Models.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "!pip install tensorflow-addons"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_h549-HrlLe_",
        "outputId": "ad503cc0-ea07-4953-f3bb-e6515bb21f3d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting tensorflow-addons\n",
            "  Downloading tensorflow_addons-0.15.0-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (1.1 MB)\n",
            "\u001b[?25l\r\u001b[K     |▎                               | 10 kB 32.9 MB/s eta 0:00:01\r\u001b[K     |▋                               | 20 kB 36.8 MB/s eta 0:00:01\r\u001b[K     |▉                               | 30 kB 42.4 MB/s eta 0:00:01\r\u001b[K     |█▏                              | 40 kB 44.9 MB/s eta 0:00:01\r\u001b[K     |█▌                              | 51 kB 33.6 MB/s eta 0:00:01\r\u001b[K     |█▊                              | 61 kB 36.3 MB/s eta 0:00:01\r\u001b[K     |██                              | 71 kB 24.4 MB/s eta 0:00:01\r\u001b[K     |██▍                             | 81 kB 26.4 MB/s eta 0:00:01\r\u001b[K     |██▋                             | 92 kB 28.3 MB/s eta 0:00:01\r\u001b[K     |███                             | 102 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███▎                            | 112 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███▌                            | 122 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███▉                            | 133 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████▏                           | 143 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████▍                           | 153 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████▊                           | 163 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████                           | 174 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████▎                          | 184 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████▋                          | 194 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████▉                          | 204 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████▏                         | 215 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████▌                         | 225 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████▊                         | 235 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████                         | 245 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████▍                        | 256 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████▋                        | 266 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████                        | 276 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████▎                       | 286 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████▌                       | 296 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████▉                       | 307 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████                       | 317 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████▍                      | 327 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████▊                      | 337 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████                      | 348 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████▎                     | 358 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████▋                     | 368 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████▉                     | 378 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████▏                    | 389 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████▌                    | 399 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████▊                    | 409 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████                    | 419 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████▍                   | 430 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████▋                   | 440 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████                   | 450 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████▎                  | 460 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████▌                  | 471 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████▉                  | 481 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████                  | 491 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████▍                 | 501 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████▊                 | 512 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████                 | 522 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████▎                | 532 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████▋                | 542 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████▉                | 552 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████▏               | 563 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████▌               | 573 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████▊               | 583 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████               | 593 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████▍              | 604 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████▋              | 614 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████              | 624 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████▏             | 634 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████▌             | 645 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████▉             | 655 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████             | 665 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████▍            | 675 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████▊            | 686 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████            | 696 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████▎           | 706 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████▋           | 716 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████▉           | 727 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████▏          | 737 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████▌          | 747 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████▊          | 757 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████          | 768 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████▎         | 778 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████▋         | 788 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████         | 798 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████▏        | 808 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████▌        | 819 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████▉        | 829 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████████        | 839 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████████▍       | 849 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████████▊       | 860 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████       | 870 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▎      | 880 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▋      | 890 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████▉      | 901 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▏     | 911 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▌     | 921 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▊     | 931 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████     | 942 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▎    | 952 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▋    | 962 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████    | 972 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▏   | 983 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▌   | 993 kB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▉   | 1.0 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████   | 1.0 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▍  | 1.0 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▊  | 1.0 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████  | 1.0 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▎ | 1.1 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▋ | 1.1 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |██████████████████████████████▉ | 1.1 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▏| 1.1 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▍| 1.1 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▊| 1.1 MB 30.0 MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 1.1 MB 30.0 MB/s \n",
            "\u001b[?25hRequirement already satisfied: typeguard>=2.7 in /usr/local/lib/python3.7/dist-packages (from tensorflow-addons) (2.7.1)\n",
            "Installing collected packages: tensorflow-addons\n",
            "Successfully installed tensorflow-addons-0.15.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "PygS3ePUXc2U"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import tensorflow as tf\n",
        "from sklearn.model_selection import train_test_split\n",
        "import warnings\n",
        "import tensorflow_addons as tfa\n",
        "warnings.filterwarnings('ignore')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sDtxy0HXXc2f"
      },
      "source": [
        "## Read datasets and labels"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TV_yNRS6Xc2q",
        "outputId": "192e9708-7fcd-4de5-d2af-d89bce4b5ab1"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(86480, 41) (103094, 41)\n"
          ]
        }
      ],
      "source": [
        "df = pd.read_csv(\"./attack_dataset.csv\") # attack dataset\n",
