VGG16-4Piece-Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "fd1efdf1",
   "metadata": {},
   "source": [
    "### Importing Necessary Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0938de48",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pickle\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.preprocessing.image import load_img, img_to_array\n",
    "\n",
    "from tensorflow.keras.layers import Input, Flatten, Dense, Dropout, GlobalAvgPool2D, BatchNormalization\n",
    "# from tensorflow.keras.applications import EfficientNetV2L\n",
    "from tensorflow.keras.regularizers import l2\n",
    "\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "\n",
    "from sklearn.metrics import precision_recall_fscore_support as score"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c954b162",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68302673",
   "metadata": {},
   "source": [
    "A detailed discussion and visualization of the data can be seen in [here](Data-Overview.ipynb)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "75b06a70",
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = pd.read_csv(\"./Data/train_labels.csv\")\n",
    "images = ['./Data/train/train/'+str(name)+'.tif' for name in labels['id']]\n",
    "labels['images'] = images\n",
    "labels = labels[['id', 'images', 'label']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "fd2fe082",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>images</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>./Data/train/train/0.tif</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>./Data/train/train/1.tif</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>./Data/train/train/2.tif</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>./Data/train/train/3.tif</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>./Data/train/train/4.tif</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15995</th>\n",
       "      <td>15995</td>\n",
       "      <td>./Data/train/train/15995.tif</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15996</th>\n",
       "      <td>15996</td>\n",
       "      <td>./Data/train/train/15996.tif</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15997</th>\n",
       "      <td>15997</td>\n",
       "      <td>./Data/train/train/15997.tif</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15998</th>\n",
       "      <td>15998</td>\n",
       "      <td>./Data/train/train/15998.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15999</th>\n",
       "      <td>15999</td>\n",
       "      <td>./Data/train/train/15999.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>16000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          id                        images  label\n",
       "0          0      ./Data/train/train/0.tif      1\n",
       "1          1      ./Data/train/train/1.tif     13\n",
       "2          2      ./Data/train/train/2.tif     13\n",
       "3          3      ./Data/train/train/3.tif     14\n",
       "4          4      ./Data/train/train/4.tif      6\n",
       "...      ...                           ...    ...\n",
       "15995  15995  ./Data/train/train/15995.tif      2\n",
       "15996  15996  ./Data/train/train/15996.tif     15\n",
       "15997  15997  ./Data/train/train/15997.tif      3\n",
       "15998  15998  ./Data/train/train/15998.tif      9\n",
       "15999  15999  ./Data/train/train/15999.tif      9\n",
       "\n",
       "[16000 rows x 3 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "276e2059",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_labels = pd.read_csv('./Data/train_labels_1.csv')\n",
    "test_labels = pd.read_csv('./Data/test_labels_1.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2aea9ea9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>images</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>./Data/train/train/0.tif</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>./Data/train/train/1.tif</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>./Data/train/train/2.tif</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>./Data/train/train/3.tif</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>./Data/train/train/4.tif</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15195</th>\n",
       "      <td>15995</td>\n",
       "      <td>./Data/train/train/15995.tif</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15196</th>\n",
       "      <td>15996</td>\n",
       "      <td>./Data/train/train/15996.tif</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15197</th>\n",
       "      <td>15997</td>\n",
       "      <td>./Data/train/train/15997.tif</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15198</th>\n",
       "      <td>15998</td>\n",
       "      <td>./Data/train/train/15998.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15199</th>\n",
       "      <td>15999</td>\n",
       "      <td>./Data/train/train/15999.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>15200 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          id                        images  label\n",
       "0          0      ./Data/train/train/0.tif      1\n",
       "1          1      ./Data/train/train/1.tif     13\n",
       "2          2      ./Data/train/train/2.tif     13\n",
       "3          3      ./Data/train/train/3.tif     14\n",
       "4          4      ./Data/train/train/4.tif      6\n",
       "...      ...                           ...    ...\n",
       "15195  15995  ./Data/train/train/15995.tif      2\n",
       "15196  15996  ./Data/train/train/15996.tif     15\n",
       "15197  15997  ./Data/train/train/15997.tif      3\n",
       "15198  15998  ./Data/train/train/15998.tif      9\n",
       "15199  15999  ./Data/train/train/15999.tif      9\n",
       "\n",
       "[15200 rows x 3 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b9ea19c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>images</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>18</td>\n",
       "      <td>./Data/train/train/18.tif</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>25</td>\n",
       "      <td>./Data/train/train/25.tif</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33</td>\n",
       "      <td>./Data/train/train/33.tif</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>41</td>\n",
       "      <td>./Data/train/train/41.tif</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>65</td>\n",
       "      <td>./Data/train/train/65.tif</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>795</th>\n",
       "      <td>15816</td>\n",
       "      <td>./Data/train/train/15816.tif</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>796</th>\n",
       "      <td>15824</td>\n",
       "      <td>./Data/train/train/15824.tif</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>797</th>\n",
       "      <td>15832</td>\n",
       "      <td>./Data/train/train/15832.tif</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>798</th>\n",
       "      <td>15863</td>\n",
       "      <td>./Data/train/train/15863.tif</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>799</th>\n",
       "      <td>15994</td>\n",
       "      <td>./Data/train/train/15994.tif</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>800 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        id                        images  label\n",
       "0       18     ./Data/train/train/18.tif     11\n",
       "1       25     ./Data/train/train/25.tif      8\n",
       "2       33     ./Data/train/train/33.tif      6\n",
       "3       41     ./Data/train/train/41.tif      1\n",
       "4       65     ./Data/train/train/65.tif     14\n",
       "..     ...                           ...    ...\n",
       "795  15816  ./Data/train/train/15816.tif      2\n",
       "796  15824  ./Data/train/train/15824.tif      2\n",
       "797  15832  ./Data/train/train/15832.tif     11\n",
       "798  15863  ./Data/train/train/15863.tif     15\n",
       "799  15994  ./Data/train/train/15994.tif      6\n",
       "\n",
       "[800 rows x 3 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "29190fd4",
   "metadata": {},
   "outputs": [],
   "source": [
    "class_labels = list(set(labels['label']))\n",
    "class_names = [\n",
    "    'letter', 'form', 'email', 'handwritten', 'advertisement', 'scientific report', 'scientific publication',\n",
    "    'specification', 'file folder', 'news article', 'budget', 'invoice', 'presentation', 'questionnaire', 'resume',\n",
    "    'memo'\n",
    "]\n",
    "label_names = pd.DataFrame({\n",
    "    'labels': class_labels,\n",
    "    'names': class_names\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2e251c1",
   "metadata": {},
   "source": [
    "## Data Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85b2ccd1",
   "metadata": {},
   "source": [
    "We have already generated the Extracted Feature Vectors from the ResNet Model now we train using that, so we load those feature vectors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "621b1741",
   "metadata": {},
   "outputs": [],
   "source": [
    "FEATURE_DIM = 512"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "fcd66022",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16000, 5, 512)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data = np.concatenate([np.transpose(np.load(f'./Data/VGG16-4P-train-precompute/train_data_precomp-{idx}.npy'), (1,0,2)) for idx in range(10)])\n",
    "np.save(f'./Data/VGG16-4P-train-precompute/whole_train_data_precomp.npy', fine_tune_data)\n",
    "fine_tune_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "37a36e2b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16000, 5, 512)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data = np.load('./Data/VGG16-4P-train-precompute/whole_train_data_precomp.npy')\n",
    "fine_tune_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "568ffcb1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16000,)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_labels = labels['label'].to_numpy()\n",
    "fine_tune_labels.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "937e1e86",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15200, 5, 512)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_train = fine_tune_data[list(train_labels['id'])]\n",
    "fine_tune_data_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "af08c830",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15200,)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_labels_train = train_labels['label'].to_numpy()\n",
    "fine_tune_labels_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "c00f0b32",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(800, 5, 512)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_test = fine_tune_data[list(test_labels['id'])]\n",
    "fine_tune_data_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "16f4bb93",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(800,)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_labels_test = test_labels['label'].to_numpy()\n",
    "fine_tune_labels_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dff1f776",
   "metadata": {},
   "source": [
    "## Fine-Tuning Models"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7380947f",
   "metadata": {},
   "source": [
    "The fine-tuning part of our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "96a2bdfe",
   "metadata": {},
   "outputs": [],
   "source": [
    "def ftmodel():\n",
    "    inputs = Input(shape=(5, FEATURE_DIM))\n",
    "    x = Flatten()(inputs)\n",
    "    x = BatchNormalization()(x)\n",
    "\n",
    "    top_dropout_rate = 0.5\n",
    "    # x = Dropout(top_dropout_rate, name='top_dropout_1')(x)\n",
    "    # x = Dense(1024, name='top_dense_1', kernel_initializer=\"he_normal\", kernel_regularizer=l2(5e-5), activation='relu')(x)\n",
    "    # x = BatchNormalization()(x)\n",
    "    x = Dropout(top_dropout_rate, name='top_dropout_2')(x)\n",
    "    x = Dense(512, name='top_dense_2', kernel_initializer=\"he_normal\", kernel_regularizer=l2(5e-5), activation='relu')(x)\n",
    "    x = BatchNormalization()(x)\n",
    "    x = Dropout(top_dropout_rate, name='top_dropout_3')(x)\n",
    "    outputs = Dense(16, activation='softmax', name='pred', kernel_regularizer=l2(5e-5))(x)\n",
    "\n",
    "    model1 = tf.keras.Model(inputs, outputs, name='VGG16-FineTune')\n",
    "    return model1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "c6574006",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M1 Max\n",
      "Model: \"VGG16-FineTune\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " input_1 (InputLayer)        [(None, 5, 512)]          0         \n",
      "                                                                 \n",
      " flatten (Flatten)           (None, 2560)              0         \n",
      "                                                                 \n",
      " batch_normalization (BatchN  (None, 2560)             10240     \n",
      " ormalization)                                                   \n",
      "                                                                 \n",
      " top_dropout_2 (Dropout)     (None, 2560)              0         \n",
      "                                                                 \n",
      " top_dense_2 (Dense)         (None, 512)               1311232   \n",
      "                                                                 \n",
      " batch_normalization_1 (Batc  (None, 512)              2048      \n",
      " hNormalization)                                                 \n",
      "                                                                 \n",
      " top_dropout_3 (Dropout)     (None, 512)               0         \n",
      "                                                                 \n",
      " pred (Dense)                (None, 16)                8208      \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 1,331,728\n",
      "Trainable params: 1,325,584\n",
      "Non-trainable params: 6,144\n",
      "_________________________________________________________________\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 17:05:43.202464: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:305] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
      "2022-10-09 17:05:43.202559: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:271] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)\n"
     ]
    }
   ],
   "source": [
    "model1 = ftmodel()\n",
    "model1.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e56505e2",
   "metadata": {},
   "source": [
    "### Model Training"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70b7e206",
   "metadata": {},
   "source": [
    "The fine-tuning will be done on the VGG16's $512D$ representation vectors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "96524578",
   "metadata": {},
   "outputs": [],
   "source": [
    "model1.compile(loss = 'sparse_categorical_crossentropy', metrics = ['accuracy'], optimizer = Adam(learning_rate = 0.0003))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "20df8897",
   "metadata": {},
   "outputs": [],
   "source": [
    "EPOCHS = 1000\n",
    "checkpoint_filepaths = ['./Models/VGG16-4P/auto-VGG16-FT-model-weight']\n",
    "model_checkpoint_callback = [tf.keras.callbacks.ModelCheckpoint(\n",
    "    filepath=checkpoint_filepath,\n",
    "    save_weights_only=True,\n",
    "    monitor = \"loss\",\n",
    "    mode='min',\n",
    "    save_best_only=True) for checkpoint_filepath in checkpoint_filepaths]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "03354278",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/1000\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 17:05:43.330045: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n",
      "2022-10-09 17:05:43.588093: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:113] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 1s 30ms/step - loss: 2.9157 - accuracy: 0.2417 - val_loss: 2.5390 - val_accuracy: 0.3162\n",
      "Epoch 2/1000\n",
      " 7/19 [==========>...................] - ETA: 0s - loss: 2.2653 - accuracy: 0.3982"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 17:05:44.217221: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:113] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 12ms/step - loss: 2.1614 - accuracy: 0.4213 - val_loss: 2.0234 - val_accuracy: 0.4212\n",
      "Epoch 3/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.8901 - accuracy: 0.4824 - val_loss: 1.7130 - val_accuracy: 0.5100\n",
      "Epoch 4/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.7489 - accuracy: 0.5173 - val_loss: 1.5601 - val_accuracy: 0.5625\n",
      "Epoch 5/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.6289 - accuracy: 0.5425 - val_loss: 1.4454 - val_accuracy: 0.6100\n",
      "Epoch 6/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.5433 - accuracy: 0.5646 - val_loss: 1.3630 - val_accuracy: 0.6362\n",
      "Epoch 7/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.4789 - accuracy: 0.5781 - val_loss: 1.2957 - val_accuracy: 0.6463\n",
      "Epoch 8/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 1.4227 - accuracy: 0.5922 - val_loss: 1.2461 - val_accuracy: 0.6662\n",
      "Epoch 9/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.3754 - accuracy: 0.6095 - val_loss: 1.2032 - val_accuracy: 0.6787\n",
      "Epoch 10/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.3381 - accuracy: 0.6174 - val_loss: 1.1687 - val_accuracy: 0.6775\n",
      "Epoch 11/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.3053 - accuracy: 0.6293 - val_loss: 1.1413 - val_accuracy: 0.6887\n",
      "Epoch 12/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.2534 - accuracy: 0.6366 - val_loss: 1.1148 - val_accuracy: 0.6987\n",
      "Epoch 13/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.2296 - accuracy: 0.6455 - val_loss: 1.1004 - val_accuracy: 0.6962\n",
      "Epoch 14/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.2082 - accuracy: 0.6530 - val_loss: 1.0767 - val_accuracy: 0.7037\n",
      "Epoch 15/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.1730 - accuracy: 0.6605 - val_loss: 1.0669 - val_accuracy: 0.7037\n",
      "Epoch 16/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 1.1604 - accuracy: 0.6639 - val_loss: 1.0462 - val_accuracy: 0.7138\n",
      "Epoch 17/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.1385 - accuracy: 0.6712 - val_loss: 1.0351 - val_accuracy: 0.7225\n",
      "Epoch 18/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.1156 - accuracy: 0.6759 - val_loss: 1.0221 - val_accuracy: 0.7150\n",
      "Epoch 19/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.1007 - accuracy: 0.6784 - val_loss: 1.0129 - val_accuracy: 0.7188\n",
      "Epoch 20/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.0734 - accuracy: 0.6884 - val_loss: 1.0050 - val_accuracy: 0.7237\n",
      "Epoch 21/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.0672 - accuracy: 0.6921 - val_loss: 1.0012 - val_accuracy: 0.7200\n",
      "Epoch 22/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.0473 - accuracy: 0.6905 - val_loss: 0.9942 - val_accuracy: 0.7350\n",
      "Epoch 23/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 1.0277 - accuracy: 0.6979 - val_loss: 0.9885 - val_accuracy: 0.7262\n",
      "Epoch 24/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 1.0183 - accuracy: 0.7017 - val_loss: 0.9810 - val_accuracy: 0.7387\n",
      "Epoch 25/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 1.0115 - accuracy: 0.7064 - val_loss: 0.9699 - val_accuracy: 0.7375\n",
      "Epoch 26/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.9899 - accuracy: 0.7102 - val_loss: 0.9712 - val_accuracy: 0.7337\n",
      "Epoch 27/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.9832 - accuracy: 0.7092 - val_loss: 0.9681 - val_accuracy: 0.7375\n",
      "Epoch 28/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.9725 - accuracy: 0.7129 - val_loss: 0.9616 - val_accuracy: 0.7362\n",
      "Epoch 29/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.9587 - accuracy: 0.7197 - val_loss: 0.9608 - val_accuracy: 0.7412\n",
      "Epoch 30/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.9480 - accuracy: 0.7234 - val_loss: 0.9494 - val_accuracy: 0.7387\n",
      "Epoch 31/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.9349 - accuracy: 0.7261 - val_loss: 0.9501 - val_accuracy: 0.7425\n",
      "Epoch 32/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.9367 - accuracy: 0.7232 - val_loss: 0.9451 - val_accuracy: 0.7362\n",
      "Epoch 33/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.9150 - accuracy: 0.7294 - val_loss: 0.9428 - val_accuracy: 0.7450\n",
      "Epoch 34/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.9140 - accuracy: 0.7308 - val_loss: 0.9447 - val_accuracy: 0.7425\n",
      "Epoch 35/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.9010 - accuracy: 0.7345 - val_loss: 0.9470 - val_accuracy: 0.7437\n",
      "Epoch 36/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.8870 - accuracy: 0.7357 - val_loss: 0.9348 - val_accuracy: 0.7387\n",
      "Epoch 37/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.8791 - accuracy: 0.7373 - val_loss: 0.9336 - val_accuracy: 0.7500\n",
      "Epoch 38/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.8781 - accuracy: 0.7390 - val_loss: 0.9334 - val_accuracy: 0.7475\n",
      "Epoch 39/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.8699 - accuracy: 0.7433 - val_loss: 0.9333 - val_accuracy: 0.7512\n",
      "Epoch 40/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.8538 - accuracy: 0.7493 - val_loss: 0.9278 - val_accuracy: 0.7537\n",
      "Epoch 41/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.8404 - accuracy: 0.7547 - val_loss: 0.9228 - val_accuracy: 0.7500\n",
      "Epoch 42/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.8456 - accuracy: 0.7495 - val_loss: 0.9280 - val_accuracy: 0.7487\n",
      "Epoch 43/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.8336 - accuracy: 0.7537 - val_loss: 0.9228 - val_accuracy: 0.7525\n",
      "Epoch 44/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.8305 - accuracy: 0.7588 - val_loss: 0.9212 - val_accuracy: 0.7537\n",
      "Epoch 45/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.8225 - accuracy: 0.7589 - val_loss: 0.9169 - val_accuracy: 0.7525\n",
