,\n",
    "    y=info_val_loss,\n",
    "    ax=ax\n",
    "#     data=model_info,\n",
    "#     legend='full'\n",
    ")\n",
    "\n",
    "plt.title(\n",
    "    \"Loss variation\",\n",
    "    fontsize=20\n",
    ")\n",
    "plt.xlabel(\n",
    "    \"Epochs\",\n",
    "    fontsize=15\n",
    ")\n",
    "plt.ylabel(\n",
    "    \"Loss\",\n",
    "    fontsize=15\n",
    ")\n",
    "fig1.legend(\n",
    "    labels=['loss', 'val_loss']\n",
    ")\n",
    "plt.savefig('model1_loss')\n",
    "# plt.show()\n",
    "\n",
    "# accuracy graphs\n",
    "fig1 = plt.figure(figsize=(12, 12))\n",
    "ax = fig1.add_subplot(1, 1, 1)\n",
    "sns.lineplot(\n",
    "    x=n_vector,\n",
    "    y=info_acc\n",
    "#     data=model_info\n",
    ")\n",
    "\n",
    "sns.lineplot(\n",
    "    x=n_vector,\n",
    "    y=info_val_acc\n",
    "#     data=model_info\n",
    ")\n",
    "\n",
    "plt.title(\n",
    "    \"Accuracy variation\",\n",
    "    fontsize=20\n",
    ")\n",
    "plt.xlabel(\n",
    "    \"Epochs\",\n",
    "    fontsize=15\n",
    ")\n",
    "plt.ylabel(\n",
    "    \"Accuracy\",\n",
    "    fontsize=15\n",
    ")\n",
    "fig1.legend(\n",
    "    labels=['accuracy', 'val_accuracy']\n",
    ")\n",
    "\n",
    "plt.savefig('model1_accuracy')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total 1104\n",
      "drwxrwxr-x 14 ec2-user ec2-user   4096 Jan 10 18:46 \u001b[0m\u001b[01;34m.\u001b[0m/\n",
      "drwxrwxr-x 15 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34m..\u001b[0m/\n",
      "drwxrwxr-x  5 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mData\u001b[0m/\n",
      "drwxrwxr-x  4 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mDatabase\u001b[0m/\n",
      "drwxrwxr-x  5 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mData_equalized\u001b[0m/\n",
      "drwxrwxr-x  3 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mData_partial_excluded\u001b[0m/\n",
      "drwxrwxr-x  3 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mData_Suspect\u001b[0m/\n",
      "drwxrwxr-x  7 ec2-user ec2-user   4096 Jan  8 17:32 \u001b[01;34mefficientnet_keras_transfer_learning\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1487 Jan  8 17:29 evaluation_test.py\n",
      "drwxrwxr-x  8 ec2-user ec2-user   4096 Jan 10 18:48 \u001b[01;34m.git\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1212 Jan  8 17:29 .gitignore\n",
      "-rw-rw-r--  1 ec2-user ec2-user   5038 Jan  8 17:29 Graphs.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user  23748 Jan  8 17:29 Interpreter.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan 10 18:35 \u001b[01;34m.ipynb_checkpoints\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user  60401 Jan 10 18:46 \u001b[01;35mmodel1_accuracy.png\u001b[0m\n",
      "-rw-rw-r--  1 ec2-user ec2-user  55346 Jan 10 18:46 \u001b[01;35mmodel1_loss.png\u001b[0m\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1302 Jan  8 17:29 pre_proc.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mProcessing\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   5092 Jan  8 17:29 Processor.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:32 \u001b[01;34m__pycache__\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1926 Jan  8 17:29 README.md\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mreferences\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user    231 Jan  8 17:29 requirements.txt\n",
      "-rw-rw-r--  1 ec2-user ec2-user 280532 Jan  9 16:10 \u001b[01;35mresnet_model.png\u001b[0m\n",
      "-rw-rw-r--  1 ec2-user ec2-user  81895 Jan  9 01:21 \u001b[01;35msimplified_model.png\u001b[0m\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1314 Jan  8 17:29 test_interpreter.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user   3009 Jan  8 17:29 test_processor.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user 248929 Jan  8 21:20 train_model.ipynb\n",
      "-rw-rw-r--  1 ec2-user ec2-user   2025 Jan  8 17:29 train_start.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user 181755 Jan 10 18:46 train_transf_net.ipynb\n",
      "-rw-rw-r--  1 ec2-user ec2-user  81132 Jan  8 17:29 Untitled.ipynb\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mWindowOpt\u001b[0m/\n",
      "[Errno 2] No such file or directory: 'eye-Image-Analysis/'\n",
      "/home/ec2-user/SageMaker/eye-Image-Analysis/eye-Image-Analysis/eye-Image-Analysis\n",
      "total 1104\n",
      "drwxrwxr-x 14 ec2-user ec2-user   4096 Jan 10 18:46 \u001b[0m\u001b[01;34m.\u001b[0m/\n",
      "drwxrwxr-x 15 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34m..\u001b[0m/\n",
      "drwxrwxr-x  5 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mData\u001b[0m/\n",
      "drwxrwxr-x  4 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mDatabase\u001b[0m/\n",