        "attack_classes = pd.read_csv(\"./attack_labels.csv\") # labels\n",
        "bonafide = pd.read_csv('./bonafide_dataset_20191121.csv.gz') # bonafide dataset\n",
        "print(df.shape, bonafide.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8toVJbI9Xc21"
      },
      "source": [
        "### Attack labels"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iZlOSW1JXc27",
        "outputId": "59f03a7f-ff9b-4297-f955-957bef2697f6"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['nmap_tcp_syn', 'nmap_tcp_conn', 'nmap_tcp_null', 'nmap_tcp_xmas',\n",
              "       'nmap_tcp_fin', 'nmap_tcp_ack', 'nmap_tcp_window',\n",
              "       'nmap_tcp_maimon', 'unicornscan_tcp_syn', 'unicornscan_tcp_conn',\n",
              "       'unicornscan_tcp_null', 'unicornscan_tcp_xmas',\n",
              "       'unicornscan_tcp_fxmas', 'unicornscan_tcp_fin',\n",
              "       'unicornscan_tcp_ack', 'hping_tcp_syn', 'hping_tcp_null',\n",
              "       'hping_tcp_xmas', 'hping_tcp_fin', 'hping_tcp_ack', 'zmap_tcp_syn',\n",
              "       'masscan_tcp_syn', 'nmap_ping_scan', 'nmap_vvv', 'nmap_connect',\n",
              "       'nmap_fast', 'nmap_servinfo', 'nmap_reason', 'nmap_open',\n",
              "       'nmap_top10', 'nmap_fragv', 'nmap_mtu', 'nmap_spoof_idle',\n",
              "       'nmap_decoy', 'nmap_noping', 'nmap_udping', 'nmap_ack_syn',\n",
              "       'nmap_syn', 'nmap_ipproto', 'nmap_udp', 'nmap_null', 'nmap_fin',\n",
              "       'nmap_stealth', 'nmap_spoof_mac', 'nmap_fake_srcprt',\n",
              "       'nmap_data_length', 'nmap_bad_checksum', 'nmap_random_host',\n",
              "       'nmap_iflist', 'nmap_fw_bypass', 'nmap_firewalk',\n",
              "       'nmap_http_methods', 'nmap_rpcinfo', 'nmap_banner',\n",
              "       'nmap_ajp_methods', 'nmap_ms_sql', 'nmap_vhosts',\n",
              "       'nmap_http_headers', 'nmap_http_tamper', 'nmap_db2_das_info',\n",
              "       'nmap_enip_info', 'nmap_smb_stat', 'nmap_waf_detect',\n",
              "       'nmap_waf_fingerprint', 'nmap_smb_enum', 'nmap_affiliate_id',\n",
              "       'nmap_cics_info'], dtype=object)"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ],
      "source": [
        "attack_classes['label'].unique()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xeJr3GYZXc2_"
      },
      "source": [
        "### Create **attack dataset** with labels defined by  *ip_src*"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 246
        },
        "id": "eSUNNQXnXc3B",
        "outputId": "8df91115-266d-478a-d3d1-59e89b374266"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-47c3d264-6098-4104-b624-f8ffef1da569\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>frame_info.encap_type</th>\n",
              "      <th>frame_info.time</th>\n",
              "      <th>frame_info.time_epoch</th>\n",
              "      <th>frame_info.number</th>\n",
              "      <th>frame_info.len</th>\n",
              "      <th>frame_info.cap_len</th>\n",
              "      <th>eth.type</th>\n",
              "      <th>ip.version</th>\n",
              "      <th>ip.hdr_len</th>\n",
              "      <th>ip.tos</th>\n",
              "      <th>ip.id</th>\n",
              "      <th>ip.flags</th>\n",
              "      <th>ip.flags.rb</th>\n",
              "      <th>ip.flags.df</th>\n",
              "      <th>ip.flags.mf</th>\n",
              "      <th>ip.frag_offset</th>\n",
              "      <th>ip.ttl</th>\n",
              "      <th>ip.proto</th>\n",
              "      <th>ip.checksum</th>\n",
              "      <th>ip.src</th>\n",
              "      <th>ip.dst</th>\n",
              "      <th>ip.len</th>\n",
              "      <th>ip.dsfield</th>\n",
              "      <th>tcp.srcport</th>\n",
              "      <th>tcp.dstport</th>\n",
              "      <th>tcp.seq</th>\n",
              "      <th>tcp.ack</th>\n",
              "      <th>tcp.len</th>\n",
              "      <th>tcp.hdr_len</th>\n",
              "      <th>tcp.flags</th>\n",
              "      <th>tcp.flags.fin</th>\n",
              "      <th>tcp.flags.syn</th>\n",
              "      <th>tcp.flags.reset</th>\n",
              "      <th>tcp.flags.push</th>\n",
              "      <th>tcp.flags.ack</th>\n",
              "      <th>tcp.flags.urg</th>\n",
              "      <th>tcp.flags.cwr</th>\n",
              "      <th>tcp.window_size</th>\n",
              "      <th>tcp.checksum</th>\n",
              "      <th>tcp.urgent_pointer</th>\n",
              "      <th>tcp.options.mss_val</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>Dec 31, 1969 21:03:41.953641000 -03</td>\n",
              "      <td>221.953641</td>\n",
              "      <td>20</td>\n",
              "      <td>58</td>\n",
              "      <td>58</td>\n",
              "      <td>0x00000800</td>\n",
              "      <td>4</td>\n",
              "      <td>20</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0x0000b2c5</td>\n",
              "      <td>0x00000000</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>46</td>\n",
              "      <td>6</td>\n",
              "      <td>0x000019e0</td>\n",
              "      <td>172.16.0.3</td>\n",
              "      <td>10.10.10.10</td>\n",
              "      <td>44</td>\n",
              "      <td>0x00000000</td>\n",
              "      <td>62218</td>\n",
              "      <td>8888</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>24</td>\n",
              "      <td>0x00000002</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1024</td>\n",
              "      <td>0x00005cb1</td>\n",
              "      <td>0</td>\n",
              "      <td>1460.0</td>\n",
              "      <td>nmap_tcp_syn</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>Dec 31, 1969 21:03:41.953762000 -03</td>\n",
              "      <td>221.953762</td>\n",
              "      <td>21</td>\n",
              "      <td>58</td>\n",
              "      <td>58</td>\n",
              "      <td>0x00000800</td>\n",
              "      <td>4</td>\n",
              "      <td>20</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0x0000d3a4</td>\n",
              "      <td>0x00000000</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>37</td>\n",
              "      <td>6</td>\n",
              "      <td>0x00000201</td>\n",
              "      <td>172.16.0.3</td>\n",
              "      <td>10.10.10.10</td>\n",
              "      <td>44</td>\n",
              "      <td>0x00000000</td>\n",
              "      <td>62218</td>\n",
              "      <td>587</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>24</td>\n",
              "      <td>0x00000002</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1024</td>\n",
              "      <td>0x00007d1e</td>\n",
              "      <td>0</td>\n",
              "      <td>1460.0</td>\n",
              "      <td>nmap_tcp_syn</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-47c3d264-6098-4104-b624-f8ffef1da569')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-47c3d264-6098-4104-b624-f8ffef1da569 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-47c3d264-6098-4104-b624-f8ffef1da569');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   frame_info.encap_type  ...         label\n",
              "0                      1  ...  nmap_tcp_syn\n",
              "1                      1  ...  nmap_tcp_syn\n",
              "\n",
              "[2 rows x 42 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ],
      "source": [
        "df_labeled = df.merge(attack_classes, how='inner', left_on='ip.src', right_on='ip')\n",
        "df_labeled.drop(['ip'], axis=1, inplace=True)\n",
        "df_labeled.head(2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ig8coXcQXc3E"
      },
      "outputs": [],
      "source": [
        "bonafide['label'] = \"bonafide\" # create column label on bonafide dataset"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EjpawwyxXc3G"
      },
      "source": [
        "## Comparing _bonafide_ and _attack_ datasets\n",
        "> Check if both dataset have the same columns/features/attributes and calculate the balance (%) of bonafide and attack packets"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4tptYLf4Xc3N",
        "outputId": "bb706456-ef64-4a0c-8b88-4bbe15163d67"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Total examples of 140163 with 0.26 of attack and 0.74 bonafide packets\n"
          ]
        }
      ],
      "source": [
        "if (df_labeled.columns == bonafide.columns).all():\n",
        "    examples_attack = df_labeled.shape[0]\n",
        "    examples_bonafide = bonafide.shape[0]\n",
        "    total = examples_attack+examples_bonafide\n",
        "    print('Total examples of {0} with {1:0.2f} of attack and {2:0.2f} bonafide packets'.format(total, examples_attack/total, examples_bonafide/total))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KGeskRvtXc3T"
      },
      "source": [
        "## Pre-processing\n",
        "> Required to convert features that were extracted from `pcap` as hexadecimal\n",
        "> Filled the values with NaN with Zero"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "KPrHy86WXc3Y"
      },
      "outputs": [],
      "source": [
        "fields = ['eth.type', 'ip.id', 'ip.flags', 'ip.checksum', 'ip.dsfield', 'tcp.flags', 'tcp.checksum']\n",
        "\n",
        "for field in fields:\n",
        "    df_labeled[field] = df_labeled[field].apply(lambda x: int(str(x), 16))\n",
        "    "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "SpQMNRysXc3c"
      },
      "outputs": [],
      "source": [
        "bonafide = bonafide.fillna(0)\n",
        "for field in fields:\n",