      "Epoch 46/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.8204 - accuracy: 0.7572 - val_loss: 0.9091 - val_accuracy: 0.7537\n",
      "Epoch 47/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.7965 - accuracy: 0.7651 - val_loss: 0.9170 - val_accuracy: 0.7500\n",
      "Epoch 48/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.7938 - accuracy: 0.7656 - val_loss: 0.9104 - val_accuracy: 0.7450\n",
      "Epoch 49/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.7995 - accuracy: 0.7620 - val_loss: 0.9156 - val_accuracy: 0.7512\n",
      "Epoch 50/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.7842 - accuracy: 0.7720 - val_loss: 0.9203 - val_accuracy: 0.7475\n",
      "Epoch 51/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.7852 - accuracy: 0.7661 - val_loss: 0.9187 - val_accuracy: 0.7525\n",
      "Epoch 52/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.7732 - accuracy: 0.7716 - val_loss: 0.9222 - val_accuracy: 0.7462\n",
      "Epoch 53/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.7609 - accuracy: 0.7738 - val_loss: 0.9183 - val_accuracy: 0.7500\n",
      "Epoch 54/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.7546 - accuracy: 0.7789 - val_loss: 0.9207 - val_accuracy: 0.7462\n",
      "Epoch 55/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.7543 - accuracy: 0.7764 - val_loss: 0.9137 - val_accuracy: 0.7550\n",
      "Epoch 56/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.7452 - accuracy: 0.7835 - val_loss: 0.9167 - val_accuracy: 0.7475\n",
      "Epoch 57/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.7367 - accuracy: 0.7860 - val_loss: 0.9072 - val_accuracy: 0.7487\n",
      "Epoch 58/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.7484 - accuracy: 0.7834 - val_loss: 0.9062 - val_accuracy: 0.7537\n",
      "Epoch 59/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 13ms/step - loss: 0.7287 - accuracy: 0.7838 - val_loss: 0.9114 - val_accuracy: 0.7537\n",
      "Epoch 60/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.7259 - accuracy: 0.7874 - val_loss: 0.9060 - val_accuracy: 0.7487\n",
      "Epoch 61/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.7279 - accuracy: 0.7877 - val_loss: 0.9166 - val_accuracy: 0.7475\n",
      "Epoch 62/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.7128 - accuracy: 0.7954 - val_loss: 0.9096 - val_accuracy: 0.7512\n",
      "Epoch 63/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.7083 - accuracy: 0.7935 - val_loss: 0.9044 - val_accuracy: 0.7600\n",
      "Epoch 64/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.7150 - accuracy: 0.7897 - val_loss: 0.9106 - val_accuracy: 0.7562\n",
      "Epoch 65/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.7077 - accuracy: 0.7909 - val_loss: 0.9020 - val_accuracy: 0.7562\n",
      "Epoch 66/1000\n",
      "19/19 [==============================] - 0s 17ms/step - loss: 0.6941 - accuracy: 0.7964 - val_loss: 0.8961 - val_accuracy: 0.7612\n",
      "Epoch 67/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6917 - accuracy: 0.7992 - val_loss: 0.8983 - val_accuracy: 0.7587\n",
      "Epoch 68/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6897 - accuracy: 0.8007 - val_loss: 0.8985 - val_accuracy: 0.7587\n",
      "Epoch 69/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.6766 - accuracy: 0.8047 - val_loss: 0.9005 - val_accuracy: 0.7587\n",
      "Epoch 70/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.6834 - accuracy: 0.8014 - val_loss: 0.8987 - val_accuracy: 0.7612\n",
      "Epoch 71/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.6778 - accuracy: 0.8032 - val_loss: 0.9018 - val_accuracy: 0.7612\n",
      "Epoch 72/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.6660 - accuracy: 0.8059 - val_loss: 0.9014 - val_accuracy: 0.7600\n",
      "Epoch 73/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.6684 - accuracy: 0.8022 - val_loss: 0.9098 - val_accuracy: 0.7600\n",
      "Epoch 74/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6561 - accuracy: 0.8074 - val_loss: 0.9145 - val_accuracy: 0.7612\n",
      "Epoch 75/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6446 - accuracy: 0.8128 - val_loss: 0.9048 - val_accuracy: 0.7587\n",
      "Epoch 76/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.6374 - accuracy: 0.8175 - val_loss: 0.9054 - val_accuracy: 0.7637\n",
      "Epoch 77/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.6474 - accuracy: 0.8076 - val_loss: 0.8991 - val_accuracy: 0.7587\n",
      "Epoch 78/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.6432 - accuracy: 0.8111 - val_loss: 0.9025 - val_accuracy: 0.7600\n",
      "Epoch 79/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.6449 - accuracy: 0.8113 - val_loss: 0.9019 - val_accuracy: 0.7612\n",
      "Epoch 80/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.6373 - accuracy: 0.8177 - val_loss: 0.8984 - val_accuracy: 0.7612\n",
      "Epoch 81/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6278 - accuracy: 0.8151 - val_loss: 0.9094 - val_accuracy: 0.7512\n",
      "Epoch 82/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6258 - accuracy: 0.8157 - val_loss: 0.9011 - val_accuracy: 0.7600\n",
      "Epoch 83/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.6086 - accuracy: 0.8222 - val_loss: 0.9063 - val_accuracy: 0.7575\n",
      "Epoch 84/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.6164 - accuracy: 0.8197 - val_loss: 0.9018 - val_accuracy: 0.7650\n",
      "Epoch 85/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6080 - accuracy: 0.8240 - val_loss: 0.9046 - val_accuracy: 0.7662\n",
      "Epoch 86/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6064 - accuracy: 0.8211 - val_loss: 0.8962 - val_accuracy: 0.7700\n",
      "Epoch 87/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.6009 - accuracy: 0.8269 - val_loss: 0.9023 - val_accuracy: 0.7662\n",
      "Epoch 88/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.5934 - accuracy: 0.8301 - val_loss: 0.8882 - val_accuracy: 0.7738\n",
      "Epoch 89/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5982 - accuracy: 0.8259 - val_loss: 0.8920 - val_accuracy: 0.7637\n",
      "Epoch 90/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5938 - accuracy: 0.8263 - val_loss: 0.8960 - val_accuracy: 0.7587\n",
      "Epoch 91/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.5915 - accuracy: 0.8291 - val_loss: 0.8927 - val_accuracy: 0.7675\n",
      "Epoch 92/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.5876 - accuracy: 0.8300 - val_loss: 0.8968 - val_accuracy: 0.7725\n",
      "Epoch 93/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5752 - accuracy: 0.8374 - val_loss: 0.8970 - val_accuracy: 0.7650\n",
      "Epoch 94/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5806 - accuracy: 0.8314 - val_loss: 0.8905 - val_accuracy: 0.7713\n",
      "Epoch 95/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5768 - accuracy: 0.8341 - val_loss: 0.8985 - val_accuracy: 0.7688\n",
      "Epoch 96/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.5748 - accuracy: 0.8344 - val_loss: 0.8996 - val_accuracy: 0.7612\n",
      "Epoch 97/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.5681 - accuracy: 0.8357 - val_loss: 0.9022 - val_accuracy: 0.7662\n",
      "Epoch 98/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5712 - accuracy: 0.8353 - val_loss: 0.8975 - val_accuracy: 0.7650\n",
      "Epoch 99/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.5718 - accuracy: 0.8351 - val_loss: 0.8966 - val_accuracy: 0.7675\n",
      "Epoch 100/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5597 - accuracy: 0.8418 - val_loss: 0.9102 - val_accuracy: 0.7575\n",
      "Epoch 101/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5660 - accuracy: 0.8353 - val_loss: 0.8999 - val_accuracy: 0.7587\n",
      "Epoch 102/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5513 - accuracy: 0.8393 - val_loss: 0.8966 - val_accuracy: 0.7562\n",
      "Epoch 103/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5473 - accuracy: 0.8409 - val_loss: 0.8944 - val_accuracy: 0.7612\n",
      "Epoch 104/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5448 - accuracy: 0.8451 - val_loss: 0.9003 - val_accuracy: 0.7688\n",
      "Epoch 105/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.5368 - accuracy: 0.8475 - val_loss: 0.8935 - val_accuracy: 0.7688\n",
      "Epoch 106/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5381 - accuracy: 0.8454 - val_loss: 0.8923 - val_accuracy: 0.7713\n",
      "Epoch 107/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5407 - accuracy: 0.8454 - val_loss: 0.8992 - val_accuracy: 0.7625\n",
      "Epoch 108/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5361 - accuracy: 0.8451 - val_loss: 0.8937 - val_accuracy: 0.7675\n",
      "Epoch 109/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5244 - accuracy: 0.8499 - val_loss: 0.8928 - val_accuracy: 0.7625\n",
      "Epoch 110/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5252 - accuracy: 0.8493 - val_loss: 0.9001 - val_accuracy: 0.7675\n",
      "Epoch 111/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.5227 - accuracy: 0.8473 - val_loss: 0.8946 - val_accuracy: 0.7612\n",
      "Epoch 112/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5103 - accuracy: 0.8561 - val_loss: 0.8968 - val_accuracy: 0.7650\n",
      "Epoch 113/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5125 - accuracy: 0.8551 - val_loss: 0.8897 - val_accuracy: 0.7675\n",
      "Epoch 114/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5120 - accuracy: 0.8559 - val_loss: 0.8979 - val_accuracy: 0.7612\n",
      "Epoch 115/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5047 - accuracy: 0.8563 - val_loss: 0.8903 - val_accuracy: 0.7662\n",
      "Epoch 116/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 9ms/step - loss: 0.5072 - accuracy: 0.8551 - val_loss: 0.8902 - val_accuracy: 0.7650\n",
      "Epoch 117/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.5004 - accuracy: 0.8563 - val_loss: 0.8989 - val_accuracy: 0.7675\n",
      "Epoch 118/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4982 - accuracy: 0.8588 - val_loss: 0.8913 - val_accuracy: 0.7700\n",
      "Epoch 119/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4831 - accuracy: 0.8634 - val_loss: 0.9019 - val_accuracy: 0.7612\n",
      "Epoch 120/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.4868 - accuracy: 0.8634 - val_loss: 0.8936 - val_accuracy: 0.7625\n",
      "Epoch 121/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.4882 - accuracy: 0.8599 - val_loss: 0.8954 - val_accuracy: 0.7650\n",
      "Epoch 122/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.4874 - accuracy: 0.8618 - val_loss: 0.9055 - val_accuracy: 0.7612\n",
      "Epoch 123/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4802 - accuracy: 0.8656 - val_loss: 0.9042 - val_accuracy: 0.7675\n",
      "Epoch 124/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4724 - accuracy: 0.8687 - val_loss: 0.9032 - val_accuracy: 0.7625\n",
      "Epoch 125/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.4761 - accuracy: 0.8674 - val_loss: 0.8995 - val_accuracy: 0.7650\n",
      "Epoch 126/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4712 - accuracy: 0.8678 - val_loss: 0.9022 - val_accuracy: 0.7650\n",
      "Epoch 127/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4676 - accuracy: 0.8683 - val_loss: 0.8913 - val_accuracy: 0.7675\n",
      "Epoch 128/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4619 - accuracy: 0.8691 - val_loss: 0.8931 - val_accuracy: 0.7700\n",
      "Epoch 129/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.4725 - accuracy: 0.8662 - val_loss: 0.8918 - val_accuracy: 0.7725\n",
      "Epoch 130/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4557 - accuracy: 0.8713 - val_loss: 0.8916 - val_accuracy: 0.7650\n",
      "Epoch 131/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.4583 - accuracy: 0.8699 - val_loss: 0.8926 - val_accuracy: 0.7750\n",
      "Epoch 132/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.4521 - accuracy: 0.8725 - val_loss: 0.8869 - val_accuracy: 0.7750\n",
      "Epoch 133/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.4508 - accuracy: 0.8743 - val_loss: 0.8958 - val_accuracy: 0.7662\n",
      "Epoch 134/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4491 - accuracy: 0.8739 - val_loss: 0.8959 - val_accuracy: 0.7675\n",
      "Epoch 135/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.4584 - accuracy: 0.8742 - val_loss: 0.9000 - val_accuracy: 0.7662\n",
      "Epoch 136/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.4490 - accuracy: 0.8739 - val_loss: 0.9066 - val_accuracy: 0.7700\n",
      "Epoch 137/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.4393 - accuracy: 0.8785 - val_loss: 0.9125 - val_accuracy: 0.7675\n",
      "Epoch 138/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.4400 - accuracy: 0.8780 - val_loss: 0.9049 - val_accuracy: 0.7600\n",
      "Epoch 139/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.4256 - accuracy: 0.8820 - val_loss: 0.8927 - val_accuracy: 0.7688\n",
      "Epoch 140/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.4387 - accuracy: 0.8775 - val_loss: 0.9034 - val_accuracy: 0.7675\n",
      "Epoch 141/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.4299 - accuracy: 0.8809 - val_loss: 0.8941 - val_accuracy: 0.7688\n",
      "Epoch 142/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.4337 - accuracy: 0.8799 - val_loss: 0.8942 - val_accuracy: 0.7700\n",
      "Epoch 143/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.4205 - accuracy: 0.8832 - val_loss: 0.9040 - val_accuracy: 0.7662\n",
      "Epoch 144/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.4218 - accuracy: 0.8824 - val_loss: 0.9006 - val_accuracy: 0.7662\n",
      "Epoch 145/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.4178 - accuracy: 0.8850 - val_loss: 0.9059 - val_accuracy: 0.7700\n",
      "Epoch 146/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.4082 - accuracy: 0.8886 - val_loss: 0.8963 - val_accuracy: 0.7750\n",
      "Epoch 147/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.4109 - accuracy: 0.8876 - val_loss: 0.9102 - val_accuracy: 0.7650\n",
      "Epoch 148/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.4245 - accuracy: 0.8832 - val_loss: 0.9109 - val_accuracy: 0.7688\n",
      "Epoch 149/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.4115 - accuracy: 0.8855 - val_loss: 0.9146 - val_accuracy: 0.7713\n",
      "Epoch 150/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.4216 - accuracy: 0.8825 - val_loss: 0.9091 - val_accuracy: 0.7662\n",
      "Epoch 151/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.4154 - accuracy: 0.8868 - val_loss: 0.9056 - val_accuracy: 0.7637\n",
      "Epoch 152/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.4057 - accuracy: 0.8909 - val_loss: 0.9183 - val_accuracy: 0.7650\n",
      "Epoch 153/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.4050 - accuracy: 0.8864 - val_loss: 0.9119 - val_accuracy: 0.7625\n",
      "Epoch 154/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.4109 - accuracy: 0.8871 - val_loss: 0.9088 - val_accuracy: 0.7662\n",
      "Epoch 155/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.3990 - accuracy: 0.8903 - val_loss: 0.9202 - val_accuracy: 0.7637\n",
      "Epoch 156/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.4037 - accuracy: 0.8890 - val_loss: 0.9135 - val_accuracy: 0.7637\n",
      "Epoch 157/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.3930 - accuracy: 0.8915 - val_loss: 0.9138 - val_accuracy: 0.7562\n",
      "Epoch 158/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3963 - accuracy: 0.8916 - val_loss: 0.9111 - val_accuracy: 0.7587\n",
      "Epoch 159/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3919 - accuracy: 0.8924 - val_loss: 0.9066 - val_accuracy: 0.7650\n",
      "Epoch 160/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3883 - accuracy: 0.8950 - val_loss: 0.8999 - val_accuracy: 0.7700\n",
      "Epoch 161/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.3883 - accuracy: 0.8946 - val_loss: 0.9061 - val_accuracy: 0.7625\n",
      "Epoch 162/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3843 - accuracy: 0.8961 - val_loss: 0.9113 - val_accuracy: 0.7625\n",
      "Epoch 163/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3854 - accuracy: 0.8952 - val_loss: 0.9141 - val_accuracy: 0.7612\n",
      "Epoch 164/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3916 - accuracy: 0.8926 - val_loss: 0.9166 - val_accuracy: 0.7650\n",
      "Epoch 165/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3774 - accuracy: 0.8973 - val_loss: 0.9155 - val_accuracy: 0.7713\n",
      "Epoch 166/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3766 - accuracy: 0.8988 - val_loss: 0.9228 - val_accuracy: 0.7637\n",
      "Epoch 167/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3916 - accuracy: 0.8943 - val_loss: 0.9167 - val_accuracy: 0.7662\n",
      "Epoch 168/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.3771 - accuracy: 0.8989 - val_loss: 0.9188 - val_accuracy: 0.7625\n",
      "Epoch 169/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3756 - accuracy: 0.8993 - val_loss: 0.9150 - val_accuracy: 0.7612\n",
      "Epoch 170/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3701 - accuracy: 0.9005 - val_loss: 0.9263 - val_accuracy: 0.7575\n",
      "Epoch 171/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3661 - accuracy: 0.9043 - val_loss: 0.9256 - val_accuracy: 0.7575\n",
      "Epoch 172/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3726 - accuracy: 0.8983 - val_loss: 0.9184 - val_accuracy: 0.7537\n",
      "Epoch 173/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.3636 - accuracy: 0.9003 - val_loss: 0.9331 - val_accuracy: 0.7537\n",
      "Epoch 174/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3612 - accuracy: 0.9026 - val_loss: 0.9278 - val_accuracy: 0.7500\n",
      "Epoch 175/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3618 - accuracy: 0.9039 - val_loss: 0.9332 - val_accuracy: 0.7562\n",
      "Epoch 176/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.3614 - accuracy: 0.9027 - val_loss: 0.9322 - val_accuracy: 0.7637\n",
      "Epoch 177/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3650 - accuracy: 0.9038 - val_loss: 0.9259 - val_accuracy: 0.7575\n",
      "Epoch 178/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.3674 - accuracy: 0.9003 - val_loss: 0.9339 - val_accuracy: 0.7675\n",
      "Epoch 179/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3572 - accuracy: 0.9041 - val_loss: 0.9378 - val_accuracy: 0.7600\n",
      "Epoch 180/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3573 - accuracy: 0.9048 - val_loss: 0.9310 - val_accuracy: 0.7587\n",
      "Epoch 181/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3478 - accuracy: 0.9082 - val_loss: 0.9360 - val_accuracy: 0.7612\n",
      "Epoch 182/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3599 - accuracy: 0.9014 - val_loss: 0.9384 - val_accuracy: 0.7637\n",
      "Epoch 183/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3444 - accuracy: 0.9100 - val_loss: 0.9383 - val_accuracy: 0.7562\n",
      "Epoch 184/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3572 - accuracy: 0.9054 - val_loss: 0.9333 - val_accuracy: 0.7587\n",
      "Epoch 185/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.3560 - accuracy: 0.9038 - val_loss: 0.9268 - val_accuracy: 0.7625\n",
      "Epoch 186/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.3480 - accuracy: 0.9048 - val_loss: 0.9313 - val_accuracy: 0.7662\n",
      "Epoch 187/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3426 - accuracy: 0.9089 - val_loss: 0.9383 - val_accuracy: 0.7587\n",
      "Epoch 188/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3453 - accuracy: 0.9066 - val_loss: 0.9354 - val_accuracy: 0.7575\n",
      "Epoch 189/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.3519 - accuracy: 0.9059 - val_loss: 0.9265 - val_accuracy: 0.7612\n",
      "Epoch 190/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.3417 - accuracy: 0.9099 - val_loss: 0.9233 - val_accuracy: 0.7625\n",
      "Epoch 191/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3396 - accuracy: 0.9101 - val_loss: 0.9247 - val_accuracy: 0.7625\n",
      "Epoch 192/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.3425 - accuracy: 0.9109 - val_loss: 0.9218 - val_accuracy: 0.7625\n",
      "Epoch 193/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3343 - accuracy: 0.9135 - val_loss: 0.9312 - val_accuracy: 0.7575\n",
      "Epoch 194/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3352 - accuracy: 0.9132 - val_loss: 0.9295 - val_accuracy: 0.7587\n",
      "Epoch 195/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.3401 - accuracy: 0.9103 - val_loss: 0.9284 - val_accuracy: 0.7575\n",
      "Epoch 196/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3297 - accuracy: 0.9153 - val_loss: 0.9304 - val_accuracy: 0.7625\n",
      "Epoch 197/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3387 - accuracy: 0.9103 - val_loss: 0.9302 - val_accuracy: 0.7562\n",
      "Epoch 198/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.3251 - accuracy: 0.9170 - val_loss: 0.9311 - val_accuracy: 0.7562\n",
      "Epoch 199/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.3342 - accuracy: 0.9116 - val_loss: 0.9361 - val_accuracy: 0.7662\n",
      "Epoch 200/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3258 - accuracy: 0.9152 - val_loss: 0.9437 - val_accuracy: 0.7612\n",
      "Epoch 201/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3344 - accuracy: 0.9116 - val_loss: 0.9389 - val_accuracy: 0.7512\n",
      "Epoch 202/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3262 - accuracy: 0.9145 - val_loss: 0.9327 - val_accuracy: 0.7650\n",
      "Epoch 203/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3270 - accuracy: 0.9153 - val_loss: 0.9449 - val_accuracy: 0.7675\n",
      "Epoch 204/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3174 - accuracy: 0.9160 - val_loss: 0.9403 - val_accuracy: 0.7662\n",
      "Epoch 205/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3228 - accuracy: 0.9155 - val_loss: 0.9384 - val_accuracy: 0.7650\n",
      "Epoch 206/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.3157 - accuracy: 0.9181 - val_loss: 0.9357 - val_accuracy: 0.7625\n",
