GRU_Experiment2_Padding_and_Masking.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "python version:  3.7.3 (default, Mar 27 2019, 22:11:17) \n",
      "[GCC 7.3.0]\n",
      "pandas version:  0.24.2\n",
      "numpy version:  1.16.4\n",
      "scipy version:  1.3.0\n",
      "sklearn version:  0.21.2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "keras version:  2.2.4\n",
      "keras backend:  tensorflow\n",
      "tensorflow version:  1.14.0\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "print(\"python version: \", sys.version)\n",
    "\n",
    "import pandas as pd\n",
    "print(\"pandas version: \", pd.__version__)\n",
    "\n",
    "import numpy as np\n",
    "print('numpy version: ', np.__version__)\n",
    "\n",
    "import scipy as sp\n",
    "print('scipy version: ', sp.__version__)\n",
    "\n",
    "import sklearn as skl\n",
    "print('sklearn version: ', skl.__version__)\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import os\n",
    "os.environ[\"CUDA_DEVICE_ORDER\"] = \"PCI_BUS_ID\"\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"\"\n",
    "\n",
    "import keras as ks\n",
    "print('keras version: ', ks.__version__)\n",
    "print('keras backend: ', ks.backend.backend())\n",
    "\n",
    "import tensorflow as tf\n",
    "print('tensorflow version: ', tf.__version__)\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "from keras import backend as K\n",
    "from sklearn.metrics import confusion_matrix, precision_score, recall_score\n",
    "from sklearn.metrics import roc_curve, auc, f1_score, accuracy_score, precision_recall_curve\n",
    "\n",
    "import random"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(7584, 208)\n"
     ]
    }
   ],
   "source": [
    "excel_file = r'geriomop-data-from-egems-paper-UPDATED-COL-FIX-ATC_2-and-behavior-change-022819.tsv'\n",
    "df = pd.DataFrame(pd.read_csv(excel_file, sep='\\t'))\n",
    "print(df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Drop features\n",
    "f_drop = [\"Blank-1\", \"ws-sudden-stop\", \"ws-sudden-stop-drugs\", \"ws-prolonged-psychotropics\", \"mds-toilet-prgm-atmptd\"] + \\\n",
    "         ['Deleted-%d' % i for i in range(1, 13)] + ['Psychotropic-%d' % i for i in range(1, 6)] + \\\n",
    "         ['Psychotropic-%d-average-daily-dose' % i for i in range(1, 6)] + [\"Entry-discharge-type\"]\n",
    "\n",
    "# Missing value means not relevant\n",
    "f_one_hot = [\"Mds-fall-2-6-months-to-admission\", \"Mds-fall-last-month-to-admission\", \"mds-pain-last-five-days\", \\\n",
    "             \"mds-delirium-scale\", \"mds-long-term-memory-ok\", \"mds-short-term-memory-ok\", \"mds-staff-assess-pain\", \\\n",
    "             \"recent-start-other-fall-risk-rx\"] + [\"mds-pressure-ulcer-stage-%d\" % i for i in range(1, 5)]\n",
    "cate_one_hot = [['Yes', 'No', 'Unable to answer']] * 3 + [['Yes', 'No']] * 9\n",
    "\n",
    "f_period = [(\"Psychotropic-%d-Start-date\" % i, 'Psychotropic-%d-End-date' % i, 'Psychotropic-%d-Period' % i) for i in range(1, 6)]\n",
    "\n",
    "f_label = ['mds-cognitive-scale', 'mds-pain-freq-last-five-days', 'mds-pain-intensity']\n",
    "dic_label = [{'Independent': 0, 'Modified Independence': 1, 'Moderately Impaired': 2, 'Severely Impaired': 3}, \\\n",
    "             {'Continuous': 4, 'Frequent': 3, 'Occasional': 2, 'Rare': 1, 'Unable to respond': 0}, \\\n",
    "             {'None': 0, 'Mild': 1, 'Moderate': 2, 'Severe': 3, 'Very severe, horrible': 4}]\n",
    "\n",
    "f_transform = [('mds-cognitive-scale', 3, 0), ('mds-pain-freq-last-five-days', 4, 0), ('mds-pain-intensity', 4, 0)] + \\\n",
    "              [('Psychotropic-%d-Period' % i, 0, 0) for i in range(1, 6)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Encode features (originally) with no missing values\n",
    "f_drop += [\"predictor-date\", \"projected-date\", \"PPS-assessment\", \"Federal-assessment\"]\n",
    "\n",
    "f_period += [('episode-start-date', 'episode-end-date', 'episode-period'),\\\n",
    "             ('stay-start-date', 'stay-end-date', 'stay-period')]\n",
    "\n",
    "f_label += [\"cms-long-stay\", \"mds-stay-trans-short-to-long\", \"mds-antianxiety-medication\", \\\n",
    "            \"mds-antidepressant-medication\", \"mds-antipsychotic-medication\", \"mds-antibiotic-medication\",\n",
    "            \"mds-anticoagulant-medication\", \"mds-diuretic-medication\", \\\n",
    "            \"mds-hypnotic-medication\", \"mds-behavioral-symptoms\", \"mds-behavioral-symptoms-to-others\", \\\n",
    "            \"mds-dehydrated\", \"mds-depression\", \"mds-fever\", \"mds-impaired-mobility\", \"mds-impaired-transfer\", \\\n",
    "            \"mds-internal-bleeding\", \"mds-malnutrition\", \"mds-no-problem-conditions\", \"mds-vomiting\", \\\n",
    "            \"mds-impaired-walk-in-room\", \"mds-impaired-walk-in-corridor\", \"mds-impaired-locomot-unit\", \\\n",
    "            \"mds-impaired-locomot-other\", \"ws-antibiotic-anticoag-coexposure\", \\\n",
    "            \"ws-psychotropic-with-not-ordered-weight-loss\", \\\n",
    "            \"ws-diuretic-adl\", \"ws-tramadol-antidepressant-coexposure\"] + [\"mds-pain-non-verbal\"]\n",
    "dic_label += [{'Yes': 1, 'No': 0}] * 28 + [{'None/Mild':0,'Moderate/Severe':1}]\n",
    "\n",
    "f_one_hot += [\"race\", \"facility\", \"gender\", \"mds-bims-summary-ranking\", \"ws-meclizine-psych-coexposure\"]\n",
    "cate_one_hot += [['White', 'Black', 'Other'], \\\n",
    "                 ['Sugar Creek', 'Heritage Place', 'Canterbury Place', 'Senaca Place', 'Cranberry Place'], \\\n",
    "                 ['Female', 'Male'], ['Intact or Moderately Intact', 'Moderate Impairment'], \\\n",
    "                 ['No', 'meclAndPsychStartSameR', 'meclPrecedesPsychR', 'psychPrecedesMeclR', 'meclAndPsychOlderStart']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Features need imputation\n",
    "f_one_hot += [\"pneumonia\", \"uti\", \"mdro\", \"anemia\", \"septicemia\", \"constipation\", \"wound\", \"hyponatremia\", \\\n",
    "              \"hyperkalemia\", \"embolisms\", \"alzheimers\", \"anxiety\", \"depression\", \"non-alz-dimentia\", \"bipolar\", \\\n",
    "              \"parkinsons\", \"psychosis\", \"schizophrenia\", \"seizure\", \"aphasia\", \"emphysema\", \"arthritis\", \"ashd\", \\\n",
    "              \"bph\", \"cancer\", \"cerebralpalsy\", \"stroke\", \"cirrhosis\", \"comatose\", \"diabetes\", \"dysrhythmias\", \"gerd\", \\\n",
    "              \"heart-failure\", \"hemiplegia/hemiparesis\", \"hepatitis\", \"huntingtons\", \"hyperlipidemia\", \"hyperthyroidism\", \\\n",
    "              \"hypothyroidism\", \"hypertension\", \"hypotension\", \"multiple-sclerosis\", \"neurogenic-bladder\", \\\n",
    "              \"obstructive-uropathy\", \"osteoporosis\", \"paraplegia\", \"ptsd\", \"pvd\", \"quadriplegia\", \"thyroid-disorder\",\\\n",
    "              \"tourettes\", \"transient-ischemic-attack\", \"traumatic-brain-injury\", \"tuberculosis\", \"renal-failure\",\\\n",
    "              \"mds-delirium-scale\", \"mds-pain-medication\", \"mds-received-pain-tx-non-pharm\", \\\n",
    "              \"mds-received-prn-pain-medication\", \"mds-conduct-staff-assessment-mental-status\", \\\n",
    "              \"mds-pressure-ulcer-prsnt\", \"conduct_bims\", \"acute_mental_change\", \"conduct_pain_assmnt\", \\\n",
    "              \"cane_or_crutch_past_7_days\", \"walker_past_7_days\", \"wheelchair_past_7_days\", \\\n",
    "              \"limb_prosthesis_past_7_days\", \"no_listed_mobility_device\"] + \\\n",
    "              [\"balance_while_standing\", \"balance_while_walking\", \"balance_turning_around\", \\\n",
    "               \"balance_toileting\", \"balance_bed_to_chair\"] + \\\n",
    "              [\"marital-status\", \"mds-entered-from\", \"mds-excess-weight-loss\"]\n",
    "cate_one_hot += [['Yes', 'No']] * 69 + \\\n",
    "                [[\"Steady at all times\", \"able to stabilize without assistance\", \\\n",
    "                  \"able to stabilize with assistance\", \"Activity did not occur\"]] * 5 + \\\n",
    "                [[\"Never Married\", \"Married\", \"Widowed\", \"Separated\", \"Divorced\"], \\\n",
    "                 [\"Acute Hospital\", \"Psychiatric Hospital\", \"Another nursing home or swing bed\", \"Hospice\", \\\n",
    "                  \"Inpatient rehabilitation facility\",\n",
    "                  \"Community (private home/apt., board/care, assisted living, group home)\", \"Other\"],\n",
    "                 [\"Yes - prescribed\", \"Yes - not prescribed\", \"No or unknown\"]]\n",
    "\n",
    "f_label += [\"mds-urinary-incontinence\", \"functlimit_rom_upper\", \"functlimit_rom_lower\", \"does_resident_wander\"] + \\\n",
    "           [\"cam_inattention\", \"cam_disorganized_thought\", \"cam_altered_conc\", \"cam_motor_retardation\"]\n",
    "dic_label += [{\"Always continent\": 0, \"Occasionally incontinent\": 1, \"Frequently incontinent\": 2,\\\n",
    "               \"Always incontinent\": 3, \"Catheter - not rated\": np.nan}] + \\\n",
    "             [{\"No impairment\": 0, \"Impairment on one side\": 1, \"Impairment on both sides\": 2}] * 2 + \\\n",
    "             [{\"Behavior not exhibited\": 0, \"1 to 3 days\": 1, \"4 to 6 days\": 2, \"occurred daily\": 3}] + \\\n",
    "             [{\"Behavior not present\": 0, \"Behavior present, fluctuates\": 1, \"Behavior continuously present\": 2}] * 4\n",
    "\n",
    "f_transform += [(\"mds-urinary-incontinence\", 3, 0), (\"cam_inattention\", 2, 0), (\"cam_disorganized_thought\", 2, 0), \\\n",
    "                (\"cam_altered_conc\", 2, 0), (\"cam_motor_retardation\", 2, 0), (\"functlimit_rom_upper\", 2, 0), \\\n",
    "                (\"functlimit_rom_lower\", 2, 0), (\"does_resident_wander\", 3, 0)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set sdd=0 to NA\n",
    "for i in range(1, 6):\n",
    "    df.loc[df['psych_sdd_%d' % i] == 0, 'psych_sdd_%d' % i] = np.nan\n",
    "# These data were dropped in Experiment 1\n",
    "f_transform += [('psych_sdd_%d' % i, 0, 0) for i in range(1, 6)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping min effective dose and one hot encoding ATC psychotropic drugs\n",
    "# These data were not used in Experiment 1 b/c of missing values\n",
    "f_drop += [\"min_effective_dose_%d\" % i for i in range(1, 6)]\n",
    "\n",
    "atc_list = []\n",
    "for i in range(1, 6):\n",
    "    atc_list.append(df.groupby('atc_%d' % i)['min_effective_dose_%d' % i].first())\n",
    "atc_codes = pd.concat(atc_list)\n",
    "atc_codes = atc_codes[~atc_codes.index.duplicated()].index.tolist()\n",
    "f_one_hot += [\"atc_%d\" % i for i in range(1, 6)]\n",
    "cate_one_hot += [atc_codes] * 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# generate target\n",
    "target = df['outcome-Mds-fall-since-prior-assessment'].map({'None': 0, 'Yes': 1})\n",
    "dff = df.drop(['outcome-mds-fall-no-injury', 'outcome-mds-fall-minor-injury', 'outcome-mds-fall-major-injury', \\\n",
    "               'outcome-Mds-fall-since-prior-assessment', 'outcome-riskmaster-fall-incident'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# train test split\n",
    "from sklearn.model_selection import GroupShuffleSplit\n",
    "def train_test_split(df, target, groupby, radio, random_state):\n",
    "    train_idx, test_idx = next(GroupShuffleSplit(train_size=radio, test_size=1-radio, random_state=random_state).split(df, target, df[groupby]))\n",
    "    return df.iloc[train_idx], df.iloc[test_idx], target.iloc[train_idx], target.iloc[test_idx]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of patients: 3133\n",
      "Number of patients in training set: 2193\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(dff, target, 'patient-id', 0.7, 0)\n",
    "print('Total number of patients: %d' % len(df.groupby('patient-id')))\n",
    "print('Number of patients in training set: %d' % len(X_train.groupby('patient-id')))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop feature\n",
    "def drop_features(df, f):\n",
    "    df.drop(f, axis=1, inplace=True)\n",
    "    \n",
    "# one hot encoding\n",
    "def one_hot_encoding(df, features, categories):\n",
    "    for f, c in zip(features, categories):\n",
    "        df[f] = df[f].astype('category', categories=c)\n",
    "    return pd.get_dummies(df, columns=features, prefix=features)\n",
    "\n",
    "# compute date difference\n",
    "def date_diff(start, end):\n",
    "    return (pd.to_datetime(end) - pd.to_datetime(start)) / np.timedelta64(1, 'D')\n",
    "\n",
    "# generate date diff features\n",
    "def generate_period(df, f):\n",
    "    for start, end, period in f:\n",
    "        df[period] = date_diff(df[start], df[end])\n",
    "        df.drop([start, end], axis=1, inplace=True)\n",
    "        \n",
    "# label encoding\n",
    "def label_encoding(df, features, dictionaries):\n",
    "    for f, dic in zip(features, dictionaries):\n",
    "        df[f] = df[f].map(dic)\n",
    "        \n",
    "# sin&cos transform\n",
    "def col_transform(df, col, mmax, mmin, df_ref=None):\n",
    "    if(mmax <= mmin):\n",
    "        # for test set, if max&min not set, use training set value\n",
    "        if(df_ref is not None): \n",
    "            mmax = df_ref[col].max()\n",
    "            mmin = df_ref[col].max()\n",
    "        else:\n",
    "            mmax = df[col].max()\n",
    "            mmin = df[col].min()\n",
    "    angle = 0.25 * np.pi * (df[col] - mmin) / (mmax - mmin) + 0.125 * np.pi\n",
    "    return (np.cos(angle), np.sin(angle))\n",
    "\n",
    "def df_transform(df, f, df_ref=None):\n",
    "    for col, mmax, mmin in f:\n",
    "        df[col + '_x'], df[col + '_y'] = col_transform(df, col, mmax, mmin, df_ref)\n",
    "        df[col + '_x'].fillna(0, inplace=True)\n",
    "        df[col + '_y'].fillna(0, inplace=True)\n",
    "        df.drop(col, axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1min 23s, sys: 1.46 s, total: 1min 25s\n",
      "Wall time: 27.8 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "## Training set data reshaping\n",
    "generate_period(X_train, f_period)\n",
    "label_encoding(X_train, f_label, dic_label)\n",
    "X_train_original = X_train.copy() # store value range\n",
    "df_transform(X_train, f_transform)\n",
    "X_train = one_hot_encoding(X_train, f_one_hot, cate_one_hot)\n",
    "drop_features(X_train, f_drop)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_use = X_train.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index([], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "print(X_train.dtypes[X_train.dtypes == 'object'].index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>count</th>\n",
       "      <th>percentage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [count, percentage]\n",
       "Index: []"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# missing data describe\n",
    "def missing_summary(df):\n",
    "    missing_cnt = df.isna().sum()\n",
    "    missing_data = pd.concat([missing_cnt, missing_cnt/df.shape[0]], axis=1, keys=['count', 'percentage'])\n",
    "    return missing_data[missing_data['percentage'] != 0].sort_values(by='percentage', ascending=False)\n",
    "\n",
    "missing_summary(X_train_use)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 25.7 s, sys: 60.5 ms, total: 25.8 s\n",
      "Wall time: 25.8 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "## Test set data reshaping\n",
    "generate_period(X_test, f_period)\n",
    "label_encoding(X_test, f_label, dic_label)\n",
    "df_transform(X_test, f_transform, X_train_original)\n",
    "X_test = one_hot_encoding(X_test, f_one_hot, cate_one_hot)\n",
    "drop_features(X_test, f_drop)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_test_use = X_test.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index([], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "print(X_test.dtypes[X_test.dtypes == 'object'].index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>count</th>\n",
       "      <th>percentage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [count, percentage]\n",
       "Index: []"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "missing_summary(X_test_use)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5301, 524)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_use.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2283, 524)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test_use.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>patient-id</th>\n",
       "      <th>current-stay-days</th>\n",
       "      <th>cumulative-days-in-facility</th>\n",
       "      <th>cms-long-stay</th>\n",
       "      <th>mds-stay-trans-short-to-long</th>\n",
       "      <th>age</th>\n",
       "      <th>mds-antianxiety-medication</th>\n",
       "      <th>mds-antidepressant-medication</th>\n",
       "      <th>mds-antipsychotic-medication</th>\n",
       "      <th>mds-antibiotic-medication</th>\n",
       "      <th>...</th>\n",
       "      <th>atc_5_N06AB10</th>\n",
       "      <th>atc_5_N06AX05</th>\n",
       "      <th>atc_5_N06AX11</th>\n",
       "      <th>atc_5_N06AX12</th>\n",
       "      <th>atc_5_N06AX16</th>\n",
       "      <th>atc_5_N06AX21</th>\n",
       "      <th>atc_5_N06AX24</th>\n",
       "      <th>atc_5_N05AA01</th>\n",
       "      <th>atc_5_N05BA05</th>\n",
       "      <th>atc_5_N06AX23</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>10353</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>67</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>10364</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>84</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>10386</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>77</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5000</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>5003</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 524 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   patient-id  current-stay-days  cumulative-days-in-facility  cms-long-stay  \\\n",
       "0       10353                  7                            7              0   \n",
       "1       10364                  7                            7              0   \n",
       "4       10386                  7                            7              0   \n",
       "5        5000                  7                            7              0   \n",
       "6        5003                  7                            7              0   \n",
       "\n",
       "   mds-stay-trans-short-to-long  age  mds-antianxiety-medication  \\\n",
       "0                             0   67                           1   \n",
       "1                             1   84                           0   \n",
       "4                             1   77                           1   \n",
       "5                             0   90                           0   \n",
       "6                             1   90                           0   \n",
       "\n",
       "   mds-antidepressant-medication  mds-antipsychotic-medication  \\\n",
       "0                              1                             0   \n",
       "1                              0                             1   \n",
       "4                              1                             1   \n",
       "5                              1                             1   \n",
       "6                              1                             1   \n",
       "\n",
       "   mds-antibiotic-medication  ...  atc_5_N06AB10  atc_5_N06AX05  \\\n",
       "0                          0  ...              0              0   \n",
       "1                          1  ...              0              0   \n",
       "4                          0  ...              0              0   \n",
       "5                          0  ...              0              0   \n",
       "6                          1  ...              0              0   \n",
       "\n",
       "   atc_5_N06AX11  atc_5_N06AX12  atc_5_N06AX16  atc_5_N06AX21  atc_5_N06AX24  \\\n",
       "0              0              0              0              0              0   \n",
       "1              0              0              0              0              0   \n",
       "4              0              0              0              0              0   \n",
       "5              0              0              0              0              0   \n",
       "6              0              0              0              0              0   \n",
       "\n",
       "   atc_5_N05AA01  atc_5_N05BA05  atc_5_N06AX23  \n",
       "0              0              0              0  \n",
       "1              0              0              0  \n",
       "4              0              0              0  \n",
       "5              0              0              0  \n",
       "6              0              0              0  \n",
       "\n",
       "[5 rows x 524 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_use.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_use.to_csv (r'/tmp/export_train_dataframe.csv', index = False, header=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>patient-id</th>\n",
       "      <th>current-stay-days</th>\n",
       "      <th>cumulative-days-in-facility</th>\n",
       "      <th>cms-long-stay</th>\n",
       "      <th>mds-stay-trans-short-to-long</th>\n",
       "      <th>age</th>\n",
       "      <th>mds-antianxiety-medication</th>\n",
       "      <th>mds-antidepressant-medication</th>\n",
       "      <th>mds-antipsychotic-medication</th>\n",
       "      <th>mds-antibiotic-medication</th>\n",
       "      <th>...</th>\n",
       "      <th>atc_5_N06AB10</th>\n",
       "      <th>atc_5_N06AX05</th>\n",
       "      <th>atc_5_N06AX11</th>\n",
       "      <th>atc_5_N06AX12</th>\n",
       "      <th>atc_5_N06AX16</th>\n",
       "      <th>atc_5_N06AX21</th>\n",
       "      <th>atc_5_N06AX24</th>\n",
       "      <th>atc_5_N05AA01</th>\n",
       "      <th>atc_5_N05BA05</th>\n",