        "    bonafide[field] = bonafide[field].apply(lambda x: int(str(x), 16))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IkcvUxNhXc3d"
      },
      "source": [
        "### Create AB-TRAP dataset with all packets (bonafide and attack)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6V9upradXc3g"
      },
      "outputs": [],
      "source": [
        "full_data = pd.concat([bonafide, df_labeled])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dCP33m3FXc3i"
      },
      "source": [
        "### Check if there are packets with the protocol field different than TCP (value 6)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hWJF1ZVfXc3m",
        "outputId": "099e67aa-efa1-4789-abb1-101a40c204c2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "It was found and removed [11708] packets.\n"
          ]
        }
      ],
      "source": [
        "wrong_proto = full_data[full_data['ip.proto'] != 6]['label'].value_counts().values\n",
        "full_data = full_data[full_data['ip.proto'] == 6]\n",
        "print(\"It was found and removed\", wrong_proto,\"packets.\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TLFFSEHrXc3r"
      },
      "source": [
        "### Features not applicable to this study\n",
        "\n",
        "> Remove features from link layer - layer 2:\n",
        "> - frame_info.time\n",
        "> - frame_info.encap_type\n",
        "> - frame_info.time_epoch\n",
        "> - frame_info.number\n",
        "> - frame_info.len\n",
        "> - frame_info.cap_len\n",
        "> - eth.type\n",
        "\n",
        "> Remove features that are redundant or invariable\n",
        "> - ip.version - we consider only IPv4\n",
        "> - ip.proto - this study is applicable only to TCP\n",
        "> - ip.src - this attribute is removed to allow the generalization of learning (not learn past attackers)\n",
        "> - ip.dst - this attribute is removed to allow the generalization of learning (not learn past targets)\n",
        "> - ip.flags - this is removed because we use bit-set of flags\n",
        "> - tcp.flags - this is removed because we use bit-set of flags"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "WzF1OjQ2Xc3s"
      },
      "outputs": [],
      "source": [
        "full_data.drop(columns=['frame_info.time', 'frame_info.encap_type', 'frame_info.time_epoch', 'frame_info.number', \n",
        "                        'frame_info.len', 'frame_info.cap_len', 'eth.type', 'ip.flags', 'ip.src', 'ip.dst',\n",
        "                        'ip.version', 'ip.proto', 'tcp.flags'], axis=1, inplace=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pthifNLqXc3t"
      },
      "source": [
        "## Exploratory Data Analysis"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PzcqlzXlXc3v",
        "outputId": "f118f34d-eb82-47f4-b971-3e02e47521bc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 128455 entries, 1 to 37068\n",
            "Data columns (total 29 columns):\n",
            " #   Column               Non-Null Count   Dtype  \n",
            "---  ------               --------------   -----  \n",
            " 0   ip.hdr_len           128455 non-null  float64\n",
            " 1   ip.tos               91386 non-null   float64\n",
            " 2   ip.id                128455 non-null  int64  \n",
            " 3   ip.flags.rb          128455 non-null  float64\n",
            " 4   ip.flags.df          128455 non-null  float64\n",
            " 5   ip.flags.mf          128455 non-null  float64\n",
            " 6   ip.frag_offset       128455 non-null  float64\n",
            " 7   ip.ttl               128455 non-null  float64\n",
            " 8   ip.checksum          128455 non-null  int64  \n",
            " 9   ip.len               128455 non-null  float64\n",
            " 10  ip.dsfield           128455 non-null  int64  \n",
            " 11  tcp.srcport          128455 non-null  float64\n",
            " 12  tcp.dstport          128455 non-null  float64\n",
            " 13  tcp.seq              128455 non-null  float64\n",
            " 14  tcp.ack              128455 non-null  float64\n",
            " 15  tcp.len              128455 non-null  float64\n",
            " 16  tcp.hdr_len          128455 non-null  float64\n",
            " 17  tcp.flags.fin        128455 non-null  float64\n",
            " 18  tcp.flags.syn        128455 non-null  float64\n",
            " 19  tcp.flags.reset      128455 non-null  float64\n",
            " 20  tcp.flags.push       128455 non-null  float64\n",
            " 21  tcp.flags.ack        128455 non-null  float64\n",
            " 22  tcp.flags.urg        128455 non-null  float64\n",
            " 23  tcp.flags.cwr        128455 non-null  float64\n",
            " 24  tcp.window_size      128455 non-null  float64\n",
            " 25  tcp.checksum         128455 non-null  int64  \n",
            " 26  tcp.urgent_pointer   128455 non-null  float64\n",
            " 27  tcp.options.mss_val  115134 non-null  float64\n",
            " 28  label                128455 non-null  object \n",
            "dtypes: float64(24), int64(4), object(1)\n",
            "memory usage: 29.4+ MB\n"
          ]
        }
      ],
      "source": [
        "full_data.info()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ja86Zl9_Xc3w",
        "outputId": "18b5f0e4-c6c3-46a5-ff25-894babd31c83"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-b8a73969-8221-442f-9474-5ed38abb9f4e\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>ip.hdr_len</th>\n",
              "      <th>ip.tos</th>\n",
              "      <th>ip.id</th>\n",
              "      <th>ip.flags.rb</th>\n",
              "      <th>ip.flags.df</th>\n",
              "      <th>ip.flags.mf</th>\n",
              "      <th>ip.frag_offset</th>\n",
              "      <th>ip.ttl</th>\n",
              "      <th>ip.checksum</th>\n",
              "      <th>ip.len</th>\n",
              "      <th>ip.dsfield</th>\n",
              "      <th>tcp.srcport</th>\n",
              "      <th>tcp.dstport</th>\n",
              "      <th>tcp.seq</th>\n",
              "      <th>tcp.ack</th>\n",
              "      <th>tcp.len</th>\n",
              "      <th>tcp.hdr_len</th>\n",
              "      <th>tcp.flags.fin</th>\n",
              "      <th>tcp.flags.syn</th>\n",
              "      <th>tcp.flags.reset</th>\n",
              "      <th>tcp.flags.push</th>\n",
              "      <th>tcp.flags.ack</th>\n",
              "      <th>tcp.flags.urg</th>\n",
              "      <th>tcp.flags.cwr</th>\n",
              "      <th>tcp.window_size</th>\n",
              "      <th>tcp.checksum</th>\n",
              "      <th>tcp.urgent_pointer</th>\n",
              "      <th>tcp.options.mss_val</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>128455.0</td>\n",
              "      <td>91386.0</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.0</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.0</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>1.284550e+05</td>\n",
              "      <td>1.284550e+05</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>1.284550e+05</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>128455.000000</td>\n",
              "      <td>115134.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>31527.863213</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.563388</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.086209</td>\n",
              "      <td>117.020287</td>\n",
              "      <td>32240.725328</td>\n",
              "      <td>562.270780</td>\n",
              "      <td>1.992293</td>\n",
              "      <td>28216.955751</td>\n",
              "      <td>26468.931314</td>\n",
              "      <td>3.553362e+07</td>\n",
              "      <td>3.624237e+07</td>\n",
              "      <td>515.677405</td>\n",
              "      <td>27.283049</td>\n",
              "      <td>0.068327</td>\n",
              "      <td>0.387435</td>\n",
              "      <td>0.019991</td>\n",
              "      <td>0.083850</td>\n",
              "      <td>0.548449</td>\n",
              "      <td>0.031365</td>\n",
              "      <td>0.008509</td>\n",
              "      <td>8.907874e+03</td>\n",
              "      <td>32449.793539</td>\n",
              "      <td>0.678744</td>\n",
              "      <td>359.404329</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>19605.220369</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.495968</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.406187</td>\n",
              "      <td>85.963373</td>\n",
              "      <td>19031.578341</td>\n",
              "      <td>676.602078</td>\n",
              "      <td>7.723105</td>\n",
              "      <td>24983.803880</td>\n",
              "      <td>24657.064388</td>\n",
              "      <td>1.055214e+08</td>\n",
              "      <td>2.299113e+08</td>\n",
              "      <td>674.104446</td>\n",
              "      <td>7.455100</td>\n",
              "      <td>0.252308</td>\n",
              "      <td>0.487166</td>\n",
              "      <td>0.139971</td>\n",
              "      <td>0.277164</td>\n",
              "      <td>0.497649</td>\n",
              "      <td>0.174303</td>\n",
              "      <td>0.091850</td>\n",
              "      <td>7.518775e+04</td>\n",
              "      <td>19183.397913</td>\n",
              "      <td>171.836116</td>\n",
              "      <td>636.972162</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>24.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>21.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000e+00</td>\n",
              "      <td>0.000000e+00</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>20.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000e+00</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>14285.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>56.000000</td>\n",