      "Epoch 207/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.3229 - accuracy: 0.9155 - val_loss: 0.9408 - val_accuracy: 0.7662\n",
      "Epoch 208/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3175 - accuracy: 0.9192 - val_loss: 0.9439 - val_accuracy: 0.7713\n",
      "Epoch 209/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3118 - accuracy: 0.9176 - val_loss: 0.9405 - val_accuracy: 0.7637\n",
      "Epoch 210/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3237 - accuracy: 0.9161 - val_loss: 0.9433 - val_accuracy: 0.7675\n",
      "Epoch 211/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3136 - accuracy: 0.9179 - val_loss: 0.9595 - val_accuracy: 0.7600\n",
      "Epoch 212/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3074 - accuracy: 0.9220 - val_loss: 0.9349 - val_accuracy: 0.7612\n",
      "Epoch 213/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.3128 - accuracy: 0.9168 - val_loss: 0.9528 - val_accuracy: 0.7625\n",
      "Epoch 214/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3084 - accuracy: 0.9199 - val_loss: 0.9426 - val_accuracy: 0.7688\n",
      "Epoch 215/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3087 - accuracy: 0.9202 - val_loss: 0.9560 - val_accuracy: 0.7612\n",
      "Epoch 216/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3059 - accuracy: 0.9188 - val_loss: 0.9469 - val_accuracy: 0.7625\n",
      "Epoch 217/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.3074 - accuracy: 0.9215 - val_loss: 0.9428 - val_accuracy: 0.7612\n",
      "Epoch 218/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3097 - accuracy: 0.9206 - val_loss: 0.9434 - val_accuracy: 0.7650\n",
      "Epoch 219/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.3082 - accuracy: 0.9216 - val_loss: 0.9452 - val_accuracy: 0.7625\n",
      "Epoch 220/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.3014 - accuracy: 0.9241 - val_loss: 0.9514 - val_accuracy: 0.7675\n",
      "Epoch 221/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3071 - accuracy: 0.9221 - val_loss: 0.9582 - val_accuracy: 0.7625\n",
      "Epoch 222/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.3052 - accuracy: 0.9208 - val_loss: 0.9563 - val_accuracy: 0.7612\n",
      "Epoch 223/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3087 - accuracy: 0.9210 - val_loss: 0.9482 - val_accuracy: 0.7612\n",
      "Epoch 224/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2989 - accuracy: 0.9230 - val_loss: 0.9622 - val_accuracy: 0.7625\n",
      "Epoch 225/1000\n",
      "19/19 [==============================] - 0s 19ms/step - loss: 0.2955 - accuracy: 0.9247 - val_loss: 0.9578 - val_accuracy: 0.7587\n",
      "Epoch 226/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2987 - accuracy: 0.9244 - val_loss: 0.9674 - val_accuracy: 0.7612\n",
      "Epoch 227/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.3000 - accuracy: 0.9249 - val_loss: 0.9655 - val_accuracy: 0.7587\n",
      "Epoch 228/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2914 - accuracy: 0.9277 - val_loss: 0.9739 - val_accuracy: 0.7612\n",
      "Epoch 229/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2956 - accuracy: 0.9255 - val_loss: 0.9609 - val_accuracy: 0.7625\n",
      "Epoch 230/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2965 - accuracy: 0.9232 - val_loss: 0.9665 - val_accuracy: 0.7612\n",
      "Epoch 231/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2963 - accuracy: 0.9263 - val_loss: 0.9664 - val_accuracy: 0.7650\n",
      "Epoch 232/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2902 - accuracy: 0.9228 - val_loss: 0.9537 - val_accuracy: 0.7637\n",
      "Epoch 233/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2933 - accuracy: 0.9251 - val_loss: 0.9623 - val_accuracy: 0.7587\n",
      "Epoch 234/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2880 - accuracy: 0.9273 - val_loss: 0.9739 - val_accuracy: 0.7575\n",
      "Epoch 235/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2896 - accuracy: 0.9282 - val_loss: 0.9777 - val_accuracy: 0.7562\n",
      "Epoch 236/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.3032 - accuracy: 0.9228 - val_loss: 0.9727 - val_accuracy: 0.7600\n",
      "Epoch 237/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2842 - accuracy: 0.9303 - val_loss: 0.9597 - val_accuracy: 0.7637\n",
      "Epoch 238/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2975 - accuracy: 0.9226 - val_loss: 0.9618 - val_accuracy: 0.7600\n",
      "Epoch 239/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2853 - accuracy: 0.9286 - val_loss: 0.9577 - val_accuracy: 0.7625\n",
      "Epoch 240/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2905 - accuracy: 0.9283 - val_loss: 0.9640 - val_accuracy: 0.7612\n",
      "Epoch 241/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2825 - accuracy: 0.9314 - val_loss: 0.9734 - val_accuracy: 0.7625\n",
      "Epoch 242/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2877 - accuracy: 0.9295 - val_loss: 0.9561 - val_accuracy: 0.7587\n",
      "Epoch 243/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2913 - accuracy: 0.9259 - val_loss: 0.9854 - val_accuracy: 0.7600\n",
      "Epoch 244/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2846 - accuracy: 0.9282 - val_loss: 0.9744 - val_accuracy: 0.7625\n",
      "Epoch 245/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2912 - accuracy: 0.9261 - val_loss: 0.9777 - val_accuracy: 0.7600\n",
      "Epoch 246/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2859 - accuracy: 0.9276 - val_loss: 0.9731 - val_accuracy: 0.7700\n",
      "Epoch 247/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2864 - accuracy: 0.9282 - val_loss: 0.9697 - val_accuracy: 0.7637\n",
      "Epoch 248/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2754 - accuracy: 0.9311 - val_loss: 0.9690 - val_accuracy: 0.7625\n",
      "Epoch 249/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2801 - accuracy: 0.9307 - val_loss: 0.9675 - val_accuracy: 0.7612\n",
      "Epoch 250/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2811 - accuracy: 0.9301 - val_loss: 0.9777 - val_accuracy: 0.7637\n",
      "Epoch 251/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2803 - accuracy: 0.9314 - val_loss: 0.9832 - val_accuracy: 0.7625\n",
      "Epoch 252/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2769 - accuracy: 0.9318 - val_loss: 0.9821 - val_accuracy: 0.7637\n",
      "Epoch 253/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.2752 - accuracy: 0.9339 - val_loss: 0.9808 - val_accuracy: 0.7650\n",
      "Epoch 254/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2818 - accuracy: 0.9303 - val_loss: 0.9889 - val_accuracy: 0.7637\n",
      "Epoch 255/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2682 - accuracy: 0.9335 - val_loss: 0.9850 - val_accuracy: 0.7637\n",
      "Epoch 256/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2718 - accuracy: 0.9359 - val_loss: 0.9818 - val_accuracy: 0.7637\n",
      "Epoch 257/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2761 - accuracy: 0.9316 - val_loss: 0.9812 - val_accuracy: 0.7612\n",
      "Epoch 258/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2766 - accuracy: 0.9310 - val_loss: 0.9881 - val_accuracy: 0.7612\n",
      "Epoch 259/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2687 - accuracy: 0.9331 - val_loss: 0.9921 - val_accuracy: 0.7537\n",
      "Epoch 260/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2694 - accuracy: 0.9341 - val_loss: 0.9818 - val_accuracy: 0.7575\n",
      "Epoch 261/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2656 - accuracy: 0.9346 - val_loss: 0.9879 - val_accuracy: 0.7662\n",
      "Epoch 262/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2672 - accuracy: 0.9341 - val_loss: 0.9825 - val_accuracy: 0.7675\n",
      "Epoch 263/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2656 - accuracy: 0.9358 - val_loss: 0.9813 - val_accuracy: 0.7612\n",
      "Epoch 264/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2660 - accuracy: 0.9363 - val_loss: 0.9781 - val_accuracy: 0.7662\n",
      "Epoch 265/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2666 - accuracy: 0.9332 - val_loss: 0.9824 - val_accuracy: 0.7600\n",
      "Epoch 266/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2634 - accuracy: 0.9342 - val_loss: 0.9900 - val_accuracy: 0.7612\n",
      "Epoch 267/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2658 - accuracy: 0.9337 - val_loss: 0.9856 - val_accuracy: 0.7612\n",
      "Epoch 268/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2673 - accuracy: 0.9330 - val_loss: 0.9768 - val_accuracy: 0.7637\n",
      "Epoch 269/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2586 - accuracy: 0.9369 - val_loss: 0.9794 - val_accuracy: 0.7612\n",
      "Epoch 270/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2657 - accuracy: 0.9356 - val_loss: 0.9938 - val_accuracy: 0.7662\n",
      "Epoch 271/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2585 - accuracy: 0.9364 - val_loss: 0.9841 - val_accuracy: 0.7625\n",
      "Epoch 272/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2592 - accuracy: 0.9375 - val_loss: 0.9911 - val_accuracy: 0.7625\n",
      "Epoch 273/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2636 - accuracy: 0.9348 - val_loss: 0.9952 - val_accuracy: 0.7525\n",
      "Epoch 274/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2537 - accuracy: 0.9413 - val_loss: 0.9921 - val_accuracy: 0.7562\n",
      "Epoch 275/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2633 - accuracy: 0.9367 - val_loss: 0.9878 - val_accuracy: 0.7587\n",
      "Epoch 276/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2581 - accuracy: 0.9348 - val_loss: 0.9924 - val_accuracy: 0.7587\n",
      "Epoch 277/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2578 - accuracy: 0.9378 - val_loss: 0.9981 - val_accuracy: 0.7600\n",
      "Epoch 278/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2572 - accuracy: 0.9376 - val_loss: 1.0014 - val_accuracy: 0.7625\n",
      "Epoch 279/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2601 - accuracy: 0.9377 - val_loss: 1.0053 - val_accuracy: 0.7625\n",
      "Epoch 280/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2597 - accuracy: 0.9359 - val_loss: 0.9984 - val_accuracy: 0.7637\n",
      "Epoch 281/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2565 - accuracy: 0.9388 - val_loss: 0.9880 - val_accuracy: 0.7688\n",
      "Epoch 282/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2566 - accuracy: 0.9385 - val_loss: 0.9933 - val_accuracy: 0.7637\n",
      "Epoch 283/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.2519 - accuracy: 0.9386 - val_loss: 0.9887 - val_accuracy: 0.7650\n",
      "Epoch 284/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2554 - accuracy: 0.9396 - val_loss: 0.9962 - val_accuracy: 0.7637\n",
      "Epoch 285/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2596 - accuracy: 0.9391 - val_loss: 1.0037 - val_accuracy: 0.7587\n",
      "Epoch 286/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2480 - accuracy: 0.9424 - val_loss: 0.9951 - val_accuracy: 0.7662\n",
      "Epoch 287/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2507 - accuracy: 0.9408 - val_loss: 0.9937 - val_accuracy: 0.7650\n",
      "Epoch 288/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2551 - accuracy: 0.9368 - val_loss: 1.0037 - val_accuracy: 0.7650\n",
      "Epoch 289/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2569 - accuracy: 0.9398 - val_loss: 1.0118 - val_accuracy: 0.7625\n",
      "Epoch 290/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2500 - accuracy: 0.9411 - val_loss: 0.9986 - val_accuracy: 0.7675\n",
      "Epoch 291/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.2459 - accuracy: 0.9407 - val_loss: 0.9987 - val_accuracy: 0.7688\n",
      "Epoch 292/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2608 - accuracy: 0.9369 - val_loss: 0.9935 - val_accuracy: 0.7700\n",
      "Epoch 293/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.2421 - accuracy: 0.9419 - val_loss: 1.0077 - val_accuracy: 0.7688\n",
      "Epoch 294/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2565 - accuracy: 0.9378 - val_loss: 1.0033 - val_accuracy: 0.7637\n",
      "Epoch 295/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2477 - accuracy: 0.9388 - val_loss: 1.0069 - val_accuracy: 0.7763\n",
      "Epoch 296/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2435 - accuracy: 0.9428 - val_loss: 1.0015 - val_accuracy: 0.7788\n",
      "Epoch 297/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2479 - accuracy: 0.9414 - val_loss: 0.9986 - val_accuracy: 0.7650\n",
      "Epoch 298/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2433 - accuracy: 0.9409 - val_loss: 1.0048 - val_accuracy: 0.7637\n",
      "Epoch 299/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2467 - accuracy: 0.9413 - val_loss: 0.9898 - val_accuracy: 0.7688\n",
      "Epoch 300/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2458 - accuracy: 0.9426 - val_loss: 0.9973 - val_accuracy: 0.7625\n",
      "Epoch 301/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2381 - accuracy: 0.9441 - val_loss: 0.9947 - val_accuracy: 0.7662\n",
      "Epoch 302/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2381 - accuracy: 0.9480 - val_loss: 0.9915 - val_accuracy: 0.7637\n",
      "Epoch 303/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2364 - accuracy: 0.9443 - val_loss: 0.9916 - val_accuracy: 0.7637\n",
      "Epoch 304/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2418 - accuracy: 0.9429 - val_loss: 1.0073 - val_accuracy: 0.7612\n",
      "Epoch 305/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2430 - accuracy: 0.9426 - val_loss: 1.0080 - val_accuracy: 0.7600\n",
      "Epoch 306/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2377 - accuracy: 0.9435 - val_loss: 1.0050 - val_accuracy: 0.7650\n",
      "Epoch 307/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2400 - accuracy: 0.9446 - val_loss: 1.0105 - val_accuracy: 0.7625\n",
      "Epoch 308/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2418 - accuracy: 0.9422 - val_loss: 1.0115 - val_accuracy: 0.7575\n",
      "Epoch 309/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2327 - accuracy: 0.9478 - val_loss: 1.0016 - val_accuracy: 0.7587\n",
      "Epoch 310/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2401 - accuracy: 0.9438 - val_loss: 1.0075 - val_accuracy: 0.7587\n",
      "Epoch 311/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2391 - accuracy: 0.9431 - val_loss: 1.0122 - val_accuracy: 0.7575\n",
      "Epoch 312/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2344 - accuracy: 0.9452 - val_loss: 1.0095 - val_accuracy: 0.7587\n",
      "Epoch 313/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2333 - accuracy: 0.9461 - val_loss: 1.0224 - val_accuracy: 0.7738\n",
      "Epoch 314/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2447 - accuracy: 0.9404 - val_loss: 1.0081 - val_accuracy: 0.7662\n",
      "Epoch 315/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.2353 - accuracy: 0.9440 - val_loss: 1.0114 - val_accuracy: 0.7700\n",
      "Epoch 316/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2374 - accuracy: 0.9438 - val_loss: 1.0034 - val_accuracy: 0.7700\n",
      "Epoch 317/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2311 - accuracy: 0.9473 - val_loss: 1.0020 - val_accuracy: 0.7662\n",
      "Epoch 318/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2349 - accuracy: 0.9461 - val_loss: 1.0130 - val_accuracy: 0.7688\n",
      "Epoch 319/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2354 - accuracy: 0.9468 - val_loss: 1.0167 - val_accuracy: 0.7700\n",
      "Epoch 320/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2349 - accuracy: 0.9449 - val_loss: 1.0065 - val_accuracy: 0.7700\n",
      "Epoch 321/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2320 - accuracy: 0.9459 - val_loss: 1.0088 - val_accuracy: 0.7562\n",
      "Epoch 322/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2313 - accuracy: 0.9457 - val_loss: 1.0100 - val_accuracy: 0.7625\n",
      "Epoch 323/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2374 - accuracy: 0.9437 - val_loss: 1.0218 - val_accuracy: 0.7637\n",
      "Epoch 324/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2315 - accuracy: 0.9457 - val_loss: 1.0222 - val_accuracy: 0.7625\n",
      "Epoch 325/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2341 - accuracy: 0.9448 - val_loss: 1.0098 - val_accuracy: 0.7612\n",
      "Epoch 326/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2324 - accuracy: 0.9475 - val_loss: 1.0024 - val_accuracy: 0.7587\n",
      "Epoch 327/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2361 - accuracy: 0.9437 - val_loss: 1.0066 - val_accuracy: 0.7637\n",
      "Epoch 328/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2280 - accuracy: 0.9478 - val_loss: 1.0092 - val_accuracy: 0.7612\n",
      "Epoch 329/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2331 - accuracy: 0.9482 - val_loss: 1.0104 - val_accuracy: 0.7575\n",
      "Epoch 330/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2385 - accuracy: 0.9454 - val_loss: 1.0108 - val_accuracy: 0.7587\n",
      "Epoch 331/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.2248 - accuracy: 0.9495 - val_loss: 1.0133 - val_accuracy: 0.7587\n",
      "Epoch 332/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2271 - accuracy: 0.9468 - val_loss: 1.0067 - val_accuracy: 0.7600\n",
      "Epoch 333/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2277 - accuracy: 0.9477 - val_loss: 1.0199 - val_accuracy: 0.7562\n",
      "Epoch 334/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2237 - accuracy: 0.9493 - val_loss: 1.0270 - val_accuracy: 0.7587\n",
      "Epoch 335/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2321 - accuracy: 0.9464 - val_loss: 1.0211 - val_accuracy: 0.7550\n",
      "Epoch 336/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2324 - accuracy: 0.9472 - val_loss: 1.0308 - val_accuracy: 0.7537\n",
      "Epoch 337/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2241 - accuracy: 0.9488 - val_loss: 1.0209 - val_accuracy: 0.7525\n",
      "Epoch 338/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2395 - accuracy: 0.9424 - val_loss: 1.0261 - val_accuracy: 0.7550\n",
      "Epoch 339/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2304 - accuracy: 0.9476 - val_loss: 1.0196 - val_accuracy: 0.7587\n",
      "Epoch 340/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2229 - accuracy: 0.9517 - val_loss: 1.0180 - val_accuracy: 0.7562\n",
      "Epoch 341/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2325 - accuracy: 0.9472 - val_loss: 1.0151 - val_accuracy: 0.7562\n",
      "Epoch 342/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2325 - accuracy: 0.9463 - val_loss: 1.0093 - val_accuracy: 0.7637\n",
      "Epoch 343/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2290 - accuracy: 0.9483 - val_loss: 1.0265 - val_accuracy: 0.7600\n",
      "Epoch 344/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2248 - accuracy: 0.9485 - val_loss: 1.0160 - val_accuracy: 0.7675\n",
      "Epoch 345/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.2208 - accuracy: 0.9491 - val_loss: 1.0190 - val_accuracy: 0.7587\n",
      "Epoch 346/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2236 - accuracy: 0.9489 - val_loss: 1.0191 - val_accuracy: 0.7612\n",
      "Epoch 347/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2336 - accuracy: 0.9444 - val_loss: 1.0150 - val_accuracy: 0.7587\n",
      "Epoch 348/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2207 - accuracy: 0.9505 - val_loss: 1.0219 - val_accuracy: 0.7675\n",
      "Epoch 349/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2250 - accuracy: 0.9485 - val_loss: 1.0210 - val_accuracy: 0.7612\n",
      "Epoch 350/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2185 - accuracy: 0.9530 - val_loss: 1.0172 - val_accuracy: 0.7600\n",
      "Epoch 351/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2213 - accuracy: 0.9520 - val_loss: 1.0108 - val_accuracy: 0.7600\n",
      "Epoch 352/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2208 - accuracy: 0.9499 - val_loss: 1.0211 - val_accuracy: 0.7612\n",
      "Epoch 353/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2273 - accuracy: 0.9488 - val_loss: 1.0131 - val_accuracy: 0.7612\n",
      "Epoch 354/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2244 - accuracy: 0.9505 - val_loss: 1.0135 - val_accuracy: 0.7562\n",
      "Epoch 355/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2225 - accuracy: 0.9486 - val_loss: 1.0202 - val_accuracy: 0.7587\n",
      "Epoch 356/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.2181 - accuracy: 0.9511 - val_loss: 1.0231 - val_accuracy: 0.7587\n",
      "Epoch 357/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2259 - accuracy: 0.9488 - val_loss: 1.0171 - val_accuracy: 0.7625\n",
      "Epoch 358/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2137 - accuracy: 0.9530 - val_loss: 1.0034 - val_accuracy: 0.7625\n",
      "Epoch 359/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2127 - accuracy: 0.9501 - val_loss: 1.0072 - val_accuracy: 0.7600\n",
      "Epoch 360/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2189 - accuracy: 0.9507 - val_loss: 1.0146 - val_accuracy: 0.7550\n",
      "Epoch 361/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2157 - accuracy: 0.9537 - val_loss: 1.0129 - val_accuracy: 0.7575\n",
      "Epoch 362/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2194 - accuracy: 0.9526 - val_loss: 1.0107 - val_accuracy: 0.7637\n",
      "Epoch 363/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2296 - accuracy: 0.9472 - val_loss: 1.0156 - val_accuracy: 0.7637\n",
      "Epoch 364/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2157 - accuracy: 0.9519 - val_loss: 1.0162 - val_accuracy: 0.7625\n",
      "Epoch 365/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2188 - accuracy: 0.9504 - val_loss: 1.0189 - val_accuracy: 0.7575\n",
      "Epoch 366/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2166 - accuracy: 0.9511 - val_loss: 1.0083 - val_accuracy: 0.7625\n",
      "Epoch 367/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2223 - accuracy: 0.9497 - val_loss: 1.0161 - val_accuracy: 0.7625\n",