       "      <th>atc_5_N06AX23</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>10379</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>84</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>10385</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>85</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>5032</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>85</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>5032</td>\n",
       "      <td>8</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>86</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>5032</td>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>86</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 524 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    patient-id  current-stay-days  cumulative-days-in-facility  cms-long-stay  \\\n",
       "2        10379                  7                            7              0   \n",
       "3        10385                  7                            7              0   \n",
       "14        5032                  7                            7              0   \n",
       "15        5032                  8                            8              0   \n",
       "16        5032                  7                            7              0   \n",
       "\n",
       "    mds-stay-trans-short-to-long  age  mds-antianxiety-medication  \\\n",
       "2                              0   84                           0   \n",
       "3                              1   85                           0   \n",
       "14                             0   85                           1   \n",
       "15                             0   86                           1   \n",
       "16                             0   86                           1   \n",
       "\n",
       "    mds-antidepressant-medication  mds-antipsychotic-medication  \\\n",
       "2                               1                             0   \n",
       "3                               1                             0   \n",
       "14                              1                             0   \n",
       "15                              1                             0   \n",
       "16                              1                             0   \n",
       "\n",
       "    mds-antibiotic-medication  ...  atc_5_N06AB10  atc_5_N06AX05  \\\n",
       "2                           0  ...              0              0   \n",
       "3                           1  ...              0              0   \n",
       "14                          1  ...              0              0   \n",
       "15                          1  ...              0              0   \n",
       "16                          1  ...              0              0   \n",
       "\n",
       "    atc_5_N06AX11  atc_5_N06AX12  atc_5_N06AX16  atc_5_N06AX21  atc_5_N06AX24  \\\n",
       "2               0              0              0              0              0   \n",
       "3               0              0              0              0              0   \n",
       "14              0              0              0              0              0   \n",
       "15              0              0              0              0              0   \n",
       "16              0              0              0              0              0   \n",
       "\n",
       "    atc_5_N05AA01  atc_5_N05BA05  atc_5_N06AX23  \n",
       "2               0              0              0  \n",
       "3               0              0              0  \n",
       "14              0              0              0  \n",
       "15              0              0              0  \n",
       "16              0              0              0  \n",
       "\n",
       "[5 rows x 524 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test_use.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The main difference in this experiment from experiment 1 \n",
    "# is that values are imputed so there are no missing values to drop. \n",
    "# This adds a bunch of features\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "from keras.models import Sequential\n",
    "from keras.layers import LSTM, GRU, Dense, Dropout, Masking, TimeDistributed\n",
    "from numpy.random import seed\n",
    "from tensorflow import set_random_seed\n",
    "from copy import deepcopy\n",
    "from keras.preprocessing.sequence import pad_sequences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "def scaler_transform(X_train_use, X_test_use):    \n",
    "    scaler = MinMaxScaler(feature_range=(0, 1)).fit(X_train_use.drop('patient-id', axis=1))\n",
    "    X_train_transformed = pd.DataFrame(scaler.transform(X_train_use.drop('patient-id', axis=1)), \\\n",
    "                                       columns=X_train_use.columns[1:], index=X_train_use.index)                                   \n",
    "    X_train_transformed['patient-id'] = X_train_use['patient-id']\n",
    "    X_test_transformed = pd.DataFrame(scaler.transform(X_test_use.drop('patient-id', axis=1)), \\\n",
    "                                       columns=X_test_use.columns[1:], index=X_test_use.index) \n",
    "    X_test_transformed['patient-id'] = X_test_use['patient-id']\n",
    "    return X_train_transformed, X_test_transformed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "def add_previous_outcome(X, Y):\n",
    "    X = X.copy()\n",
    "    x1, x2 = [], []\n",
    "    pids = X['patient-id'].unique()\n",
    "    for pid in pids:\n",
    "        df = Y[X[X['patient-id']==pid].index]\n",
    "        if(df.shape[0] == 1):\n",
    "            x1.append(pd.Series([0]))\n",
    "            x2.append(pd.Series([0]))\n",
    "        else:\n",
    "            # input 1=(1, 0), 0=(0, 1), NA=(0, 0)\n",
    "            x1.append(pd.Series([0]))\n",
    "            x1.append(df[:-1])\n",
    "            x2.append(pd.Series([0]))\n",
    "            x2.append(df[:-1].map({0: 1, 1: 0}))\n",
    "    X['input_outcome_x'] = pd.Series(pd.concat(x1, ignore_index=True).tolist(), index=X.index)\n",
    "    X['input_outcome_y'] = pd.Series(pd.concat(x2, ignore_index=True).tolist(), index=X.index)\n",
    "    return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "## This step converts the outcome to a point in x,y coordinates where fall (1) = (1,0); no fall (0) = (0,1); and NA = (0,0) \n",
    "X_train_use = add_previous_outcome(X_train_use, y_train)\n",
    "X_test_use = add_previous_outcome(X_test_use, y_test)\n",
    "\n",
    "## This step transforms the data to range between 0 and 1\n",
    "X_train_transformed, X_test_transformed = scaler_transform(X_train_use, X_test_use)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_nested_list(X, Y):\n",
    "    x_list, y_list = [], [] \n",
    "    pids = X['patient-id'].unique()\n",
    "    for pid in pids:\n",
    "        df = X[X['patient-id']==pid]\n",
    "        y_list.append(Y[df.index].values)\n",
    "        x_list.append([record for record in df.drop('patient-id', axis=1).values])\n",
    "    return x_list, y_list\n",
    "\n",
    "def mygenerator(x_list, y_list=None):\n",
    "    if(y_list is not None):\n",
    "        while True:\n",
    "            for x, y in zip(x_list, y_list):\n",
    "                yield np.array(x).reshape((len(x), 1, x[0].shape[0])), y\n",
    "    else:\n",
    "        while True:\n",
    "            for x in x_list:\n",
    "                yield np.array(x).reshape((len(x), 1, x[0].shape[0])) \n",
    "                \n",
    "def fit_model(model, epoch, X_train, y_train, X_valid=None, y_valid=None, verbose=0):\n",
    "    history = model.fit(X_train, y_train, epochs=epoch, verbose=verbose)\n",
    "    if(verbose != 0):\n",
    "        plot_history(history)\n",
    "    return history\n",
    "\n",
    "def plot_roc_curve(fpr, tpr):\n",
    "    plt.figure()\n",
    "    plt.plot(fpr, tpr, color='darkorange', label='ROC curve (area = %0.2f)' % auc(fpr, tpr))\n",
    "    plt.plot([0, 1], [0, 1], color='navy', linestyle='--')\n",
    "    plt.xlim([0.0, 1.0])\n",
    "    plt.ylim([0.0, 1.05])\n",
    "    plt.xlabel('False Positive Rate')\n",
    "    plt.ylabel('True Positive Rate')\n",
    "    plt.legend(loc=\"lower right\")\n",
    "    plt.show()\n",
    "    \n",
    "def plot_history(history, loss=None, val_loss=None):\n",
    "    if(history is not None):\n",
    "        loss = history.history['loss']\n",
    "    plt.figure()\n",
    "    plt.plot(loss)\n",
    "    plt.title('model loss')\n",
    "    plt.ylabel('loss')\n",
    "    plt.xlabel('epoch')\n",
    "    if('val_loss' in history.history):\n",
    "        val_loss = history.history['val_loss']\n",
    "        plt.plot(val_loss)\n",
    "    plt.legend(['train', 'test'], loc='upper right')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pad_data(X, maxlen):\n",
    "    return pad_sequences(X, maxlen=maxlen, dtype='float64', padding='post')\n",
    "\n",
    "def pad_all(X_train, y_train, X_test, y_test, maxlen):\n",
    "    return pad_data(X_train, maxlen), pad_data(y_train, maxlen), pad_data(X_test, maxlen), pad_data(y_test, maxlen)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def auroc(y_predict, y_true, plot=False):\n",
    "    fpr, tpr, thresholds = roc_curve(y_true, y_predict, pos_label=1)    \n",
    "    if(plot):\n",
    "        plot_roc_curve(fpr, tpr)\n",
    "    return auc(fpr, tpr)\n",
    "\n",
    "\n",
    "def find_best_f1(y_predict, y_true):\n",
    "    fpr, tpr, thresholds = roc_curve(y_true, y_predict, pos_label=1) \n",
    "    max_f1, th = 0, 0\n",
    "    for threshold in thresholds:\n",
    "        f1 = f1_score(y_true, y_predict > threshold)\n",
    "        if(f1 > max_f1):\n",
    "            max_f1 = f1\n",
    "            th = threshold\n",
    "    return max_f1, th\n",
    "\n",
    "def other_metrics(y_predict, y_true, threshold):\n",
    "    y_threshold = y_predict > threshold\n",
    "    tn, fp, fn, tp = confusion_matrix(y_true, y_threshold).ravel()\n",
    "    return precision_score(y_true, y_threshold), recall_score(y_true, y_threshold),\\\n",
    "            accuracy_score(y_true, y_threshold), float(tn) / (tn + fp)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GroupKFold\n",
    "def kfold_validation(unit, ratio, epoch, X, y, groupby, n_features, k=5, verbose=0, max_timestamps=20):\n",
    "    auroc_list = []\n",
    "    other_metric_list = []\n",
    "    for train_idx, valid_idx in GroupKFold(n_splits=5).split(X, y, X[groupby]):\n",
    "        X_train, X_valid, y_train, y_valid = X.iloc[train_idx], X.iloc[valid_idx], y.iloc[train_idx], y.iloc[valid_idx]\n",
    "        X_train, y_train = generate_nested_list(X_train, y_train)\n",
    "        X_valid, y_valid = generate_nested_list(X_valid, y_valid)\n",
    "        X_train_pad, y_train_pad, X_valid_pad, y_valid_pad = pad_all(X_train, y_train, X_valid, y_valid, max_timestamps)\n",
    "        y_train_pad = y_train_pad.reshape(y_train_pad.shape[0], y_train_pad.shape[1], 1)\n",
    "        seed(1)\n",
    "        set_random_seed(2)\n",
    "        model = build_model(unit, ratio, max_timestamps, n_features)\n",
    "        fit_model(model, epoch, X_train_pad, y_train_pad, verbose=verbose)\n",
    "        y_valid_len = [len(y) for y in y_valid]\n",
    "        y_predict = model.predict(X_valid_pad)\n",
    "        y_predict = truncate_prediction(y_predict, y_valid_len)\n",
    "        y_true = np.concatenate(y_valid)\n",
    "        auroc_list.append(auroc(y_predict, y_true))\n",
    "        f1, th = find_best_f1(y_predict, y_true)\n",
    "        precision, recall, accuracy, specificity = other_metrics(y_predict, y_true, th)\n",
    "        other_metric_list.append((precision, recall, accuracy, specificity,f1,th))\n",
    "    return (auroc_list, other_metric_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# shape: (num_of_patients, num_of_records, num_of_features)\n",
    "X_train_list, y_train_list = generate_nested_list(X_train_transformed, y_train)\n",
    "X_test_list, y_test_list = generate_nested_list(X_test_transformed, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "## comparing this to the experiment 1 transformed dataset to confirm that this has \n",
    "## diagnoses and ATC codes while the other does not\n",
    "X_train_transformed.to_csv (r'/tmp/exp2_train_transformed_df.csv', index = False, header=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5301, 526)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_transformed.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2283, 526)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test_transformed.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19\n"
     ]
    }
   ],
   "source": [
    "max_timestamps = df.groupby(['patient-id']).size().max()\n",
    "print(max_timestamps)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "525\n"
     ]
    }
   ],
   "source": [
    "n_features = X_train_list[0][0].shape[0]\n",
    "print(n_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "## function to build an LSTM\n",
    "from keras import regularizers\n",
    "from keras.optimizers import Adagrad, Adadelta\n",
    "def build_model(units, dropout_ratio, n_timesteps, n_dimensions):\n",
    "    model = Sequential()\n",
    "    model.add(Masking(input_shape=(n_timesteps, n_dimensions)))\n",
    "    model.add(GRU(units, return_sequences=True, recurrent_dropout=dropout_ratio, dropout=0.5))\n",
    "    model.add(TimeDistributed(Dense(1, activation='sigmoid')))\n",
    "    model.compile(loss='binary_crossentropy', optimizer='adagrad', metrics=['binary_accuracy'])\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "def truncate_prediction(y_, y_len):\n",
    "    y_concat = []\n",
    "    for i, y in enumerate(y_):\n",
    "        y_concat.append(np.concatenate(y[:y_len[i], :]))\n",
    "    y_concat = np.concatenate(y_concat)\n",
    "    return y_concat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "## the data is padded so that the final shape is as follows with the outcome data array having the same\n",
    "## shape but only one value\n",
    "#  array([[patient 1 data record 1],  \n",
    "#         [patient 1 data record 2],\n",
    "#         ...\n",
    "#         [patient 1 data record 19],\n",
    "#        ],\n",
    "#        [[patient 2 data record 1],  \n",
    "#         [patient 2 data record 2],\n",
    "#         ...\n",
    "#         [patient 2 data record 19],\n",
    "#        ],\n",
    "#        ...\n",
    "#       )\n",
    "# \n",
    "X_train_pad, y_train_pad, X_test_pad, y_test_pad = pad_all(X_train_list, y_train_list, X_test_list, y_test_list, max_timestamps)\n",
    "y_train_pad = y_train_pad.reshape((2193, 19, 1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Logging before flag parsing goes to stderr.\n",
      "W0423 09:50:02.462745 140353037133632 deprecation_wrapper.py:119] From /usr/local/anaconda3/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:74: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.\n",
      "\n",
      "W0423 09:50:02.465331 140353037133632 deprecation_wrapper.py:119] From /usr/local/anaconda3/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:517: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.\n",
      "\n",
      "W0423 09:50:02.479107 140353037133632 deprecation_wrapper.py:119] From /usr/local/anaconda3/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:4138: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.\n",
      "\n",
      "W0423 09:50:02.568729 140353037133632 deprecation_wrapper.py:119] From /usr/local/anaconda3/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:133: The name tf.placeholder_with_default is deprecated. Please use tf.compat.v1.placeholder_with_default instead.\n",
      "\n",
      "W0423 09:50:02.578809 140353037133632 deprecation.py:506] From /usr/local/anaconda3/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
      "W0423 09:50:03.071312 140353037133632 deprecation.py:323] From /usr/local/anaconda3/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:2974: add_dispatch_support.<locals>.wrapper (from tensorflow.python.ops.array_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.where in 2.0, which has the same broadcast rule as np.where\n",
      "W0423 09:50:03.134570 140353037133632 deprecation_wrapper.py:119] From /usr/local/anaconda3/lib/python3.7/site-packages/keras/optimizers.py:790: The name tf.train.Optimizer is deprecated. Please use tf.compat.v1.train.Optimizer instead.\n",
      "\n",
      "W0423 09:50:03.152289 140353037133632 deprecation_wrapper.py:119] From /usr/local/anaconda3/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:3376: The name tf.log is deprecated. Please use tf.math.log instead.\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/175\n",
      "2193/2193 [==============================] - 4s 2ms/step - loss: 0.4620 - binary_accuracy: 0.8175\n",
      "Epoch 2/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4344 - binary_accuracy: 0.8207\n",
      "Epoch 3/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4194 - binary_accuracy: 0.8257\n",
      "Epoch 4/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4211 - binary_accuracy: 0.8245\n",
      "Epoch 5/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4201 - binary_accuracy: 0.8232\n",
      "Epoch 6/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.4166 - binary_accuracy: 0.8230\n",
      "Epoch 7/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4151 - binary_accuracy: 0.8244\n",
      "Epoch 8/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4130 - binary_accuracy: 0.8270\n",
      "Epoch 9/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.4061 - binary_accuracy: 0.8267\n",
      "Epoch 10/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.4111 - binary_accuracy: 0.8265\n",
      "Epoch 11/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4065 - binary_accuracy: 0.8279\n",
      "Epoch 12/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4068 - binary_accuracy: 0.8300\n",
      "Epoch 13/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.4002 - binary_accuracy: 0.8333\n",
      "Epoch 14/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.4050 - binary_accuracy: 0.8312\n",
      "Epoch 15/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.4028 - binary_accuracy: 0.8294\n",
      "Epoch 16/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3988 - binary_accuracy: 0.8330\n",
      "Epoch 17/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4018 - binary_accuracy: 0.8316\n",
      "Epoch 18/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4011 - binary_accuracy: 0.8329\n",
      "Epoch 19/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4018 - binary_accuracy: 0.8315\n",
      "Epoch 20/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3994 - binary_accuracy: 0.8302\n",
      "Epoch 21/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3989 - binary_accuracy: 0.8330\n",
      "Epoch 22/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.4040 - binary_accuracy: 0.8280\n",
      "Epoch 23/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3957 - binary_accuracy: 0.8338\n",
      "Epoch 24/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.4027 - binary_accuracy: 0.8312\n",
      "Epoch 25/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3964 - binary_accuracy: 0.8333\n",
      "Epoch 26/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3922 - binary_accuracy: 0.8365\n",
      "Epoch 27/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3906 - binary_accuracy: 0.8364\n",
      "Epoch 28/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3972 - binary_accuracy: 0.8321\n",
      "Epoch 29/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3961 - binary_accuracy: 0.8302\n",
      "Epoch 30/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3968 - binary_accuracy: 0.8342\n",
      "Epoch 31/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3879 - binary_accuracy: 0.8377\n",
      "Epoch 32/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3875 - binary_accuracy: 0.8341\n",
      "Epoch 33/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3971 - binary_accuracy: 0.8311\n",
      "Epoch 34/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3946 - binary_accuracy: 0.8334\n",
      "Epoch 35/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3909 - binary_accuracy: 0.8357\n",
      "Epoch 36/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3890 - binary_accuracy: 0.8373\n",
      "Epoch 37/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3865 - binary_accuracy: 0.8362\n",
      "Epoch 38/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3893 - binary_accuracy: 0.8319\n",
      "Epoch 39/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3907 - binary_accuracy: 0.8325\n",
      "Epoch 40/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3910 - binary_accuracy: 0.8373\n",
      "Epoch 41/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3906 - binary_accuracy: 0.8359\n",
      "Epoch 42/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3884 - binary_accuracy: 0.8380\n",
      "Epoch 43/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3904 - binary_accuracy: 0.8344\n",
      "Epoch 44/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3836 - binary_accuracy: 0.8374\n",
      "Epoch 45/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3874 - binary_accuracy: 0.8348\n",
      "Epoch 46/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3900 - binary_accuracy: 0.8278\n",
      "Epoch 47/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3855 - binary_accuracy: 0.8397\n",
      "Epoch 48/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3831 - binary_accuracy: 0.8417\n",
      "Epoch 49/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3824 - binary_accuracy: 0.8386\n",
      "Epoch 50/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3889 - binary_accuracy: 0.8338\n",
      "Epoch 51/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3845 - binary_accuracy: 0.8409\n",
      "Epoch 52/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3901 - binary_accuracy: 0.8336\n",
      "Epoch 53/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3831 - binary_accuracy: 0.8407\n",