              "      <td>15366.500000</td>\n",
              "      <td>40.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>443.000000</td>\n",
              "      <td>1038.000000</td>\n",
              "      <td>0.000000e+00</td>\n",
              "      <td>0.000000e+00</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>20.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>2.430000e+02</td>\n",
              "      <td>15436.500000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>31282.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>62.000000</td>\n",
              "      <td>31893.000000</td>\n",
              "      <td>52.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>34995.000000</td>\n",
              "      <td>16993.000000</td>\n",
              "      <td>1.000000e+00</td>\n",
              "      <td>1.000000e+00</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>24.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.024000e+03</td>\n",
              "      <td>32507.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>48886.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>242.000000</td>\n",
              "      <td>48786.000000</td>\n",
              "      <td>1426.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>53023.000000</td>\n",
              "      <td>51244.000000</td>\n",
              "      <td>6.877558e+06</td>\n",
              "      <td>1.277000e+03</td>\n",
              "      <td>1374.000000</td>\n",
              "      <td>32.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.024000e+03</td>\n",
              "      <td>48897.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>65535.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>254.000000</td>\n",
              "      <td>65534.000000</td>\n",
              "      <td>5981.000000</td>\n",
              "      <td>224.000000</td>\n",
              "      <td>65534.000000</td>\n",
              "      <td>65535.000000</td>\n",
              "      <td>4.294967e+09</td>\n",
              "      <td>4.294967e+09</td>\n",
              "      <td>5941.000000</td>\n",
              "      <td>60.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.644544e+07</td>\n",
              "      <td>65533.000000</td>\n",
              "      <td>46335.000000</td>\n",
              "      <td>8961.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b8a73969-8221-442f-9474-5ed38abb9f4e')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-b8a73969-8221-442f-9474-5ed38abb9f4e button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-b8a73969-8221-442f-9474-5ed38abb9f4e');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "       ip.hdr_len   ip.tos  ...  tcp.urgent_pointer  tcp.options.mss_val\n",
              "count    128455.0  91386.0  ...       128455.000000        115134.000000\n",
              "mean         20.0      0.0  ...            0.678744           359.404329\n",
              "std           0.0      0.0  ...          171.836116           636.972162\n",
              "min          20.0      0.0  ...            0.000000             0.000000\n",
              "25%          20.0      0.0  ...            0.000000             0.000000\n",
              "50%          20.0      0.0  ...            0.000000             0.000000\n",
              "75%          20.0      0.0  ...            0.000000             0.000000\n",
              "max          20.0      0.0  ...        46335.000000          8961.000000\n",
              "\n",
              "[8 rows x 28 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ],
      "source": [
        "full_data.describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XoQcceWWXc3x",
        "outputId": "eb0c3940-6080-4cfd-f7c3-45af35b1cc5d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ip.hdr_len              True\n",
              "ip.tos                  True\n",
              "ip.id                  False\n",
              "ip.flags.rb             True\n",
              "ip.flags.df            False\n",
              "ip.flags.mf             True\n",
              "ip.frag_offset         False\n",
              "ip.ttl                 False\n",
              "ip.checksum            False\n",
              "ip.len                 False\n",
              "ip.dsfield             False\n",
              "tcp.srcport            False\n",
              "tcp.dstport            False\n",
              "tcp.seq                False\n",
              "tcp.ack                False\n",
              "tcp.len                False\n",
              "tcp.hdr_len            False\n",
              "tcp.flags.fin          False\n",
              "tcp.flags.syn          False\n",
              "tcp.flags.reset        False\n",
              "tcp.flags.push         False\n",
              "tcp.flags.ack          False\n",
              "tcp.flags.urg          False\n",
              "tcp.flags.cwr          False\n",
              "tcp.window_size        False\n",
              "tcp.checksum           False\n",
              "tcp.urgent_pointer     False\n",
              "tcp.options.mss_val    False\n",
              "dtype: bool"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ],
      "source": [
        "# check features with zero variance (not useful for learning)\n",
        "(full_data.var() == 0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wn9zJjPFXc30"
      },
      "outputs": [],
      "source": [
        "# remove columns with zero variance\n",
        "full_data.drop(columns=['ip.hdr_len', 'ip.tos', 'ip.flags.rb', \n",
        "                        'ip.flags.mf', 'ip.frag_offset'], axis=1, inplace=True)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qVYZPpPMXc33"
      },
      "source": [
        "### Univariate features analysis (boxplot)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZEkHtj14Xc34",
        "outputId": "5df0e610-8031-4a57-b8f4-df146a71d388"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['ip.id', 'ip.flags.df', 'ip.ttl', 'ip.checksum', 'ip.len', 'ip.dsfield',\n",
              "       'tcp.srcport', 'tcp.dstport', 'tcp.seq', 'tcp.ack', 'tcp.len',\n",
              "       'tcp.hdr_len', 'tcp.flags.fin', 'tcp.flags.syn', 'tcp.flags.reset',\n",
              "       'tcp.flags.push', 'tcp.flags.ack', 'tcp.flags.urg', 'tcp.flags.cwr',\n",
              "       'tcp.window_size', 'tcp.checksum', 'tcp.urgent_pointer',\n",
              "       'tcp.options.mss_val', 'label'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ],
      "source": [
        "full_data.columns"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rHVIADLVXc37",
        "outputId": "d9e488b4-e39f-4dc7-a144-18529c1a0633"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 7200x7200 with 23 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "_columns = 3\n",
        "lines = int(full_data.shape[1]/_columns)+1\n",
        "\n",
        "plt.figure(figsize = [100, 100])\n",
        "\n",
        "i = 1\n",
        "for column in full_data.columns.values:\n",
        "    if column != \"label\":\n",
        "        plt.subplot(lines, _columns, i)\n",
        "        full_data.boxplot([column]);\n",
        "        i += 1"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fCxLDLc3Xc39"
      },
      "source": [
        "### Bivariate Analysis - Linear correlation (absolute threshold of 0.5)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u6w3so4nXc4A",
        "outputId": "a9a1c9ea-5dbf-4313-b0eb-f359f27b7cb0"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x1152 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "import seaborn as sns\n",
        "high_corr = full_data.corr().abs().round(2)\n",
        "high_corr_var = high_corr[high_corr>0.5]\n",
        "plt.figure(figsize = (20,16))\n",
        "sns.heatmap(high_corr_var, xticklabels=high_corr_var.columns, yticklabels=high_corr_var.columns, annot=True);"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TvMLKK_9Xc4B"
      },
      "source": [
        "## Machine Learning - Binary Classification _(attack x bonafide)_"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DmW2j9JJXc4C",
        "outputId": "4567ed6b-1b79-4bd2-d1a6-8bc7d04cb344"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "bonafide                 91386\n",
              "nmap_servinfo             2016\n",
              "nmap_connect              2008\n",
              "nmap_mtu                  2006\n",
              "nmap_open                 2006\n",
              "nmap_reason               2006\n",
              "nmap_vvv                  2002\n",
              "nmap_spoof_idle           1417\n",
              "zmap_tcp_syn              1250\n",
              "unicornscan_tcp_conn      1026\n",
              "unicornscan_tcp_syn       1017\n",
              "unicornscan_tcp_null      1014\n",
              "unicornscan_tcp_xmas      1014\n",
              "unicornscan_tcp_fxmas     1014\n",
              "unicornscan_tcp_fin       1014\n",
              "unicornscan_tcp_ack       1014\n",
              "nmap_fragv                1003\n",
              "masscan_tcp_syn           1003\n",
              "nmap_tcp_conn             1002\n",
              "hping_tcp_syn             1001\n",
              "nmap_tcp_fin              1001\n",
              "nmap_tcp_syn              1001\n",
              "nmap_tcp_null             1001\n",
              "nmap_tcp_xmas             1001\n",
              "nmap_tcp_maimon           1000\n",
              "nmap_tcp_ack              1000\n",
              "nmap_tcp_window           1000\n",
              "hping_tcp_fin             1000\n",
              "hping_tcp_xmas            1000\n",
              "hping_tcp_null            1000\n",
              "hping_tcp_ack             1000\n",
              "nmap_fast                  206\n",