      "Epoch 368/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2192 - accuracy: 0.9495 - val_loss: 1.0212 - val_accuracy: 0.7700\n",
      "Epoch 369/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2186 - accuracy: 0.9494 - val_loss: 1.0232 - val_accuracy: 0.7625\n",
      "Epoch 370/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2191 - accuracy: 0.9514 - val_loss: 1.0236 - val_accuracy: 0.7612\n",
      "Epoch 371/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2181 - accuracy: 0.9503 - val_loss: 1.0199 - val_accuracy: 0.7662\n",
      "Epoch 372/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2152 - accuracy: 0.9513 - val_loss: 1.0142 - val_accuracy: 0.7562\n",
      "Epoch 373/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2133 - accuracy: 0.9538 - val_loss: 1.0100 - val_accuracy: 0.7562\n",
      "Epoch 374/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2189 - accuracy: 0.9503 - val_loss: 1.0266 - val_accuracy: 0.7537\n",
      "Epoch 375/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2139 - accuracy: 0.9536 - val_loss: 1.0173 - val_accuracy: 0.7550\n",
      "Epoch 376/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.2110 - accuracy: 0.9551 - val_loss: 1.0234 - val_accuracy: 0.7562\n",
      "Epoch 377/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2138 - accuracy: 0.9524 - val_loss: 1.0240 - val_accuracy: 0.7612\n",
      "Epoch 378/1000\n",
      "19/19 [==============================] - 0s 18ms/step - loss: 0.2080 - accuracy: 0.9548 - val_loss: 1.0232 - val_accuracy: 0.7575\n",
      "Epoch 379/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2152 - accuracy: 0.9531 - val_loss: 1.0214 - val_accuracy: 0.7537\n",
      "Epoch 380/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2114 - accuracy: 0.9551 - val_loss: 1.0292 - val_accuracy: 0.7500\n",
      "Epoch 381/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2087 - accuracy: 0.9543 - val_loss: 1.0234 - val_accuracy: 0.7562\n",
      "Epoch 382/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2141 - accuracy: 0.9526 - val_loss: 1.0216 - val_accuracy: 0.7562\n",
      "Epoch 383/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2079 - accuracy: 0.9563 - val_loss: 1.0341 - val_accuracy: 0.7550\n",
      "Epoch 384/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2053 - accuracy: 0.9545 - val_loss: 1.0361 - val_accuracy: 0.7537\n",
      "Epoch 385/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2125 - accuracy: 0.9535 - val_loss: 1.0392 - val_accuracy: 0.7550\n",
      "Epoch 386/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2131 - accuracy: 0.9547 - val_loss: 1.0315 - val_accuracy: 0.7575\n",
      "Epoch 387/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2229 - accuracy: 0.9499 - val_loss: 1.0366 - val_accuracy: 0.7650\n",
      "Epoch 388/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2146 - accuracy: 0.9520 - val_loss: 1.0408 - val_accuracy: 0.7650\n",
      "Epoch 389/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2125 - accuracy: 0.9530 - val_loss: 1.0278 - val_accuracy: 0.7650\n",
      "Epoch 390/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2138 - accuracy: 0.9537 - val_loss: 1.0258 - val_accuracy: 0.7662\n",
      "Epoch 391/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2041 - accuracy: 0.9544 - val_loss: 1.0377 - val_accuracy: 0.7612\n",
      "Epoch 392/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2046 - accuracy: 0.9555 - val_loss: 1.0233 - val_accuracy: 0.7600\n",
      "Epoch 393/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2070 - accuracy: 0.9552 - val_loss: 1.0337 - val_accuracy: 0.7650\n",
      "Epoch 394/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2100 - accuracy: 0.9543 - val_loss: 1.0265 - val_accuracy: 0.7612\n",
      "Epoch 395/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2065 - accuracy: 0.9547 - val_loss: 1.0235 - val_accuracy: 0.7612\n",
      "Epoch 396/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2120 - accuracy: 0.9530 - val_loss: 1.0320 - val_accuracy: 0.7625\n",
      "Epoch 397/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2051 - accuracy: 0.9551 - val_loss: 1.0284 - val_accuracy: 0.7600\n",
      "Epoch 398/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2081 - accuracy: 0.9549 - val_loss: 1.0236 - val_accuracy: 0.7612\n",
      "Epoch 399/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2063 - accuracy: 0.9547 - val_loss: 1.0145 - val_accuracy: 0.7612\n",
      "Epoch 400/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2053 - accuracy: 0.9540 - val_loss: 1.0324 - val_accuracy: 0.7625\n",
      "Epoch 401/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2108 - accuracy: 0.9553 - val_loss: 1.0367 - val_accuracy: 0.7637\n",
      "Epoch 402/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2072 - accuracy: 0.9555 - val_loss: 1.0426 - val_accuracy: 0.7612\n",
      "Epoch 403/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2046 - accuracy: 0.9561 - val_loss: 1.0454 - val_accuracy: 0.7650\n",
      "Epoch 404/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2132 - accuracy: 0.9509 - val_loss: 1.0435 - val_accuracy: 0.7625\n",
      "Epoch 405/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2073 - accuracy: 0.9544 - val_loss: 1.0357 - val_accuracy: 0.7625\n",
      "Epoch 406/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2055 - accuracy: 0.9566 - val_loss: 1.0402 - val_accuracy: 0.7637\n",
      "Epoch 407/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2047 - accuracy: 0.9546 - val_loss: 1.0519 - val_accuracy: 0.7550\n",
      "Epoch 408/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2108 - accuracy: 0.9547 - val_loss: 1.0588 - val_accuracy: 0.7525\n",
      "Epoch 409/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.2067 - accuracy: 0.9544 - val_loss: 1.0477 - val_accuracy: 0.7562\n",
      "Epoch 410/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2138 - accuracy: 0.9544 - val_loss: 1.0462 - val_accuracy: 0.7525\n",
      "Epoch 411/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2149 - accuracy: 0.9518 - val_loss: 1.0489 - val_accuracy: 0.7575\n",
      "Epoch 412/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2085 - accuracy: 0.9551 - val_loss: 1.0560 - val_accuracy: 0.7562\n",
      "Epoch 413/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1989 - accuracy: 0.9580 - val_loss: 1.0612 - val_accuracy: 0.7525\n",
      "Epoch 414/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2045 - accuracy: 0.9556 - val_loss: 1.0615 - val_accuracy: 0.7562\n",
      "Epoch 415/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2058 - accuracy: 0.9561 - val_loss: 1.0582 - val_accuracy: 0.7487\n",
      "Epoch 416/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1999 - accuracy: 0.9555 - val_loss: 1.0776 - val_accuracy: 0.7550\n",
      "Epoch 417/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1996 - accuracy: 0.9579 - val_loss: 1.0704 - val_accuracy: 0.7575\n",
      "Epoch 418/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2065 - accuracy: 0.9545 - val_loss: 1.0638 - val_accuracy: 0.7537\n",
      "Epoch 419/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2030 - accuracy: 0.9574 - val_loss: 1.0731 - val_accuracy: 0.7525\n",
      "Epoch 420/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1959 - accuracy: 0.9576 - val_loss: 1.0774 - val_accuracy: 0.7525\n",
      "Epoch 421/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2024 - accuracy: 0.9560 - val_loss: 1.0739 - val_accuracy: 0.7512\n",
      "Epoch 422/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2044 - accuracy: 0.9574 - val_loss: 1.0734 - val_accuracy: 0.7562\n",
      "Epoch 423/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.1959 - accuracy: 0.9588 - val_loss: 1.0589 - val_accuracy: 0.7600\n",
      "Epoch 424/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1997 - accuracy: 0.9568 - val_loss: 1.0483 - val_accuracy: 0.7600\n",
      "Epoch 425/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2040 - accuracy: 0.9557 - val_loss: 1.0533 - val_accuracy: 0.7575\n",
      "Epoch 426/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2120 - accuracy: 0.9534 - val_loss: 1.0482 - val_accuracy: 0.7600\n",
      "Epoch 427/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2018 - accuracy: 0.9566 - val_loss: 1.0550 - val_accuracy: 0.7637\n",
      "Epoch 428/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2054 - accuracy: 0.9556 - val_loss: 1.0554 - val_accuracy: 0.7575\n",
      "Epoch 429/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1954 - accuracy: 0.9593 - val_loss: 1.0560 - val_accuracy: 0.7587\n",
      "Epoch 430/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.2031 - accuracy: 0.9558 - val_loss: 1.0507 - val_accuracy: 0.7612\n",
      "Epoch 431/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2049 - accuracy: 0.9572 - val_loss: 1.0502 - val_accuracy: 0.7600\n",
      "Epoch 432/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2043 - accuracy: 0.9565 - val_loss: 1.0530 - val_accuracy: 0.7625\n",
      "Epoch 433/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1965 - accuracy: 0.9594 - val_loss: 1.0569 - val_accuracy: 0.7600\n",
      "Epoch 434/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2023 - accuracy: 0.9566 - val_loss: 1.0476 - val_accuracy: 0.7575\n",
      "Epoch 435/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2000 - accuracy: 0.9580 - val_loss: 1.0566 - val_accuracy: 0.7650\n",
      "Epoch 436/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1986 - accuracy: 0.9576 - val_loss: 1.0615 - val_accuracy: 0.7650\n",
      "Epoch 437/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.2015 - accuracy: 0.9566 - val_loss: 1.0700 - val_accuracy: 0.7612\n",
      "Epoch 438/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1972 - accuracy: 0.9586 - val_loss: 1.0796 - val_accuracy: 0.7612\n",
      "Epoch 439/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2001 - accuracy: 0.9582 - val_loss: 1.0489 - val_accuracy: 0.7725\n",
      "Epoch 440/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.2028 - accuracy: 0.9567 - val_loss: 1.0416 - val_accuracy: 0.7700\n",
      "Epoch 441/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1961 - accuracy: 0.9591 - val_loss: 1.0519 - val_accuracy: 0.7662\n",
      "Epoch 442/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1991 - accuracy: 0.9603 - val_loss: 1.0560 - val_accuracy: 0.7675\n",
      "Epoch 443/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2010 - accuracy: 0.9575 - val_loss: 1.0637 - val_accuracy: 0.7637\n",
      "Epoch 444/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1969 - accuracy: 0.9593 - val_loss: 1.0568 - val_accuracy: 0.7600\n",
      "Epoch 445/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1985 - accuracy: 0.9589 - val_loss: 1.0674 - val_accuracy: 0.7587\n",
      "Epoch 446/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2000 - accuracy: 0.9563 - val_loss: 1.0725 - val_accuracy: 0.7587\n",
      "Epoch 447/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2000 - accuracy: 0.9583 - val_loss: 1.0785 - val_accuracy: 0.7525\n",
      "Epoch 448/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.2003 - accuracy: 0.9564 - val_loss: 1.0829 - val_accuracy: 0.7562\n",
      "Epoch 449/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1936 - accuracy: 0.9588 - val_loss: 1.0854 - val_accuracy: 0.7612\n",
      "Epoch 450/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1936 - accuracy: 0.9599 - val_loss: 1.0939 - val_accuracy: 0.7550\n",
      "Epoch 451/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2016 - accuracy: 0.9589 - val_loss: 1.0841 - val_accuracy: 0.7600\n",
      "Epoch 452/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 11ms/step - loss: 0.2024 - accuracy: 0.9575 - val_loss: 1.0625 - val_accuracy: 0.7625\n",
      "Epoch 453/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1927 - accuracy: 0.9613 - val_loss: 1.0490 - val_accuracy: 0.7625\n",
      "Epoch 454/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1947 - accuracy: 0.9592 - val_loss: 1.0621 - val_accuracy: 0.7575\n",
      "Epoch 455/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1933 - accuracy: 0.9579 - val_loss: 1.0550 - val_accuracy: 0.7625\n",
      "Epoch 456/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1982 - accuracy: 0.9581 - val_loss: 1.0529 - val_accuracy: 0.7662\n",
      "Epoch 457/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1981 - accuracy: 0.9580 - val_loss: 1.0641 - val_accuracy: 0.7575\n",
      "Epoch 458/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.1930 - accuracy: 0.9605 - val_loss: 1.0677 - val_accuracy: 0.7562\n",
      "Epoch 459/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1920 - accuracy: 0.9598 - val_loss: 1.0704 - val_accuracy: 0.7587\n",
      "Epoch 460/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1921 - accuracy: 0.9607 - val_loss: 1.0620 - val_accuracy: 0.7575\n",
      "Epoch 461/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1938 - accuracy: 0.9609 - val_loss: 1.0668 - val_accuracy: 0.7612\n",
      "Epoch 462/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1960 - accuracy: 0.9599 - val_loss: 1.0766 - val_accuracy: 0.7600\n",
      "Epoch 463/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1960 - accuracy: 0.9603 - val_loss: 1.0700 - val_accuracy: 0.7525\n",
      "Epoch 464/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1899 - accuracy: 0.9582 - val_loss: 1.0751 - val_accuracy: 0.7512\n",
      "Epoch 465/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1904 - accuracy: 0.9623 - val_loss: 1.0810 - val_accuracy: 0.7562\n",
      "Epoch 466/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1957 - accuracy: 0.9583 - val_loss: 1.0856 - val_accuracy: 0.7550\n",
      "Epoch 467/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1966 - accuracy: 0.9607 - val_loss: 1.0802 - val_accuracy: 0.7562\n",
      "Epoch 468/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.1913 - accuracy: 0.9602 - val_loss: 1.0842 - val_accuracy: 0.7537\n",
      "Epoch 469/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1975 - accuracy: 0.9583 - val_loss: 1.0859 - val_accuracy: 0.7537\n",
      "Epoch 470/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1923 - accuracy: 0.9609 - val_loss: 1.0903 - val_accuracy: 0.7550\n",
      "Epoch 471/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1917 - accuracy: 0.9598 - val_loss: 1.0928 - val_accuracy: 0.7525\n",
      "Epoch 472/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1987 - accuracy: 0.9600 - val_loss: 1.1023 - val_accuracy: 0.7500\n",
      "Epoch 473/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.1930 - accuracy: 0.9586 - val_loss: 1.1100 - val_accuracy: 0.7525\n",
      "Epoch 474/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1878 - accuracy: 0.9603 - val_loss: 1.0899 - val_accuracy: 0.7537\n",
      "Epoch 475/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1943 - accuracy: 0.9603 - val_loss: 1.0935 - val_accuracy: 0.7600\n",
      "Epoch 476/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1932 - accuracy: 0.9628 - val_loss: 1.0798 - val_accuracy: 0.7625\n",
      "Epoch 477/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.1947 - accuracy: 0.9592 - val_loss: 1.0904 - val_accuracy: 0.7637\n",
      "Epoch 478/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1904 - accuracy: 0.9613 - val_loss: 1.0945 - val_accuracy: 0.7562\n",
      "Epoch 479/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1960 - accuracy: 0.9588 - val_loss: 1.0875 - val_accuracy: 0.7612\n",
      "Epoch 480/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1861 - accuracy: 0.9628 - val_loss: 1.0930 - val_accuracy: 0.7562\n",
      "Epoch 481/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1893 - accuracy: 0.9597 - val_loss: 1.1039 - val_accuracy: 0.7512\n",
      "Epoch 482/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1858 - accuracy: 0.9617 - val_loss: 1.1108 - val_accuracy: 0.7525\n",
      "Epoch 483/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1981 - accuracy: 0.9591 - val_loss: 1.1169 - val_accuracy: 0.7537\n",
      "Epoch 484/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1867 - accuracy: 0.9628 - val_loss: 1.1162 - val_accuracy: 0.7512\n",
      "Epoch 485/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1909 - accuracy: 0.9609 - val_loss: 1.1103 - val_accuracy: 0.7612\n",
      "Epoch 486/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1922 - accuracy: 0.9607 - val_loss: 1.1106 - val_accuracy: 0.7600\n",
      "Epoch 487/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1859 - accuracy: 0.9636 - val_loss: 1.1133 - val_accuracy: 0.7562\n",
      "Epoch 488/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1929 - accuracy: 0.9604 - val_loss: 1.1196 - val_accuracy: 0.7487\n",
      "Epoch 489/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1948 - accuracy: 0.9613 - val_loss: 1.1080 - val_accuracy: 0.7500\n",
      "Epoch 490/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1866 - accuracy: 0.9608 - val_loss: 1.0884 - val_accuracy: 0.7525\n",
      "Epoch 491/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1875 - accuracy: 0.9619 - val_loss: 1.0959 - val_accuracy: 0.7550\n",
      "Epoch 492/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1923 - accuracy: 0.9604 - val_loss: 1.1102 - val_accuracy: 0.7537\n",
      "Epoch 493/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1834 - accuracy: 0.9607 - val_loss: 1.1058 - val_accuracy: 0.7562\n",
      "Epoch 494/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1903 - accuracy: 0.9607 - val_loss: 1.1150 - val_accuracy: 0.7575\n",
      "Epoch 495/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1914 - accuracy: 0.9600 - val_loss: 1.1284 - val_accuracy: 0.7600\n",
      "Epoch 496/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1891 - accuracy: 0.9611 - val_loss: 1.1291 - val_accuracy: 0.7625\n",
      "Epoch 497/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1918 - accuracy: 0.9597 - val_loss: 1.1232 - val_accuracy: 0.7587\n",
      "Epoch 498/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1913 - accuracy: 0.9627 - val_loss: 1.1127 - val_accuracy: 0.7662\n",
      "Epoch 499/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1922 - accuracy: 0.9601 - val_loss: 1.1154 - val_accuracy: 0.7512\n",
      "Epoch 500/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1918 - accuracy: 0.9601 - val_loss: 1.1077 - val_accuracy: 0.7462\n",
      "Epoch 501/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1934 - accuracy: 0.9607 - val_loss: 1.1160 - val_accuracy: 0.7462\n",
      "Epoch 502/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1854 - accuracy: 0.9634 - val_loss: 1.1175 - val_accuracy: 0.7500\n",
      "Epoch 503/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1864 - accuracy: 0.9632 - val_loss: 1.1186 - val_accuracy: 0.7450\n",
      "Epoch 504/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1877 - accuracy: 0.9622 - val_loss: 1.1114 - val_accuracy: 0.7487\n",
      "Epoch 505/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1842 - accuracy: 0.9616 - val_loss: 1.1042 - val_accuracy: 0.7500\n",
      "Epoch 506/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1834 - accuracy: 0.9631 - val_loss: 1.0975 - val_accuracy: 0.7462\n",
      "Epoch 507/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1894 - accuracy: 0.9624 - val_loss: 1.0984 - val_accuracy: 0.7537\n",
      "Epoch 508/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1908 - accuracy: 0.9607 - val_loss: 1.1030 - val_accuracy: 0.7475\n",
      "Epoch 509/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1901 - accuracy: 0.9609 - val_loss: 1.0975 - val_accuracy: 0.7512\n",
      "Epoch 510/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1897 - accuracy: 0.9626 - val_loss: 1.1112 - val_accuracy: 0.7562\n",
      "Epoch 511/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1901 - accuracy: 0.9627 - val_loss: 1.1051 - val_accuracy: 0.7562\n",
      "Epoch 512/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1857 - accuracy: 0.9634 - val_loss: 1.1012 - val_accuracy: 0.7550\n",
      "Epoch 513/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1824 - accuracy: 0.9651 - val_loss: 1.0872 - val_accuracy: 0.7637\n",
      "Epoch 514/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1854 - accuracy: 0.9628 - val_loss: 1.0844 - val_accuracy: 0.7650\n",
      "Epoch 515/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.1889 - accuracy: 0.9605 - val_loss: 1.0770 - val_accuracy: 0.7612\n",
      "Epoch 516/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1847 - accuracy: 0.9634 - val_loss: 1.0914 - val_accuracy: 0.7600\n",
      "Epoch 517/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1815 - accuracy: 0.9635 - val_loss: 1.0956 - val_accuracy: 0.7587\n",
      "Epoch 518/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1889 - accuracy: 0.9616 - val_loss: 1.1049 - val_accuracy: 0.7475\n",
      "Epoch 519/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1835 - accuracy: 0.9622 - val_loss: 1.0957 - val_accuracy: 0.7587\n",
      "Epoch 520/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1838 - accuracy: 0.9623 - val_loss: 1.0859 - val_accuracy: 0.7637\n",
      "Epoch 521/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1890 - accuracy: 0.9616 - val_loss: 1.0754 - val_accuracy: 0.7550\n",
      "Epoch 522/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1829 - accuracy: 0.9646 - val_loss: 1.0637 - val_accuracy: 0.7650\n",
      "Epoch 523/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1831 - accuracy: 0.9658 - val_loss: 1.0679 - val_accuracy: 0.7550\n",
      "Epoch 524/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1888 - accuracy: 0.9621 - val_loss: 1.0624 - val_accuracy: 0.7562\n",
      "Epoch 525/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1804 - accuracy: 0.9653 - val_loss: 1.0751 - val_accuracy: 0.7550\n",
      "Epoch 526/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1946 - accuracy: 0.9611 - val_loss: 1.0723 - val_accuracy: 0.7600\n",
      "Epoch 527/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1857 - accuracy: 0.9622 - val_loss: 1.0747 - val_accuracy: 0.7562\n",
      "Epoch 528/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1905 - accuracy: 0.9629 - val_loss: 1.0810 - val_accuracy: 0.7512\n",