      "Epoch 54/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3883 - binary_accuracy: 0.8323\n",
      "Epoch 55/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3909 - binary_accuracy: 0.8323\n",
      "Epoch 56/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3885 - binary_accuracy: 0.8327\n",
      "Epoch 57/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3829 - binary_accuracy: 0.8373\n",
      "Epoch 58/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3903 - binary_accuracy: 0.8340\n",
      "Epoch 59/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3853 - binary_accuracy: 0.8348\n",
      "Epoch 60/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3859 - binary_accuracy: 0.8403\n",
      "Epoch 61/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3855 - binary_accuracy: 0.8375\n",
      "Epoch 62/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3811 - binary_accuracy: 0.8396\n",
      "Epoch 63/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3824 - binary_accuracy: 0.8355\n",
      "Epoch 64/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3796 - binary_accuracy: 0.8379\n",
      "Epoch 65/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3780 - binary_accuracy: 0.8374\n",
      "Epoch 66/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3831 - binary_accuracy: 0.8361\n",
      "Epoch 67/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3840 - binary_accuracy: 0.8390\n",
      "Epoch 68/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3841 - binary_accuracy: 0.8347\n",
      "Epoch 69/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3775 - binary_accuracy: 0.8408\n",
      "Epoch 70/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3780 - binary_accuracy: 0.8376\n",
      "Epoch 71/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3785 - binary_accuracy: 0.8394\n",
      "Epoch 72/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3813 - binary_accuracy: 0.8361\n",
      "Epoch 73/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3819 - binary_accuracy: 0.8417\n",
      "Epoch 74/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3748 - binary_accuracy: 0.8442\n",
      "Epoch 75/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3757 - binary_accuracy: 0.8444\n",
      "Epoch 76/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3812 - binary_accuracy: 0.8373\n",
      "Epoch 77/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3782 - binary_accuracy: 0.8399\n",
      "Epoch 78/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3728 - binary_accuracy: 0.8413\n",
      "Epoch 79/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3775 - binary_accuracy: 0.8381\n",
      "Epoch 80/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3760 - binary_accuracy: 0.8400\n",
      "Epoch 81/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3776 - binary_accuracy: 0.8388\n",
      "Epoch 82/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3766 - binary_accuracy: 0.8399\n",
      "Epoch 83/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3791 - binary_accuracy: 0.8362\n",
      "Epoch 84/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3735 - binary_accuracy: 0.8400\n",
      "Epoch 85/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3760 - binary_accuracy: 0.8381\n",
      "Epoch 86/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3778 - binary_accuracy: 0.8402\n",
      "Epoch 87/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3745 - binary_accuracy: 0.8427\n",
      "Epoch 88/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3725 - binary_accuracy: 0.8432\n",
      "Epoch 89/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3761 - binary_accuracy: 0.8393\n",
      "Epoch 90/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3749 - binary_accuracy: 0.8396\n",
      "Epoch 91/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3746 - binary_accuracy: 0.8443\n",
      "Epoch 92/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3780 - binary_accuracy: 0.8386\n",
      "Epoch 93/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3696 - binary_accuracy: 0.8457\n",
      "Epoch 94/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3758 - binary_accuracy: 0.8413\n",
      "Epoch 95/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3777 - binary_accuracy: 0.8390\n",
      "Epoch 96/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3756 - binary_accuracy: 0.8360\n",
      "Epoch 97/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3701 - binary_accuracy: 0.8419\n",
      "Epoch 98/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3747 - binary_accuracy: 0.8419\n",
      "Epoch 99/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3731 - binary_accuracy: 0.8424\n",
      "Epoch 100/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3680 - binary_accuracy: 0.8434\n",
      "Epoch 101/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3673 - binary_accuracy: 0.8452\n",
      "Epoch 102/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3707 - binary_accuracy: 0.8425\n",
      "Epoch 103/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3652 - binary_accuracy: 0.8445\n",
      "Epoch 104/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3706 - binary_accuracy: 0.8443\n",
      "Epoch 105/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3671 - binary_accuracy: 0.8409\n",
      "Epoch 106/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3740 - binary_accuracy: 0.8435\n",
      "Epoch 107/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3751 - binary_accuracy: 0.8382\n",
      "Epoch 108/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3704 - binary_accuracy: 0.8437\n",
      "Epoch 109/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3742 - binary_accuracy: 0.8442\n",
      "Epoch 110/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3690 - binary_accuracy: 0.8418\n",
      "Epoch 111/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3688 - binary_accuracy: 0.8443\n",
      "Epoch 112/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3723 - binary_accuracy: 0.8411\n",
      "Epoch 113/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3651 - binary_accuracy: 0.8439\n",
      "Epoch 114/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3761 - binary_accuracy: 0.8377\n",
      "Epoch 115/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3726 - binary_accuracy: 0.8415\n",
      "Epoch 116/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3621 - binary_accuracy: 0.8458\n",
      "Epoch 117/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3654 - binary_accuracy: 0.8487\n",
      "Epoch 118/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3646 - binary_accuracy: 0.8444\n",
      "Epoch 119/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3702 - binary_accuracy: 0.8413\n",
      "Epoch 120/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3651 - binary_accuracy: 0.8449\n",
      "Epoch 121/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3609 - binary_accuracy: 0.8471\n",
      "Epoch 122/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3685 - binary_accuracy: 0.8405\n",
      "Epoch 123/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3696 - binary_accuracy: 0.8355\n",
      "Epoch 124/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3659 - binary_accuracy: 0.8422\n",
      "Epoch 125/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3666 - binary_accuracy: 0.8428\n",
      "Epoch 126/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3682 - binary_accuracy: 0.8468\n",
      "Epoch 127/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3617 - binary_accuracy: 0.8471\n",
      "Epoch 128/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3648 - binary_accuracy: 0.8438\n",
      "Epoch 129/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3664 - binary_accuracy: 0.8433\n",
      "Epoch 130/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3647 - binary_accuracy: 0.8471\n",
      "Epoch 131/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3618 - binary_accuracy: 0.8480\n",
      "Epoch 132/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3558 - binary_accuracy: 0.8462\n",
      "Epoch 133/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3614 - binary_accuracy: 0.8468\n",
      "Epoch 134/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3579 - binary_accuracy: 0.8472\n",
      "Epoch 135/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3589 - binary_accuracy: 0.8444\n",
      "Epoch 136/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3632 - binary_accuracy: 0.8470\n",
      "Epoch 137/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3639 - binary_accuracy: 0.8447\n",
      "Epoch 138/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3640 - binary_accuracy: 0.8448\n",
      "Epoch 139/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3652 - binary_accuracy: 0.8464\n",
      "Epoch 140/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3599 - binary_accuracy: 0.8484\n",
      "Epoch 141/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3633 - binary_accuracy: 0.8434\n",
      "Epoch 142/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3633 - binary_accuracy: 0.8463\n",
      "Epoch 143/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3600 - binary_accuracy: 0.8416\n",
      "Epoch 144/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3647 - binary_accuracy: 0.8443\n",
      "Epoch 145/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3631 - binary_accuracy: 0.8486\n",
      "Epoch 146/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3575 - binary_accuracy: 0.8492\n",
      "Epoch 147/175\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3659 - binary_accuracy: 0.8462\n",
      "Epoch 148/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3611 - binary_accuracy: 0.8475\n",
      "Epoch 149/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3569 - binary_accuracy: 0.8467\n",
      "Epoch 150/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3585 - binary_accuracy: 0.8495\n",
      "Epoch 151/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3611 - binary_accuracy: 0.8462\n",
      "Epoch 152/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3555 - binary_accuracy: 0.8508\n",
      "Epoch 153/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3618 - binary_accuracy: 0.8443\n",
      "Epoch 154/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3607 - binary_accuracy: 0.8448\n",
      "Epoch 155/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3589 - binary_accuracy: 0.8485\n",
      "Epoch 156/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3595 - binary_accuracy: 0.8478\n",
      "Epoch 157/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3549 - binary_accuracy: 0.8513\n",
      "Epoch 158/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3599 - binary_accuracy: 0.8420\n",
      "Epoch 159/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3597 - binary_accuracy: 0.8475\n",
      "Epoch 160/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3553 - binary_accuracy: 0.8462\n",
      "Epoch 161/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3568 - binary_accuracy: 0.8502\n",
      "Epoch 162/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3634 - binary_accuracy: 0.8402\n",
      "Epoch 163/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3572 - binary_accuracy: 0.8466\n",
      "Epoch 164/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3554 - binary_accuracy: 0.8514\n",
      "Epoch 165/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3597 - binary_accuracy: 0.8417\n",
      "Epoch 166/175\n",
      "2193/2193 [==============================] - 3s 1ms/step - loss: 0.3598 - binary_accuracy: 0.8468\n",
      "Epoch 167/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3560 - binary_accuracy: 0.8501\n",
      "Epoch 168/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3567 - binary_accuracy: 0.8494\n",
      "Epoch 169/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3579 - binary_accuracy: 0.8475\n",
      "Epoch 170/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3616 - binary_accuracy: 0.8468\n",
      "Epoch 171/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3573 - binary_accuracy: 0.8444\n",
      "Epoch 172/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3511 - binary_accuracy: 0.8520\n",
      "Epoch 173/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3576 - binary_accuracy: 0.8498\n",
      "Epoch 174/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3539 - binary_accuracy: 0.8518\n",
      "Epoch 175/175\n",
      "2193/2193 [==============================] - 2s 1ms/step - loss: 0.3528 - binary_accuracy: 0.8495\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "auroc:  0.7031970538667999\n",
      "f1:  0.4170073589533933\n",
      "threshold:  0.15926597\n",
      "precision:  0.3227848101265823\n",
      "recall:  0.5889145496535797\n",
      "accuracy:  0.6876916338151555\n",
      "specificity:  0.7108108108108108\n",
      "CPU times: user 28min 28s, sys: 5min 23s, total: 33min 51s\n",
      "Wall time: 7min 24s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "### A single run w/out cross-validation\n",
    "# with previous outcome added to the features\n",
    "# outcome is added before scaling\n",
    "seed(1)\n",
    "set_random_seed(2)\n",
    "model = build_model(32, 0.5, max_timestamps, n_features)\n",
    "fit_model(model, 175, X_train_pad, y_train_pad, verbose=1)\n",
    "y_predict = model.predict(X_test_pad)\n",
    "y_test_len = [len(y) for y in y_test_list]\n",
    "y_predict = truncate_prediction(y_predict, y_test_len)\n",
    "y_true = np.concatenate(y_test_list)\n",
    "print('auroc: ', auroc(y_predict, y_true))\n",
    "f1, th = find_best_f1(y_predict, y_true)\n",
    "print('f1: ', f1);\n",
    "print('threshold: ', th)\n",
    "precision, recall, accuracy, specificity = other_metrics(y_predict, y_true, th)\n",
    "print('precision: ', precision)\n",
    "print('recall: ', recall)\n",
    "print('accuracy: ', accuracy)\n",
    "print('specificity: ', specificity)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/225\n",
      "1755/1755 [==============================] - 4s 2ms/step - loss: 0.4657 - binary_accuracy: 0.8153\n",
      "Epoch 2/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4331 - binary_accuracy: 0.8225\n",
      "Epoch 3/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4284 - binary_accuracy: 0.8203\n",
      "Epoch 4/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4163 - binary_accuracy: 0.8264\n",
      "Epoch 5/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4201 - binary_accuracy: 0.8221\n",
      "Epoch 6/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4187 - binary_accuracy: 0.8250\n",
      "Epoch 7/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4163 - binary_accuracy: 0.8243\n",
      "Epoch 8/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4092 - binary_accuracy: 0.8273\n",
      "Epoch 9/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4128 - binary_accuracy: 0.8223\n",
      "Epoch 10/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4042 - binary_accuracy: 0.8242\n",
      "Epoch 11/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4066 - binary_accuracy: 0.8309\n",
      "Epoch 12/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4084 - binary_accuracy: 0.8228\n",
      "Epoch 13/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4081 - binary_accuracy: 0.8265\n",
      "Epoch 14/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4118 - binary_accuracy: 0.8242\n",
      "Epoch 15/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4031 - binary_accuracy: 0.8322\n",
      "Epoch 16/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4000 - binary_accuracy: 0.8305\n",
      "Epoch 17/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3953 - binary_accuracy: 0.8327\n",
      "Epoch 18/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4014 - binary_accuracy: 0.8310\n",
      "Epoch 19/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4021 - binary_accuracy: 0.8302\n",
      "Epoch 20/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4022 - binary_accuracy: 0.8357\n",
      "Epoch 21/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4013 - binary_accuracy: 0.8284\n",
      "Epoch 22/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3986 - binary_accuracy: 0.8309\n",
      "Epoch 23/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3967 - binary_accuracy: 0.8334\n",
      "Epoch 24/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3934 - binary_accuracy: 0.8372\n",
      "Epoch 25/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3990 - binary_accuracy: 0.8297\n",
      "Epoch 26/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3972 - binary_accuracy: 0.8303\n",
      "Epoch 27/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3925 - binary_accuracy: 0.8342\n",
      "Epoch 28/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3941 - binary_accuracy: 0.8332\n",
      "Epoch 29/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3884 - binary_accuracy: 0.8361\n",
      "Epoch 30/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3926 - binary_accuracy: 0.8292\n",
      "Epoch 31/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3935 - binary_accuracy: 0.8384\n",
      "Epoch 32/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3975 - binary_accuracy: 0.8309\n",
      "Epoch 33/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3921 - binary_accuracy: 0.8350\n",
      "Epoch 34/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3895 - binary_accuracy: 0.8356\n",
      "Epoch 35/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3917 - binary_accuracy: 0.8381\n",
      "Epoch 36/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3857 - binary_accuracy: 0.8386\n",
      "Epoch 37/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3954 - binary_accuracy: 0.8355\n",
      "Epoch 38/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3910 - binary_accuracy: 0.8357\n",
      "Epoch 39/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3804 - binary_accuracy: 0.8382\n",
      "Epoch 40/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3940 - binary_accuracy: 0.8334\n",
      "Epoch 41/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3902 - binary_accuracy: 0.8342\n",
      "Epoch 42/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3880 - binary_accuracy: 0.8366\n",
      "Epoch 43/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3935 - binary_accuracy: 0.8305\n",
      "Epoch 44/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3845 - binary_accuracy: 0.8375\n",
      "Epoch 45/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3866 - binary_accuracy: 0.8442\n",
      "Epoch 46/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3870 - binary_accuracy: 0.8416\n",
      "Epoch 47/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3867 - binary_accuracy: 0.8355\n",
      "Epoch 48/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3829 - binary_accuracy: 0.8359\n",
      "Epoch 49/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3822 - binary_accuracy: 0.8378\n",
      "Epoch 50/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3827 - binary_accuracy: 0.8374\n",
      "Epoch 51/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3853 - binary_accuracy: 0.8387\n",
      "Epoch 52/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3856 - binary_accuracy: 0.8417\n",
      "Epoch 53/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3872 - binary_accuracy: 0.8375\n",
      "Epoch 54/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3864 - binary_accuracy: 0.8399\n",
      "Epoch 55/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3831 - binary_accuracy: 0.8395\n",
      "Epoch 56/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3846 - binary_accuracy: 0.8358\n",
      "Epoch 57/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3828 - binary_accuracy: 0.8409\n",
      "Epoch 58/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3834 - binary_accuracy: 0.8351\n",
      "Epoch 59/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3795 - binary_accuracy: 0.8417\n",
      "Epoch 60/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3852 - binary_accuracy: 0.8324\n",
      "Epoch 61/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3855 - binary_accuracy: 0.8364\n",
      "Epoch 62/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3767 - binary_accuracy: 0.8438\n",
      "Epoch 63/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3819 - binary_accuracy: 0.8338\n",
      "Epoch 64/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3778 - binary_accuracy: 0.8392\n",
      "Epoch 65/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3838 - binary_accuracy: 0.8354\n",
      "Epoch 66/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3816 - binary_accuracy: 0.8388\n",
      "Epoch 67/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3788 - binary_accuracy: 0.8408\n",
      "Epoch 68/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3784 - binary_accuracy: 0.8387\n",
      "Epoch 69/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3783 - binary_accuracy: 0.8368\n",
      "Epoch 70/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3805 - binary_accuracy: 0.8373\n",
      "Epoch 71/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3801 - binary_accuracy: 0.8401\n",
      "Epoch 72/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3702 - binary_accuracy: 0.8431\n",
      "Epoch 73/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3783 - binary_accuracy: 0.8464\n",
      "Epoch 74/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3793 - binary_accuracy: 0.8417\n",
      "Epoch 75/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3805 - binary_accuracy: 0.8396\n",
      "Epoch 76/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3794 - binary_accuracy: 0.8364\n",
      "Epoch 77/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3799 - binary_accuracy: 0.8401\n",
      "Epoch 78/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3701 - binary_accuracy: 0.8386\n",