      "drwxrwxr-x  5 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mData_equalized\u001b[0m/\n",
      "drwxrwxr-x  3 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mData_partial_excluded\u001b[0m/\n",
      "drwxrwxr-x  3 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mData_Suspect\u001b[0m/\n",
      "drwxrwxr-x  7 ec2-user ec2-user   4096 Jan  8 17:32 \u001b[01;34mefficientnet_keras_transfer_learning\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1487 Jan  8 17:29 evaluation_test.py\n",
      "drwxrwxr-x  8 ec2-user ec2-user   4096 Jan 10 18:48 \u001b[01;34m.git\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1212 Jan  8 17:29 .gitignore\n",
      "-rw-rw-r--  1 ec2-user ec2-user   5038 Jan  8 17:29 Graphs.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user  23748 Jan  8 17:29 Interpreter.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan 10 18:35 \u001b[01;34m.ipynb_checkpoints\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user  60401 Jan 10 18:46 \u001b[01;35mmodel1_accuracy.png\u001b[0m\n",
      "-rw-rw-r--  1 ec2-user ec2-user  55346 Jan 10 18:46 \u001b[01;35mmodel1_loss.png\u001b[0m\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1302 Jan  8 17:29 pre_proc.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mProcessing\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   5092 Jan  8 17:29 Processor.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:32 \u001b[01;34m__pycache__\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1926 Jan  8 17:29 README.md\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mreferences\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user    231 Jan  8 17:29 requirements.txt\n",
      "-rw-rw-r--  1 ec2-user ec2-user 280532 Jan  9 16:10 \u001b[01;35mresnet_model.png\u001b[0m\n",
      "-rw-rw-r--  1 ec2-user ec2-user  81895 Jan  9 01:21 \u001b[01;35msimplified_model.png\u001b[0m\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1314 Jan  8 17:29 test_interpreter.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user   3009 Jan  8 17:29 test_processor.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user 248929 Jan  8 21:20 train_model.ipynb\n",
      "-rw-rw-r--  1 ec2-user ec2-user   2025 Jan  8 17:29 train_start.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user 181755 Jan 10 18:46 train_transf_net.ipynb\n",
      "-rw-rw-r--  1 ec2-user ec2-user  81132 Jan  8 17:29 Untitled.ipynb\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[01;34mWindowOpt\u001b[0m/\n"
     ]
    }
   ],
   "source": [
    "%ls -la\n",
    "%cd eye-Image-Analysis/\n",
    "%ls -la"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/ec2-user/SageMaker/eye-Image-Analysis/eye-Image-Analysis\n",
      "total 444\n",
      "drwxrwxr-x 15 ec2-user ec2-user   4096 Jan  8 17:29 \u001b[0m\u001b[01;34m.\u001b[0m/\n",
      "drwxrwxr-x 16 ec2-user ec2-user   4096 Jan 10 18:05 \u001b[01;34m..\u001b[0m/\n",
      "drwxrwxr-x  5 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mData\u001b[0m/\n",
      "drwxrwxr-x  4 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mDatabase\u001b[0m/\n",
      "drwxrwxr-x  5 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mData_equalized\u001b[0m/\n",
      "drwxrwxr-x  3 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mData_partial_excluded\u001b[0m/\n",
      "drwxrwxr-x  3 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mData_Suspect\u001b[0m/\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mefficientnet_keras_transfer_learning\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1487 Jan  1 13:28 evaluation_test.py\n",
      "drwxrwxr-x 14 ec2-user ec2-user   4096 Jan 10 18:46 \u001b[01;34meye-Image-Analysis\u001b[0m/\n",
      "drwxrwxr-x  8 ec2-user ec2-user   4096 Jan 10 18:31 \u001b[01;34m.git\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1212 Jan  1 13:28 .gitignore\n",
      "-rw-rw-r--  1 ec2-user ec2-user   4829 Jan  1 13:28 Graphs.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user  24421 Jan  1 13:28 Interpreter.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  8 17:31 \u001b[01;34m.ipynb_checkpoints\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1302 Jan  1 13:28 pre_proc.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mProcessing\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   5092 Jan  1 13:28 Processor.py\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  6 14:52 \u001b[01;34m__pycache__\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1926 Jan  1 13:28 README.md\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mreferences\u001b[0m/\n",
      "-rw-rw-r--  1 ec2-user ec2-user    231 Jan  1 13:28 requirements.txt\n",
      "-rw-rw-r--  1 ec2-user ec2-user   1314 Jan  1 13:28 test_interpreter.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user   3009 Jan  1 13:28 test_processor.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user 212687 Jan  1 13:28 train_model.ipynb\n",