      "Epoch 529/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1832 - accuracy: 0.9639 - val_loss: 1.0950 - val_accuracy: 0.7537\n",
      "Epoch 530/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1786 - accuracy: 0.9657 - val_loss: 1.0981 - val_accuracy: 0.7487\n",
      "Epoch 531/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1869 - accuracy: 0.9645 - val_loss: 1.1021 - val_accuracy: 0.7550\n",
      "Epoch 532/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1874 - accuracy: 0.9607 - val_loss: 1.1051 - val_accuracy: 0.7550\n",
      "Epoch 533/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1879 - accuracy: 0.9629 - val_loss: 1.0933 - val_accuracy: 0.7587\n",
      "Epoch 534/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1891 - accuracy: 0.9627 - val_loss: 1.0944 - val_accuracy: 0.7512\n",
      "Epoch 535/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1867 - accuracy: 0.9634 - val_loss: 1.0932 - val_accuracy: 0.7525\n",
      "Epoch 536/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1867 - accuracy: 0.9627 - val_loss: 1.0858 - val_accuracy: 0.7562\n",
      "Epoch 537/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1862 - accuracy: 0.9619 - val_loss: 1.0960 - val_accuracy: 0.7537\n",
      "Epoch 538/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1840 - accuracy: 0.9624 - val_loss: 1.0892 - val_accuracy: 0.7600\n",
      "Epoch 539/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1818 - accuracy: 0.9655 - val_loss: 1.0991 - val_accuracy: 0.7587\n",
      "Epoch 540/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1864 - accuracy: 0.9643 - val_loss: 1.0980 - val_accuracy: 0.7600\n",
      "Epoch 541/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1820 - accuracy: 0.9645 - val_loss: 1.0964 - val_accuracy: 0.7600\n",
      "Epoch 542/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1816 - accuracy: 0.9650 - val_loss: 1.0896 - val_accuracy: 0.7650\n",
      "Epoch 543/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1881 - accuracy: 0.9614 - val_loss: 1.0932 - val_accuracy: 0.7612\n",
      "Epoch 544/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1830 - accuracy: 0.9647 - val_loss: 1.0964 - val_accuracy: 0.7625\n",
      "Epoch 545/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1789 - accuracy: 0.9646 - val_loss: 1.1010 - val_accuracy: 0.7600\n",
      "Epoch 546/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1840 - accuracy: 0.9653 - val_loss: 1.1131 - val_accuracy: 0.7575\n",
      "Epoch 547/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1737 - accuracy: 0.9673 - val_loss: 1.1010 - val_accuracy: 0.7550\n",
      "Epoch 548/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1789 - accuracy: 0.9643 - val_loss: 1.1045 - val_accuracy: 0.7587\n",
      "Epoch 549/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1817 - accuracy: 0.9663 - val_loss: 1.1060 - val_accuracy: 0.7587\n",
      "Epoch 550/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1810 - accuracy: 0.9647 - val_loss: 1.1081 - val_accuracy: 0.7612\n",
      "Epoch 551/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1854 - accuracy: 0.9644 - val_loss: 1.0978 - val_accuracy: 0.7675\n",
      "Epoch 552/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1806 - accuracy: 0.9648 - val_loss: 1.1107 - val_accuracy: 0.7650\n",
      "Epoch 553/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1742 - accuracy: 0.9661 - val_loss: 1.1103 - val_accuracy: 0.7600\n",
      "Epoch 554/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1898 - accuracy: 0.9618 - val_loss: 1.1145 - val_accuracy: 0.7625\n",
      "Epoch 555/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1816 - accuracy: 0.9634 - val_loss: 1.1140 - val_accuracy: 0.7612\n",
      "Epoch 556/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1838 - accuracy: 0.9623 - val_loss: 1.1276 - val_accuracy: 0.7575\n",
      "Epoch 557/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1816 - accuracy: 0.9640 - val_loss: 1.1298 - val_accuracy: 0.7612\n",
      "Epoch 558/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1810 - accuracy: 0.9650 - val_loss: 1.1065 - val_accuracy: 0.7625\n",
      "Epoch 559/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1808 - accuracy: 0.9636 - val_loss: 1.1118 - val_accuracy: 0.7587\n",
      "Epoch 560/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1819 - accuracy: 0.9645 - val_loss: 1.1176 - val_accuracy: 0.7575\n",
      "Epoch 561/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1813 - accuracy: 0.9641 - val_loss: 1.1197 - val_accuracy: 0.7562\n",
      "Epoch 562/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1772 - accuracy: 0.9655 - val_loss: 1.1029 - val_accuracy: 0.7575\n",
      "Epoch 563/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1868 - accuracy: 0.9620 - val_loss: 1.1049 - val_accuracy: 0.7562\n",
      "Epoch 564/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1870 - accuracy: 0.9620 - val_loss: 1.0962 - val_accuracy: 0.7562\n",
      "Epoch 565/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1817 - accuracy: 0.9652 - val_loss: 1.1009 - val_accuracy: 0.7637\n",
      "Epoch 566/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1772 - accuracy: 0.9662 - val_loss: 1.1142 - val_accuracy: 0.7675\n",
      "Epoch 567/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1831 - accuracy: 0.9632 - val_loss: 1.1117 - val_accuracy: 0.7637\n",
      "Epoch 568/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1782 - accuracy: 0.9682 - val_loss: 1.1271 - val_accuracy: 0.7625\n",
      "Epoch 569/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1823 - accuracy: 0.9643 - val_loss: 1.1274 - val_accuracy: 0.7612\n",
      "Epoch 570/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1835 - accuracy: 0.9651 - val_loss: 1.1210 - val_accuracy: 0.7625\n",
      "Epoch 571/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1836 - accuracy: 0.9649 - val_loss: 1.1153 - val_accuracy: 0.7700\n",
      "Epoch 572/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1796 - accuracy: 0.9632 - val_loss: 1.1101 - val_accuracy: 0.7650\n",
      "Epoch 573/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1793 - accuracy: 0.9655 - val_loss: 1.1091 - val_accuracy: 0.7612\n",
      "Epoch 574/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1773 - accuracy: 0.9657 - val_loss: 1.1053 - val_accuracy: 0.7637\n",
      "Epoch 575/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1750 - accuracy: 0.9670 - val_loss: 1.1063 - val_accuracy: 0.7650\n",
      "Epoch 576/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1778 - accuracy: 0.9674 - val_loss: 1.1100 - val_accuracy: 0.7713\n",
      "Epoch 577/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1810 - accuracy: 0.9661 - val_loss: 1.1229 - val_accuracy: 0.7713\n",
      "Epoch 578/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1761 - accuracy: 0.9659 - val_loss: 1.1114 - val_accuracy: 0.7650\n",
      "Epoch 579/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1809 - accuracy: 0.9639 - val_loss: 1.0966 - val_accuracy: 0.7637\n",
      "Epoch 580/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1773 - accuracy: 0.9643 - val_loss: 1.1070 - val_accuracy: 0.7650\n",
      "Epoch 581/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1796 - accuracy: 0.9663 - val_loss: 1.1181 - val_accuracy: 0.7713\n",
      "Epoch 582/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1772 - accuracy: 0.9661 - val_loss: 1.0979 - val_accuracy: 0.7725\n",
      "Epoch 583/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1806 - accuracy: 0.9644 - val_loss: 1.1093 - val_accuracy: 0.7662\n",
      "Epoch 584/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1783 - accuracy: 0.9671 - val_loss: 1.0996 - val_accuracy: 0.7738\n",
      "Epoch 585/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1787 - accuracy: 0.9661 - val_loss: 1.1124 - val_accuracy: 0.7725\n",
      "Epoch 586/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1815 - accuracy: 0.9633 - val_loss: 1.1276 - val_accuracy: 0.7700\n",
      "Epoch 587/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1781 - accuracy: 0.9664 - val_loss: 1.1140 - val_accuracy: 0.7625\n",
      "Epoch 588/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1777 - accuracy: 0.9676 - val_loss: 1.1221 - val_accuracy: 0.7637\n",
      "Epoch 589/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1837 - accuracy: 0.9635 - val_loss: 1.1288 - val_accuracy: 0.7600\n",
      "Epoch 590/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1770 - accuracy: 0.9655 - val_loss: 1.1198 - val_accuracy: 0.7612\n",
      "Epoch 591/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1737 - accuracy: 0.9689 - val_loss: 1.1302 - val_accuracy: 0.7650\n",
      "Epoch 592/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1766 - accuracy: 0.9668 - val_loss: 1.1290 - val_accuracy: 0.7562\n",
      "Epoch 593/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1700 - accuracy: 0.9689 - val_loss: 1.1283 - val_accuracy: 0.7575\n",
      "Epoch 594/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1737 - accuracy: 0.9678 - val_loss: 1.1280 - val_accuracy: 0.7625\n",
      "Epoch 595/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1810 - accuracy: 0.9645 - val_loss: 1.1291 - val_accuracy: 0.7637\n",
      "Epoch 596/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1771 - accuracy: 0.9664 - val_loss: 1.1375 - val_accuracy: 0.7625\n",
      "Epoch 597/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1728 - accuracy: 0.9660 - val_loss: 1.1332 - val_accuracy: 0.7575\n",
      "Epoch 598/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1762 - accuracy: 0.9663 - val_loss: 1.1203 - val_accuracy: 0.7537\n",
      "Epoch 599/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1749 - accuracy: 0.9665 - val_loss: 1.1162 - val_accuracy: 0.7550\n",
      "Epoch 600/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1770 - accuracy: 0.9672 - val_loss: 1.1270 - val_accuracy: 0.7575\n",
      "Epoch 601/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1783 - accuracy: 0.9657 - val_loss: 1.1195 - val_accuracy: 0.7575\n",
      "Epoch 602/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1734 - accuracy: 0.9682 - val_loss: 1.1213 - val_accuracy: 0.7637\n",
      "Epoch 603/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1693 - accuracy: 0.9678 - val_loss: 1.1249 - val_accuracy: 0.7625\n",
      "Epoch 604/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1772 - accuracy: 0.9668 - val_loss: 1.1197 - val_accuracy: 0.7600\n",
      "Epoch 605/1000\n",
      "19/19 [==============================] - 0s 9ms/step - loss: 0.1795 - accuracy: 0.9668 - val_loss: 1.1325 - val_accuracy: 0.7587\n",
      "Epoch 606/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1798 - accuracy: 0.9652 - val_loss: 1.1316 - val_accuracy: 0.7650\n",
      "Epoch 607/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1746 - accuracy: 0.9665 - val_loss: 1.1116 - val_accuracy: 0.7562\n",
      "Epoch 608/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1844 - accuracy: 0.9653 - val_loss: 1.1131 - val_accuracy: 0.7662\n",
      "Epoch 609/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1750 - accuracy: 0.9664 - val_loss: 1.1131 - val_accuracy: 0.7688\n",
      "Epoch 610/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1731 - accuracy: 0.9674 - val_loss: 1.0965 - val_accuracy: 0.7700\n",
      "Epoch 611/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1746 - accuracy: 0.9664 - val_loss: 1.1089 - val_accuracy: 0.7637\n",
      "Epoch 612/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1740 - accuracy: 0.9673 - val_loss: 1.1125 - val_accuracy: 0.7637\n",
      "Epoch 613/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1809 - accuracy: 0.9642 - val_loss: 1.1108 - val_accuracy: 0.7650\n",
      "Epoch 614/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1826 - accuracy: 0.9656 - val_loss: 1.1084 - val_accuracy: 0.7725\n",
      "Epoch 615/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1778 - accuracy: 0.9646 - val_loss: 1.0994 - val_accuracy: 0.7662\n",
      "Epoch 616/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1756 - accuracy: 0.9670 - val_loss: 1.1130 - val_accuracy: 0.7637\n",
      "Epoch 617/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1732 - accuracy: 0.9670 - val_loss: 1.1144 - val_accuracy: 0.7612\n",
      "Epoch 618/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1793 - accuracy: 0.9669 - val_loss: 1.1116 - val_accuracy: 0.7637\n",
      "Epoch 619/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1804 - accuracy: 0.9670 - val_loss: 1.1203 - val_accuracy: 0.7600\n",
      "Epoch 620/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1718 - accuracy: 0.9680 - val_loss: 1.1200 - val_accuracy: 0.7600\n",
      "Epoch 621/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1791 - accuracy: 0.9635 - val_loss: 1.1155 - val_accuracy: 0.7637\n",
      "Epoch 622/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1812 - accuracy: 0.9666 - val_loss: 1.0966 - val_accuracy: 0.7612\n",
      "Epoch 623/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1725 - accuracy: 0.9674 - val_loss: 1.0937 - val_accuracy: 0.7650\n",
      "Epoch 624/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1804 - accuracy: 0.9650 - val_loss: 1.0840 - val_accuracy: 0.7625\n",
      "Epoch 625/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1742 - accuracy: 0.9673 - val_loss: 1.0959 - val_accuracy: 0.7675\n",
      "Epoch 626/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1790 - accuracy: 0.9655 - val_loss: 1.0913 - val_accuracy: 0.7688\n",
      "Epoch 627/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1807 - accuracy: 0.9632 - val_loss: 1.0896 - val_accuracy: 0.7775\n",
      "Epoch 628/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1749 - accuracy: 0.9680 - val_loss: 1.0911 - val_accuracy: 0.7688\n",
      "Epoch 629/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1773 - accuracy: 0.9655 - val_loss: 1.0847 - val_accuracy: 0.7600\n",
      "Epoch 630/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1762 - accuracy: 0.9670 - val_loss: 1.1001 - val_accuracy: 0.7662\n",
      "Epoch 631/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1749 - accuracy: 0.9686 - val_loss: 1.0824 - val_accuracy: 0.7688\n",
      "Epoch 632/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1738 - accuracy: 0.9658 - val_loss: 1.0670 - val_accuracy: 0.7662\n",
      "Epoch 633/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1733 - accuracy: 0.9688 - val_loss: 1.0778 - val_accuracy: 0.7650\n",
      "Epoch 634/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1745 - accuracy: 0.9668 - val_loss: 1.0870 - val_accuracy: 0.7637\n",
      "Epoch 635/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1791 - accuracy: 0.9653 - val_loss: 1.0864 - val_accuracy: 0.7637\n",
      "Epoch 636/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1750 - accuracy: 0.9672 - val_loss: 1.0932 - val_accuracy: 0.7550\n",
      "Epoch 637/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1761 - accuracy: 0.9672 - val_loss: 1.0983 - val_accuracy: 0.7612\n",
      "Epoch 638/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1747 - accuracy: 0.9673 - val_loss: 1.0950 - val_accuracy: 0.7637\n",
      "Epoch 639/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1740 - accuracy: 0.9666 - val_loss: 1.0962 - val_accuracy: 0.7625\n",
      "Epoch 640/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1720 - accuracy: 0.9691 - val_loss: 1.0929 - val_accuracy: 0.7537\n",
      "Epoch 641/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1754 - accuracy: 0.9673 - val_loss: 1.0926 - val_accuracy: 0.7625\n",
      "Epoch 642/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1742 - accuracy: 0.9665 - val_loss: 1.0808 - val_accuracy: 0.7675\n",
      "Epoch 643/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1681 - accuracy: 0.9704 - val_loss: 1.0836 - val_accuracy: 0.7612\n",
      "Epoch 644/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1745 - accuracy: 0.9681 - val_loss: 1.0815 - val_accuracy: 0.7650\n",
      "Epoch 645/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1754 - accuracy: 0.9679 - val_loss: 1.0876 - val_accuracy: 0.7650\n",
      "Epoch 646/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1746 - accuracy: 0.9680 - val_loss: 1.1033 - val_accuracy: 0.7587\n",
      "Epoch 647/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1755 - accuracy: 0.9670 - val_loss: 1.1072 - val_accuracy: 0.7600\n",
      "Epoch 648/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1824 - accuracy: 0.9661 - val_loss: 1.1102 - val_accuracy: 0.7675\n",
      "Epoch 649/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1786 - accuracy: 0.9673 - val_loss: 1.1101 - val_accuracy: 0.7650\n",
      "Epoch 650/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1719 - accuracy: 0.9682 - val_loss: 1.1185 - val_accuracy: 0.7650\n",
      "Epoch 651/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1754 - accuracy: 0.9648 - val_loss: 1.1197 - val_accuracy: 0.7650\n",
      "Epoch 652/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1755 - accuracy: 0.9664 - val_loss: 1.1365 - val_accuracy: 0.7625\n",
      "Epoch 653/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1744 - accuracy: 0.9663 - val_loss: 1.1350 - val_accuracy: 0.7662\n",
      "Epoch 654/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1725 - accuracy: 0.9678 - val_loss: 1.1484 - val_accuracy: 0.7600\n",
      "Epoch 655/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1690 - accuracy: 0.9697 - val_loss: 1.1298 - val_accuracy: 0.7650\n",
      "Epoch 656/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1662 - accuracy: 0.9697 - val_loss: 1.1208 - val_accuracy: 0.7637\n",
      "Epoch 657/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1738 - accuracy: 0.9669 - val_loss: 1.1243 - val_accuracy: 0.7675\n",
      "Epoch 658/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1710 - accuracy: 0.9681 - val_loss: 1.1177 - val_accuracy: 0.7650\n",
      "Epoch 659/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1789 - accuracy: 0.9646 - val_loss: 1.1248 - val_accuracy: 0.7637\n",
      "Epoch 660/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1681 - accuracy: 0.9682 - val_loss: 1.1259 - val_accuracy: 0.7650\n",
      "Epoch 661/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1760 - accuracy: 0.9680 - val_loss: 1.1193 - val_accuracy: 0.7700\n",
      "Epoch 662/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1783 - accuracy: 0.9669 - val_loss: 1.1095 - val_accuracy: 0.7662\n",
      "Epoch 663/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1785 - accuracy: 0.9663 - val_loss: 1.1106 - val_accuracy: 0.7700\n",
      "Epoch 664/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1687 - accuracy: 0.9699 - val_loss: 1.1163 - val_accuracy: 0.7700\n",
      "Epoch 665/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1728 - accuracy: 0.9682 - val_loss: 1.1199 - val_accuracy: 0.7637\n",
      "Epoch 666/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1696 - accuracy: 0.9696 - val_loss: 1.1170 - val_accuracy: 0.7612\n",
      "Epoch 667/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1714 - accuracy: 0.9671 - val_loss: 1.1339 - val_accuracy: 0.7713\n",
      "Epoch 668/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1729 - accuracy: 0.9683 - val_loss: 1.1388 - val_accuracy: 0.7650\n",
      "Epoch 669/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1690 - accuracy: 0.9684 - val_loss: 1.1384 - val_accuracy: 0.7650\n",
      "Epoch 670/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1744 - accuracy: 0.9666 - val_loss: 1.1220 - val_accuracy: 0.7625\n",
      "Epoch 671/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1685 - accuracy: 0.9697 - val_loss: 1.1319 - val_accuracy: 0.7625\n",
      "Epoch 672/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1690 - accuracy: 0.9704 - val_loss: 1.1231 - val_accuracy: 0.7650\n",
      "Epoch 673/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1751 - accuracy: 0.9680 - val_loss: 1.1272 - val_accuracy: 0.7650\n",
      "Epoch 674/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1660 - accuracy: 0.9695 - val_loss: 1.1228 - val_accuracy: 0.7637\n",
      "Epoch 675/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1727 - accuracy: 0.9692 - val_loss: 1.1291 - val_accuracy: 0.7700\n",
      "Epoch 676/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1768 - accuracy: 0.9674 - val_loss: 1.1148 - val_accuracy: 0.7675\n",
      "Epoch 677/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1698 - accuracy: 0.9702 - val_loss: 1.1205 - val_accuracy: 0.7650\n",
      "Epoch 678/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1710 - accuracy: 0.9684 - val_loss: 1.1299 - val_accuracy: 0.7587\n",
      "Epoch 679/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1626 - accuracy: 0.9697 - val_loss: 1.1260 - val_accuracy: 0.7575\n",
      "Epoch 680/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1696 - accuracy: 0.9693 - val_loss: 1.1297 - val_accuracy: 0.7562\n",
      "Epoch 681/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1670 - accuracy: 0.9714 - val_loss: 1.1349 - val_accuracy: 0.7637\n",
      "Epoch 682/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1782 - accuracy: 0.9672 - val_loss: 1.1424 - val_accuracy: 0.7587\n",
      "Epoch 683/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1677 - accuracy: 0.9705 - val_loss: 1.1311 - val_accuracy: 0.7562\n",
      "Epoch 684/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1711 - accuracy: 0.9668 - val_loss: 1.1369 - val_accuracy: 0.7587\n",
      "Epoch 685/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1746 - accuracy: 0.9657 - val_loss: 1.1393 - val_accuracy: 0.7562\n",
      "Epoch 686/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1716 - accuracy: 0.9691 - val_loss: 1.1410 - val_accuracy: 0.7550\n",
      "Epoch 687/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1683 - accuracy: 0.9698 - val_loss: 1.1434 - val_accuracy: 0.7587\n",
      "Epoch 688/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1642 - accuracy: 0.9701 - val_loss: 1.1446 - val_accuracy: 0.7550\n",
      "Epoch 689/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1777 - accuracy: 0.9663 - val_loss: 1.1397 - val_accuracy: 0.7562\n",
      "Epoch 690/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1731 - accuracy: 0.9682 - val_loss: 1.1292 - val_accuracy: 0.7587\n",