      "Epoch 79/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3711 - binary_accuracy: 0.8456\n",
      "Epoch 80/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3761 - binary_accuracy: 0.8451\n",
      "Epoch 81/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3786 - binary_accuracy: 0.8414\n",
      "Epoch 82/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3694 - binary_accuracy: 0.8451\n",
      "Epoch 83/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3712 - binary_accuracy: 0.8435\n",
      "Epoch 84/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3730 - binary_accuracy: 0.8415\n",
      "Epoch 85/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3685 - binary_accuracy: 0.8421\n",
      "Epoch 86/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3711 - binary_accuracy: 0.8468\n",
      "Epoch 87/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3751 - binary_accuracy: 0.8359\n",
      "Epoch 88/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3695 - binary_accuracy: 0.8437\n",
      "Epoch 89/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3685 - binary_accuracy: 0.8411\n",
      "Epoch 90/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3732 - binary_accuracy: 0.8433\n",
      "Epoch 91/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3696 - binary_accuracy: 0.8483\n",
      "Epoch 92/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3689 - binary_accuracy: 0.8447\n",
      "Epoch 93/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3751 - binary_accuracy: 0.8398\n",
      "Epoch 94/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3710 - binary_accuracy: 0.8404\n",
      "Epoch 95/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3721 - binary_accuracy: 0.8430\n",
      "Epoch 96/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3765 - binary_accuracy: 0.8388\n",
      "Epoch 97/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3684 - binary_accuracy: 0.8464\n",
      "Epoch 98/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3718 - binary_accuracy: 0.8381\n",
      "Epoch 99/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3703 - binary_accuracy: 0.8426\n",
      "Epoch 100/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3686 - binary_accuracy: 0.8426\n",
      "Epoch 101/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3705 - binary_accuracy: 0.8430\n",
      "Epoch 102/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3735 - binary_accuracy: 0.8451\n",
      "Epoch 103/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3770 - binary_accuracy: 0.8402\n",
      "Epoch 104/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3701 - binary_accuracy: 0.8416\n",
      "Epoch 105/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3648 - binary_accuracy: 0.8453\n",
      "Epoch 106/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3667 - binary_accuracy: 0.8470\n",
      "Epoch 107/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3695 - binary_accuracy: 0.8469\n",
      "Epoch 108/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3591 - binary_accuracy: 0.8484\n",
      "Epoch 109/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3658 - binary_accuracy: 0.8418\n",
      "Epoch 110/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3675 - binary_accuracy: 0.8457\n",
      "Epoch 111/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3638 - binary_accuracy: 0.8449\n",
      "Epoch 112/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3623 - binary_accuracy: 0.8450\n",
      "Epoch 113/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3586 - binary_accuracy: 0.8485\n",
      "Epoch 114/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3660 - binary_accuracy: 0.8422\n",
      "Epoch 115/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3654 - binary_accuracy: 0.8433\n",
      "Epoch 116/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3687 - binary_accuracy: 0.8463\n",
      "Epoch 117/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3641 - binary_accuracy: 0.8416\n",
      "Epoch 118/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3627 - binary_accuracy: 0.8471\n",
      "Epoch 119/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3595 - binary_accuracy: 0.8442\n",
      "Epoch 120/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3654 - binary_accuracy: 0.8404\n",
      "Epoch 121/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3678 - binary_accuracy: 0.8462\n",
      "Epoch 122/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3557 - binary_accuracy: 0.8501\n",
      "Epoch 123/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3639 - binary_accuracy: 0.8462\n",
      "Epoch 124/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3619 - binary_accuracy: 0.8485\n",
      "Epoch 125/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3590 - binary_accuracy: 0.8488\n",
      "Epoch 126/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3600 - binary_accuracy: 0.8530\n",
      "Epoch 127/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3594 - binary_accuracy: 0.8432\n",
      "Epoch 128/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3582 - binary_accuracy: 0.8553\n",
      "Epoch 129/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3553 - binary_accuracy: 0.8485\n",
      "Epoch 130/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3627 - binary_accuracy: 0.8447\n",
      "Epoch 131/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3605 - binary_accuracy: 0.8465\n",
      "Epoch 132/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3571 - binary_accuracy: 0.8513\n",
      "Epoch 133/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3543 - binary_accuracy: 0.8486\n",
      "Epoch 134/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3605 - binary_accuracy: 0.8480\n",
      "Epoch 135/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3558 - binary_accuracy: 0.8480\n",
      "Epoch 136/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3556 - binary_accuracy: 0.8485\n",
      "Epoch 137/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3646 - binary_accuracy: 0.8485\n",
      "Epoch 138/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3653 - binary_accuracy: 0.8477\n",
      "Epoch 139/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3564 - binary_accuracy: 0.8469\n",
      "Epoch 140/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3619 - binary_accuracy: 0.8504\n",
      "Epoch 141/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3589 - binary_accuracy: 0.8484\n",
      "Epoch 142/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3628 - binary_accuracy: 0.8447\n",
      "Epoch 143/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3533 - binary_accuracy: 0.8522\n",
      "Epoch 144/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3568 - binary_accuracy: 0.8433\n",
      "Epoch 145/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3531 - binary_accuracy: 0.8539\n",
      "Epoch 146/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3545 - binary_accuracy: 0.8500\n",
      "Epoch 147/225\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3566 - binary_accuracy: 0.8469\n",
      "Epoch 148/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3534 - binary_accuracy: 0.8550\n",
      "Epoch 149/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3601 - binary_accuracy: 0.8511\n",
      "Epoch 150/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3623 - binary_accuracy: 0.8475\n",
      "Epoch 151/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3535 - binary_accuracy: 0.8512\n",
      "Epoch 152/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3536 - binary_accuracy: 0.8538\n",
      "Epoch 153/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3523 - binary_accuracy: 0.8474\n",
      "Epoch 154/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3587 - binary_accuracy: 0.8462\n",
      "Epoch 155/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3539 - binary_accuracy: 0.8522\n",
      "Epoch 156/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3541 - binary_accuracy: 0.8453\n",
      "Epoch 157/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3527 - binary_accuracy: 0.8544\n",
      "Epoch 158/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3525 - binary_accuracy: 0.8504\n",
      "Epoch 159/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3524 - binary_accuracy: 0.8526\n",
      "Epoch 160/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3567 - binary_accuracy: 0.8458\n",
      "Epoch 161/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3561 - binary_accuracy: 0.8484\n",
      "Epoch 162/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3545 - binary_accuracy: 0.8465\n",
      "Epoch 163/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3501 - binary_accuracy: 0.8543\n",
      "Epoch 164/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3533 - binary_accuracy: 0.8519\n",
      "Epoch 165/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3531 - binary_accuracy: 0.8546\n",
      "Epoch 166/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3527 - binary_accuracy: 0.8451\n",
      "Epoch 167/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3512 - binary_accuracy: 0.8491\n",
      "Epoch 168/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3478 - binary_accuracy: 0.8530\n",
      "Epoch 169/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3559 - binary_accuracy: 0.8498\n",
      "Epoch 170/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3528 - binary_accuracy: 0.8501\n",
      "Epoch 171/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3494 - binary_accuracy: 0.8528\n",
      "Epoch 172/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3499 - binary_accuracy: 0.8527\n",
      "Epoch 173/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3514 - binary_accuracy: 0.8544\n",
      "Epoch 174/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3528 - binary_accuracy: 0.8540\n",
      "Epoch 175/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3493 - binary_accuracy: 0.8501\n",
      "Epoch 176/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3498 - binary_accuracy: 0.8477\n",
      "Epoch 177/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3512 - binary_accuracy: 0.8496\n",
      "Epoch 178/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3452 - binary_accuracy: 0.8542\n",
      "Epoch 179/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3507 - binary_accuracy: 0.8513\n",
      "Epoch 180/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3466 - binary_accuracy: 0.8524\n",
      "Epoch 181/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3418 - binary_accuracy: 0.8550\n",
      "Epoch 182/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3434 - binary_accuracy: 0.8535\n",
      "Epoch 183/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3511 - binary_accuracy: 0.8531\n",
      "Epoch 184/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3458 - binary_accuracy: 0.8535\n",
      "Epoch 185/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3432 - binary_accuracy: 0.8569\n",
      "Epoch 186/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3413 - binary_accuracy: 0.8559\n",
      "Epoch 187/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3485 - binary_accuracy: 0.8552\n",
      "Epoch 188/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3454 - binary_accuracy: 0.8568\n",
      "Epoch 189/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3505 - binary_accuracy: 0.8562\n",
      "Epoch 190/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3472 - binary_accuracy: 0.8496\n",
      "Epoch 191/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3451 - binary_accuracy: 0.8528\n",
      "Epoch 192/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3420 - binary_accuracy: 0.8525\n",
      "Epoch 193/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3457 - binary_accuracy: 0.8536\n",
      "Epoch 194/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3437 - binary_accuracy: 0.8522\n",
      "Epoch 195/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3470 - binary_accuracy: 0.8552\n",
      "Epoch 196/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3519 - binary_accuracy: 0.8527\n",
      "Epoch 197/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3435 - binary_accuracy: 0.8518\n",
      "Epoch 198/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3440 - binary_accuracy: 0.8543\n",
      "Epoch 199/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3425 - binary_accuracy: 0.8568\n",
      "Epoch 200/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3391 - binary_accuracy: 0.8541\n",
      "Epoch 201/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3465 - binary_accuracy: 0.8528\n",
      "Epoch 202/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3427 - binary_accuracy: 0.8581\n",
      "Epoch 203/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3407 - binary_accuracy: 0.8557\n",
      "Epoch 204/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3447 - binary_accuracy: 0.8544\n",
      "Epoch 205/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3493 - binary_accuracy: 0.8539\n",
      "Epoch 206/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3437 - binary_accuracy: 0.8544\n",
      "Epoch 207/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3411 - binary_accuracy: 0.8595\n",
      "Epoch 208/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3431 - binary_accuracy: 0.8565\n",
      "Epoch 209/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3351 - binary_accuracy: 0.8594\n",
      "Epoch 210/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3410 - binary_accuracy: 0.8558\n",
      "Epoch 211/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3423 - binary_accuracy: 0.8587\n",
      "Epoch 212/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3433 - binary_accuracy: 0.8563\n",
      "Epoch 213/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3413 - binary_accuracy: 0.8515\n",
      "Epoch 214/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3453 - binary_accuracy: 0.8513\n",
      "Epoch 215/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3359 - binary_accuracy: 0.8582\n",
      "Epoch 216/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3361 - binary_accuracy: 0.8588\n",
      "Epoch 217/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3299 - binary_accuracy: 0.8602\n",
      "Epoch 218/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3394 - binary_accuracy: 0.8506\n",
      "Epoch 219/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3404 - binary_accuracy: 0.8539\n",
      "Epoch 220/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3372 - binary_accuracy: 0.8585\n",
      "Epoch 221/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3375 - binary_accuracy: 0.8578\n",
      "Epoch 222/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3341 - binary_accuracy: 0.8619\n",
      "Epoch 223/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3384 - binary_accuracy: 0.8593\n",
      "Epoch 224/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3394 - binary_accuracy: 0.8577\n",
      "Epoch 225/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3410 - binary_accuracy: 0.8518\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/225\n",
      "1755/1755 [==============================] - 4s 2ms/step - loss: 0.4703 - binary_accuracy: 0.8206\n",
      "Epoch 2/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4385 - binary_accuracy: 0.8238\n",
      "Epoch 3/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4278 - binary_accuracy: 0.8242\n",
      "Epoch 4/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4253 - binary_accuracy: 0.8259\n",
      "Epoch 5/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4207 - binary_accuracy: 0.8234\n",
      "Epoch 6/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4164 - binary_accuracy: 0.8271\n",
      "Epoch 7/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4175 - binary_accuracy: 0.8248\n",
      "Epoch 8/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4097 - binary_accuracy: 0.8282\n",
      "Epoch 9/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4119 - binary_accuracy: 0.8276\n",
      "Epoch 10/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4080 - binary_accuracy: 0.8299\n",
      "Epoch 11/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4100 - binary_accuracy: 0.8245\n",
      "Epoch 12/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4067 - binary_accuracy: 0.8277\n",
      "Epoch 13/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4079 - binary_accuracy: 0.8305\n",
      "Epoch 14/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4065 - binary_accuracy: 0.8278\n",
      "Epoch 15/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3995 - binary_accuracy: 0.8343\n",
      "Epoch 16/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4052 - binary_accuracy: 0.8276\n",
      "Epoch 17/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4033 - binary_accuracy: 0.8335\n",
      "Epoch 18/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4021 - binary_accuracy: 0.8331\n",
      "Epoch 19/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4003 - binary_accuracy: 0.8276\n",
      "Epoch 20/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3983 - binary_accuracy: 0.8343\n",
      "Epoch 21/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3972 - binary_accuracy: 0.8353\n",
      "Epoch 22/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3974 - binary_accuracy: 0.8355\n",
      "Epoch 23/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3961 - binary_accuracy: 0.8309\n",
      "Epoch 24/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3949 - binary_accuracy: 0.8312\n",
      "Epoch 25/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3875 - binary_accuracy: 0.8393\n",
      "Epoch 26/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3885 - binary_accuracy: 0.8338\n",
      "Epoch 27/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.4017 - binary_accuracy: 0.8308\n",
      "Epoch 28/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3934 - binary_accuracy: 0.8340\n",
      "Epoch 29/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3872 - binary_accuracy: 0.8362\n",
      "Epoch 30/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3860 - binary_accuracy: 0.8399\n",
      "Epoch 31/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3921 - binary_accuracy: 0.8406\n",
      "Epoch 32/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3923 - binary_accuracy: 0.8411\n",
      "Epoch 33/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3865 - binary_accuracy: 0.8388\n",
      "Epoch 34/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3892 - binary_accuracy: 0.8362\n",
      "Epoch 35/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3882 - binary_accuracy: 0.8348\n",
      "Epoch 36/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3908 - binary_accuracy: 0.8398\n",
      "Epoch 37/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3883 - binary_accuracy: 0.8375\n",
      "Epoch 38/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3843 - binary_accuracy: 0.8376\n",
      "Epoch 39/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3887 - binary_accuracy: 0.8382\n",
      "Epoch 40/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3809 - binary_accuracy: 0.8369\n",
      "Epoch 41/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3866 - binary_accuracy: 0.8373\n",
      "Epoch 42/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3904 - binary_accuracy: 0.8316\n",
      "Epoch 43/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3869 - binary_accuracy: 0.8382\n",
      "Epoch 44/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3799 - binary_accuracy: 0.8417\n",
      "Epoch 45/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3812 - binary_accuracy: 0.8400\n",
      "Epoch 46/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3844 - binary_accuracy: 0.8418\n",
      "Epoch 47/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3835 - binary_accuracy: 0.8390\n",
      "Epoch 48/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3836 - binary_accuracy: 0.8416\n",
      "Epoch 49/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3865 - binary_accuracy: 0.8358\n",
      "Epoch 50/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3880 - binary_accuracy: 0.8371\n",
      "Epoch 51/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3748 - binary_accuracy: 0.8441\n",
      "Epoch 52/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3824 - binary_accuracy: 0.8389\n",
      "Epoch 53/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3807 - binary_accuracy: 0.8406\n",
      "Epoch 54/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3754 - binary_accuracy: 0.8413\n",
      "Epoch 55/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3743 - binary_accuracy: 0.8421\n",
      "Epoch 56/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3766 - binary_accuracy: 0.8420\n",
      "Epoch 57/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3847 - binary_accuracy: 0.8413\n",
      "Epoch 58/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3759 - binary_accuracy: 0.8384\n",
      "Epoch 59/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3835 - binary_accuracy: 0.8346\n",
      "Epoch 60/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3843 - binary_accuracy: 0.8387\n",
      "Epoch 61/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3771 - binary_accuracy: 0.8455\n",
      "Epoch 62/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3748 - binary_accuracy: 0.8403\n",
      "Epoch 63/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3738 - binary_accuracy: 0.8426\n",
      "Epoch 64/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3761 - binary_accuracy: 0.8401\n",
      "Epoch 65/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3788 - binary_accuracy: 0.8375\n",
      "Epoch 66/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3763 - binary_accuracy: 0.8408\n",