              "nmap_top10                  26\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ],
      "source": [
        "full_data['label'].value_counts()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tcAdUjfnXc4D",
        "outputId": "a14639bb-2fc2-4089-b5b1-144e29f04bbe"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    91386\n",
              "1    37069\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ],
      "source": [
        "full_data.label[full_data.label == \"bonafide\"] = 0 # convert bonafide label to 0\n",
        "full_data.label[full_data.label != 0] = 1 # convert attack labels to 1\n",
        "full_data['label'].value_counts()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YMBa45MtXc4E",
        "outputId": "3e16e422-ed81-4109-a9df-5a8c3691edfa"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(128455, 24)"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ],
      "source": [
        "full_data.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "7-N8IPsuXc4F"
      },
      "outputs": [],
      "source": [
        "# It is removed ttl because previous attempt shows that it is learning the LAN architecture TTL=62\n",
        "# (from scan tools TTL=64 minus 2 routers in the infrastructure)\n",
        "\n",
        "# sequence, checksum and acknowledge features because they are random\n",
        "\n",
        "# removed source and destination ports to be agnostic regarding the service ports\n",
        "\n",
        "# removed tcp.options.mss_val because it is dificult to be retrieved as LKM\n",
        "\n",
        "# removed tcp.window_size to allow running iperf on ESP32 and to reduce false positives\n",
        "\n",
        "full_data.drop(columns=[\"ip.checksum\", \"ip.ttl\", \"tcp.checksum\", \"tcp.dstport\", \"tcp.seq\", \"tcp.srcport\", \n",
        "                        \"tcp.ack\", \"tcp.options.mss_val\", \"tcp.window_size\"], axis=1, inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 488
        },
        "id": "5Cxoe13fXc4G",
        "outputId": "65d1bc05-5107-4798-d74d-289783a70417"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-d1442b93-4f1f-433c-9db0-50bfb799b226\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>ip.id</th>\n",
              "      <th>ip.flags.df</th>\n",
              "      <th>ip.len</th>\n",
              "      <th>ip.dsfield</th>\n",
              "      <th>tcp.len</th>\n",
              "      <th>tcp.hdr_len</th>\n",
              "      <th>tcp.flags.fin</th>\n",
              "      <th>tcp.flags.syn</th>\n",
              "      <th>tcp.flags.reset</th>\n",
              "      <th>tcp.flags.push</th>\n",
              "      <th>tcp.flags.ack</th>\n",
              "      <th>tcp.flags.urg</th>\n",
              "      <th>tcp.flags.cwr</th>\n",
              "      <th>tcp.urgent_pointer</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>46834</td>\n",
              "      <td>0.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3793</td>\n",
              "      <td>1.0</td>\n",
              "      <td>537.0</td>\n",
              "      <td>8</td>\n",
              "      <td>485.0</td>\n",
              "      <td>32.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>60.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>8559</td>\n",
              "      <td>1.0</td>\n",
              "      <td>54.0</td>\n",
              "      <td>0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>32.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>54321</td>\n",
              "      <td>0.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>8</td>\n",
              "      <td>0.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37064</th>\n",
              "      <td>19724</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37065</th>\n",
              "      <td>15843</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37066</th>\n",
              "      <td>55863</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37067</th>\n",
              "      <td>41278</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37068</th>\n",
              "      <td>30240</td>\n",
              "      <td>0.0</td>\n",
              "      <td>28.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>24.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>128455 rows × 15 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d1442b93-4f1f-433c-9db0-50bfb799b226')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-d1442b93-4f1f-433c-9db0-50bfb799b226 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-d1442b93-4f1f-433c-9db0-50bfb799b226');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "       ip.id  ip.flags.df  ip.len  ...  tcp.flags.cwr  tcp.urgent_pointer  label\n",
              "1      46834          0.0    40.0  ...            0.0                 0.0      0\n",
              "3       3793          1.0   537.0  ...            0.0                 0.0      0\n",
              "4          0          1.0    80.0  ...            0.0                 0.0      0\n",
              "5       8559          1.0    54.0  ...            0.0                 0.0      0\n",
              "6      54321          0.0    40.0  ...            0.0                 0.0      0\n",
              "...      ...          ...     ...  ...            ...                 ...    ...\n",
              "37064  19724          0.0    28.0  ...            0.0                 0.0      1\n",
              "37065  15843          0.0    28.0  ...            0.0                 0.0      1\n",
              "37066  55863          0.0    28.0  ...            0.0                 0.0      1\n",
              "37067  41278          0.0    28.0  ...            0.0                 0.0      1\n",
              "37068  30240          0.0    28.0  ...            0.0                 0.0      1\n",
              "\n",
              "[128455 rows x 15 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ],
      "source": [
        "full_data"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8BipwzS2Xc4J",
        "outputId": "33f79021-d698-47d6-f812-b3f22c54e740"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(128455, 14) (128455,)\n"
          ]
        }
      ],
      "source": [
        "from sklearn.tree import DecisionTreeClassifier\n",
        "\n",
        "algorithms = {\n",
        "    \"DT\" : (DecisionTreeClassifier(), {\n",
        "        \"criterion\": (\"gini\", \"entropy\"), \n",
        "        \"max_depth\": [x for x in range(1,21)],\n",
        "        \"class_weight\": (None, \"balanced\")\n",
        "    }),\n",
        "}\n",
        "\n",
        "full_data = full_data.fillna(0)\n",
        "X = full_data.drop(columns = [\"label\"])\n",
        "y = full_data.label\n",
        "\n",
        "print(X.shape, y.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 270
        },
        "id": "vKVU-nhCXc4L",
        "outputId": "ecb63a2b-24d5-4a0d-9e39-7c9b12d26ab5"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-292b2b1d-39f9-4fe6-8b9c-0df09ba6dc3b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>ip.id</th>\n",
              "      <th>ip.flags.df</th>\n",
              "      <th>ip.len</th>\n",
              "      <th>ip.dsfield</th>\n",
              "      <th>tcp.len</th>\n",
              "      <th>tcp.hdr_len</th>\n",
              "      <th>tcp.flags.fin</th>\n",
              "      <th>tcp.flags.syn</th>\n",
              "      <th>tcp.flags.reset</th>\n",
              "      <th>tcp.flags.push</th>\n",
              "      <th>tcp.flags.ack</th>\n",
              "      <th>tcp.flags.urg</th>\n",
              "      <th>tcp.flags.cwr</th>\n",
              "      <th>tcp.urgent_pointer</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>46834</td>\n",
              "      <td>0.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3793</td>\n",
              "      <td>1.0</td>\n",
              "      <td>537.0</td>\n",
              "      <td>8</td>\n",
              "      <td>485.0</td>\n",
              "      <td>32.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>80.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>60.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>8559</td>\n",
              "      <td>1.0</td>\n",
              "      <td>54.0</td>\n",
              "      <td>0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>32.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>54321</td>\n",
              "      <td>0.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>8</td>\n",
              "      <td>0.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-292b2b1d-39f9-4fe6-8b9c-0df09ba6dc3b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-292b2b1d-39f9-4fe6-8b9c-0df09ba6dc3b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-292b2b1d-39f9-4fe6-8b9c-0df09ba6dc3b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   ip.id  ip.flags.df  ip.len  ...  tcp.flags.urg  tcp.flags.cwr  tcp.urgent_pointer\n",
              "1  46834          0.0    40.0  ...            0.0            0.0                 0.0\n",
              "3   3793          1.0   537.0  ...            0.0            0.0                 0.0\n",
              "4      0          1.0    80.0  ...            0.0            0.0                 0.0\n",
              "5   8559          1.0    54.0  ...            0.0            0.0                 0.0\n",
              "6  54321          0.0    40.0  ...            0.0            0.0                 0.0\n",
              "\n",
              "[5 rows x 14 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ],
      "source": [
        "X.head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 270
        },
        "id": "RUVo0VRMXc4M",