      "Epoch 691/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1751 - accuracy: 0.9680 - val_loss: 1.1341 - val_accuracy: 0.7600\n",
      "Epoch 692/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1629 - accuracy: 0.9716 - val_loss: 1.1155 - val_accuracy: 0.7650\n",
      "Epoch 693/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1712 - accuracy: 0.9681 - val_loss: 1.1246 - val_accuracy: 0.7650\n",
      "Epoch 694/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1681 - accuracy: 0.9690 - val_loss: 1.1158 - val_accuracy: 0.7650\n",
      "Epoch 695/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1723 - accuracy: 0.9684 - val_loss: 1.1161 - val_accuracy: 0.7612\n",
      "Epoch 696/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1722 - accuracy: 0.9682 - val_loss: 1.1238 - val_accuracy: 0.7562\n",
      "Epoch 697/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1662 - accuracy: 0.9693 - val_loss: 1.1137 - val_accuracy: 0.7612\n",
      "Epoch 698/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1736 - accuracy: 0.9677 - val_loss: 1.1198 - val_accuracy: 0.7650\n",
      "Epoch 699/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1663 - accuracy: 0.9696 - val_loss: 1.1162 - val_accuracy: 0.7637\n",
      "Epoch 700/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1693 - accuracy: 0.9685 - val_loss: 1.1170 - val_accuracy: 0.7612\n",
      "Epoch 701/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1720 - accuracy: 0.9686 - val_loss: 1.1283 - val_accuracy: 0.7662\n",
      "Epoch 702/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1644 - accuracy: 0.9723 - val_loss: 1.1275 - val_accuracy: 0.7637\n",
      "Epoch 703/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1739 - accuracy: 0.9675 - val_loss: 1.1332 - val_accuracy: 0.7675\n",
      "Epoch 704/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1681 - accuracy: 0.9701 - val_loss: 1.1355 - val_accuracy: 0.7637\n",
      "Epoch 705/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1676 - accuracy: 0.9689 - val_loss: 1.1342 - val_accuracy: 0.7662\n",
      "Epoch 706/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1729 - accuracy: 0.9686 - val_loss: 1.1297 - val_accuracy: 0.7688\n",
      "Epoch 707/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1711 - accuracy: 0.9688 - val_loss: 1.1339 - val_accuracy: 0.7713\n",
      "Epoch 708/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1675 - accuracy: 0.9705 - val_loss: 1.1471 - val_accuracy: 0.7612\n",
      "Epoch 709/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1675 - accuracy: 0.9689 - val_loss: 1.1382 - val_accuracy: 0.7662\n",
      "Epoch 710/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1727 - accuracy: 0.9670 - val_loss: 1.1396 - val_accuracy: 0.7700\n",
      "Epoch 711/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1660 - accuracy: 0.9711 - val_loss: 1.1347 - val_accuracy: 0.7688\n",
      "Epoch 712/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1647 - accuracy: 0.9711 - val_loss: 1.1284 - val_accuracy: 0.7700\n",
      "Epoch 713/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1701 - accuracy: 0.9694 - val_loss: 1.1394 - val_accuracy: 0.7675\n",
      "Epoch 714/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1726 - accuracy: 0.9694 - val_loss: 1.1480 - val_accuracy: 0.7625\n",
      "Epoch 715/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1737 - accuracy: 0.9686 - val_loss: 1.1431 - val_accuracy: 0.7600\n",
      "Epoch 716/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1700 - accuracy: 0.9684 - val_loss: 1.1466 - val_accuracy: 0.7612\n",
      "Epoch 717/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1684 - accuracy: 0.9709 - val_loss: 1.1490 - val_accuracy: 0.7600\n",
      "Epoch 718/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1687 - accuracy: 0.9690 - val_loss: 1.1490 - val_accuracy: 0.7600\n",
      "Epoch 719/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1676 - accuracy: 0.9696 - val_loss: 1.1421 - val_accuracy: 0.7625\n",
      "Epoch 720/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1625 - accuracy: 0.9706 - val_loss: 1.1268 - val_accuracy: 0.7637\n",
      "Epoch 721/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1719 - accuracy: 0.9680 - val_loss: 1.1307 - val_accuracy: 0.7562\n",
      "Epoch 722/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1735 - accuracy: 0.9695 - val_loss: 1.1417 - val_accuracy: 0.7637\n",
      "Epoch 723/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1692 - accuracy: 0.9695 - val_loss: 1.1521 - val_accuracy: 0.7562\n",
      "Epoch 724/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1673 - accuracy: 0.9700 - val_loss: 1.1576 - val_accuracy: 0.7562\n",
      "Epoch 725/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1645 - accuracy: 0.9711 - val_loss: 1.1539 - val_accuracy: 0.7625\n",
      "Epoch 726/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1701 - accuracy: 0.9686 - val_loss: 1.1560 - val_accuracy: 0.7637\n",
      "Epoch 727/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1639 - accuracy: 0.9701 - val_loss: 1.1506 - val_accuracy: 0.7625\n",
      "Epoch 728/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1634 - accuracy: 0.9701 - val_loss: 1.1423 - val_accuracy: 0.7662\n",
      "Epoch 729/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1681 - accuracy: 0.9698 - val_loss: 1.1406 - val_accuracy: 0.7650\n",
      "Epoch 730/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1678 - accuracy: 0.9691 - val_loss: 1.1467 - val_accuracy: 0.7650\n",
      "Epoch 731/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1672 - accuracy: 0.9703 - val_loss: 1.1415 - val_accuracy: 0.7625\n",
      "Epoch 732/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1708 - accuracy: 0.9691 - val_loss: 1.1483 - val_accuracy: 0.7612\n",
      "Epoch 733/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1670 - accuracy: 0.9694 - val_loss: 1.1356 - val_accuracy: 0.7612\n",
      "Epoch 734/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1640 - accuracy: 0.9709 - val_loss: 1.1208 - val_accuracy: 0.7612\n",
      "Epoch 735/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1684 - accuracy: 0.9703 - val_loss: 1.1188 - val_accuracy: 0.7637\n",
      "Epoch 736/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1733 - accuracy: 0.9655 - val_loss: 1.1182 - val_accuracy: 0.7763\n",
      "Epoch 737/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1707 - accuracy: 0.9699 - val_loss: 1.1083 - val_accuracy: 0.7650\n",
      "Epoch 738/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1666 - accuracy: 0.9697 - val_loss: 1.1245 - val_accuracy: 0.7675\n",
      "Epoch 739/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1699 - accuracy: 0.9707 - val_loss: 1.1347 - val_accuracy: 0.7587\n",
      "Epoch 740/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1677 - accuracy: 0.9685 - val_loss: 1.1312 - val_accuracy: 0.7625\n",
      "Epoch 741/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1715 - accuracy: 0.9688 - val_loss: 1.1384 - val_accuracy: 0.7637\n",
      "Epoch 742/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1711 - accuracy: 0.9695 - val_loss: 1.1303 - val_accuracy: 0.7637\n",
      "Epoch 743/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1681 - accuracy: 0.9703 - val_loss: 1.1395 - val_accuracy: 0.7575\n",
      "Epoch 744/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1738 - accuracy: 0.9670 - val_loss: 1.1348 - val_accuracy: 0.7688\n",
      "Epoch 745/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1623 - accuracy: 0.9730 - val_loss: 1.1287 - val_accuracy: 0.7637\n",
      "Epoch 746/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1680 - accuracy: 0.9702 - val_loss: 1.1267 - val_accuracy: 0.7600\n",
      "Epoch 747/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1696 - accuracy: 0.9699 - val_loss: 1.1190 - val_accuracy: 0.7675\n",
      "Epoch 748/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1692 - accuracy: 0.9691 - val_loss: 1.1251 - val_accuracy: 0.7713\n",
      "Epoch 749/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1649 - accuracy: 0.9719 - val_loss: 1.1318 - val_accuracy: 0.7688\n",
      "Epoch 750/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1619 - accuracy: 0.9722 - val_loss: 1.1293 - val_accuracy: 0.7612\n",
      "Epoch 751/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1690 - accuracy: 0.9690 - val_loss: 1.1363 - val_accuracy: 0.7612\n",
      "Epoch 752/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1728 - accuracy: 0.9676 - val_loss: 1.1460 - val_accuracy: 0.7587\n",
      "Epoch 753/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1672 - accuracy: 0.9699 - val_loss: 1.1615 - val_accuracy: 0.7625\n",
      "Epoch 754/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1635 - accuracy: 0.9693 - val_loss: 1.1541 - val_accuracy: 0.7612\n",
      "Epoch 755/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1703 - accuracy: 0.9696 - val_loss: 1.1529 - val_accuracy: 0.7637\n",
      "Epoch 756/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1639 - accuracy: 0.9707 - val_loss: 1.1598 - val_accuracy: 0.7612\n",
      "Epoch 757/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1734 - accuracy: 0.9670 - val_loss: 1.1448 - val_accuracy: 0.7700\n",
      "Epoch 758/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1692 - accuracy: 0.9703 - val_loss: 1.1342 - val_accuracy: 0.7675\n",
      "Epoch 759/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1679 - accuracy: 0.9706 - val_loss: 1.1313 - val_accuracy: 0.7650\n",
      "Epoch 760/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1663 - accuracy: 0.9707 - val_loss: 1.1357 - val_accuracy: 0.7675\n",
      "Epoch 761/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1710 - accuracy: 0.9682 - val_loss: 1.1310 - val_accuracy: 0.7600\n",
      "Epoch 762/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1623 - accuracy: 0.9728 - val_loss: 1.1306 - val_accuracy: 0.7625\n",
      "Epoch 763/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1631 - accuracy: 0.9717 - val_loss: 1.1374 - val_accuracy: 0.7662\n",
      "Epoch 764/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1653 - accuracy: 0.9702 - val_loss: 1.1337 - val_accuracy: 0.7675\n",
      "Epoch 765/1000\n",
      "19/19 [==============================] - 0s 19ms/step - loss: 0.1593 - accuracy: 0.9730 - val_loss: 1.1356 - val_accuracy: 0.7637\n",
      "Epoch 766/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1679 - accuracy: 0.9697 - val_loss: 1.1394 - val_accuracy: 0.7612\n",
      "Epoch 767/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1658 - accuracy: 0.9710 - val_loss: 1.1378 - val_accuracy: 0.7587\n",
      "Epoch 768/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1656 - accuracy: 0.9705 - val_loss: 1.1417 - val_accuracy: 0.7637\n",
      "Epoch 769/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1709 - accuracy: 0.9679 - val_loss: 1.1375 - val_accuracy: 0.7600\n",
      "Epoch 770/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1655 - accuracy: 0.9712 - val_loss: 1.1444 - val_accuracy: 0.7562\n",
      "Epoch 771/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1626 - accuracy: 0.9699 - val_loss: 1.1539 - val_accuracy: 0.7600\n",
      "Epoch 772/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1634 - accuracy: 0.9706 - val_loss: 1.1413 - val_accuracy: 0.7587\n",
      "Epoch 773/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1622 - accuracy: 0.9719 - val_loss: 1.1517 - val_accuracy: 0.7612\n",
      "Epoch 774/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1676 - accuracy: 0.9694 - val_loss: 1.1704 - val_accuracy: 0.7650\n",
      "Epoch 775/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1659 - accuracy: 0.9713 - val_loss: 1.1710 - val_accuracy: 0.7600\n",
      "Epoch 776/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1641 - accuracy: 0.9728 - val_loss: 1.1676 - val_accuracy: 0.7587\n",
      "Epoch 777/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1594 - accuracy: 0.9739 - val_loss: 1.1543 - val_accuracy: 0.7612\n",
      "Epoch 778/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1647 - accuracy: 0.9720 - val_loss: 1.1472 - val_accuracy: 0.7662\n",
      "Epoch 779/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1612 - accuracy: 0.9724 - val_loss: 1.1505 - val_accuracy: 0.7675\n",
      "Epoch 780/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1684 - accuracy: 0.9701 - val_loss: 1.1554 - val_accuracy: 0.7650\n",
      "Epoch 781/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1637 - accuracy: 0.9725 - val_loss: 1.1780 - val_accuracy: 0.7625\n",
      "Epoch 782/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1676 - accuracy: 0.9699 - val_loss: 1.1765 - val_accuracy: 0.7575\n",
      "Epoch 783/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1677 - accuracy: 0.9686 - val_loss: 1.1778 - val_accuracy: 0.7587\n",
      "Epoch 784/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1626 - accuracy: 0.9735 - val_loss: 1.1898 - val_accuracy: 0.7600\n",
      "Epoch 785/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1662 - accuracy: 0.9702 - val_loss: 1.1774 - val_accuracy: 0.7662\n",
      "Epoch 786/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1659 - accuracy: 0.9716 - val_loss: 1.1690 - val_accuracy: 0.7650\n",
      "Epoch 787/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1584 - accuracy: 0.9728 - val_loss: 1.1567 - val_accuracy: 0.7550\n",
      "Epoch 788/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1628 - accuracy: 0.9721 - val_loss: 1.1475 - val_accuracy: 0.7587\n",
      "Epoch 789/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1633 - accuracy: 0.9714 - val_loss: 1.1473 - val_accuracy: 0.7587\n",
      "Epoch 790/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1602 - accuracy: 0.9726 - val_loss: 1.1520 - val_accuracy: 0.7575\n",
      "Epoch 791/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1655 - accuracy: 0.9715 - val_loss: 1.1474 - val_accuracy: 0.7575\n",
      "Epoch 792/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1676 - accuracy: 0.9709 - val_loss: 1.1424 - val_accuracy: 0.7600\n",
      "Epoch 793/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1604 - accuracy: 0.9726 - val_loss: 1.1516 - val_accuracy: 0.7587\n",
      "Epoch 794/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1695 - accuracy: 0.9693 - val_loss: 1.1576 - val_accuracy: 0.7525\n",
      "Epoch 795/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1712 - accuracy: 0.9698 - val_loss: 1.1616 - val_accuracy: 0.7550\n",
      "Epoch 796/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1647 - accuracy: 0.9707 - val_loss: 1.1592 - val_accuracy: 0.7562\n",
      "Epoch 797/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1636 - accuracy: 0.9710 - val_loss: 1.1496 - val_accuracy: 0.7537\n",
      "Epoch 798/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1641 - accuracy: 0.9713 - val_loss: 1.1524 - val_accuracy: 0.7575\n",
      "Epoch 799/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1611 - accuracy: 0.9717 - val_loss: 1.1572 - val_accuracy: 0.7575\n",
      "Epoch 800/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1668 - accuracy: 0.9720 - val_loss: 1.1583 - val_accuracy: 0.7562\n",
      "Epoch 801/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1636 - accuracy: 0.9722 - val_loss: 1.1544 - val_accuracy: 0.7575\n",
      "Epoch 802/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1639 - accuracy: 0.9722 - val_loss: 1.1643 - val_accuracy: 0.7650\n",
      "Epoch 803/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1646 - accuracy: 0.9707 - val_loss: 1.1723 - val_accuracy: 0.7587\n",
      "Epoch 804/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1636 - accuracy: 0.9735 - val_loss: 1.1632 - val_accuracy: 0.7650\n",
      "Epoch 805/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1631 - accuracy: 0.9727 - val_loss: 1.1561 - val_accuracy: 0.7637\n",
      "Epoch 806/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1607 - accuracy: 0.9718 - val_loss: 1.1534 - val_accuracy: 0.7587\n",
      "Epoch 807/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1654 - accuracy: 0.9705 - val_loss: 1.1492 - val_accuracy: 0.7562\n",
      "Epoch 808/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1717 - accuracy: 0.9691 - val_loss: 1.1463 - val_accuracy: 0.7600\n",
      "Epoch 809/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1659 - accuracy: 0.9697 - val_loss: 1.1644 - val_accuracy: 0.7575\n",
      "Epoch 810/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1624 - accuracy: 0.9715 - val_loss: 1.1552 - val_accuracy: 0.7575\n",
      "Epoch 811/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1669 - accuracy: 0.9703 - val_loss: 1.1544 - val_accuracy: 0.7600\n",
      "Epoch 812/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1634 - accuracy: 0.9714 - val_loss: 1.1410 - val_accuracy: 0.7600\n",
      "Epoch 813/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1701 - accuracy: 0.9712 - val_loss: 1.1398 - val_accuracy: 0.7550\n",
      "Epoch 814/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1655 - accuracy: 0.9708 - val_loss: 1.1322 - val_accuracy: 0.7562\n",
      "Epoch 815/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1618 - accuracy: 0.9714 - val_loss: 1.1275 - val_accuracy: 0.7637\n",
      "Epoch 816/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1635 - accuracy: 0.9724 - val_loss: 1.1332 - val_accuracy: 0.7600\n",
      "Epoch 817/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1632 - accuracy: 0.9721 - val_loss: 1.1502 - val_accuracy: 0.7612\n",
      "Epoch 818/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1615 - accuracy: 0.9714 - val_loss: 1.1467 - val_accuracy: 0.7487\n",
      "Epoch 819/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1648 - accuracy: 0.9694 - val_loss: 1.1577 - val_accuracy: 0.7550\n",
      "Epoch 820/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1632 - accuracy: 0.9716 - val_loss: 1.1651 - val_accuracy: 0.7525\n",
      "Epoch 821/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1598 - accuracy: 0.9740 - val_loss: 1.1671 - val_accuracy: 0.7612\n",
      "Epoch 822/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1728 - accuracy: 0.9691 - val_loss: 1.1760 - val_accuracy: 0.7525\n",
      "Epoch 823/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1614 - accuracy: 0.9733 - val_loss: 1.1865 - val_accuracy: 0.7537\n",
      "Epoch 824/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1638 - accuracy: 0.9718 - val_loss: 1.1783 - val_accuracy: 0.7550\n",
      "Epoch 825/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1659 - accuracy: 0.9713 - val_loss: 1.1731 - val_accuracy: 0.7600\n",
      "Epoch 826/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1623 - accuracy: 0.9711 - val_loss: 1.1655 - val_accuracy: 0.7637\n",
      "Epoch 827/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1608 - accuracy: 0.9728 - val_loss: 1.1528 - val_accuracy: 0.7587\n",
      "Epoch 828/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1652 - accuracy: 0.9707 - val_loss: 1.1338 - val_accuracy: 0.7587\n",
      "Epoch 829/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1624 - accuracy: 0.9724 - val_loss: 1.1357 - val_accuracy: 0.7512\n",
      "Epoch 830/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1639 - accuracy: 0.9709 - val_loss: 1.1444 - val_accuracy: 0.7550\n",
      "Epoch 831/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1612 - accuracy: 0.9710 - val_loss: 1.1622 - val_accuracy: 0.7525\n",
      "Epoch 832/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1616 - accuracy: 0.9723 - val_loss: 1.1574 - val_accuracy: 0.7575\n",
      "Epoch 833/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1633 - accuracy: 0.9709 - val_loss: 1.1477 - val_accuracy: 0.7575\n",
      "Epoch 834/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1616 - accuracy: 0.9730 - val_loss: 1.1623 - val_accuracy: 0.7612\n",
      "Epoch 835/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1654 - accuracy: 0.9726 - val_loss: 1.1607 - val_accuracy: 0.7637\n",
      "Epoch 836/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1633 - accuracy: 0.9707 - val_loss: 1.1507 - val_accuracy: 0.7600\n",
      "Epoch 837/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1619 - accuracy: 0.9724 - val_loss: 1.1626 - val_accuracy: 0.7575\n",
      "Epoch 838/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1629 - accuracy: 0.9720 - val_loss: 1.1668 - val_accuracy: 0.7625\n",
      "Epoch 839/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1631 - accuracy: 0.9722 - val_loss: 1.1765 - val_accuracy: 0.7662\n",
      "Epoch 840/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1656 - accuracy: 0.9724 - val_loss: 1.1668 - val_accuracy: 0.7650\n",
      "Epoch 841/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1574 - accuracy: 0.9726 - val_loss: 1.1739 - val_accuracy: 0.7675\n",
      "Epoch 842/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1620 - accuracy: 0.9720 - val_loss: 1.1806 - val_accuracy: 0.7562\n",
      "Epoch 843/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.1605 - accuracy: 0.9739 - val_loss: 1.1742 - val_accuracy: 0.7550\n",
      "Epoch 844/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1621 - accuracy: 0.9722 - val_loss: 1.1852 - val_accuracy: 0.7525\n",
      "Epoch 845/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1649 - accuracy: 0.9714 - val_loss: 1.1840 - val_accuracy: 0.7675\n",
      "Epoch 846/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1643 - accuracy: 0.9718 - val_loss: 1.1729 - val_accuracy: 0.7587\n",
      "Epoch 847/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1641 - accuracy: 0.9716 - val_loss: 1.1676 - val_accuracy: 0.7688\n",
      "Epoch 848/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1622 - accuracy: 0.9728 - val_loss: 1.1854 - val_accuracy: 0.7637\n",
      "Epoch 849/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1598 - accuracy: 0.9723 - val_loss: 1.1943 - val_accuracy: 0.7587\n",
      "Epoch 850/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1590 - accuracy: 0.9739 - val_loss: 1.1852 - val_accuracy: 0.7512\n",
      "Epoch 851/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1608 - accuracy: 0.9724 - val_loss: 1.1867 - val_accuracy: 0.7575\n",