      "Epoch 67/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3710 - binary_accuracy: 0.8411\n",
      "Epoch 68/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3730 - binary_accuracy: 0.8454\n",
      "Epoch 69/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3801 - binary_accuracy: 0.8376\n",
      "Epoch 70/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3757 - binary_accuracy: 0.8388\n",
      "Epoch 71/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3760 - binary_accuracy: 0.8439\n",
      "Epoch 72/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3729 - binary_accuracy: 0.8410\n",
      "Epoch 73/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3728 - binary_accuracy: 0.8403\n",
      "Epoch 74/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3706 - binary_accuracy: 0.8443\n",
      "Epoch 75/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3698 - binary_accuracy: 0.8468\n",
      "Epoch 76/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3748 - binary_accuracy: 0.8409\n",
      "Epoch 77/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3760 - binary_accuracy: 0.8381\n",
      "Epoch 78/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3696 - binary_accuracy: 0.8411\n",
      "Epoch 79/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3685 - binary_accuracy: 0.8482\n",
      "Epoch 80/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3739 - binary_accuracy: 0.8450\n",
      "Epoch 81/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3771 - binary_accuracy: 0.8401\n",
      "Epoch 82/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3663 - binary_accuracy: 0.8449\n",
      "Epoch 83/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3702 - binary_accuracy: 0.8418\n",
      "Epoch 84/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3706 - binary_accuracy: 0.8436\n",
      "Epoch 85/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3702 - binary_accuracy: 0.8430\n",
      "Epoch 86/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3652 - binary_accuracy: 0.8479\n",
      "Epoch 87/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3651 - binary_accuracy: 0.8453\n",
      "Epoch 88/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3696 - binary_accuracy: 0.8470\n",
      "Epoch 89/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3633 - binary_accuracy: 0.8446\n",
      "Epoch 90/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3727 - binary_accuracy: 0.8410\n",
      "Epoch 91/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3696 - binary_accuracy: 0.8433\n",
      "Epoch 92/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3673 - binary_accuracy: 0.8460\n",
      "Epoch 93/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3614 - binary_accuracy: 0.8440\n",
      "Epoch 94/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3679 - binary_accuracy: 0.8450\n",
      "Epoch 95/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3639 - binary_accuracy: 0.8466\n",
      "Epoch 96/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3667 - binary_accuracy: 0.8439\n",
      "Epoch 97/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3664 - binary_accuracy: 0.8426\n",
      "Epoch 98/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3655 - binary_accuracy: 0.8420\n",
      "Epoch 99/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3684 - binary_accuracy: 0.8412\n",
      "Epoch 100/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3632 - binary_accuracy: 0.8492\n",
      "Epoch 101/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3577 - binary_accuracy: 0.8527\n",
      "Epoch 102/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3683 - binary_accuracy: 0.8419\n",
      "Epoch 103/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3639 - binary_accuracy: 0.8429\n",
      "Epoch 104/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3631 - binary_accuracy: 0.8479\n",
      "Epoch 105/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3593 - binary_accuracy: 0.8463\n",
      "Epoch 106/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3628 - binary_accuracy: 0.8472\n",
      "Epoch 107/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3631 - binary_accuracy: 0.8464\n",
      "Epoch 108/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3662 - binary_accuracy: 0.8455\n",
      "Epoch 109/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3631 - binary_accuracy: 0.8476\n",
      "Epoch 110/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3615 - binary_accuracy: 0.8428\n",
      "Epoch 111/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3534 - binary_accuracy: 0.8505\n",
      "Epoch 112/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3558 - binary_accuracy: 0.8488\n",
      "Epoch 113/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3653 - binary_accuracy: 0.8425\n",
      "Epoch 114/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3586 - binary_accuracy: 0.8465\n",
      "Epoch 115/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3607 - binary_accuracy: 0.8475\n",
      "Epoch 116/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3627 - binary_accuracy: 0.8456\n",
      "Epoch 117/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3633 - binary_accuracy: 0.8478\n",
      "Epoch 118/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3616 - binary_accuracy: 0.8488\n",
      "Epoch 119/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3594 - binary_accuracy: 0.8490\n",
      "Epoch 120/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3576 - binary_accuracy: 0.8463\n",
      "Epoch 121/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3640 - binary_accuracy: 0.8483\n",
      "Epoch 122/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3595 - binary_accuracy: 0.8487\n",
      "Epoch 123/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3643 - binary_accuracy: 0.8463\n",
      "Epoch 124/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3510 - binary_accuracy: 0.8495\n",
      "Epoch 125/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3569 - binary_accuracy: 0.8483\n",
      "Epoch 126/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3595 - binary_accuracy: 0.8455\n",
      "Epoch 127/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3570 - binary_accuracy: 0.8527\n",
      "Epoch 128/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3570 - binary_accuracy: 0.8504\n",
      "Epoch 129/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3573 - binary_accuracy: 0.8429\n",
      "Epoch 130/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3560 - binary_accuracy: 0.8507\n",
      "Epoch 131/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3525 - binary_accuracy: 0.8513\n",
      "Epoch 132/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3597 - binary_accuracy: 0.8500\n",
      "Epoch 133/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3534 - binary_accuracy: 0.8487\n",
      "Epoch 134/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3532 - binary_accuracy: 0.8520\n",
      "Epoch 135/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3540 - binary_accuracy: 0.8491\n",
      "Epoch 136/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3488 - binary_accuracy: 0.8547\n",
      "Epoch 137/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3504 - binary_accuracy: 0.8524\n",
      "Epoch 138/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3510 - binary_accuracy: 0.8519\n",
      "Epoch 139/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3536 - binary_accuracy: 0.8510\n",
      "Epoch 140/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3493 - binary_accuracy: 0.8448\n",
      "Epoch 141/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3527 - binary_accuracy: 0.8514\n",
      "Epoch 142/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3505 - binary_accuracy: 0.8525\n",
      "Epoch 143/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3444 - binary_accuracy: 0.8524\n",
      "Epoch 144/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3468 - binary_accuracy: 0.8508\n",
      "Epoch 145/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3453 - binary_accuracy: 0.8537\n",
      "Epoch 146/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3546 - binary_accuracy: 0.8491\n",
      "Epoch 147/225\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3563 - binary_accuracy: 0.8508\n",
      "Epoch 148/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3450 - binary_accuracy: 0.8512\n",
      "Epoch 149/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3520 - binary_accuracy: 0.8521\n",
      "Epoch 150/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3520 - binary_accuracy: 0.8499\n",
      "Epoch 151/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3578 - binary_accuracy: 0.8494\n",
      "Epoch 152/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3454 - binary_accuracy: 0.8536\n",
      "Epoch 153/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3496 - binary_accuracy: 0.8529\n",
      "Epoch 154/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3546 - binary_accuracy: 0.8492\n",
      "Epoch 155/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3461 - binary_accuracy: 0.8560\n",
      "Epoch 156/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3480 - binary_accuracy: 0.8550\n",
      "Epoch 157/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3450 - binary_accuracy: 0.8520\n",
      "Epoch 158/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3498 - binary_accuracy: 0.8512\n",
      "Epoch 159/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3492 - binary_accuracy: 0.8508\n",
      "Epoch 160/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3498 - binary_accuracy: 0.8520\n",
      "Epoch 161/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3428 - binary_accuracy: 0.8529\n",
      "Epoch 162/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3442 - binary_accuracy: 0.8547\n",
      "Epoch 163/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3473 - binary_accuracy: 0.8532\n",
      "Epoch 164/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3502 - binary_accuracy: 0.8525\n",
      "Epoch 165/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3429 - binary_accuracy: 0.8517\n",
      "Epoch 166/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3431 - binary_accuracy: 0.8567\n",
      "Epoch 167/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3498 - binary_accuracy: 0.8497\n",
      "Epoch 168/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3415 - binary_accuracy: 0.8543\n",
      "Epoch 169/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3503 - binary_accuracy: 0.8504\n",
      "Epoch 170/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3447 - binary_accuracy: 0.8533\n",
      "Epoch 171/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3428 - binary_accuracy: 0.8539\n",
      "Epoch 172/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3461 - binary_accuracy: 0.8541\n",
      "Epoch 173/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3457 - binary_accuracy: 0.8497\n",
      "Epoch 174/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3424 - binary_accuracy: 0.8558\n",
      "Epoch 175/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3418 - binary_accuracy: 0.8560\n",
      "Epoch 176/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3401 - binary_accuracy: 0.8522\n",
      "Epoch 177/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3437 - binary_accuracy: 0.8555\n",
      "Epoch 178/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3462 - binary_accuracy: 0.8526\n",
      "Epoch 179/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3490 - binary_accuracy: 0.8537\n",
      "Epoch 180/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3441 - binary_accuracy: 0.8570\n",
      "Epoch 181/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3402 - binary_accuracy: 0.8488\n",
      "Epoch 182/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3443 - binary_accuracy: 0.8498\n",
      "Epoch 183/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3484 - binary_accuracy: 0.8526\n",
      "Epoch 184/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3335 - binary_accuracy: 0.8575\n",
      "Epoch 185/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3403 - binary_accuracy: 0.8543\n",
      "Epoch 186/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3400 - binary_accuracy: 0.8549\n",
      "Epoch 187/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3378 - binary_accuracy: 0.8561\n",
      "Epoch 188/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3402 - binary_accuracy: 0.8583\n",
      "Epoch 189/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3402 - binary_accuracy: 0.8615\n",
      "Epoch 190/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3415 - binary_accuracy: 0.8535\n",
      "Epoch 191/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3410 - binary_accuracy: 0.8525\n",
      "Epoch 192/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3372 - binary_accuracy: 0.8533\n",
      "Epoch 193/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3473 - binary_accuracy: 0.8501\n",
      "Epoch 194/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3381 - binary_accuracy: 0.8563\n",
      "Epoch 195/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3440 - binary_accuracy: 0.8542\n",
      "Epoch 196/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3339 - binary_accuracy: 0.8603\n",
      "Epoch 197/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3352 - binary_accuracy: 0.8570\n",
      "Epoch 198/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3335 - binary_accuracy: 0.8587\n",
      "Epoch 199/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3389 - binary_accuracy: 0.8555\n",
      "Epoch 200/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3339 - binary_accuracy: 0.8579\n",
      "Epoch 201/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3371 - binary_accuracy: 0.8567\n",
      "Epoch 202/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3440 - binary_accuracy: 0.8568\n",
      "Epoch 203/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3342 - binary_accuracy: 0.8594\n",
      "Epoch 204/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3350 - binary_accuracy: 0.8605\n",
      "Epoch 205/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3404 - binary_accuracy: 0.8523\n",
      "Epoch 206/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3410 - binary_accuracy: 0.8544\n",
      "Epoch 207/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3397 - binary_accuracy: 0.8526\n",
      "Epoch 208/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3344 - binary_accuracy: 0.8589\n",
      "Epoch 209/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3354 - binary_accuracy: 0.8587\n",
      "Epoch 210/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3384 - binary_accuracy: 0.8551\n",
      "Epoch 211/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3333 - binary_accuracy: 0.8633\n",
      "Epoch 212/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3288 - binary_accuracy: 0.8596\n",
      "Epoch 213/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3303 - binary_accuracy: 0.8585\n",
      "Epoch 214/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3373 - binary_accuracy: 0.8582\n",
      "Epoch 215/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3349 - binary_accuracy: 0.8644\n",
      "Epoch 216/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3418 - binary_accuracy: 0.8558\n",
      "Epoch 217/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3319 - binary_accuracy: 0.8591\n",
      "Epoch 218/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3357 - binary_accuracy: 0.8584\n",
      "Epoch 219/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3332 - binary_accuracy: 0.8552\n",
      "Epoch 220/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3339 - binary_accuracy: 0.8614\n",
      "Epoch 221/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3341 - binary_accuracy: 0.8597\n",
      "Epoch 222/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3312 - binary_accuracy: 0.8608\n",
      "Epoch 223/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3350 - binary_accuracy: 0.8553\n",
      "Epoch 224/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3353 - binary_accuracy: 0.8572\n",
      "Epoch 225/225\n",
      "1755/1755 [==============================] - 2s 1ms/step - loss: 0.3269 - binary_accuracy: 0.8614\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/225\n",
      "1754/1754 [==============================] - 4s 2ms/step - loss: 0.4670 - binary_accuracy: 0.8167\n",
      "Epoch 2/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4358 - binary_accuracy: 0.8216\n",
      "Epoch 3/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4269 - binary_accuracy: 0.8207\n",
      "Epoch 4/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4171 - binary_accuracy: 0.8265\n",
      "Epoch 5/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4131 - binary_accuracy: 0.8279\n",
      "Epoch 6/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4150 - binary_accuracy: 0.8241\n",
      "Epoch 7/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4099 - binary_accuracy: 0.8260\n",
      "Epoch 8/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4099 - binary_accuracy: 0.8267\n",
      "Epoch 9/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4059 - binary_accuracy: 0.8243\n",
      "Epoch 10/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3978 - binary_accuracy: 0.8332\n",
      "Epoch 11/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4003 - binary_accuracy: 0.8290\n",
      "Epoch 12/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3987 - binary_accuracy: 0.8309\n",
      "Epoch 13/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4018 - binary_accuracy: 0.8291\n",
      "Epoch 14/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4004 - binary_accuracy: 0.8333\n",
      "Epoch 15/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3957 - binary_accuracy: 0.8318\n",
      "Epoch 16/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3964 - binary_accuracy: 0.8352\n",
      "Epoch 17/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3973 - binary_accuracy: 0.8330\n",
      "Epoch 18/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3921 - binary_accuracy: 0.8348\n",
      "Epoch 19/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3947 - binary_accuracy: 0.8331\n",
      "Epoch 20/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3891 - binary_accuracy: 0.8350\n",
      "Epoch 21/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3905 - binary_accuracy: 0.8337\n",
      "Epoch 22/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3901 - binary_accuracy: 0.8371\n",
      "Epoch 23/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3917 - binary_accuracy: 0.8348\n",
      "Epoch 24/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3921 - binary_accuracy: 0.8294\n",
      "Epoch 25/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3867 - binary_accuracy: 0.8363\n",
      "Epoch 26/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3933 - binary_accuracy: 0.8305\n",
      "Epoch 27/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3896 - binary_accuracy: 0.8369\n",
      "Epoch 28/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3891 - binary_accuracy: 0.8399\n",
      "Epoch 29/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3900 - binary_accuracy: 0.8347\n",
      "Epoch 30/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3867 - binary_accuracy: 0.8319\n",
      "Epoch 31/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3853 - binary_accuracy: 0.8332\n",
      "Epoch 32/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3843 - binary_accuracy: 0.8385\n",
      "Epoch 33/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3827 - binary_accuracy: 0.8404\n",
      "Epoch 34/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3907 - binary_accuracy: 0.8371\n",
      "Epoch 35/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3806 - binary_accuracy: 0.8443\n",
      "Epoch 36/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3802 - binary_accuracy: 0.8431\n",
      "Epoch 37/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3818 - binary_accuracy: 0.8364\n",
      "Epoch 38/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3837 - binary_accuracy: 0.8380\n",
      "Epoch 39/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3790 - binary_accuracy: 0.8404\n",
      "Epoch 40/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3823 - binary_accuracy: 0.8403\n",
      "Epoch 41/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3775 - binary_accuracy: 0.8396\n",
      "Epoch 42/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3881 - binary_accuracy: 0.8346\n",
      "Epoch 43/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3854 - binary_accuracy: 0.8386\n",
      "Epoch 44/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3817 - binary_accuracy: 0.8377\n",
      "Epoch 45/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3813 - binary_accuracy: 0.8380\n",
      "Epoch 46/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3736 - binary_accuracy: 0.8421\n",
      "Epoch 47/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3779 - binary_accuracy: 0.8426\n",
      "Epoch 48/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3819 - binary_accuracy: 0.8429\n",
      "Epoch 49/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3754 - binary_accuracy: 0.8436\n",
      "Epoch 50/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3808 - binary_accuracy: 0.8375\n",
      "Epoch 51/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3789 - binary_accuracy: 0.8444\n",
      "Epoch 52/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3778 - binary_accuracy: 0.8407\n",
      "Epoch 53/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3800 - binary_accuracy: 0.8389\n",