      "-rw-rw-r--  1 ec2-user ec2-user   2042 Jan  1 13:28 train_start.py\n",
      "-rw-rw-r--  1 ec2-user ec2-user  23895 Jan  1 13:28 train_transf_net.ipynb\n",
      "-rw-rw-r--  1 ec2-user ec2-user  81132 Jan  1 13:28 Untitled.ipynb\n",
      "drwxrwxr-x  2 ec2-user ec2-user   4096 Jan  1 13:28 \u001b[01;34mWindowOpt\u001b[0m/\n"
     ]
    }
   ],
   "source": [
    "%cd ..\n",
    "%ls -la"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cloning into 'efficientnet_keras_transfer_learning'...\n",
      "remote: Enumerating objects: 171, done.\u001b[K\n",
      "remote: Counting objects: 100% (171/171), done.\u001b[K\n",
      "remote: Compressing objects: 100% (74/74), done.\u001b[K\n",
      "remote: Total 171 (delta 98), reused 161 (delta 93), pack-reused 0\u001b[K\n",
      "Receiving objects: 100% (171/171), 5.44 MiB | 4.64 MiB/s, done.\n",
      "Resolving deltas: 100% (98/98), done.\n"
     ]
    }
   ],
   "source": [
    "!git clone https://github.com/Tony607/efficientnet_keras_transfer_learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "from Interpreter import Interpreter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "TARGET_SIZE = (224, 224)\n",
    "BATCH_SIZE = 4\n",
    "EPOCHS = 100\n",
    "IMAGE_SHAPE = (224, 224, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 706 images belonging to 2 classes.\n",
      "Found 140 images belonging to 2 classes.\n",
      "Found 94 images belonging to 2 classes.\n"
     ]
    }
   ],
   "source": [
    "inter = Interpreter(\n",
    "    BATCH_SIZE,\n",
    "    IMAGE_SHAPE,\n",
    "    EPOCHS,\n",
    "    TARGET_SIZE\n",
    ")\n",
    "\n",
    "train_images, validation_images, test_images = inter.split_data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> Train EfficientNetB0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages/tensorflow/python/ops/resource_variable_ops.py:435: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "WARNING:tensorflow:From /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages/tensorflow/python/framework/function.py:1007: calling Graph.create_op (from tensorflow.python.framework.ops) with compute_shapes is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Shapes are always computed; don't use the compute_shapes as it has no effect.\n",
      "WARNING:tensorflow:From /home/ec2-user/SageMaker/eye-Image-Analysis/eye-Image-Analysis/eye-Image-Analysis/efficientnet_keras_transfer_learning/efficientnet/layers.py:30: div (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Deprecated in favor of operator or tf.math.divide.\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "efficientnet-b0 (Model)      (None, 7, 7, 1280)        4048988   \n",
      "=================================================================\n",
      "Total params: 4,048,988\n",
      "Trainable params: 4,006,972\n",
      "Non-trainable params: 42,016\n",
      "_________________________________________________________________\n",
      "WARNING:tensorflow:From /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages/tensorflow/python/keras/layers/core.py:143: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "efficientnet-b0 (Model)      (None, 7, 7, 1280)        4048988   \n",
      "_________________________________________________________________\n",
      "gap (GlobalMaxPooling2D)     (None, 1280)              0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 1280)              0         \n",
      "_________________________________________________________________\n",
      "dropout_out (Dropout)        (None, 1280)              0         \n",
      "=================================================================\n",
      "Total params: 4,048,988\n",
      "Trainable params: 4,006,972\n",
      "Non-trainable params: 42,016\n",
      "_________________________________________________________________\n",
      "Net after add:\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "efficientnet-b0 (Model)      (None, 7, 7, 1280)        4048988   \n",
      "_________________________________________________________________\n",
      "gap (GlobalMaxPooling2D)     (None, 1280)              0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 1280)              0         \n",
      "_________________________________________________________________\n",
      "dropout_out (Dropout)        (None, 1280)              0         \n",
      "_________________________________________________________________\n",
      "fc1 (Dense)                  (None, 256)               327936    \n",
      "_________________________________________________________________\n",
      "fc_out (Dense)               (None, 2)                 514       \n",