      "Epoch 852/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1613 - accuracy: 0.9738 - val_loss: 1.1726 - val_accuracy: 0.7587\n",
      "Epoch 853/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1633 - accuracy: 0.9714 - val_loss: 1.1730 - val_accuracy: 0.7575\n",
      "Epoch 854/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1617 - accuracy: 0.9726 - val_loss: 1.1669 - val_accuracy: 0.7625\n",
      "Epoch 855/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1663 - accuracy: 0.9707 - val_loss: 1.1706 - val_accuracy: 0.7600\n",
      "Epoch 856/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1644 - accuracy: 0.9713 - val_loss: 1.1677 - val_accuracy: 0.7550\n",
      "Epoch 857/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1606 - accuracy: 0.9734 - val_loss: 1.1556 - val_accuracy: 0.7650\n",
      "Epoch 858/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1610 - accuracy: 0.9735 - val_loss: 1.1575 - val_accuracy: 0.7575\n",
      "Epoch 859/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1647 - accuracy: 0.9719 - val_loss: 1.1613 - val_accuracy: 0.7600\n",
      "Epoch 860/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1571 - accuracy: 0.9733 - val_loss: 1.1522 - val_accuracy: 0.7575\n",
      "Epoch 861/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1660 - accuracy: 0.9703 - val_loss: 1.1511 - val_accuracy: 0.7550\n",
      "Epoch 862/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1656 - accuracy: 0.9699 - val_loss: 1.1468 - val_accuracy: 0.7550\n",
      "Epoch 863/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1655 - accuracy: 0.9714 - val_loss: 1.1495 - val_accuracy: 0.7612\n",
      "Epoch 864/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1625 - accuracy: 0.9709 - val_loss: 1.1541 - val_accuracy: 0.7550\n",
      "Epoch 865/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1602 - accuracy: 0.9735 - val_loss: 1.1591 - val_accuracy: 0.7625\n",
      "Epoch 866/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1633 - accuracy: 0.9721 - val_loss: 1.1517 - val_accuracy: 0.7625\n",
      "Epoch 867/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1613 - accuracy: 0.9724 - val_loss: 1.1585 - val_accuracy: 0.7600\n",
      "Epoch 868/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1675 - accuracy: 0.9715 - val_loss: 1.1575 - val_accuracy: 0.7587\n",
      "Epoch 869/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1608 - accuracy: 0.9718 - val_loss: 1.1402 - val_accuracy: 0.7575\n",
      "Epoch 870/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1645 - accuracy: 0.9712 - val_loss: 1.1585 - val_accuracy: 0.7550\n",
      "Epoch 871/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1624 - accuracy: 0.9720 - val_loss: 1.1711 - val_accuracy: 0.7587\n",
      "Epoch 872/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1625 - accuracy: 0.9736 - val_loss: 1.1608 - val_accuracy: 0.7650\n",
      "Epoch 873/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1592 - accuracy: 0.9732 - val_loss: 1.1603 - val_accuracy: 0.7637\n",
      "Epoch 874/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1686 - accuracy: 0.9695 - val_loss: 1.1546 - val_accuracy: 0.7650\n",
      "Epoch 875/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1609 - accuracy: 0.9735 - val_loss: 1.1494 - val_accuracy: 0.7662\n",
      "Epoch 876/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1656 - accuracy: 0.9708 - val_loss: 1.1369 - val_accuracy: 0.7637\n",
      "Epoch 877/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1650 - accuracy: 0.9713 - val_loss: 1.1368 - val_accuracy: 0.7650\n",
      "Epoch 878/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1695 - accuracy: 0.9694 - val_loss: 1.1538 - val_accuracy: 0.7625\n",
      "Epoch 879/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1599 - accuracy: 0.9716 - val_loss: 1.1585 - val_accuracy: 0.7612\n",
      "Epoch 880/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1631 - accuracy: 0.9718 - val_loss: 1.1649 - val_accuracy: 0.7637\n",
      "Epoch 881/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1642 - accuracy: 0.9740 - val_loss: 1.1594 - val_accuracy: 0.7612\n",
      "Epoch 882/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1547 - accuracy: 0.9736 - val_loss: 1.1698 - val_accuracy: 0.7637\n",
      "Epoch 883/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1568 - accuracy: 0.9747 - val_loss: 1.1798 - val_accuracy: 0.7650\n",
      "Epoch 884/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1602 - accuracy: 0.9724 - val_loss: 1.1785 - val_accuracy: 0.7675\n",
      "Epoch 885/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1616 - accuracy: 0.9742 - val_loss: 1.1756 - val_accuracy: 0.7625\n",
      "Epoch 886/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1575 - accuracy: 0.9753 - val_loss: 1.1742 - val_accuracy: 0.7650\n",
      "Epoch 887/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1586 - accuracy: 0.9737 - val_loss: 1.1762 - val_accuracy: 0.7625\n",
      "Epoch 888/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1623 - accuracy: 0.9707 - val_loss: 1.1678 - val_accuracy: 0.7675\n",
      "Epoch 889/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1606 - accuracy: 0.9738 - val_loss: 1.1724 - val_accuracy: 0.7650\n",
      "Epoch 890/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1586 - accuracy: 0.9736 - val_loss: 1.1572 - val_accuracy: 0.7600\n",
      "Epoch 891/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1610 - accuracy: 0.9728 - val_loss: 1.1675 - val_accuracy: 0.7612\n",
      "Epoch 892/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1614 - accuracy: 0.9710 - val_loss: 1.1623 - val_accuracy: 0.7688\n",
      "Epoch 893/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1616 - accuracy: 0.9725 - val_loss: 1.1572 - val_accuracy: 0.7688\n",
      "Epoch 894/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1598 - accuracy: 0.9726 - val_loss: 1.1577 - val_accuracy: 0.7675\n",
      "Epoch 895/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1681 - accuracy: 0.9694 - val_loss: 1.1638 - val_accuracy: 0.7738\n",
      "Epoch 896/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1594 - accuracy: 0.9738 - val_loss: 1.1630 - val_accuracy: 0.7688\n",
      "Epoch 897/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1601 - accuracy: 0.9726 - val_loss: 1.1617 - val_accuracy: 0.7650\n",
      "Epoch 898/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1577 - accuracy: 0.9727 - val_loss: 1.1792 - val_accuracy: 0.7625\n",
      "Epoch 899/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1585 - accuracy: 0.9734 - val_loss: 1.1785 - val_accuracy: 0.7612\n",
      "Epoch 900/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1562 - accuracy: 0.9744 - val_loss: 1.1760 - val_accuracy: 0.7600\n",
      "Epoch 901/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1624 - accuracy: 0.9728 - val_loss: 1.1695 - val_accuracy: 0.7688\n",
      "Epoch 902/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1587 - accuracy: 0.9727 - val_loss: 1.1684 - val_accuracy: 0.7662\n",
      "Epoch 903/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1598 - accuracy: 0.9720 - val_loss: 1.1725 - val_accuracy: 0.7625\n",
      "Epoch 904/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1617 - accuracy: 0.9725 - val_loss: 1.1741 - val_accuracy: 0.7587\n",
      "Epoch 905/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1537 - accuracy: 0.9755 - val_loss: 1.1891 - val_accuracy: 0.7675\n",
      "Epoch 906/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1665 - accuracy: 0.9721 - val_loss: 1.1957 - val_accuracy: 0.7650\n",
      "Epoch 907/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1598 - accuracy: 0.9730 - val_loss: 1.1775 - val_accuracy: 0.7612\n",
      "Epoch 908/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1660 - accuracy: 0.9707 - val_loss: 1.1706 - val_accuracy: 0.7650\n",
      "Epoch 909/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1561 - accuracy: 0.9747 - val_loss: 1.1663 - val_accuracy: 0.7662\n",
      "Epoch 910/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1629 - accuracy: 0.9703 - val_loss: 1.1623 - val_accuracy: 0.7688\n",
      "Epoch 911/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1590 - accuracy: 0.9721 - val_loss: 1.1612 - val_accuracy: 0.7637\n",
      "Epoch 912/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1564 - accuracy: 0.9742 - val_loss: 1.1507 - val_accuracy: 0.7625\n",
      "Epoch 913/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1597 - accuracy: 0.9724 - val_loss: 1.1369 - val_accuracy: 0.7662\n",
      "Epoch 914/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1598 - accuracy: 0.9730 - val_loss: 1.1281 - val_accuracy: 0.7675\n",
      "Epoch 915/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1621 - accuracy: 0.9724 - val_loss: 1.1292 - val_accuracy: 0.7675\n",
      "Epoch 916/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1529 - accuracy: 0.9757 - val_loss: 1.1535 - val_accuracy: 0.7612\n",
      "Epoch 917/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1621 - accuracy: 0.9724 - val_loss: 1.1643 - val_accuracy: 0.7637\n",
      "Epoch 918/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1623 - accuracy: 0.9726 - val_loss: 1.1630 - val_accuracy: 0.7625\n",
      "Epoch 919/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1611 - accuracy: 0.9722 - val_loss: 1.1419 - val_accuracy: 0.7600\n",
      "Epoch 920/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1618 - accuracy: 0.9714 - val_loss: 1.1497 - val_accuracy: 0.7612\n",
      "Epoch 921/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1635 - accuracy: 0.9716 - val_loss: 1.1489 - val_accuracy: 0.7587\n",
      "Epoch 922/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1563 - accuracy: 0.9733 - val_loss: 1.1531 - val_accuracy: 0.7587\n",
      "Epoch 923/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1587 - accuracy: 0.9722 - val_loss: 1.1614 - val_accuracy: 0.7537\n",
      "Epoch 924/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1579 - accuracy: 0.9731 - val_loss: 1.1703 - val_accuracy: 0.7537\n",
      "Epoch 925/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1585 - accuracy: 0.9744 - val_loss: 1.1795 - val_accuracy: 0.7600\n",
      "Epoch 926/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1603 - accuracy: 0.9736 - val_loss: 1.1725 - val_accuracy: 0.7612\n",
      "Epoch 927/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1566 - accuracy: 0.9736 - val_loss: 1.1662 - val_accuracy: 0.7600\n",
      "Epoch 928/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1602 - accuracy: 0.9731 - val_loss: 1.1598 - val_accuracy: 0.7612\n",
      "Epoch 929/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1582 - accuracy: 0.9718 - val_loss: 1.1629 - val_accuracy: 0.7550\n",
      "Epoch 930/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1599 - accuracy: 0.9716 - val_loss: 1.1537 - val_accuracy: 0.7575\n",
      "Epoch 931/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1611 - accuracy: 0.9712 - val_loss: 1.1546 - val_accuracy: 0.7600\n",
      "Epoch 932/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1575 - accuracy: 0.9737 - val_loss: 1.1675 - val_accuracy: 0.7662\n",
      "Epoch 933/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1597 - accuracy: 0.9721 - val_loss: 1.1780 - val_accuracy: 0.7662\n",
      "Epoch 934/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1615 - accuracy: 0.9726 - val_loss: 1.1678 - val_accuracy: 0.7650\n",
      "Epoch 935/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1600 - accuracy: 0.9724 - val_loss: 1.1679 - val_accuracy: 0.7650\n",
      "Epoch 936/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1550 - accuracy: 0.9741 - val_loss: 1.1693 - val_accuracy: 0.7600\n",
      "Epoch 937/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1579 - accuracy: 0.9725 - val_loss: 1.1713 - val_accuracy: 0.7675\n",
      "Epoch 938/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1607 - accuracy: 0.9707 - val_loss: 1.1634 - val_accuracy: 0.7688\n",
      "Epoch 939/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1592 - accuracy: 0.9738 - val_loss: 1.1522 - val_accuracy: 0.7713\n",
      "Epoch 940/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1567 - accuracy: 0.9738 - val_loss: 1.1479 - val_accuracy: 0.7688\n",
      "Epoch 941/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1553 - accuracy: 0.9753 - val_loss: 1.1499 - val_accuracy: 0.7675\n",
      "Epoch 942/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1601 - accuracy: 0.9720 - val_loss: 1.1787 - val_accuracy: 0.7612\n",
      "Epoch 943/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1516 - accuracy: 0.9757 - val_loss: 1.1839 - val_accuracy: 0.7625\n",
      "Epoch 944/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1596 - accuracy: 0.9724 - val_loss: 1.1650 - val_accuracy: 0.7600\n",
      "Epoch 945/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1583 - accuracy: 0.9748 - val_loss: 1.1677 - val_accuracy: 0.7662\n",
      "Epoch 946/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1578 - accuracy: 0.9734 - val_loss: 1.1957 - val_accuracy: 0.7612\n",
      "Epoch 947/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1624 - accuracy: 0.9718 - val_loss: 1.1919 - val_accuracy: 0.7637\n",
      "Epoch 948/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1625 - accuracy: 0.9722 - val_loss: 1.1942 - val_accuracy: 0.7587\n",
      "Epoch 949/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1577 - accuracy: 0.9741 - val_loss: 1.1922 - val_accuracy: 0.7587\n",
      "Epoch 950/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1556 - accuracy: 0.9752 - val_loss: 1.1808 - val_accuracy: 0.7625\n",
      "Epoch 951/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1542 - accuracy: 0.9755 - val_loss: 1.1760 - val_accuracy: 0.7650\n",
      "Epoch 952/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1551 - accuracy: 0.9739 - val_loss: 1.1690 - val_accuracy: 0.7600\n",
      "Epoch 953/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1605 - accuracy: 0.9725 - val_loss: 1.1636 - val_accuracy: 0.7625\n",
      "Epoch 954/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1650 - accuracy: 0.9729 - val_loss: 1.1563 - val_accuracy: 0.7750\n",
      "Epoch 955/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1566 - accuracy: 0.9740 - val_loss: 1.1685 - val_accuracy: 0.7700\n",
      "Epoch 956/1000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1609 - accuracy: 0.9730 - val_loss: 1.1647 - val_accuracy: 0.7688\n",
      "Epoch 957/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1600 - accuracy: 0.9726 - val_loss: 1.1546 - val_accuracy: 0.7675\n",
      "Epoch 958/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1608 - accuracy: 0.9728 - val_loss: 1.1694 - val_accuracy: 0.7713\n",
      "Epoch 959/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1606 - accuracy: 0.9716 - val_loss: 1.1693 - val_accuracy: 0.7612\n",
      "Epoch 960/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1573 - accuracy: 0.9725 - val_loss: 1.1629 - val_accuracy: 0.7575\n",
      "Epoch 961/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1569 - accuracy: 0.9724 - val_loss: 1.1649 - val_accuracy: 0.7637\n",
      "Epoch 962/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1574 - accuracy: 0.9746 - val_loss: 1.1755 - val_accuracy: 0.7650\n",
      "Epoch 963/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1596 - accuracy: 0.9739 - val_loss: 1.1612 - val_accuracy: 0.7662\n",
      "Epoch 964/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1533 - accuracy: 0.9743 - val_loss: 1.1683 - val_accuracy: 0.7700\n",
      "Epoch 965/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1614 - accuracy: 0.9724 - val_loss: 1.1760 - val_accuracy: 0.7637\n",
      "Epoch 966/1000\n",
      "19/19 [==============================] - 0s 15ms/step - loss: 0.1619 - accuracy: 0.9728 - val_loss: 1.1692 - val_accuracy: 0.7725\n",
      "Epoch 967/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1517 - accuracy: 0.9751 - val_loss: 1.1634 - val_accuracy: 0.7688\n",
      "Epoch 968/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1571 - accuracy: 0.9738 - val_loss: 1.1691 - val_accuracy: 0.7662\n",
      "Epoch 969/1000\n",
      "19/19 [==============================] - 0s 13ms/step - loss: 0.1581 - accuracy: 0.9739 - val_loss: 1.1642 - val_accuracy: 0.7675\n",
      "Epoch 970/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1605 - accuracy: 0.9738 - val_loss: 1.1741 - val_accuracy: 0.7688\n",
      "Epoch 971/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1576 - accuracy: 0.9726 - val_loss: 1.1668 - val_accuracy: 0.7637\n",
      "Epoch 972/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1564 - accuracy: 0.9753 - val_loss: 1.1569 - val_accuracy: 0.7650\n",
      "Epoch 973/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1660 - accuracy: 0.9726 - val_loss: 1.1578 - val_accuracy: 0.7725\n",
      "Epoch 974/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1586 - accuracy: 0.9731 - val_loss: 1.1662 - val_accuracy: 0.7725\n",
      "Epoch 975/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.1572 - accuracy: 0.9747 - val_loss: 1.1646 - val_accuracy: 0.7738\n",
      "Epoch 976/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1578 - accuracy: 0.9742 - val_loss: 1.1653 - val_accuracy: 0.7750\n",
      "Epoch 977/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1585 - accuracy: 0.9724 - val_loss: 1.1705 - val_accuracy: 0.7738\n",
      "Epoch 978/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1575 - accuracy: 0.9736 - val_loss: 1.1617 - val_accuracy: 0.7700\n",
      "Epoch 979/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1580 - accuracy: 0.9736 - val_loss: 1.1651 - val_accuracy: 0.7650\n",
      "Epoch 980/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1618 - accuracy: 0.9720 - val_loss: 1.1664 - val_accuracy: 0.7650\n",
      "Epoch 981/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1581 - accuracy: 0.9749 - val_loss: 1.1727 - val_accuracy: 0.7713\n",
      "Epoch 982/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1648 - accuracy: 0.9723 - val_loss: 1.1720 - val_accuracy: 0.7600\n",
      "Epoch 983/1000\n",
      "19/19 [==============================] - 0s 14ms/step - loss: 0.1619 - accuracy: 0.9734 - val_loss: 1.1777 - val_accuracy: 0.7587\n",
      "Epoch 984/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1565 - accuracy: 0.9753 - val_loss: 1.1737 - val_accuracy: 0.7662\n",
      "Epoch 985/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1561 - accuracy: 0.9751 - val_loss: 1.1678 - val_accuracy: 0.7713\n",
      "Epoch 986/1000\n",
      "19/19 [==============================] - 0s 10ms/step - loss: 0.1608 - accuracy: 0.9727 - val_loss: 1.1680 - val_accuracy: 0.7662\n",
      "Epoch 987/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1546 - accuracy: 0.9755 - val_loss: 1.1691 - val_accuracy: 0.7662\n",
      "Epoch 988/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.1544 - accuracy: 0.9745 - val_loss: 1.1655 - val_accuracy: 0.7600\n",
      "Epoch 989/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1549 - accuracy: 0.9756 - val_loss: 1.1642 - val_accuracy: 0.7650\n",
      "Epoch 990/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.1503 - accuracy: 0.9763 - val_loss: 1.1657 - val_accuracy: 0.7700\n",
      "Epoch 991/1000\n",
      "19/19 [==============================] - 0s 11ms/step - loss: 0.1552 - accuracy: 0.9745 - val_loss: 1.1619 - val_accuracy: 0.7662\n",
      "Epoch 992/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1540 - accuracy: 0.9764 - val_loss: 1.1552 - val_accuracy: 0.7637\n",
      "Epoch 993/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1602 - accuracy: 0.9729 - val_loss: 1.1518 - val_accuracy: 0.7725\n",
      "Epoch 994/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1546 - accuracy: 0.9744 - val_loss: 1.1553 - val_accuracy: 0.7612\n",
      "Epoch 995/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1548 - accuracy: 0.9743 - val_loss: 1.1473 - val_accuracy: 0.7675\n",
      "Epoch 996/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1574 - accuracy: 0.9734 - val_loss: 1.1543 - val_accuracy: 0.7625\n",
      "Epoch 997/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1556 - accuracy: 0.9741 - val_loss: 1.1544 - val_accuracy: 0.7650\n",
      "Epoch 998/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1590 - accuracy: 0.9755 - val_loss: 1.1474 - val_accuracy: 0.7688\n",
      "Epoch 999/1000\n",
      "19/19 [==============================] - 0s 12ms/step - loss: 0.1576 - accuracy: 0.9725 - val_loss: 1.1406 - val_accuracy: 0.7675\n",
      "Epoch 1000/1000\n",
      "19/19 [==============================] - 0s 16ms/step - loss: 0.1546 - accuracy: 0.9746 - val_loss: 1.1465 - val_accuracy: 0.7688\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.training.tracking.util.CheckpointLoadStatus at 0x2f9986520>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history = model1.fit(fine_tune_data_train, fine_tune_labels_train, epochs=EPOCHS, batch_size=800, validation_data=(fine_tune_data_test, fine_tune_labels_test), validation_batch_size=800, callbacks=[model_checkpoint_callback])\n",
    "model1.save_weights(\"./Models/VGG16-4P/effnet-4p-model-weights\")\n",
    "model1.load_weights('./Models/VGG16-4P/effnet-4p-model-weights')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "e7b10c51",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.training.tracking.util.CheckpointLoadStatus at 0x29b43d340>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model1.load_weights('./Models/VGG16-4P/auto-VGG16-FT-model-weight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9116a6a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_train_loss(history, model_name):  \n",
    "    plt.figure(figsize=(8,6))\n",
    "    plt.plot(history.history['loss'])\n",
    "    plt.plot(history.history['accuracy'])\n",
    "    plt.plot(history.history['val_loss'])\n",
    "    plt.plot(history.history['val_accuracy'])\n",
    "    plt.title(model_name+' loss')\n",
    "    plt.ylabel('loss')\n",
    "    plt.xlabel('epochs')\n",
    "    plt.legend(['train loss', 'train accuracy', 'val loss', 'val accuracy'], loc='upper left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "c55b5870",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_train_loss(history, 'Fine-Tuned VGG16-4Piece Model')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03390fa6",