      "Epoch 54/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3759 - binary_accuracy: 0.8443\n",
      "Epoch 55/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3772 - binary_accuracy: 0.8409\n",
      "Epoch 56/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3777 - binary_accuracy: 0.8431\n",
      "Epoch 57/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3739 - binary_accuracy: 0.8426\n",
      "Epoch 58/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3800 - binary_accuracy: 0.8379\n",
      "Epoch 59/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3715 - binary_accuracy: 0.8402\n",
      "Epoch 60/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3774 - binary_accuracy: 0.8401\n",
      "Epoch 61/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3801 - binary_accuracy: 0.8385\n",
      "Epoch 62/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3721 - binary_accuracy: 0.8414\n",
      "Epoch 63/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3726 - binary_accuracy: 0.8436\n",
      "Epoch 64/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3766 - binary_accuracy: 0.8409\n",
      "Epoch 65/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3749 - binary_accuracy: 0.8362\n",
      "Epoch 66/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3672 - binary_accuracy: 0.8453\n",
      "Epoch 67/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3780 - binary_accuracy: 0.8392\n",
      "Epoch 68/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3739 - binary_accuracy: 0.8404\n",
      "Epoch 69/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3764 - binary_accuracy: 0.8400\n",
      "Epoch 70/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3701 - binary_accuracy: 0.8437\n",
      "Epoch 71/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3734 - binary_accuracy: 0.8397\n",
      "Epoch 72/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3719 - binary_accuracy: 0.8434\n",
      "Epoch 73/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3716 - binary_accuracy: 0.8470\n",
      "Epoch 74/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3720 - binary_accuracy: 0.8387\n",
      "Epoch 75/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3725 - binary_accuracy: 0.8408\n",
      "Epoch 76/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3695 - binary_accuracy: 0.8405\n",
      "Epoch 77/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3688 - binary_accuracy: 0.8423\n",
      "Epoch 78/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3701 - binary_accuracy: 0.8442\n",
      "Epoch 79/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3729 - binary_accuracy: 0.8424\n",
      "Epoch 80/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3670 - binary_accuracy: 0.8446\n",
      "Epoch 81/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3688 - binary_accuracy: 0.8441\n",
      "Epoch 82/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3682 - binary_accuracy: 0.8418\n",
      "Epoch 83/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3689 - binary_accuracy: 0.8457\n",
      "Epoch 84/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3671 - binary_accuracy: 0.8448\n",
      "Epoch 85/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3656 - binary_accuracy: 0.8453\n",
      "Epoch 86/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3648 - binary_accuracy: 0.8426\n",
      "Epoch 87/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3679 - binary_accuracy: 0.8471\n",
      "Epoch 88/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3704 - binary_accuracy: 0.8474\n",
      "Epoch 89/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3715 - binary_accuracy: 0.8425\n",
      "Epoch 90/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3671 - binary_accuracy: 0.8475\n",
      "Epoch 91/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3633 - binary_accuracy: 0.8432\n",
      "Epoch 92/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3671 - binary_accuracy: 0.8450\n",
      "Epoch 93/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3643 - binary_accuracy: 0.8454\n",
      "Epoch 94/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3627 - binary_accuracy: 0.8480\n",
      "Epoch 95/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3637 - binary_accuracy: 0.8470\n",
      "Epoch 96/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3695 - binary_accuracy: 0.8429\n",
      "Epoch 97/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3667 - binary_accuracy: 0.8449\n",
      "Epoch 98/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3616 - binary_accuracy: 0.8532\n",
      "Epoch 99/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3664 - binary_accuracy: 0.8425\n",
      "Epoch 100/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3631 - binary_accuracy: 0.8488\n",
      "Epoch 101/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3614 - binary_accuracy: 0.8495\n",
      "Epoch 102/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3625 - binary_accuracy: 0.8456\n",
      "Epoch 103/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3619 - binary_accuracy: 0.8493\n",
      "Epoch 104/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3592 - binary_accuracy: 0.8490\n",
      "Epoch 105/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3642 - binary_accuracy: 0.8429\n",
      "Epoch 106/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3626 - binary_accuracy: 0.8482\n",
      "Epoch 107/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3634 - binary_accuracy: 0.8448\n",
      "Epoch 108/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3553 - binary_accuracy: 0.8453\n",
      "Epoch 109/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3607 - binary_accuracy: 0.8435\n",
      "Epoch 110/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3591 - binary_accuracy: 0.8487\n",
      "Epoch 111/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3637 - binary_accuracy: 0.8478\n",
      "Epoch 112/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3652 - binary_accuracy: 0.8460\n",
      "Epoch 113/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3600 - binary_accuracy: 0.8504\n",
      "Epoch 114/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3548 - binary_accuracy: 0.8532\n",
      "Epoch 115/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3569 - binary_accuracy: 0.8507\n",
      "Epoch 116/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3572 - binary_accuracy: 0.8487\n",
      "Epoch 117/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3593 - binary_accuracy: 0.8498\n",
      "Epoch 118/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3518 - binary_accuracy: 0.8547\n",
      "Epoch 119/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3526 - binary_accuracy: 0.8518\n",
      "Epoch 120/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3526 - binary_accuracy: 0.8493\n",
      "Epoch 121/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3598 - binary_accuracy: 0.8473\n",
      "Epoch 122/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3600 - binary_accuracy: 0.8461\n",
      "Epoch 123/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3470 - binary_accuracy: 0.8552\n",
      "Epoch 124/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3549 - binary_accuracy: 0.8486\n",
      "Epoch 125/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3577 - binary_accuracy: 0.8539\n",
      "Epoch 126/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3577 - binary_accuracy: 0.8502\n",
      "Epoch 127/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3543 - binary_accuracy: 0.8530\n",
      "Epoch 128/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3564 - binary_accuracy: 0.8532\n",
      "Epoch 129/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3535 - binary_accuracy: 0.8495\n",
      "Epoch 130/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3537 - binary_accuracy: 0.8514\n",
      "Epoch 131/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3526 - binary_accuracy: 0.8510\n",
      "Epoch 132/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3574 - binary_accuracy: 0.8493\n",
      "Epoch 133/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3488 - binary_accuracy: 0.8557\n",
      "Epoch 134/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3577 - binary_accuracy: 0.8430\n",
      "Epoch 135/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3580 - binary_accuracy: 0.8445\n",
      "Epoch 136/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3524 - binary_accuracy: 0.8547\n",
      "Epoch 137/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3521 - binary_accuracy: 0.8520\n",
      "Epoch 138/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3515 - binary_accuracy: 0.8534\n",
      "Epoch 139/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3554 - binary_accuracy: 0.8496\n",
      "Epoch 140/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3522 - binary_accuracy: 0.8500\n",
      "Epoch 141/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3536 - binary_accuracy: 0.8548\n",
      "Epoch 142/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3494 - binary_accuracy: 0.8571\n",
      "Epoch 143/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3451 - binary_accuracy: 0.8538\n",
      "Epoch 144/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3487 - binary_accuracy: 0.8526\n",
      "Epoch 145/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3521 - binary_accuracy: 0.8494\n",
      "Epoch 146/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3523 - binary_accuracy: 0.8540\n",
      "Epoch 147/225\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3475 - binary_accuracy: 0.8538\n",
      "Epoch 148/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3500 - binary_accuracy: 0.8503\n",
      "Epoch 149/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3522 - binary_accuracy: 0.8491\n",
      "Epoch 150/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3496 - binary_accuracy: 0.8549\n",
      "Epoch 151/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3533 - binary_accuracy: 0.8483\n",
      "Epoch 152/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3537 - binary_accuracy: 0.8468\n",
      "Epoch 153/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3466 - binary_accuracy: 0.8493\n",
      "Epoch 154/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3415 - binary_accuracy: 0.8566\n",
      "Epoch 155/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3497 - binary_accuracy: 0.8507\n",
      "Epoch 156/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3547 - binary_accuracy: 0.8473\n",
      "Epoch 157/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3577 - binary_accuracy: 0.8481\n",
      "Epoch 158/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3432 - binary_accuracy: 0.8530\n",
      "Epoch 159/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3467 - binary_accuracy: 0.8523\n",
      "Epoch 160/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3431 - binary_accuracy: 0.8521\n",
      "Epoch 161/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3437 - binary_accuracy: 0.8548\n",
      "Epoch 162/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3437 - binary_accuracy: 0.8581\n",
      "Epoch 163/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3440 - binary_accuracy: 0.8573\n",
      "Epoch 164/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3504 - binary_accuracy: 0.8510\n",
      "Epoch 165/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3464 - binary_accuracy: 0.8537\n",
      "Epoch 166/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3461 - binary_accuracy: 0.8535\n",
      "Epoch 167/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3459 - binary_accuracy: 0.8556\n",
      "Epoch 168/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3457 - binary_accuracy: 0.8525\n",
      "Epoch 169/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3497 - binary_accuracy: 0.8519\n",
      "Epoch 170/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3457 - binary_accuracy: 0.8536\n",
      "Epoch 171/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3449 - binary_accuracy: 0.8508\n",
      "Epoch 172/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3430 - binary_accuracy: 0.8592\n",
      "Epoch 173/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3475 - binary_accuracy: 0.8513\n",
      "Epoch 174/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3478 - binary_accuracy: 0.8539\n",
      "Epoch 175/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3498 - binary_accuracy: 0.8495\n",
      "Epoch 176/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3455 - binary_accuracy: 0.8519\n",
      "Epoch 177/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3418 - binary_accuracy: 0.8575\n",
      "Epoch 178/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3406 - binary_accuracy: 0.8586\n",
      "Epoch 179/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3425 - binary_accuracy: 0.8535\n",
      "Epoch 180/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3394 - binary_accuracy: 0.8579\n",
      "Epoch 181/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3441 - binary_accuracy: 0.8589\n",
      "Epoch 182/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3412 - binary_accuracy: 0.8573\n",
      "Epoch 183/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3416 - binary_accuracy: 0.8555\n",
      "Epoch 184/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3444 - binary_accuracy: 0.8544\n",
      "Epoch 185/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3411 - binary_accuracy: 0.8539\n",
      "Epoch 186/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3422 - binary_accuracy: 0.8552\n",
      "Epoch 187/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3424 - binary_accuracy: 0.8491\n",
      "Epoch 188/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3410 - binary_accuracy: 0.8558\n",
      "Epoch 189/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3406 - binary_accuracy: 0.8578\n",
      "Epoch 190/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3432 - binary_accuracy: 0.8571\n",
      "Epoch 191/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3376 - binary_accuracy: 0.8571\n",
      "Epoch 192/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3435 - binary_accuracy: 0.8554\n",
      "Epoch 193/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3387 - binary_accuracy: 0.8576\n",
      "Epoch 194/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3373 - binary_accuracy: 0.8612\n",
      "Epoch 195/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3446 - binary_accuracy: 0.8492\n",
      "Epoch 196/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3483 - binary_accuracy: 0.8525\n",
      "Epoch 197/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3329 - binary_accuracy: 0.8583\n",
      "Epoch 198/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3411 - binary_accuracy: 0.8601\n",
      "Epoch 199/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3385 - binary_accuracy: 0.8588\n",
      "Epoch 200/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3389 - binary_accuracy: 0.8520\n",
      "Epoch 201/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3407 - binary_accuracy: 0.8592\n",
      "Epoch 202/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3333 - binary_accuracy: 0.8590\n",
      "Epoch 203/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3352 - binary_accuracy: 0.8586\n",
      "Epoch 204/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3393 - binary_accuracy: 0.8532\n",
      "Epoch 205/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3343 - binary_accuracy: 0.8584\n",
      "Epoch 206/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3347 - binary_accuracy: 0.8577\n",
      "Epoch 207/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3365 - binary_accuracy: 0.8573\n",
      "Epoch 208/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3365 - binary_accuracy: 0.8581\n",
      "Epoch 209/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3405 - binary_accuracy: 0.8533\n",
      "Epoch 210/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3374 - binary_accuracy: 0.8585\n",
      "Epoch 211/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3334 - binary_accuracy: 0.8609\n",
      "Epoch 212/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3375 - binary_accuracy: 0.8568\n",
      "Epoch 213/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3376 - binary_accuracy: 0.8614\n",
      "Epoch 214/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3313 - binary_accuracy: 0.8603\n",
      "Epoch 215/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3328 - binary_accuracy: 0.8606\n",
      "Epoch 216/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3377 - binary_accuracy: 0.8580\n",
      "Epoch 217/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3341 - binary_accuracy: 0.8580\n",
      "Epoch 218/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3361 - binary_accuracy: 0.8530\n",
      "Epoch 219/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3386 - binary_accuracy: 0.8566\n",
      "Epoch 220/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3309 - binary_accuracy: 0.8576\n",
      "Epoch 221/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3402 - binary_accuracy: 0.8567\n",
      "Epoch 222/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3350 - binary_accuracy: 0.8608\n",
      "Epoch 223/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3336 - binary_accuracy: 0.8584\n",
      "Epoch 224/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3312 - binary_accuracy: 0.8628\n",
      "Epoch 225/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3372 - binary_accuracy: 0.8580\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/225\n",
      "1754/1754 [==============================] - 4s 2ms/step - loss: 0.4719 - binary_accuracy: 0.8120\n",
      "Epoch 2/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4394 - binary_accuracy: 0.8193\n",
      "Epoch 3/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4295 - binary_accuracy: 0.8209\n",
      "Epoch 4/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4270 - binary_accuracy: 0.8185\n",
      "Epoch 5/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4244 - binary_accuracy: 0.8206\n",
      "Epoch 6/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4234 - binary_accuracy: 0.8230\n",
      "Epoch 7/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4214 - binary_accuracy: 0.8226\n",
      "Epoch 8/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4145 - binary_accuracy: 0.8219\n",
      "Epoch 9/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4127 - binary_accuracy: 0.8256\n",
      "Epoch 10/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4085 - binary_accuracy: 0.8300\n",
      "Epoch 11/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4116 - binary_accuracy: 0.8243\n",
      "Epoch 12/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4074 - binary_accuracy: 0.8274\n",
      "Epoch 13/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4044 - binary_accuracy: 0.8293\n",
      "Epoch 14/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4031 - binary_accuracy: 0.8312\n",
      "Epoch 15/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4064 - binary_accuracy: 0.8281\n",
      "Epoch 16/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4020 - binary_accuracy: 0.8333\n",
      "Epoch 17/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4031 - binary_accuracy: 0.8258\n",
      "Epoch 18/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4034 - binary_accuracy: 0.8309\n",
      "Epoch 19/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4050 - binary_accuracy: 0.8277\n",
      "Epoch 20/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3974 - binary_accuracy: 0.8337\n",
      "Epoch 21/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3957 - binary_accuracy: 0.8341\n",
      "Epoch 22/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3974 - binary_accuracy: 0.8287\n",
      "Epoch 23/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4029 - binary_accuracy: 0.8331\n",
      "Epoch 24/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4004 - binary_accuracy: 0.8303\n",
      "Epoch 25/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3956 - binary_accuracy: 0.8299\n",
      "Epoch 26/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3904 - binary_accuracy: 0.8364\n",
      "Epoch 27/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3966 - binary_accuracy: 0.8344\n",
      "Epoch 28/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3975 - binary_accuracy: 0.8287\n",
      "Epoch 29/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3974 - binary_accuracy: 0.8360\n",
      "Epoch 30/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3907 - binary_accuracy: 0.8352\n",
      "Epoch 31/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3976 - binary_accuracy: 0.8333\n",
      "Epoch 32/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3911 - binary_accuracy: 0.8378\n",
      "Epoch 33/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3931 - binary_accuracy: 0.8334\n",
      "Epoch 34/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3949 - binary_accuracy: 0.8311\n",
      "Epoch 35/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3883 - binary_accuracy: 0.8375\n",
      "Epoch 36/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3909 - binary_accuracy: 0.8308\n",
      "Epoch 37/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3859 - binary_accuracy: 0.8378\n",
      "Epoch 38/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3930 - binary_accuracy: 0.8321\n",
      "Epoch 39/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3925 - binary_accuracy: 0.8327\n",
      "Epoch 40/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3908 - binary_accuracy: 0.8326\n",
      "Epoch 41/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3871 - binary_accuracy: 0.8307\n",