      "=================================================================\n",
      "Total params: 4,377,438\n",
      "Trainable params: 4,335,422\n",
      "Non-trainable params: 42,016\n",
      "_________________________________________________________________\n",
      "WARNING:tensorflow:From /home/ec2-user/anaconda3/envs/amazonei_tensorflow_p36/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Epoch 1/100\n",
      "35/35 [==============================] - 13s 370ms/step - loss: 0.6977 - acc: 0.5000\n",
      "177/177 [==============================] - 360s 2s/step - loss: 1.0342 - acc: 0.5170 - val_loss: 0.6977 - val_acc: 0.5000\n",
      "Epoch 2/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 0.6994 - acc: 0.5000\n",
      "177/177 [==============================] - 355s 2s/step - loss: 1.0646 - acc: 0.5142 - val_loss: 0.6994 - val_acc: 0.5000\n",
      "Epoch 3/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 0.7052 - acc: 0.5071\n",
      "177/177 [==============================] - 354s 2s/step - loss: 1.1141 - acc: 0.4873 - val_loss: 0.7052 - val_acc: 0.5071\n",
      "Epoch 4/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.2656 - acc: 0.5000\n",
      "177/177 [==============================] - 354s 2s/step - loss: 1.0843 - acc: 0.5255 - val_loss: 1.2656 - val_acc: 0.5000\n",
      "Epoch 5/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.9046 - acc: 0.4929\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.9863 - acc: 0.5354 - val_loss: 1.9046 - val_acc: 0.4929\n",
      "Epoch 6/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.6511 - acc: 0.5071\n",
      "177/177 [==============================] - 355s 2s/step - loss: 1.0026 - acc: 0.5071 - val_loss: 1.6511 - val_acc: 0.5071\n",
      "Epoch 7/100\n",
      "35/35 [==============================] - 12s 345ms/step - loss: 1.5540 - acc: 0.5500\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.9200 - acc: 0.5212 - val_loss: 1.5540 - val_acc: 0.5500\n",
      "Epoch 8/100\n",
      "35/35 [==============================] - 14s 388ms/step - loss: 1.7777 - acc: 0.4857\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.8996 - acc: 0.5467 - val_loss: 1.7777 - val_acc: 0.4857\n",
      "Epoch 9/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.4850 - acc: 0.5429\n",
      "177/177 [==============================] - 370s 2s/step - loss: 0.9565 - acc: 0.5397 - val_loss: 1.4850 - val_acc: 0.5429\n",
      "Epoch 10/100\n",
      "35/35 [==============================] - 12s 357ms/step - loss: 1.7448 - acc: 0.5071\n",
      "177/177 [==============================] - 362s 2s/step - loss: 0.9806 - acc: 0.4703 - val_loss: 1.7448 - val_acc: 0.5071\n",
      "Epoch 11/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.7236 - acc: 0.5143\n",
      "177/177 [==============================] - 360s 2s/step - loss: 0.8831 - acc: 0.5340 - val_loss: 1.7236 - val_acc: 0.5143\n",
      "Epoch 12/100\n",
      "35/35 [==============================] - 12s 347ms/step - loss: 1.7089 - acc: 0.5143\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.9379 - acc: 0.4943 - val_loss: 1.7089 - val_acc: 0.5143\n",
      "Epoch 13/100\n",
      "35/35 [==============================] - 13s 383ms/step - loss: 1.6378 - acc: 0.5643\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.9211 - acc: 0.5142 - val_loss: 1.6378 - val_acc: 0.5643\n",
      "Epoch 14/100\n",
      "35/35 [==============================] - 14s 392ms/step - loss: 1.9452 - acc: 0.5000\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.8634 - acc: 0.5326 - val_loss: 1.9452 - val_acc: 0.5000\n",
      "Epoch 15/100\n",
      "35/35 [==============================] - 14s 387ms/step - loss: 1.7653 - acc: 0.5143\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.8797 - acc: 0.5283 - val_loss: 1.7653 - val_acc: 0.5143\n",
      "Epoch 16/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.6470 - acc: 0.5214\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.8683 - acc: 0.5482 - val_loss: 1.6470 - val_acc: 0.5214\n",
      "Epoch 17/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.4801 - acc: 0.5214\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.8752 - acc: 0.5283 - val_loss: 1.4801 - val_acc: 0.5214\n",
      "Epoch 18/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.1215 - acc: 0.5571\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.9039 - acc: 0.5028 - val_loss: 1.1215 - val_acc: 0.5571\n",
      "Epoch 19/100\n",
      "35/35 [==============================] - 12s 354ms/step - loss: 1.6753 - acc: 0.5286\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.8899 - acc: 0.5113 - val_loss: 1.6753 - val_acc: 0.5286\n",
      "Epoch 20/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.4486 - acc: 0.5286\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.8137 - acc: 0.5425 - val_loss: 1.4486 - val_acc: 0.5286\n",