   "metadata": {},
   "source": [
    "### Model Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48701107",
   "metadata": {},
   "source": [
    "#### On Train Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0594069",
   "metadata": {},
   "source": [
    "Now, after the model is being trained, we will look at the performance of the model on the training set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "cf7ded38",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " 54/475 [==>...........................] - ETA: 1s"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 17:09:31.466043: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:113] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "475/475 [==============================] - 1s 3ms/step\n"
     ]
    }
   ],
   "source": [
    "train_preds = model1.predict(fine_tune_data_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "f2a28269",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_pred_labels = np.argmax(train_preds, axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "768775b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1, 13, 13, ...,  3,  9,  9])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_pred_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "eff52a3d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15197"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(train_pred_labels == train_labels['label'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3fd403fa",
   "metadata": {},
   "source": [
    "Looking at the misclassified images to get a better idea about what features the model is unable to capture."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "8f594a33",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>images</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9198</th>\n",
       "      <td>9682</td>\n",
       "      <td>./Data/train/train/9682.tif</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12255</th>\n",
       "      <td>12896</td>\n",
       "      <td>./Data/train/train/12896.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14569</th>\n",
       "      <td>15333</td>\n",
       "      <td>./Data/train/train/15333.tif</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          id                        images  label\n",
       "9198    9682   ./Data/train/train/9682.tif     12\n",
       "12255  12896  ./Data/train/train/12896.tif      9\n",
       "14569  15333  ./Data/train/train/15333.tif      7"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mis_train_data = train_labels[train_pred_labels != train_labels['label']]\n",
    "mis_train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "2ea8f949",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>precision</th>\n",
       "      <th>recall</th>\n",
       "      <th>fscore</th>\n",
       "      <th>support</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.997959</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998979</td>\n",
       "      <td>978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.998926</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.999463</td>\n",
       "      <td>930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998905</td>\n",
       "      <td>0.999452</td>\n",
       "      <td>913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998951</td>\n",
       "      <td>0.999475</td>\n",
       "      <td>953</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998976</td>\n",
       "      <td>0.999488</td>\n",
       "      <td>977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>943</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>945</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    precision    recall    fscore  support\n",
       "0    0.997959  1.000000  0.998979      978\n",
       "1    1.000000  1.000000  1.000000      925\n",
       "2    1.000000  1.000000  1.000000      925\n",
       "3    1.000000  1.000000  1.000000      947\n",
       "4    0.998926  1.000000  0.999463      930\n",
       "5    1.000000  1.000000  1.000000      962\n",
       "6    1.000000  1.000000  1.000000      998\n",
       "7    1.000000  0.998905  0.999452      913\n",
       "8    1.000000  1.000000  1.000000     1005\n",
       "9    1.000000  0.998951  0.999475      953\n",
       "10   1.000000  1.000000  1.000000      916\n",
       "11   1.000000  1.000000  1.000000      955\n",
       "12   1.000000  0.998976  0.999488      977\n",
       "13   1.000000  1.000000  1.000000      943\n",
       "14   1.000000  1.000000  1.000000      928\n",
       "15   1.000000  1.000000  1.000000      945"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision, recall, fscore, support = score(train_labels['label'], train_pred_labels)\n",
    "pd.DataFrame({\n",
    "    'precision': precision,\n",
    "    'recall': recall,\n",
    "    'fscore': fscore,\n",
    "    'support': support\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "659ae313",
   "metadata": {},
   "source": [
    "#### On Train-Test Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "0d1e8f85",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25/25 [==============================] - 0s 4ms/step\n"
     ]
    }
   ],
   "source": [
    "test_preds = model1.predict(fine_tune_data_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "49233509",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pred_labels = np.argmax(test_preds, axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "8f5cba39",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([11,  8,  6,  7,  1,  6, 13, 12, 14,  1,  6, 15,  0, 14, 13, 11,  0,\n",
       "        6,  6, 12,  8, 14,  3,  5,  6,  0,  3, 14,  3,  3, 11,  1,  6,  2,\n",
       "        1,  8,  4, 10, 11,  9,  6, 12, 14,  5,  5, 14,  3,  0, 13,  4,  0,\n",
       "       10, 13, 15, 11, 14, 15,  9,  9,  2, 15,  9, 10, 11,  6,  8,  8,  0,\n",
       "        6,  3,  4,  4, 12,  0,  8,  6,  7,  0, 12,  0, 11, 15, 12, 12,  9,\n",
       "        8,  1, 13, 13, 13, 10, 11,  4, 10,  1, 13, 11,  8, 13, 10,  3, 10,\n",
       "       12,  9, 11, 14,  8,  2,  5,  8,  7,  7,  9,  4, 12,  9, 12,  1, 11,\n",
       "        6, 10,  1,  0, 10,  1, 10, 13,  8, 15,  7,  3, 11,  6,  1,  6,  9,\n",
       "       14, 10,  7, 14, 10,  8,  4, 12,  8, 13,  8,  9,  2, 13, 15, 15,  4,\n",
       "       14,  4,  9,  4, 14,  7,  7,  8,  4,  4, 12, 13,  8, 15,  3,  8,  8,\n",
       "       13,  6,  3, 11,  7, 12, 10, 14, 14,  8,  6, 10, 13, 11,  6, 13,  4,\n",
       "       14, 11,  7, 15,  0,  3,  0, 10, 13, 12, 10, 11, 11,  5,  1,  9,  1,\n",
       "        9,  5, 13, 15, 12,  1,  4,  8,  7,  6, 13, 12,  4, 12,  8, 15,  9,\n",
       "       13, 15, 13,  7,  7,  2,  2,  9, 11,  7,  0,  3, 12,  3,  1,  2, 12,\n",
       "        4,  7, 12, 11, 12,  8, 13, 10, 12, 15, 11,  0, 11, 13, 15, 15,  8,\n",
       "        5, 10, 13,  5,  0,  9, 14,  6, 13,  5,  0, 11, 15, 11, 11, 14,  3,\n",
       "        0, 11,  4,  3,  1,  4,  6,  2,  0,  8, 10, 10,  8,  5,  7,  8, 13,\n",
       "        6,  8, 15,  0,  2,  3,  4, 10,  1,  8, 13,  0,  7, 15, 10, 10,  5,\n",
       "        3, 13, 11, 10,  9,  2, 11,  3, 15,  6,  5,  4, 14, 11, 13,  2, 15,\n",
       "        9, 15, 14,  7,  0, 15,  3, 11,  0, 15, 15,  1, 10,  6, 13, 15, 15,\n",
       "       13,  0,  3,  3,  2,  8,  9,  0, 11,  3,  3,  2, 12, 14,  4,  0,  3,\n",
       "        2, 12,  1,  3,  0,  5, 11, 15,  7, 15, 12,  3,  2, 12, 15,  3,  6,\n",
       "        0, 14, 14,  0,  5,  6, 10,  6, 12, 13,  6,  9,  8,  7, 10,  1,  2,\n",
       "       11, 14,  2, 14, 15,  4,  0,  1,  2, 10,  1,  0, 14,  6,  9,  0,  2,\n",
       "        7,  3, 14,  8, 13, 11,  4,  8, 15,  1,  4,  7, 13, 14,  5, 11,  8,\n",
       "        0, 15, 11,  0, 10,  3,  7,  2,  5, 13, 15, 11, 11, 13,  3, 11,  8,\n",
       "       12,  0,  3, 11, 15,  9,  6, 14,  7,  8,  4,  2, 13, 12,  5, 13, 11,\n",
       "        9,  8, 10, 14, 12,  0, 15,  1,  9,  9,  0, 12,  7, 14,  9, 15, 11,\n",
       "        5, 10,  4,  8,  6,  3,  7,  8,  6, 10,  1, 10,  7, 15, 12, 11, 10,\n",
       "        2,  3,  1, 10, 13,  7,  8,  1,  9,  1,  6,  7, 15,  0,  8,  7,  8,\n",
       "       12,  1,  0, 12, 11, 11, 10,  7,  9,  1, 14,  3,  5, 14,  6, 14,  7,\n",
       "        1, 14,  6, 14,  9, 13,  4, 14,  1, 11,  2, 13,  7,  2,  0,  8,  8,\n",
       "        0,  8,  6,  5, 14,  3,  6, 11,  8, 14,  6,  9,  1,  9, 11,  0,  0,\n",
       "        3, 10,  1, 10, 15,  2, 14, 15, 14,  9,  8,  1,  9, 13,  1, 15, 12,\n",
       "        9, 11,  6, 10, 12,  8, 10, 14, 14,  9,  0, 15,  8, 15,  8,  6,  5,\n",
       "        9,  5,  0,  0, 14,  5,  3,  2, 12,  7, 15, 11, 10, 11, 13,  1, 11,\n",
       "       10, 13, 15,  4,  4,  9, 11,  0, 14, 13,  4,  8,  5,  9, 11,  4, 13,\n",
       "       13,  0,  1, 15, 13,  7,  6, 12,  8,  7,  6, 15,  5,  9,  4,  1,  0,\n",
       "        1, 13,  6,  6,  3,  7,  7,  5,  5, 12,  6,  8,  0, 12,  7,  7,  0,\n",
       "       12,  6, 12,  0,  3, 11,  2,  3,  5,  4, 13,  1, 11, 10, 14, 15,  8,\n",
       "       11, 12,  1,  7,  2,  5,  8,  8, 11,  6, 11,  2, 10, 13,  7,  6,  3,\n",
       "       13,  3,  2, 14,  0,  8, 12, 12,  8, 13, 13, 14,  4,  9,  2,  5, 14,\n",
       "       12, 13, 13,  4, 12, 15, 10,  0,  8, 11, 15,  8,  4,  9,  2,  8,  0,\n",
       "        1,  2, 15, 15,  0,  8,  9, 11, 15,  8,  9,  7,  4, 11, 15, 11,  3,\n",
       "        7,  0,  3,  8, 12,  2, 12, 15,  7, 10, 13, 12, 12,  6, 14,  0,  7,\n",
       "        5,  3, 12, 13,  5,  7,  2,  8, 10, 13,  2, 15,  8, 14, 12, 14, 12,\n",
       "        3,  5,  8, 12,  4, 11, 15,  0,  9,  9,  5, 13, 10,  2,  2, 11, 10,\n",
       "        6])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "ee078822",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "616"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(test_pred_labels == test_labels['label'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "c3db9b5c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>images</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>41</td>\n",
       "      <td>./Data/train/train/41.tif</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>65</td>\n",
       "      <td>./Data/train/train/65.tif</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>67</td>\n",
       "      <td>./Data/train/train/67.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>100</td>\n",
       "      <td>./Data/train/train/100.tif</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>179</td>\n",
       "      <td>./Data/train/train/179.tif</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>776</th>\n",
       "      <td>15528</td>\n",
       "      <td>./Data/train/train/15528.tif</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>779</th>\n",
       "      <td>15568</td>\n",
       "      <td>./Data/train/train/15568.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>785</th>\n",
       "      <td>15621</td>\n",
       "      <td>./Data/train/train/15621.tif</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>788</th>\n",
       "      <td>15687</td>\n",
       "      <td>./Data/train/train/15687.tif</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>798</th>\n",
       "      <td>15863</td>\n",
       "      <td>./Data/train/train/15863.tif</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>184 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        id                        images  label\n",
       "3       41     ./Data/train/train/41.tif      1\n",
       "4       65     ./Data/train/train/65.tif     14\n",
       "5       67     ./Data/train/train/67.tif      9\n",
       "7      100    ./Data/train/train/100.tif     10\n",
       "12     179    ./Data/train/train/179.tif      7\n",
       "..     ...                           ...    ...\n",
       "776  15528  ./Data/train/train/15528.tif      5\n",
       "779  15568  ./Data/train/train/15568.tif      9\n",
       "785  15621  ./Data/train/train/15621.tif     15\n",
       "788  15687  ./Data/train/train/15687.tif     12\n",
       "798  15863  ./Data/train/train/15863.tif     15\n",
       "\n",
       "[184 rows x 3 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mis_train_data = test_labels[test_pred_labels != test_labels['label']]\n",
    "mis_train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "b538b721",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>precision</th>\n",
       "      <th>recall</th>\n",
       "      <th>fscore</th>\n",
       "      <th>support</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.689655</td>\n",
       "      <td>0.740741</td>\n",
       "      <td>0.714286</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.595238</td>\n",
       "      <td>0.568182</td>\n",
       "      <td>0.581395</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.973684</td>\n",
       "      <td>0.860465</td>\n",
       "      <td>0.913580</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.934783</td>\n",
       "      <td>0.934783</td>\n",
       "      <td>0.934783</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.868421</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>0.891892</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.647059</td>\n",
       "      <td>0.440000</td>\n",
       "      <td>0.523810</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.795918</td>\n",
       "      <td>0.847826</td>\n",
       "      <td>0.821053</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.872340</td>\n",
       "      <td>0.788462</td>\n",
       "      <td>0.828283</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.818182</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.900000</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.800000</td>\n",
       "      <td>0.765957</td>\n",
       "      <td>0.782609</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.795918</td>\n",
       "      <td>0.672414</td>\n",
       "      <td>0.728972</td>\n",
       "      <td>58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.687500</td>\n",
       "      <td>0.785714</td>\n",
       "      <td>0.733333</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.618182</td>\n",
       "      <td>0.790698</td>\n",
       "      <td>0.693878</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.783333</td>\n",
       "      <td>0.734375</td>\n",
       "      <td>0.758065</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.960784</td>\n",
       "      <td>0.765625</td>\n",
       "      <td>0.852174</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.568966</td>\n",
       "      <td>0.767442</td>\n",
       "      <td>0.653465</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    precision    recall    fscore  support\n",
       "0    0.689655  0.740741  0.714286       54\n",
       "1    0.595238  0.568182  0.581395       44\n",
       "2    0.973684  0.860465  0.913580       43\n",
       "3    0.934783  0.934783  0.934783       46\n",
       "4    0.868421  0.916667  0.891892       36\n",
       "5    0.647059  0.440000  0.523810       50\n",
       "6    0.795918  0.847826  0.821053       46\n",
       "7    0.872340  0.788462  0.828283       52\n",
       "8    0.818182  1.000000  0.900000       54\n",
       "9    0.800000  0.765957  0.782609       47\n",
       "10   0.795918  0.672414  0.728972       58\n",
       "11   0.687500  0.785714  0.733333       56\n",
       "12   0.618182  0.790698  0.693878       43\n",
       "13   0.783333  0.734375  0.758065       64\n",
       "14   0.960784  0.765625  0.852174       64\n",
       "15   0.568966  0.767442  0.653465       43"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision, recall, fscore, support = score(test_labels['label'], test_pred_labels)\n",
    "pd.DataFrame({\n",
    "    'precision': precision,\n",
    "    'recall': recall,\n",
    "    'fscore': fscore,\n",
    "    'support': support\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "ed4db76c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7694735070390781"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(fscore)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "067c5096",
   "metadata": {},
   "source": [
    "#### On Validation Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e665591",
   "metadata": {},
   "source": [
    "Generating Predictions for the Validation Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "4b085c0c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>images</th>\n",
       "      <th>id</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>850</th>\n",
       "      <td>./Data/validation/validation/17801.tif</td>\n",
       "      <td>17801</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>898</th>\n",
       "      <td>./Data/validation/validation/17802.tif</td>\n",
       "      <td>17802</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>852</th>\n",
       "      <td>./Data/validation/validation/17803.tif</td>\n",
       "      <td>17803</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>798</th>\n",
       "      <td>./Data/validation/validation/17804.tif</td>\n",
       "      <td>17804</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>750</th>\n",
       "      <td>./Data/validation/validation/17805.tif</td>\n",
       "      <td>17805</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>./Data/validation/validation/18696.tif</td>\n",
       "      <td>18696</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>211</th>\n",
       "      <td>./Data/validation/validation/18697.tif</td>\n",
       "      <td>18697</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>760</th>\n",
       "      <td>./Data/validation/validation/18698.tif</td>\n",
       "      <td>18698</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>787</th>\n",
       "      <td>./Data/validation/validation/18699.tif</td>\n",
       "      <td>18699</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>891</th>\n",
       "      <td>./Data/validation/validation/18700.tif</td>\n",
       "      <td>18700</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>900 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     images     id  label\n",
       "850  ./Data/validation/validation/17801.tif  17801     -1\n",
       "898  ./Data/validation/validation/17802.tif  17802     -1\n",
       "852  ./Data/validation/validation/17803.tif  17803     -1\n",
       "798  ./Data/validation/validation/17804.tif  17804     -1\n",
       "750  ./Data/validation/validation/17805.tif  17805     -1\n",
       "..                                      ...    ...    ...\n",
       "246  ./Data/validation/validation/18696.tif  18696     -1\n",
       "211  ./Data/validation/validation/18697.tif  18697     -1\n",
       "760  ./Data/validation/validation/18698.tif  18698     -1\n",
       "787  ./Data/validation/validation/18699.tif  18699     -1\n",
       "891  ./Data/validation/validation/18700.tif  18700     -1\n",
       "\n",
       "[900 rows x 3 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation_data = pd.DataFrame({'images':['./Data/validation/validation/'+name for name in os.listdir('./Data/validation/validation/')]})\n",
    "validation_data['id'] = [name.split('.')[0] for name in os.listdir('./Data/validation/validation/')]\n",
    "validation_data['label'] = -1 # Simply added to prevent re-writing code\n",
    "validation_data.sort_values(by=['id'], inplace=True)\n",
    "validation_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "53635253",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(900, 5, 512)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_validation = np.load('./Data/VGG16-4P-validation-precompute/whole_validation_data_precomp.npy')\n",
    "fine_tune_data_validation.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "db7d1256",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "29/29 [==============================] - 0s 3ms/step\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 17:09:33.239632: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:113] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    }
   ],
   "source": [
    "valid_preds = model1.predict(fine_tune_data_validation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "18b17d2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "valid_pred_labels = np.argmax(valid_preds, axis=-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2bece594",
   "metadata": {},
   "source": [
    "### Submission"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "304c47de",
   "metadata": {},
   "source": [
    "For Kaggle Submission"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "e7ca9d19",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17801</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17802</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>17803</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>17804</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>17805</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>895</th>\n",
       "      <td>18696</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>896</th>\n",
       "      <td>18697</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>897</th>\n",
       "      <td>18698</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>898</th>\n",
       "      <td>18699</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>899</th>\n",
       "      <td>18700</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>900 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        id  label\n",
       "0    17801      4\n",
       "1    17802      6\n",
       "2    17803      4\n",
       "3    17804      8\n",
       "4    17805      3\n",
       "..     ...    ...\n",
       "895  18696      7\n",
       "896  18697     12\n",
       "897  18698      8\n",
       "898  18699      5\n",
       "899  18700      4\n",
       "\n",
       "[900 rows x 2 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submission_data = pd.DataFrame({'id':list(range(17801, 18701))})\n",
    "submission_data['label'] = valid_pred_labels\n",
    "submission_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "6eea7d9c",
   "metadata": {},
   "outputs": [],
   "source": [
    "submission_data.to_csv('./Submissions/submission-vgg16-4p.csv', index=False)"
   ]
  }
 ],
 "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": "3.8.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}