      "Epoch 42/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3911 - binary_accuracy: 0.8307\n",
      "Epoch 43/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3902 - binary_accuracy: 0.8307\n",
      "Epoch 44/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3839 - binary_accuracy: 0.8339\n",
      "Epoch 45/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3874 - binary_accuracy: 0.8365\n",
      "Epoch 46/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3865 - binary_accuracy: 0.8357\n",
      "Epoch 47/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3895 - binary_accuracy: 0.8331\n",
      "Epoch 48/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3812 - binary_accuracy: 0.8402\n",
      "Epoch 49/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3867 - binary_accuracy: 0.8325\n",
      "Epoch 50/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3837 - binary_accuracy: 0.8363\n",
      "Epoch 51/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3860 - binary_accuracy: 0.8382\n",
      "Epoch 52/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3838 - binary_accuracy: 0.8396\n",
      "Epoch 53/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3817 - binary_accuracy: 0.8364\n",
      "Epoch 54/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3834 - binary_accuracy: 0.8357\n",
      "Epoch 55/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3855 - binary_accuracy: 0.8342\n",
      "Epoch 56/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3874 - binary_accuracy: 0.8327\n",
      "Epoch 57/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3819 - binary_accuracy: 0.8332\n",
      "Epoch 58/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3756 - binary_accuracy: 0.8428\n",
      "Epoch 59/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3784 - binary_accuracy: 0.8374\n",
      "Epoch 60/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3831 - binary_accuracy: 0.8392\n",
      "Epoch 61/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3791 - binary_accuracy: 0.8393\n",
      "Epoch 62/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3838 - binary_accuracy: 0.8348\n",
      "Epoch 63/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3844 - binary_accuracy: 0.8374\n",
      "Epoch 64/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3852 - binary_accuracy: 0.8370\n",
      "Epoch 65/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3797 - binary_accuracy: 0.8368\n",
      "Epoch 66/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3849 - binary_accuracy: 0.8334\n",
      "Epoch 67/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3789 - binary_accuracy: 0.8360\n",
      "Epoch 68/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3723 - binary_accuracy: 0.8387\n",
      "Epoch 69/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3841 - binary_accuracy: 0.8368\n",
      "Epoch 70/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3797 - binary_accuracy: 0.8367\n",
      "Epoch 71/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3775 - binary_accuracy: 0.8407\n",
      "Epoch 72/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3746 - binary_accuracy: 0.8389\n",
      "Epoch 73/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3723 - binary_accuracy: 0.8422\n",
      "Epoch 74/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3748 - binary_accuracy: 0.8363\n",
      "Epoch 75/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3802 - binary_accuracy: 0.8359\n",
      "Epoch 76/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3797 - binary_accuracy: 0.8310\n",
      "Epoch 77/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3731 - binary_accuracy: 0.8419\n",
      "Epoch 78/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3748 - binary_accuracy: 0.8392\n",
      "Epoch 79/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3723 - binary_accuracy: 0.8417\n",
      "Epoch 80/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3743 - binary_accuracy: 0.8390\n",
      "Epoch 81/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3826 - binary_accuracy: 0.8339\n",
      "Epoch 82/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3736 - binary_accuracy: 0.8424\n",
      "Epoch 83/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3717 - binary_accuracy: 0.8415\n",
      "Epoch 84/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3704 - binary_accuracy: 0.8395\n",
      "Epoch 85/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3680 - binary_accuracy: 0.8380\n",
      "Epoch 86/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3767 - binary_accuracy: 0.8403\n",
      "Epoch 87/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3633 - binary_accuracy: 0.8507\n",
      "Epoch 88/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3710 - binary_accuracy: 0.8448\n",
      "Epoch 89/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3750 - binary_accuracy: 0.8404\n",
      "Epoch 90/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3708 - binary_accuracy: 0.8396\n",
      "Epoch 91/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3680 - binary_accuracy: 0.8442\n",
      "Epoch 92/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3735 - binary_accuracy: 0.8380\n",
      "Epoch 93/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3719 - binary_accuracy: 0.8438\n",
      "Epoch 94/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3695 - binary_accuracy: 0.8408\n",
      "Epoch 95/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3660 - binary_accuracy: 0.8431\n",
      "Epoch 96/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3727 - binary_accuracy: 0.8428\n",
      "Epoch 97/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3709 - binary_accuracy: 0.8417\n",
      "Epoch 98/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3686 - binary_accuracy: 0.8433\n",
      "Epoch 99/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3638 - binary_accuracy: 0.8479\n",
      "Epoch 100/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3721 - binary_accuracy: 0.8406\n",
      "Epoch 101/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3732 - binary_accuracy: 0.8425\n",
      "Epoch 102/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3666 - binary_accuracy: 0.8441\n",
      "Epoch 103/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3714 - binary_accuracy: 0.8449\n",
      "Epoch 104/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3650 - binary_accuracy: 0.8447\n",
      "Epoch 105/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3715 - binary_accuracy: 0.8406\n",
      "Epoch 106/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3701 - binary_accuracy: 0.8453\n",
      "Epoch 107/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3741 - binary_accuracy: 0.8431\n",
      "Epoch 108/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3631 - binary_accuracy: 0.8441\n",
      "Epoch 109/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3629 - binary_accuracy: 0.8425\n",
      "Epoch 110/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3683 - binary_accuracy: 0.8455\n",
      "Epoch 111/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3682 - binary_accuracy: 0.8415\n",
      "Epoch 112/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3734 - binary_accuracy: 0.8429\n",
      "Epoch 113/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3622 - binary_accuracy: 0.8462\n",
      "Epoch 114/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3644 - binary_accuracy: 0.8488\n",
      "Epoch 115/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3649 - binary_accuracy: 0.8441\n",
      "Epoch 116/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3645 - binary_accuracy: 0.8449\n",
      "Epoch 117/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3583 - binary_accuracy: 0.8512\n",
      "Epoch 118/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3669 - binary_accuracy: 0.8428\n",
      "Epoch 119/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3604 - binary_accuracy: 0.8455\n",
      "Epoch 120/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3593 - binary_accuracy: 0.8516\n",
      "Epoch 121/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3618 - binary_accuracy: 0.8462\n",
      "Epoch 122/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3622 - binary_accuracy: 0.8475\n",
      "Epoch 123/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3652 - binary_accuracy: 0.8434\n",
      "Epoch 124/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3626 - binary_accuracy: 0.8479\n",
      "Epoch 125/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3580 - binary_accuracy: 0.8542\n",
      "Epoch 126/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3618 - binary_accuracy: 0.8467\n",
      "Epoch 127/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3600 - binary_accuracy: 0.8446\n",
      "Epoch 128/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3631 - binary_accuracy: 0.8461\n",
      "Epoch 129/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3608 - binary_accuracy: 0.8481\n",
      "Epoch 130/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3610 - binary_accuracy: 0.8433\n",
      "Epoch 131/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3596 - binary_accuracy: 0.8479\n",
      "Epoch 132/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3538 - binary_accuracy: 0.8487\n",
      "Epoch 133/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3550 - binary_accuracy: 0.8483\n",
      "Epoch 134/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3582 - binary_accuracy: 0.8521\n",
      "Epoch 135/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3588 - binary_accuracy: 0.8453\n",
      "Epoch 136/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3579 - binary_accuracy: 0.8480\n",
      "Epoch 137/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3520 - binary_accuracy: 0.8535\n",
      "Epoch 138/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3519 - binary_accuracy: 0.8505\n",
      "Epoch 139/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3574 - binary_accuracy: 0.8431\n",
      "Epoch 140/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3606 - binary_accuracy: 0.8470\n",
      "Epoch 141/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3518 - binary_accuracy: 0.8490\n",
      "Epoch 142/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3555 - binary_accuracy: 0.8488\n",
      "Epoch 143/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3522 - binary_accuracy: 0.8532\n",
      "Epoch 144/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3551 - binary_accuracy: 0.8491\n",
      "Epoch 145/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3559 - binary_accuracy: 0.8474\n",
      "Epoch 146/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3549 - binary_accuracy: 0.8472\n",
      "Epoch 147/225\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3566 - binary_accuracy: 0.8504\n",
      "Epoch 148/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3533 - binary_accuracy: 0.8525\n",
      "Epoch 149/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3557 - binary_accuracy: 0.8477\n",
      "Epoch 150/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3557 - binary_accuracy: 0.8467\n",
      "Epoch 151/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3556 - binary_accuracy: 0.8528\n",
      "Epoch 152/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3534 - binary_accuracy: 0.8527\n",
      "Epoch 153/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3498 - binary_accuracy: 0.8488\n",
      "Epoch 154/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3543 - binary_accuracy: 0.8464\n",
      "Epoch 155/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3499 - binary_accuracy: 0.8528\n",
      "Epoch 156/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3580 - binary_accuracy: 0.8478\n",
      "Epoch 157/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3516 - binary_accuracy: 0.8519\n",
      "Epoch 158/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3524 - binary_accuracy: 0.8517\n",
      "Epoch 159/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3514 - binary_accuracy: 0.8496\n",
      "Epoch 160/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3604 - binary_accuracy: 0.8469\n",
      "Epoch 161/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3469 - binary_accuracy: 0.8539\n",
      "Epoch 162/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3493 - binary_accuracy: 0.8563\n",
      "Epoch 163/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3479 - binary_accuracy: 0.8510\n",
      "Epoch 164/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3520 - binary_accuracy: 0.8518\n",
      "Epoch 165/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3480 - binary_accuracy: 0.8513\n",
      "Epoch 166/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3467 - binary_accuracy: 0.8511\n",
      "Epoch 167/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3548 - binary_accuracy: 0.8460\n",
      "Epoch 168/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3478 - binary_accuracy: 0.8559\n",
      "Epoch 169/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3499 - binary_accuracy: 0.8468\n",
      "Epoch 170/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3503 - binary_accuracy: 0.8531\n",
      "Epoch 171/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3497 - binary_accuracy: 0.8505\n",
      "Epoch 172/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3482 - binary_accuracy: 0.8511\n",
      "Epoch 173/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3531 - binary_accuracy: 0.8551\n",
      "Epoch 174/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3497 - binary_accuracy: 0.8493\n",
      "Epoch 175/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3469 - binary_accuracy: 0.8521\n",
      "Epoch 176/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3475 - binary_accuracy: 0.8536\n",
      "Epoch 177/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3484 - binary_accuracy: 0.8536\n",
      "Epoch 178/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3504 - binary_accuracy: 0.8491\n",
      "Epoch 179/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3458 - binary_accuracy: 0.8578\n",
      "Epoch 180/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3503 - binary_accuracy: 0.8510\n",
      "Epoch 181/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3487 - binary_accuracy: 0.8540\n",
      "Epoch 182/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3534 - binary_accuracy: 0.8529\n",
      "Epoch 183/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3451 - binary_accuracy: 0.8535\n",
      "Epoch 184/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3493 - binary_accuracy: 0.8509\n",
      "Epoch 185/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3479 - binary_accuracy: 0.8531\n",
      "Epoch 186/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3477 - binary_accuracy: 0.8525\n",
      "Epoch 187/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3382 - binary_accuracy: 0.8611\n",
      "Epoch 188/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3350 - binary_accuracy: 0.8592\n",
      "Epoch 189/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3442 - binary_accuracy: 0.8570\n",
      "Epoch 190/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3429 - binary_accuracy: 0.8544\n",
      "Epoch 191/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3473 - binary_accuracy: 0.8491\n",
      "Epoch 192/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3493 - binary_accuracy: 0.8534\n",
      "Epoch 193/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3495 - binary_accuracy: 0.8514\n",
      "Epoch 194/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3427 - binary_accuracy: 0.8536\n",
      "Epoch 195/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3465 - binary_accuracy: 0.8519\n",
      "Epoch 196/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3441 - binary_accuracy: 0.8530\n",
      "Epoch 197/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3479 - binary_accuracy: 0.8502\n",
      "Epoch 198/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3408 - binary_accuracy: 0.8556\n",
      "Epoch 199/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3505 - binary_accuracy: 0.8503\n",
      "Epoch 200/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3425 - binary_accuracy: 0.8564\n",
      "Epoch 201/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3471 - binary_accuracy: 0.8543\n",
      "Epoch 202/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3375 - binary_accuracy: 0.8588\n",
      "Epoch 203/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3473 - binary_accuracy: 0.8577\n",
      "Epoch 204/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3395 - binary_accuracy: 0.8530\n",
      "Epoch 205/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3431 - binary_accuracy: 0.8576\n",
      "Epoch 206/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3445 - binary_accuracy: 0.8553\n",
      "Epoch 207/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3468 - binary_accuracy: 0.8535\n",
      "Epoch 208/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3429 - binary_accuracy: 0.8536\n",
      "Epoch 209/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3435 - binary_accuracy: 0.8509\n",
      "Epoch 210/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3436 - binary_accuracy: 0.8514\n",
      "Epoch 211/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3425 - binary_accuracy: 0.8548\n",
      "Epoch 212/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3369 - binary_accuracy: 0.8579\n",
      "Epoch 213/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3456 - binary_accuracy: 0.8522\n",
      "Epoch 214/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3420 - binary_accuracy: 0.8548\n",
      "Epoch 215/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3443 - binary_accuracy: 0.8518\n",
      "Epoch 216/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3320 - binary_accuracy: 0.8589\n",
      "Epoch 217/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3418 - binary_accuracy: 0.8567\n",
      "Epoch 218/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3348 - binary_accuracy: 0.8567\n",
      "Epoch 219/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3381 - binary_accuracy: 0.8573\n",
      "Epoch 220/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3398 - binary_accuracy: 0.8555\n",
      "Epoch 221/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3402 - binary_accuracy: 0.8601\n",
      "Epoch 222/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3438 - binary_accuracy: 0.8591\n",
      "Epoch 223/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3436 - binary_accuracy: 0.8552\n",
      "Epoch 224/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3357 - binary_accuracy: 0.8553\n",
      "Epoch 225/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3393 - binary_accuracy: 0.8538\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/225\n",
      "1754/1754 [==============================] - 4s 2ms/step - loss: 0.4755 - binary_accuracy: 0.8112\n",
      "Epoch 2/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4413 - binary_accuracy: 0.8188\n",
      "Epoch 3/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4329 - binary_accuracy: 0.8193\n",
      "Epoch 4/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4248 - binary_accuracy: 0.8234\n",
      "Epoch 5/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4296 - binary_accuracy: 0.8204\n",
      "Epoch 6/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4136 - binary_accuracy: 0.8235\n",
      "Epoch 7/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4147 - binary_accuracy: 0.8228\n",
      "Epoch 8/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4122 - binary_accuracy: 0.8247\n",
      "Epoch 9/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4106 - binary_accuracy: 0.8267\n",
      "Epoch 10/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4136 - binary_accuracy: 0.8219\n",
      "Epoch 11/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4139 - binary_accuracy: 0.8229\n",
      "Epoch 12/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4117 - binary_accuracy: 0.8261\n",
      "Epoch 13/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4051 - binary_accuracy: 0.8274\n",
      "Epoch 14/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4089 - binary_accuracy: 0.8285\n",
      "Epoch 15/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4004 - binary_accuracy: 0.8297\n",
      "Epoch 16/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4032 - binary_accuracy: 0.8324\n",
      "Epoch 17/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3990 - binary_accuracy: 0.8330\n",
      "Epoch 18/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4043 - binary_accuracy: 0.8305\n",
      "Epoch 19/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3952 - binary_accuracy: 0.8334\n",
      "Epoch 20/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4023 - binary_accuracy: 0.8311\n",
      "Epoch 21/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3996 - binary_accuracy: 0.8333\n",
      "Epoch 22/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4000 - binary_accuracy: 0.8307\n",
      "Epoch 23/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4018 - binary_accuracy: 0.8310\n",
      "Epoch 24/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3971 - binary_accuracy: 0.8294\n",
      "Epoch 25/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3950 - binary_accuracy: 0.8343\n",
      "Epoch 26/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3951 - binary_accuracy: 0.8338\n",
      "Epoch 27/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3983 - binary_accuracy: 0.8267\n",
      "Epoch 28/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3903 - binary_accuracy: 0.8345\n",
      "Epoch 29/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.4016 - binary_accuracy: 0.8313\n",