        "outputId": "2a05074d-ce0c-4879-d5fa-6b70d09a562c"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-4b143e7c-b4ef-43eb-b3f4-6f3f00a9133a\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>ip.id</th>\n",
              "      <th>ip.flags.df</th>\n",
              "      <th>ip.len</th>\n",
              "      <th>ip.dsfield</th>\n",
              "      <th>tcp.len</th>\n",
              "      <th>tcp.hdr_len</th>\n",
              "      <th>tcp.flags.fin</th>\n",
              "      <th>tcp.flags.syn</th>\n",
              "      <th>tcp.flags.reset</th>\n",
              "      <th>tcp.flags.push</th>\n",
              "      <th>tcp.flags.ack</th>\n",
              "      <th>tcp.flags.urg</th>\n",
              "      <th>tcp.flags.cwr</th>\n",
              "      <th>tcp.urgent_pointer</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>46834</td>\n",
              "      <td>0</td>\n",
              "      <td>40</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>20</td>\n",
              "      <td>0</td>\n",
              "      <td>1</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>3793</td>\n",
              "      <td>1</td>\n",
              "      <td>537</td>\n",
              "      <td>8</td>\n",
              "      <td>485</td>\n",
              "      <td>32</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>80</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>60</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>8559</td>\n",
              "      <td>1</td>\n",
              "      <td>54</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>32</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>54321</td>\n",
              "      <td>0</td>\n",
              "      <td>40</td>\n",
              "      <td>8</td>\n",
              "      <td>0</td>\n",
              "      <td>20</td>\n",
              "      <td>0</td>\n",
              "      <td>1</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",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4b143e7c-b4ef-43eb-b3f4-6f3f00a9133a')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-4b143e7c-b4ef-43eb-b3f4-6f3f00a9133a button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-4b143e7c-b4ef-43eb-b3f4-6f3f00a9133a');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   ip.id  ip.flags.df  ip.len  ...  tcp.flags.urg  tcp.flags.cwr  tcp.urgent_pointer\n",
              "1  46834            0      40  ...              0              0                   0\n",
              "3   3793            1     537  ...              0              0                   0\n",
              "4      0            1      80  ...              0              0                   0\n",
              "5   8559            1      54  ...              0              0                   0\n",
              "6  54321            0      40  ...              0              0                   0\n",
              "\n",
              "[5 rows x 14 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ],
      "source": [
        "X = X.astype(int) \n",
        "X.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8Z58R4k-Xc4N"
      },
      "source": [
        "## Grid Search"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CFAIvmplXc4O",
        "outputId": "c7efcc39-9a91-4be1-ac21-8b9561dcc70f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "DT\n"
          ]
        }
      ],
      "source": [
        "from sklearn.model_selection import StratifiedKFold, GridSearchCV\n",
        "from sklearn.metrics import accuracy_score, make_scorer, f1_score, roc_auc_score\n",
        "from sklearn.preprocessing import StandardScaler, MinMaxScaler\n",
        "\n",
        "kf = StratifiedKFold(n_splits=10, shuffle=True, random_state=17) # Train, Test\n",
        "gskf = StratifiedKFold(n_splits=3, shuffle=True, random_state=17) # Validation\n",
        "perf = f1_score # can be used roc_auc_score for binary classification\n",
        "\n",
        "score = {}\n",
        "for algorithm in algorithms.keys():\n",
        "    score[algorithm] = []\n",
        "\n",
        "for algorithm, (clf, parameters) in algorithms.items():\n",
        "    print(algorithm)\n",
        "    for train, test in kf.split(X, y):\n",
        "        best = GridSearchCV(clf, parameters, cv=gskf, scoring=make_scorer(perf))\n",
        "        best.fit(X.iloc[train], y.iloc[train])\n",
        "        score[algorithm].append(perf(best.predict(X.iloc[test]), y.iloc[test]))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "scrolled": true,
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "ONLSrthaXc4R",
        "outputId": "c48489b9-5f5e-4dd5-fda1-3e1792faca0d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-c9d698f6-1fb8-429b-9e78-998ee6a34764\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>DT</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.934986</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.939454</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.934301</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.940588</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.937442</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0.934823</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>0.932457</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>0.937253</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>0.933683</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>0.935458</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c9d698f6-1fb8-429b-9e78-998ee6a34764')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-c9d698f6-1fb8-429b-9e78-998ee6a34764 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-c9d698f6-1fb8-429b-9e78-998ee6a34764');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "         DT\n",
              "0  0.934986\n",
              "1  0.939454\n",
              "2  0.934301\n",
              "3  0.940588\n",
              "4  0.937442\n",
              "5  0.934823\n",
              "6  0.932457\n",
              "7  0.937253\n",
              "8  0.933683\n",
              "9  0.935458"
            ]
          },
          "metadata": {},
          "execution_count": 29
        }
      ],
      "source": [
        "# f1-scores\n",
        "pd.DataFrame.from_dict(score)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dR_i2M57Xc4c"
      },
      "source": [
        "## Results from Grid Search\n",
        "> For conjunction of parameters, mean and std deviation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 520
        },
        "id": "2KkZm2kCXc4d",
        "outputId": "e90fa3f2-3dd7-47fc-eb6b-1d0334f7a1f5"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-42b2b156-ff7e-4df0-b5c7-0e9065b04f79\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>class_weight</th>\n",
              "      <th>criterion</th>\n",
              "      <th>max_depth</th>\n",
              "      <th>mean_test_score</th>\n",
              "      <th>std_test_score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>1</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>2</td>\n",
              "      <td>0.651701</td>\n",
              "      <td>0.002808</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>3</td>\n",
              "      <td>0.832995</td>\n",
              "      <td>0.002339</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>4</td>\n",
              "      <td>0.910205</td>\n",
              "      <td>0.002044</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>5</td>\n",
              "      <td>0.913181</td>\n",
              "      <td>0.002296</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>6</td>\n",
              "      <td>0.920202</td>\n",
              "      <td>0.001833</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>7</td>\n",
              "      <td>0.922086</td>\n",
              "      <td>0.002218</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>8</td>\n",
              "      <td>0.935903</td>\n",
              "      <td>0.001747</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>9</td>\n",
              "      <td>0.936028</td>\n",
              "      <td>0.001791</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>10</td>\n",
              "      <td>0.935776</td>\n",
              "      <td>0.001608</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>11</td>\n",
              "      <td>0.935750</td>\n",
              "      <td>0.001718</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>12</td>\n",
              "      <td>0.935476</td>\n",
              "      <td>0.001735</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>13</td>\n",
              "      <td>0.935033</td>\n",
              "      <td>0.001706</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>14</td>\n",
              "      <td>0.934226</td>\n",
              "      <td>0.001814</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>None</td>\n",
              "      <td>gini</td>\n",
              "      <td>15</td>\n",
              "      <td>0.933724</td>\n",
              "      <td>0.001451</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-42b2b156-ff7e-4df0-b5c7-0e9065b04f79')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-42b2b156-ff7e-4df0-b5c7-0e9065b04f79 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-42b2b156-ff7e-4df0-b5c7-0e9065b04f79');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ],
            "text/plain": [
              "   class_weight criterion  max_depth  mean_test_score  std_test_score\n",
              "0          None      gini          1         0.000000        0.000000\n",
              "1          None      gini          2         0.651701        0.002808\n",
              "2          None      gini          3         0.832995        0.002339\n",
              "3          None      gini          4         0.910205        0.002044\n",
              "4          None      gini          5         0.913181        0.002296\n",