      "Epoch 21/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.6196 - acc: 0.5214\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.8621 - acc: 0.5255 - val_loss: 1.6196 - val_acc: 0.5214\n",
      "Epoch 22/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.4025 - acc: 0.5643\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.8749 - acc: 0.5028 - val_loss: 1.4025 - val_acc: 0.5643\n",
      "Epoch 23/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.5974 - acc: 0.5571\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.7928 - acc: 0.5269 - val_loss: 1.5974 - val_acc: 0.5571\n",
      "Epoch 24/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.8315 - acc: 0.5214\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.8580 - acc: 0.5283 - val_loss: 1.8315 - val_acc: 0.5214\n",
      "Epoch 25/100\n",
      "35/35 [==============================] - 12s 354ms/step - loss: 1.5801 - acc: 0.5714\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.8328 - acc: 0.5340 - val_loss: 1.5801 - val_acc: 0.5714\n",
      "Epoch 26/100\n",
      "35/35 [==============================] - 12s 353ms/step - loss: 1.5183 - acc: 0.5429\n",
      "177/177 [==============================] - 356s 2s/step - loss: 0.8495 - acc: 0.5142 - val_loss: 1.5183 - val_acc: 0.5429\n",
      "Epoch 27/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.4985 - acc: 0.5643\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.7947 - acc: 0.5510 - val_loss: 1.4985 - val_acc: 0.5643\n",
      "Epoch 28/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.5267 - acc: 0.5286\n",
      "177/177 [==============================] - 356s 2s/step - loss: 0.7722 - acc: 0.5567 - val_loss: 1.5267 - val_acc: 0.5286\n",
      "Epoch 29/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.2313 - acc: 0.5357\n",
      "177/177 [==============================] - 356s 2s/step - loss: 0.8168 - acc: 0.5297 - val_loss: 1.2313 - val_acc: 0.5357\n",
      "Epoch 30/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.3673 - acc: 0.5429\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.8167 - acc: 0.5283 - val_loss: 1.3673 - val_acc: 0.5429\n",
      "Epoch 31/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.5033 - acc: 0.5429\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.8524 - acc: 0.5042 - val_loss: 1.5033 - val_acc: 0.5429\n",
      "Epoch 32/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.3914 - acc: 0.5214\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.8018 - acc: 0.5184 - val_loss: 1.3914 - val_acc: 0.5214\n",
      "Epoch 33/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.6173 - acc: 0.5357\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.8416 - acc: 0.5085 - val_loss: 1.6173 - val_acc: 0.5357\n",
      "Epoch 34/100\n",
      "35/35 [==============================] - 12s 353ms/step - loss: 1.4350 - acc: 0.5286\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.7751 - acc: 0.5496 - val_loss: 1.4350 - val_acc: 0.5286\n",
      "Epoch 35/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.5576 - acc: 0.5357\n",
      "177/177 [==============================] - 355s 2s/step - loss: 0.7823 - acc: 0.5354 - val_loss: 1.5576 - val_acc: 0.5357\n",
      "Epoch 36/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.2085 - acc: 0.5714\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.7790 - acc: 0.5212 - val_loss: 1.2085 - val_acc: 0.5714\n",
      "Epoch 37/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.4645 - acc: 0.5071\n",
      "177/177 [==============================] - 354s 2s/step - loss: 0.7864 - acc: 0.5156 - val_loss: 1.4645 - val_acc: 0.5071\n",
      "Epoch 38/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.4834 - acc: 0.5571\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7744 - acc: 0.5453 - val_loss: 1.4834 - val_acc: 0.5571\n",
      "Epoch 39/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.3494 - acc: 0.5571\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7602 - acc: 0.5340 - val_loss: 1.3494 - val_acc: 0.5571\n",
      "Epoch 40/100\n",
      "35/35 [==============================] - 12s 346ms/step - loss: 1.7563 - acc: 0.5000\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7693 - acc: 0.5581 - val_loss: 1.7563 - val_acc: 0.5000\n",
      "Epoch 41/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.4095 - acc: 0.5286\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7994 - acc: 0.5142 - val_loss: 1.4095 - val_acc: 0.5286\n",
      "Epoch 42/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.6371 - acc: 0.6000\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7580 - acc: 0.5581 - val_loss: 1.6371 - val_acc: 0.6000\n",
      "Epoch 43/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.3658 - acc: 0.5357\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7727 - acc: 0.5567 - val_loss: 1.3658 - val_acc: 0.5357\n",