      "Epoch 30/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3969 - binary_accuracy: 0.8341\n",
      "Epoch 31/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3952 - binary_accuracy: 0.8315\n",
      "Epoch 32/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3915 - binary_accuracy: 0.8353\n",
      "Epoch 33/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3909 - binary_accuracy: 0.8306\n",
      "Epoch 34/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3920 - binary_accuracy: 0.8303\n",
      "Epoch 35/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3885 - binary_accuracy: 0.8358\n",
      "Epoch 36/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3925 - binary_accuracy: 0.8323\n",
      "Epoch 37/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3961 - binary_accuracy: 0.8299\n",
      "Epoch 38/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3899 - binary_accuracy: 0.8312\n",
      "Epoch 39/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3909 - binary_accuracy: 0.8358\n",
      "Epoch 40/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3827 - binary_accuracy: 0.8385\n",
      "Epoch 41/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3914 - binary_accuracy: 0.8333\n",
      "Epoch 42/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3915 - binary_accuracy: 0.8343\n",
      "Epoch 43/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3925 - binary_accuracy: 0.8346\n",
      "Epoch 44/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3857 - binary_accuracy: 0.8374\n",
      "Epoch 45/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3847 - binary_accuracy: 0.8363\n",
      "Epoch 46/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3870 - binary_accuracy: 0.8368\n",
      "Epoch 47/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3884 - binary_accuracy: 0.8362\n",
      "Epoch 48/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3812 - binary_accuracy: 0.8378\n",
      "Epoch 49/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3856 - binary_accuracy: 0.8350\n",
      "Epoch 50/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3829 - binary_accuracy: 0.8390\n",
      "Epoch 51/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3814 - binary_accuracy: 0.8372\n",
      "Epoch 52/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3787 - binary_accuracy: 0.8354\n",
      "Epoch 53/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3844 - binary_accuracy: 0.8343\n",
      "Epoch 54/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3873 - binary_accuracy: 0.8324\n",
      "Epoch 55/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3833 - binary_accuracy: 0.8377\n",
      "Epoch 56/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3819 - binary_accuracy: 0.8359\n",
      "Epoch 57/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3776 - binary_accuracy: 0.8384\n",
      "Epoch 58/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3797 - binary_accuracy: 0.8414\n",
      "Epoch 59/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3823 - binary_accuracy: 0.8372\n",
      "Epoch 60/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3816 - binary_accuracy: 0.8348\n",
      "Epoch 61/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3823 - binary_accuracy: 0.8347\n",
      "Epoch 62/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3770 - binary_accuracy: 0.8362\n",
      "Epoch 63/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3767 - binary_accuracy: 0.8404\n",
      "Epoch 64/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3798 - binary_accuracy: 0.8393\n",
      "Epoch 65/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3797 - binary_accuracy: 0.8390\n",
      "Epoch 66/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3783 - binary_accuracy: 0.8411\n",
      "Epoch 67/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3797 - binary_accuracy: 0.8370\n",
      "Epoch 68/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3757 - binary_accuracy: 0.8394\n",
      "Epoch 69/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3847 - binary_accuracy: 0.8302\n",
      "Epoch 70/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3762 - binary_accuracy: 0.8395\n",
      "Epoch 71/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3761 - binary_accuracy: 0.8358\n",
      "Epoch 72/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3721 - binary_accuracy: 0.8393\n",
      "Epoch 73/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3816 - binary_accuracy: 0.8337\n",
      "Epoch 74/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3787 - binary_accuracy: 0.8352\n",
      "Epoch 75/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3753 - binary_accuracy: 0.8377\n",
      "Epoch 76/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3806 - binary_accuracy: 0.8372\n",
      "Epoch 77/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3726 - binary_accuracy: 0.8400\n",
      "Epoch 78/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3765 - binary_accuracy: 0.8435\n",
      "Epoch 79/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3795 - binary_accuracy: 0.8408\n",
      "Epoch 80/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3752 - binary_accuracy: 0.8396\n",
      "Epoch 81/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3755 - binary_accuracy: 0.8420\n",
      "Epoch 82/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3726 - binary_accuracy: 0.8427\n",
      "Epoch 83/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3741 - binary_accuracy: 0.8424\n",
      "Epoch 84/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3699 - binary_accuracy: 0.8376\n",
      "Epoch 85/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3742 - binary_accuracy: 0.8397\n",
      "Epoch 86/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3714 - binary_accuracy: 0.8398\n",
      "Epoch 87/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3738 - binary_accuracy: 0.8436\n",
      "Epoch 88/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3735 - binary_accuracy: 0.8399\n",
      "Epoch 89/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3704 - binary_accuracy: 0.8424\n",
      "Epoch 90/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3673 - binary_accuracy: 0.8444\n",
      "Epoch 91/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3732 - binary_accuracy: 0.8409\n",
      "Epoch 92/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3697 - binary_accuracy: 0.8415\n",
      "Epoch 93/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3754 - binary_accuracy: 0.8381\n",
      "Epoch 94/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3717 - binary_accuracy: 0.8415\n",
      "Epoch 95/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3652 - binary_accuracy: 0.8414\n",
      "Epoch 96/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3710 - binary_accuracy: 0.8334\n",
      "Epoch 97/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3727 - binary_accuracy: 0.8396\n",
      "Epoch 98/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3666 - binary_accuracy: 0.8444\n",
      "Epoch 99/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3722 - binary_accuracy: 0.8396\n",
      "Epoch 100/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3727 - binary_accuracy: 0.8334\n",
      "Epoch 101/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3648 - binary_accuracy: 0.8427\n",
      "Epoch 102/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3712 - binary_accuracy: 0.8381\n",
      "Epoch 103/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3739 - binary_accuracy: 0.8397\n",
      "Epoch 104/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3612 - binary_accuracy: 0.8480\n",
      "Epoch 105/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3699 - binary_accuracy: 0.8412\n",
      "Epoch 106/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3672 - binary_accuracy: 0.8439\n",
      "Epoch 107/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3676 - binary_accuracy: 0.8400\n",
      "Epoch 108/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3674 - binary_accuracy: 0.8417\n",
      "Epoch 109/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3652 - binary_accuracy: 0.8469\n",
      "Epoch 110/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3674 - binary_accuracy: 0.8419\n",
      "Epoch 111/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3625 - binary_accuracy: 0.8417\n",
      "Epoch 112/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3685 - binary_accuracy: 0.8438\n",
      "Epoch 113/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3628 - binary_accuracy: 0.8413\n",
      "Epoch 114/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3615 - binary_accuracy: 0.8437\n",
      "Epoch 115/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3652 - binary_accuracy: 0.8440\n",
      "Epoch 116/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3675 - binary_accuracy: 0.8432\n",
      "Epoch 117/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3661 - binary_accuracy: 0.8406\n",
      "Epoch 118/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3619 - binary_accuracy: 0.8457\n",
      "Epoch 119/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3649 - binary_accuracy: 0.8428\n",
      "Epoch 120/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3657 - binary_accuracy: 0.8413\n",
      "Epoch 121/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3705 - binary_accuracy: 0.8374\n",
      "Epoch 122/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3610 - binary_accuracy: 0.8454\n",
      "Epoch 123/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3653 - binary_accuracy: 0.8421\n",
      "Epoch 124/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3559 - binary_accuracy: 0.8488\n",
      "Epoch 125/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3641 - binary_accuracy: 0.8456\n",
      "Epoch 126/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3637 - binary_accuracy: 0.8409\n",
      "Epoch 127/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3582 - binary_accuracy: 0.8495\n",
      "Epoch 128/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3693 - binary_accuracy: 0.8371\n",
      "Epoch 129/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3638 - binary_accuracy: 0.8490\n",
      "Epoch 130/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3586 - binary_accuracy: 0.8471\n",
      "Epoch 131/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3615 - binary_accuracy: 0.8460\n",
      "Epoch 132/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3584 - binary_accuracy: 0.8490\n",
      "Epoch 133/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3643 - binary_accuracy: 0.8422\n",
      "Epoch 134/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3611 - binary_accuracy: 0.8452\n",
      "Epoch 135/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3573 - binary_accuracy: 0.8441\n",
      "Epoch 136/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3630 - binary_accuracy: 0.8409\n",
      "Epoch 137/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3631 - binary_accuracy: 0.8445\n",
      "Epoch 138/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3572 - binary_accuracy: 0.8434\n",
      "Epoch 139/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3626 - binary_accuracy: 0.8487\n",
      "Epoch 140/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3626 - binary_accuracy: 0.8419\n",
      "Epoch 141/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3578 - binary_accuracy: 0.8451\n",
      "Epoch 142/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3592 - binary_accuracy: 0.8436\n",
      "Epoch 143/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3587 - binary_accuracy: 0.8479\n",
      "Epoch 144/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3551 - binary_accuracy: 0.8479\n",
      "Epoch 145/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3601 - binary_accuracy: 0.8434\n",
      "Epoch 146/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3523 - binary_accuracy: 0.8511\n",
      "Epoch 147/225\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3596 - binary_accuracy: 0.8479\n",
      "Epoch 148/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3562 - binary_accuracy: 0.8484\n",
      "Epoch 149/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3522 - binary_accuracy: 0.8505\n",
      "Epoch 150/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3569 - binary_accuracy: 0.8483\n",
      "Epoch 151/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3515 - binary_accuracy: 0.8484\n",
      "Epoch 152/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3569 - binary_accuracy: 0.8439\n",
      "Epoch 153/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3530 - binary_accuracy: 0.8510\n",
      "Epoch 154/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3592 - binary_accuracy: 0.8460\n",
      "Epoch 155/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3561 - binary_accuracy: 0.8501\n",
      "Epoch 156/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3581 - binary_accuracy: 0.8460\n",
      "Epoch 157/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3457 - binary_accuracy: 0.8519\n",
      "Epoch 158/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3595 - binary_accuracy: 0.8465\n",
      "Epoch 159/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3618 - binary_accuracy: 0.8471\n",
      "Epoch 160/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3495 - binary_accuracy: 0.8506\n",
      "Epoch 161/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3500 - binary_accuracy: 0.8493\n",
      "Epoch 162/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3497 - binary_accuracy: 0.8543\n",
      "Epoch 163/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3533 - binary_accuracy: 0.8433\n",
      "Epoch 164/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3596 - binary_accuracy: 0.8439\n",
      "Epoch 165/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3530 - binary_accuracy: 0.8460\n",
      "Epoch 166/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3551 - binary_accuracy: 0.8436\n",
      "Epoch 167/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3484 - binary_accuracy: 0.8491\n",
      "Epoch 168/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3503 - binary_accuracy: 0.8519\n",
      "Epoch 169/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3489 - binary_accuracy: 0.8518\n",
      "Epoch 170/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3511 - binary_accuracy: 0.8452\n",
      "Epoch 171/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3556 - binary_accuracy: 0.8490\n",
      "Epoch 172/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3527 - binary_accuracy: 0.8485\n",
      "Epoch 173/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3459 - binary_accuracy: 0.8490\n",
      "Epoch 174/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3512 - binary_accuracy: 0.8484\n",
      "Epoch 175/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3489 - binary_accuracy: 0.8522\n",
      "Epoch 176/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3510 - binary_accuracy: 0.8494\n",
      "Epoch 177/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3495 - binary_accuracy: 0.8470\n",
      "Epoch 178/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3503 - binary_accuracy: 0.8471\n",
      "Epoch 179/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3499 - binary_accuracy: 0.8498\n",
      "Epoch 180/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3493 - binary_accuracy: 0.8493\n",
      "Epoch 181/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3570 - binary_accuracy: 0.8432\n",
      "Epoch 182/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3483 - binary_accuracy: 0.8523\n",
      "Epoch 183/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3495 - binary_accuracy: 0.8503\n",
      "Epoch 184/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3480 - binary_accuracy: 0.8477\n",
      "Epoch 185/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3454 - binary_accuracy: 0.8533\n",
      "Epoch 186/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3532 - binary_accuracy: 0.8468\n",
      "Epoch 187/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3441 - binary_accuracy: 0.8548\n",
      "Epoch 188/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3471 - binary_accuracy: 0.8495\n",
      "Epoch 189/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3518 - binary_accuracy: 0.8507\n",
      "Epoch 190/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3424 - binary_accuracy: 0.8557\n",
      "Epoch 191/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3483 - binary_accuracy: 0.8530\n",
      "Epoch 192/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3499 - binary_accuracy: 0.8491\n",
      "Epoch 193/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3536 - binary_accuracy: 0.8524\n",
      "Epoch 194/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3399 - binary_accuracy: 0.8591\n",
      "Epoch 195/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3445 - binary_accuracy: 0.8529\n",
      "Epoch 196/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3425 - binary_accuracy: 0.8538\n",
      "Epoch 197/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3418 - binary_accuracy: 0.8540\n",
      "Epoch 198/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3456 - binary_accuracy: 0.8480\n",
      "Epoch 199/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3429 - binary_accuracy: 0.8517\n",
      "Epoch 200/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3427 - binary_accuracy: 0.8501\n",
      "Epoch 201/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3480 - binary_accuracy: 0.8504\n",
      "Epoch 202/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3429 - binary_accuracy: 0.8511\n",
      "Epoch 203/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3524 - binary_accuracy: 0.8494\n",
      "Epoch 204/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3498 - binary_accuracy: 0.8484\n",
      "Epoch 205/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3418 - binary_accuracy: 0.8517\n",
      "Epoch 206/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3435 - binary_accuracy: 0.8526\n",
      "Epoch 207/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3411 - binary_accuracy: 0.8557\n",
      "Epoch 208/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3525 - binary_accuracy: 0.8481\n",
      "Epoch 209/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3468 - binary_accuracy: 0.8486\n",
      "Epoch 210/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3397 - binary_accuracy: 0.8511\n",
      "Epoch 211/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3448 - binary_accuracy: 0.8473\n",
      "Epoch 212/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3416 - binary_accuracy: 0.8509\n",
      "Epoch 213/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3421 - binary_accuracy: 0.8551\n",
      "Epoch 214/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3463 - binary_accuracy: 0.8515\n",
      "Epoch 215/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3409 - binary_accuracy: 0.8506\n",
      "Epoch 216/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3389 - binary_accuracy: 0.8550\n",
      "Epoch 217/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3467 - binary_accuracy: 0.8472\n",
      "Epoch 218/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3380 - binary_accuracy: 0.8535\n",
      "Epoch 219/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3461 - binary_accuracy: 0.8543\n",
      "Epoch 220/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3420 - binary_accuracy: 0.8495\n",
      "Epoch 221/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3412 - binary_accuracy: 0.8494\n",
      "Epoch 222/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3418 - binary_accuracy: 0.8581\n",
      "Epoch 223/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3380 - binary_accuracy: 0.8524\n",
      "Epoch 224/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3454 - binary_accuracy: 0.8532\n",
      "Epoch 225/225\n",
      "1754/1754 [==============================] - 2s 1ms/step - loss: 0.3421 - binary_accuracy: 0.8577\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Run 5-fold cross-validation to get average performance characteristics\n",
    "unit = 32\n",
    "ratio = 0.5\n",
    "epoch = 225\n",
    "(auroc_list, other_metric_list) = kfold_validation(unit, ratio, epoch, X_train_transformed, y_train, 'patient-id', \\\n",
    "                                                   n_features=n_features, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "average precision:0.3329845008463798\n",
      "average recall:0.5997828431790834\n",
      "average specificity:0.7283631645843947\n",
      "average F1:0.42320472896434874\n"
     ]
    }
   ],
   "source": [
    "## other_metric_list has precision, recall, accuracy, specificity, f1, threshold for metrics\n",
    "ave_prec = np.mean([x[0] for x in other_metric_list])\n",
    "print(\"average precision:\" + str(ave_prec))\n",
    "\n",
    "ave_recall = np.mean([x[1] for x in other_metric_list])\n",
    "print(\"average recall:\" + str(ave_recall))\n",
    "\n",
    "ave_specificity = np.mean([x[3] for x in other_metric_list])\n",
    "print(\"average specificity:\" + str(ave_specificity))\n",
    "\n",
    "ave_f1 = np.mean([x[4] for x in other_metric_list])\n",
    "print(\"average F1:\" + str(ave_f1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max F1: 0.4504132231404959\n"
     ]
    }
   ],
   "source": [
    "max_f1 = np.max([x[4] for x in other_metric_list])\n",
    "print(\"max F1: \" + str(max_f1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "unit: 32\tepoch: 225\tauroc_mean: 0.705206\tauroc_std: 0.015318\n"
     ]
    }
   ],
   "source": [
    "## Average metrics performance\n",
    "print(\"unit: %d\\tepoch: %d\\tauroc_mean: %f\\tauroc_std: %f\"  % (unit, epoch, np.mean(auroc_list), np.std(auroc_list)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}