              "5          None      gini          6         0.920202        0.001833\n",
              "6          None      gini          7         0.922086        0.002218\n",
              "7          None      gini          8         0.935903        0.001747\n",
              "8          None      gini          9         0.936028        0.001791\n",
              "9          None      gini         10         0.935776        0.001608\n",
              "10         None      gini         11         0.935750        0.001718\n",
              "11         None      gini         12         0.935476        0.001735\n",
              "12         None      gini         13         0.935033        0.001706\n",
              "13         None      gini         14         0.934226        0.001814\n",
              "14         None      gini         15         0.933724        0.001451"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ],
      "source": [
        "table = {\n",
        "    'class_weight': [x['class_weight'] for x in best.cv_results_['params']],\n",
        "    'criterion': [x['criterion'] for x in best.cv_results_['params']],\n",
        "    'max_depth': [x['max_depth'] for x in best.cv_results_['params']],\n",
        "    'mean_test_score': best.cv_results_['mean_test_score'],\n",
        "    'std_test_score': best.cv_results_['std_test_score']\n",
        "}\n",
        "dados = []\n",
        "for i in range(0,len(best.cv_results_['params'])):\n",
        "    dados.append([best.cv_results_['params'][i]['criterion'], \n",
        "                  best.cv_results_['params'][i]['max_depth'], best.cv_results_['mean_test_score'][i]])\n",
        "pd.DataFrame.from_dict(table).head(15)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "7CV21OHMXc4e"
      },
      "outputs": [],
      "source": [
        "gini = {}\n",
        "entropia = {}\n",
        "for valores in dados:\n",
        "    if valores[0] == \"gini\":\n",
        "        gini.update({valores[1] : valores[2]})\n",
        "    else:\n",
        "        entropia.update({valores[1] : valores[2]})"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 292
        },
        "id": "CjYRZ4lOXc4f",
        "outputId": "ecaf94dc-a165-4663-f7d9-303ebba2fb0a"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ],
      "source": [
        "lists1 = sorted(gini.items())\n",
        "lists2 = sorted(entropia.items())\n",
        "x_gini, y_gini = zip(*lists1)\n",
        "x_entropia, y_entropia = zip(*lists2)\n",
        "\n",
        "plt.ylabel('F1-score', fontsize=20)\n",
        "plt.xlabel('Max depth', fontsize=20)\n",
        "plt.plot(x_gini, y_gini, '--', label='Gini')\n",
        "plt.plot(x_entropia, y_entropia, '.-', label='Entropy')\n",
        "plt.legend(loc=\"lower right\");"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "x_train, x_test, y_train, y_test = train_test_split(X, y, test_size=0.2, stratify=y)\n",
        "x_train, x_val, y_train, y_val = train_test_split(x_train, y_train, test_size=0.2, stratify=y_train)"
      ],
      "metadata": {
        "id": "W5gvBqShe9n3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dt10 = DecisionTreeClassifier(criterion=\"entropy\", max_depth=10, class_weight=None)\n",
        "dt12 = DecisionTreeClassifier(criterion=\"entropy\", max_depth=12, class_weight=None)\n",
        "\n",
        "dt10.fit(x_train, y_train)\n",
        "dt12.fit(x_train, y_train)\n",
        "dt10_pred = dt10.predict(x_test)\n",
        "dt12_pred = dt12.predict(x_test)\n",
        "print(f1_score(y_test, dt10_pred))\n",
        "print(f1_score(y_test, dt12_pred))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "M4SaWBsVuwti",
        "outputId": "1bc5e1cf-ca8c-459c-b7b7-6a5d73dcfd09"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.9339195485268064\n",
            "0.9334996389417711\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "BinaryF1Score = tfa.metrics.F1Score(num_classes=2)\n",
        "F1Score = lambda y_true, y_pred: BinaryF1Score(tf.one_hot(tf.cast(y_true, tf.int32), 2), y_pred)\n",
        "print(F1Score(y_train, tf.one_hot(tf.cast(y_train, tf.int32), 2)))\n",
        "hidden_size = x_train.shape[1] + 2\n",
        "model = tf.keras.models.Sequential([\n",
        "    tf.keras.layers.Dense(hidden_size, activation=tf.nn.sigmoid, input_shape=(x_train.shape[1],)),\n",
        "    tf.keras.layers.Dense(hidden_size, activation=tf.nn.sigmoid),\n",
        "    tf.keras.layers.Dense(2, activation=tf.nn.softmax)\n",
        "])\n",
        "model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=1e-5),\n",
        "              loss=tf.keras.losses.CategoricalCrossentropy(),\n",
        "              metrics=[tfa.metrics.F1Score(num_classes=2)])\n",
        "model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PUdWL1ENfeFX",
        "outputId": "0af6dc2c-5678-4340-e8df-e4307d12daff"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "tf.Tensor([1. 1.], shape=(2,), dtype=float32)\n",
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " dense (Dense)               (None, 16)                240       \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 16)                272       \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 2)                 34        \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 546\n",
            "Trainable params: 546\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_ds = tf.data.Dataset.from_tensor_slices((x_train,tf.one_hot(tf.cast(y_train, tf.int32), 2))).batch(256)\n",
        "val_ds = tf.data.Dataset.from_tensor_slices((x_val, tf.one_hot(tf.cast(y_val, tf.int32), 2))).batch(1024)"
      ],
      "metadata": {
        "id": "ozN-RjBAfooo"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "history = model.fit(train_ds, validation_data=val_ds, epochs=2000)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tOgqLYFGfsPm",
        "outputId": "ea600142-053c-46df-c482-883060c2635d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/2000\n",
            "322/322 [==============================] - 3s 4ms/step - loss: 0.8139 - f1_score: 0.2239 - val_loss: 0.7995 - val_f1_score: 0.2239\n",
            "Epoch 2/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.7859 - f1_score: 0.2239 - val_loss: 0.7728 - val_f1_score: 0.2239\n",
            "Epoch 3/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.7605 - f1_score: 0.2256 - val_loss: 0.7487 - val_f1_score: 0.2296\n",
            "Epoch 4/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.7376 - f1_score: 0.2385 - val_loss: 0.7270 - val_f1_score: 0.2450\n",
            "Epoch 5/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.7171 - f1_score: 0.2516 - val_loss: 0.7077 - val_f1_score: 0.2542\n",
            "Epoch 6/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6989 - f1_score: 0.3076 - val_loss: 0.6905 - val_f1_score: 0.4212\n",
            "Epoch 7/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6827 - f1_score: 0.4243 - val_loss: 0.6752 - val_f1_score: 0.4181\n",
            "Epoch 8/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6684 - f1_score: 0.4171 - val_loss: 0.6619 - val_f1_score: 0.4157\n",
            "Epoch 9/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6559 - f1_score: 0.4157 - val_loss: 0.6502 - val_f1_score: 0.4157\n",
            "Epoch 10/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6450 - f1_score: 0.4157 - val_loss: 0.6401 - val_f1_score: 0.4157\n",
            "Epoch 11/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6357 - f1_score: 0.4157 - val_loss: 0.6315 - val_f1_score: 0.4157\n",
            "Epoch 12/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6278 - f1_score: 0.4157 - val_loss: 0.6242 - val_f1_score: 0.4157\n",
            "Epoch 13/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6211 - f1_score: 0.4157 - val_loss: 0.6182 - val_f1_score: 0.4157\n",
            "Epoch 14/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6156 - f1_score: 0.4157 - val_loss: 0.6132 - val_f1_score: 0.4157\n",
            "Epoch 15/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6111 - f1_score: 0.4157 - val_loss: 0.6093 - val_f1_score: 0.4157\n",
            "Epoch 16/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6076 - f1_score: 0.4157 - val_loss: 0.6061 - val_f1_score: 0.4157\n",
            "Epoch 17/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6048 - f1_score: 0.4157 - val_loss: 0.6037 - val_f1_score: 0.4157\n",
            "Epoch 18/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6028 - f1_score: 0.4157 - val_loss: 0.6020 - val_f1_score: 0.4157\n",
            "Epoch 19/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6012 - f1_score: 0.4157 - val_loss: 0.6007 - val_f1_score: 0.4157\n",
            "Epoch 20/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.6001 - f1_score: 0.4157 - val_loss: 0.5998 - val_f1_score: 0.4157\n",
            "Epoch 21/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5994 - f1_score: 0.4157 - val_loss: 0.5991 - val_f1_score: 0.4157\n",
            "Epoch 22/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5988 - f1_score: 0.4157 - val_loss: 0.5987 - val_f1_score: 0.4157\n",
            "Epoch 23/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5984 - f1_score: 0.4157 - val_loss: 0.5983 - val_f1_score: 0.4157\n",
            "Epoch 24/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5945 - f1_score: 0.4157 - val_loss: 0.5901 - val_f1_score: 0.4157\n",
            "Epoch 25/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5893 - f1_score: 0.4157 - val_loss: 0.5888 - val_f1_score: 0.4157\n",
            "Epoch 26/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5881 - f1_score: 0.4157 - val_loss: 0.5877 - val_f1_score: 0.4157\n",
            "Epoch 27/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5871 - f1_score: 0.4157 - val_loss: 0.5867 - val_f1_score: 0.4157\n",