      "Epoch 44/100\n",
      "35/35 [==============================] - 14s 389ms/step - loss: 1.6366 - acc: 0.5786\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7904 - acc: 0.5283 - val_loss: 1.6366 - val_acc: 0.5786\n",
      "Epoch 45/100\n",
      "35/35 [==============================] - 14s 386ms/step - loss: 1.4620 - acc: 0.5429\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7633 - acc: 0.5354 - val_loss: 1.4620 - val_acc: 0.5429\n",
      "Epoch 46/100\n",
      "35/35 [==============================] - 12s 346ms/step - loss: 1.7982 - acc: 0.5500\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7582 - acc: 0.5567 - val_loss: 1.7982 - val_acc: 0.5500\n",
      "Epoch 47/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.5273 - acc: 0.5500\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7470 - acc: 0.5510 - val_loss: 1.5273 - val_acc: 0.5500\n",
      "Epoch 48/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 2.0862 - acc: 0.5857\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7359 - acc: 0.5666 - val_loss: 2.0862 - val_acc: 0.5857\n",
      "Epoch 49/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.6126 - acc: 0.5500\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7831 - acc: 0.5411 - val_loss: 1.6126 - val_acc: 0.5500\n",
      "Epoch 50/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.5763 - acc: 0.5357\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7776 - acc: 0.5510 - val_loss: 1.5763 - val_acc: 0.5357\n",
      "Epoch 51/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.7656 - acc: 0.5071\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7603 - acc: 0.5680 - val_loss: 1.7656 - val_acc: 0.5071\n",
      "Epoch 52/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.3623 - acc: 0.5286\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7666 - acc: 0.5482 - val_loss: 1.3623 - val_acc: 0.5286\n",
      "Epoch 53/100\n",
      "35/35 [==============================] - 12s 346ms/step - loss: 1.4690 - acc: 0.6357\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7465 - acc: 0.5297 - val_loss: 1.4690 - val_acc: 0.6357\n",
      "Epoch 54/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.5387 - acc: 0.5857\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7626 - acc: 0.5382 - val_loss: 1.5387 - val_acc: 0.5857\n",
      "Epoch 55/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.2767 - acc: 0.6143\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7662 - acc: 0.5496 - val_loss: 1.2767 - val_acc: 0.6143\n",
      "Epoch 56/100\n",
      "35/35 [==============================] - 12s 345ms/step - loss: 1.4798 - acc: 0.5929\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7861 - acc: 0.5212 - val_loss: 1.4798 - val_acc: 0.5929\n",
      "Epoch 57/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.3852 - acc: 0.5786\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7354 - acc: 0.5283 - val_loss: 1.3852 - val_acc: 0.5786\n",
      "Epoch 58/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.2538 - acc: 0.6214\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7243 - acc: 0.5779 - val_loss: 1.2538 - val_acc: 0.6214\n",
      "Epoch 59/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.3447 - acc: 0.5929\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7830 - acc: 0.5156 - val_loss: 1.3447 - val_acc: 0.5929\n",
      "Epoch 60/100\n",
      "35/35 [==============================] - 12s 345ms/step - loss: 1.5597 - acc: 0.5786\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7481 - acc: 0.5326 - val_loss: 1.5597 - val_acc: 0.5786\n",
      "Epoch 61/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.8820 - acc: 0.5357\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7229 - acc: 0.5779 - val_loss: 1.8820 - val_acc: 0.5357\n",
      "Epoch 62/100\n",
      "35/35 [==============================] - 12s 346ms/step - loss: 1.4845 - acc: 0.6143\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7437 - acc: 0.5439 - val_loss: 1.4845 - val_acc: 0.6143\n",
      "Epoch 63/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.2506 - acc: 0.5929\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7135 - acc: 0.5567 - val_loss: 1.2506 - val_acc: 0.5929\n",
      "Epoch 64/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.3851 - acc: 0.6071\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7737 - acc: 0.5510 - val_loss: 1.3851 - val_acc: 0.6071\n",
      "Epoch 65/100\n",
      "35/35 [==============================] - 12s 356ms/step - loss: 1.3919 - acc: 0.6000\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7112 - acc: 0.5680 - val_loss: 1.3919 - val_acc: 0.6000\n",
      "Epoch 66/100\n",
      "35/35 [==============================] - 13s 385ms/step - loss: 1.2077 - acc: 0.6214\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7247 - acc: 0.5793 - val_loss: 1.2077 - val_acc: 0.6214\n",
      "Epoch 67/100\n",
      "35/35 [==============================] - 14s 386ms/step - loss: 1.4491 - acc: 0.5714\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7230 - acc: 0.5892 - val_loss: 1.4491 - val_acc: 0.5714\n",
      "Epoch 68/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.3574 - acc: 0.6143\n",
      "177/177 [==============================] - 350s 2s/step - loss: 0.7359 - acc: 0.5652 - val_loss: 1.3574 - val_acc: 0.6143\n",
      "Epoch 69/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.6551 - acc: 0.5929\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7464 - acc: 0.5581 - val_loss: 1.6551 - val_acc: 0.5929\n",
      "Epoch 70/100\n",
      "35/35 [==============================] - 12s 345ms/step - loss: 1.4182 - acc: 0.5786\n",
      "177/177 [==============================] - 352s 2s/step - loss: 0.7486 - acc: 0.5467 - val_loss: 1.4182 - val_acc: 0.5786\n",
      "Epoch 71/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.4200 - acc: 0.6357\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7485 - acc: 0.5397 - val_loss: 1.4200 - val_acc: 0.6357\n",
      "Epoch 72/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.2136 - acc: 0.6000\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7233 - acc: 0.5836 - val_loss: 1.2136 - val_acc: 0.6000\n",
      "Epoch 73/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.3870 - acc: 0.6500\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7058 - acc: 0.5708 - val_loss: 1.3870 - val_acc: 0.6500\n",
      "Epoch 74/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.5277 - acc: 0.6000\n",
      "177/177 [==============================] - 349s 2s/step - loss: 0.7123 - acc: 0.5666 - val_loss: 1.5277 - val_acc: 0.6000\n",
      "Epoch 75/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.3115 - acc: 0.6214\n",
      "177/177 [==============================] - 350s 2s/step - loss: 0.7273 - acc: 0.5510 - val_loss: 1.3115 - val_acc: 0.6214\n",
      "Epoch 76/100\n",
      "35/35 [==============================] - 12s 346ms/step - loss: 1.2856 - acc: 0.6071\n",
      "177/177 [==============================] - 350s 2s/step - loss: 0.6953 - acc: 0.5949 - val_loss: 1.2856 - val_acc: 0.6071\n",
      "Epoch 77/100\n",
      "35/35 [==============================] - 13s 358ms/step - loss: 1.0861 - acc: 0.6286\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7408 - acc: 0.5382 - val_loss: 1.0861 - val_acc: 0.6286\n",
      "Epoch 78/100\n",
      "35/35 [==============================] - 12s 347ms/step - loss: 1.3493 - acc: 0.5786\n",
      "177/177 [==============================] - 375s 2s/step - loss: 0.7128 - acc: 0.5935 - val_loss: 1.3493 - val_acc: 0.5786\n",
      "Epoch 79/100\n",
      "35/35 [==============================] - 12s 346ms/step - loss: 1.7437 - acc: 0.6000\n",
      "177/177 [==============================] - 350s 2s/step - loss: 0.7010 - acc: 0.5751 - val_loss: 1.7437 - val_acc: 0.6000\n",
      "Epoch 80/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.8764 - acc: 0.5643\n",
      "177/177 [==============================] - 350s 2s/step - loss: 0.7115 - acc: 0.5822 - val_loss: 1.8764 - val_acc: 0.5643\n",
      "Epoch 81/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.5451 - acc: 0.5929\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7003 - acc: 0.5765 - val_loss: 1.5451 - val_acc: 0.5929\n",
      "Epoch 82/100\n",
      "35/35 [==============================] - 12s 343ms/step - loss: 1.3768 - acc: 0.6071\n",
      "177/177 [==============================] - 350s 2s/step - loss: 0.6922 - acc: 0.5977 - val_loss: 1.3768 - val_acc: 0.6071\n",
      "Epoch 83/100\n",
      "35/35 [==============================] - 12s 347ms/step - loss: 1.1351 - acc: 0.6357\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7364 - acc: 0.5411 - val_loss: 1.1351 - val_acc: 0.6357\n",
      "Epoch 84/100\n",
      "35/35 [==============================] - 12s 349ms/step - loss: 1.1815 - acc: 0.6143\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7081 - acc: 0.5878 - val_loss: 1.1815 - val_acc: 0.6143\n",
      "Epoch 85/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 0.9732 - acc: 0.6143\n",
      "177/177 [==============================] - 350s 2s/step - loss: 0.6874 - acc: 0.5907 - val_loss: 0.9732 - val_acc: 0.6143\n",
      "Epoch 86/100\n",
      "35/35 [==============================] - 12s 346ms/step - loss: 1.0815 - acc: 0.6643\n",
      "177/177 [==============================] - 350s 2s/step - loss: 0.6862 - acc: 0.6176 - val_loss: 1.0815 - val_acc: 0.6643\n",
      "Epoch 87/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.1921 - acc: 0.6000\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.6872 - acc: 0.5907 - val_loss: 1.1921 - val_acc: 0.6000\n",
      "Epoch 88/100\n",
      "35/35 [==============================] - 13s 385ms/step - loss: 1.3799 - acc: 0.5857\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.6949 - acc: 0.5836 - val_loss: 1.3799 - val_acc: 0.5857\n",
      "Epoch 89/100\n",
      "35/35 [==============================] - 14s 386ms/step - loss: 1.4467 - acc: 0.5714\n",
      "177/177 [==============================] - 356s 2s/step - loss: 0.7115 - acc: 0.5652 - val_loss: 1.4467 - val_acc: 0.5714\n",
      "Epoch 90/100\n",
      "35/35 [==============================] - 16s 452ms/step - loss: 1.1733 - acc: 0.6500\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7215 - acc: 0.5779 - val_loss: 1.1733 - val_acc: 0.6500\n",
      "Epoch 91/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.2933 - acc: 0.6571\n",
      "177/177 [==============================] - 353s 2s/step - loss: 0.7074 - acc: 0.5751 - val_loss: 1.2933 - val_acc: 0.6571\n",
      "Epoch 92/100\n",
      "35/35 [==============================] - 12s 352ms/step - loss: 1.1475 - acc: 0.6643\n",
      "177/177 [==============================] - 358s 2s/step - loss: 0.7042 - acc: 0.5595 - val_loss: 1.1475 - val_acc: 0.6643\n",
      "Epoch 93/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.3488 - acc: 0.6429\n",
      "177/177 [==============================] - 363s 2s/step - loss: 0.6933 - acc: 0.6048 - val_loss: 1.3488 - val_acc: 0.6429\n",
      "Epoch 94/100\n",
      "35/35 [==============================] - 12s 351ms/step - loss: 1.4886 - acc: 0.6143\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.7251 - acc: 0.5765 - val_loss: 1.4886 - val_acc: 0.6143\n",
      "Epoch 95/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.6898 - acc: 0.6071\n",
      "177/177 [==============================] - 351s 2s/step - loss: 0.6946 - acc: 0.5864 - val_loss: 1.6898 - val_acc: 0.6071\n",
      "Epoch 96/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.4532 - acc: 0.6429\n",
      "177/177 [==============================] - 362s 2s/step - loss: 0.7111 - acc: 0.5609 - val_loss: 1.4532 - val_acc: 0.6429\n",
      "Epoch 97/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.6644 - acc: 0.6143\n",
      "177/177 [==============================] - 362s 2s/step - loss: 0.7099 - acc: 0.5722 - val_loss: 1.6644 - val_acc: 0.6143\n",
      "Epoch 98/100\n",
      "35/35 [==============================] - 12s 347ms/step - loss: 1.2321 - acc: 0.6071\n",
      "177/177 [==============================] - 349s 2s/step - loss: 0.6885 - acc: 0.5722 - val_loss: 1.2321 - val_acc: 0.6071\n",
      "Epoch 99/100\n",
      "35/35 [==============================] - 12s 348ms/step - loss: 1.3576 - acc: 0.5857\n",
      "177/177 [==============================] - 349s 2s/step - loss: 0.7037 - acc: 0.5652 - val_loss: 1.3576 - val_acc: 0.5857\n",
      "Epoch 100/100\n",
      "35/35 [==============================] - 12s 350ms/step - loss: 1.2964 - acc: 0.6000\n",
      "177/177 [==============================] - 365s 2s/step - loss: 0.6941 - acc: 0.5892 - val_loss: 1.2964 - val_acc: 0.6000\n"
     ]
    }
   ],
   "source": [
    "model, model_out = inter.train_efficient_net(\n",
    "    train_images,\n",
    "    test_images,\n",
    "    validation_images,\n",
    "    False,\n",
    "    0\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(model_out.history[\"loss\"])\n",
    "print(model_out.history[\"val_loss\"])\n",
    "print(model_out.history['acc'])\n",
    "print(model_out.history['val_acc'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred[pred <= 0.5] = 0\n",
    "pred[pred > 0.5] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: \n",
      "\n",
      "0.5106382978723404\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "print('Accuracy: \\n')\n",
    "print(accuracy_score(\n",
    "    test_images.classes,\n",
    "    pred[:,1]\n",
    "))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> **Seems that `transfer learning` works well if the initial and target problems are similar enough**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> Train EfficientNetB1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model, model_out = inter.train_efficient_net(\n",
    "    train_images,\n",
    "    test_images,\n",
    "    validation_images,\n",
    "    1\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> Train EfficientNetB2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model, model_out = inter.train_efficient_net(\n",
    "    train_images,\n",
    "    test_images,\n",
    "    validation_images,\n",
    "    2\n",
    ")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "conda_amazonei_tensorflow_p36",
   "language": "python",
   "name": "conda_amazonei_tensorflow_p36"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}