            "Epoch 28/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5861 - f1_score: 0.4157 - val_loss: 0.5857 - val_f1_score: 0.4157\n",
            "Epoch 29/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5852 - f1_score: 0.4157 - val_loss: 0.5848 - val_f1_score: 0.4157\n",
            "Epoch 30/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5843 - f1_score: 0.4157 - val_loss: 0.5839 - val_f1_score: 0.4157\n",
            "Epoch 31/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5833 - f1_score: 0.4157 - val_loss: 0.5830 - val_f1_score: 0.4157\n",
            "Epoch 32/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5824 - f1_score: 0.4157 - val_loss: 0.5821 - val_f1_score: 0.4157\n",
            "Epoch 33/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5815 - f1_score: 0.4157 - val_loss: 0.5812 - val_f1_score: 0.4157\n",
            "Epoch 34/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5806 - f1_score: 0.4157 - val_loss: 0.5803 - val_f1_score: 0.4157\n",
            "Epoch 35/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5797 - f1_score: 0.4157 - val_loss: 0.5794 - val_f1_score: 0.4157\n",
            "Epoch 36/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5788 - f1_score: 0.4157 - val_loss: 0.5785 - val_f1_score: 0.4157\n",
            "Epoch 37/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5779 - f1_score: 0.4157 - val_loss: 0.5775 - val_f1_score: 0.4157\n",
            "Epoch 38/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5769 - f1_score: 0.4157 - val_loss: 0.5766 - val_f1_score: 0.4157\n",
            "Epoch 39/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5760 - f1_score: 0.4157 - val_loss: 0.5757 - val_f1_score: 0.4157\n",
            "Epoch 40/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5751 - f1_score: 0.4157 - val_loss: 0.5748 - val_f1_score: 0.4157\n",
            "Epoch 41/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5742 - f1_score: 0.4157 - val_loss: 0.5739 - val_f1_score: 0.4157\n",
            "Epoch 42/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5732 - f1_score: 0.4157 - val_loss: 0.5729 - val_f1_score: 0.4157\n",
            "Epoch 43/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5723 - f1_score: 0.4157 - val_loss: 0.5720 - val_f1_score: 0.4157\n",
            "Epoch 44/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5713 - f1_score: 0.4157 - val_loss: 0.5711 - val_f1_score: 0.4157\n",
            "Epoch 45/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5704 - f1_score: 0.4157 - val_loss: 0.5702 - val_f1_score: 0.4157\n",
            "Epoch 46/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5695 - f1_score: 0.4157 - val_loss: 0.5692 - val_f1_score: 0.4157\n",
            "Epoch 47/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5685 - f1_score: 0.4157 - val_loss: 0.5683 - val_f1_score: 0.4157\n",
            "Epoch 48/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5676 - f1_score: 0.4157 - val_loss: 0.5673 - val_f1_score: 0.4157\n",
            "Epoch 49/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5666 - f1_score: 0.4157 - val_loss: 0.5664 - val_f1_score: 0.4157\n",
            "Epoch 50/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5657 - f1_score: 0.4157 - val_loss: 0.5654 - val_f1_score: 0.4157\n",
            "Epoch 51/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5647 - f1_score: 0.4157 - val_loss: 0.5645 - val_f1_score: 0.4157\n",
            "Epoch 52/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5637 - f1_score: 0.4157 - val_loss: 0.5635 - val_f1_score: 0.4157\n",
            "Epoch 53/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5627 - f1_score: 0.4157 - val_loss: 0.5625 - val_f1_score: 0.4157\n",
            "Epoch 54/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5551 - f1_score: 0.4157 - val_loss: 0.5532 - val_f1_score: 0.4157\n",
            "Epoch 55/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5520 - f1_score: 0.4157 - val_loss: 0.5513 - val_f1_score: 0.4157\n",
            "Epoch 56/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5501 - f1_score: 0.4157 - val_loss: 0.5494 - val_f1_score: 0.4157\n",
            "Epoch 57/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5483 - f1_score: 0.4157 - val_loss: 0.5477 - val_f1_score: 0.4157\n",
            "Epoch 58/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5466 - f1_score: 0.4157 - val_loss: 0.5460 - val_f1_score: 0.4157\n",
            "Epoch 59/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5449 - f1_score: 0.4157 - val_loss: 0.5443 - val_f1_score: 0.4157\n",
            "Epoch 60/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5432 - f1_score: 0.4157 - val_loss: 0.5427 - val_f1_score: 0.4157\n",
            "Epoch 61/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5415 - f1_score: 0.4157 - val_loss: 0.5410 - val_f1_score: 0.4157\n",
            "Epoch 62/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.5397 - f1_score: 0.4157 - val_loss: 0.5392 - val_f1_score: 0.4157\n",
            "Epoch 63/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.5379 - f1_score: 0.4157 - val_loss: 0.5374 - val_f1_score: 0.4157\n",
            "Epoch 64/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5361 - f1_score: 0.4157 - val_loss: 0.5356 - val_f1_score: 0.4157\n",
            "Epoch 65/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5343 - f1_score: 0.4157 - val_loss: 0.5338 - val_f1_score: 0.4157\n",
            "Epoch 66/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5325 - f1_score: 0.4157 - val_loss: 0.5319 - val_f1_score: 0.4157\n",
            "Epoch 67/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5305 - f1_score: 0.4157 - val_loss: 0.5300 - val_f1_score: 0.4157\n",
            "Epoch 68/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5284 - f1_score: 0.4157 - val_loss: 0.5278 - val_f1_score: 0.4157\n",
            "Epoch 69/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5261 - f1_score: 0.4157 - val_loss: 0.5254 - val_f1_score: 0.4157\n",
            "Epoch 70/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5234 - f1_score: 0.4157 - val_loss: 0.5221 - val_f1_score: 0.4157\n",
            "Epoch 71/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5194 - f1_score: 0.4157 - val_loss: 0.5172 - val_f1_score: 0.4157\n",
            "Epoch 72/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5144 - f1_score: 0.4157 - val_loss: 0.5137 - val_f1_score: 0.4157\n",
            "Epoch 73/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5121 - f1_score: 0.4157 - val_loss: 0.5117 - val_f1_score: 0.4157\n",
            "Epoch 74/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5102 - f1_score: 0.4157 - val_loss: 0.5098 - val_f1_score: 0.4157\n",
            "Epoch 75/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5082 - f1_score: 0.4157 - val_loss: 0.5079 - val_f1_score: 0.4157\n",
            "Epoch 76/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5064 - f1_score: 0.4157 - val_loss: 0.5060 - val_f1_score: 0.4157\n",
            "Epoch 77/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5044 - f1_score: 0.4157 - val_loss: 0.5041 - val_f1_score: 0.4157\n",
            "Epoch 78/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5026 - f1_score: 0.4157 - val_loss: 0.5023 - val_f1_score: 0.4157\n",
            "Epoch 79/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.5007 - f1_score: 0.4157 - val_loss: 0.5005 - val_f1_score: 0.4157\n",
            "Epoch 80/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4989 - f1_score: 0.4157 - val_loss: 0.4987 - val_f1_score: 0.4157\n",
            "Epoch 81/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4971 - f1_score: 0.4157 - val_loss: 0.4969 - val_f1_score: 0.4157\n",
            "Epoch 82/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4953 - f1_score: 0.4157 - val_loss: 0.4951 - val_f1_score: 0.4157\n",
            "Epoch 83/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4935 - f1_score: 0.4157 - val_loss: 0.4933 - val_f1_score: 0.4157\n",
            "Epoch 84/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4917 - f1_score: 0.4157 - val_loss: 0.4916 - val_f1_score: 0.4157\n",
            "Epoch 85/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4900 - f1_score: 0.4157 - val_loss: 0.4898 - val_f1_score: 0.4157\n",
            "Epoch 86/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4882 - f1_score: 0.4157 - val_loss: 0.4881 - val_f1_score: 0.4157\n",
            "Epoch 87/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4864 - f1_score: 0.4157 - val_loss: 0.4863 - val_f1_score: 0.4157\n",
            "Epoch 88/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4847 - f1_score: 0.4157 - val_loss: 0.4846 - val_f1_score: 0.4157\n",
            "Epoch 89/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4828 - f1_score: 0.4157 - val_loss: 0.4827 - val_f1_score: 0.4157\n",
            "Epoch 90/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4810 - f1_score: 0.4157 - val_loss: 0.4808 - val_f1_score: 0.4157\n",
            "Epoch 91/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.4791 - f1_score: 0.4157 - val_loss: 0.4788 - val_f1_score: 0.4157\n",
            "Epoch 92/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.4772 - f1_score: 0.4157 - val_loss: 0.4770 - val_f1_score: 0.4157\n",
            "Epoch 93/2000\n",
            "322/322 [==============================] - 2s 6ms/step - loss: 0.4753 - f1_score: 0.4157 - val_loss: 0.4750 - val_f1_score: 0.4157\n",
            "Epoch 94/2000\n",
            "322/322 [==============================] - 2s 6ms/step - loss: 0.4735 - f1_score: 0.4157 - val_loss: 0.4732 - val_f1_score: 0.4157\n",
            "Epoch 95/2000\n",
            "322/322 [==============================] - 2s 6ms/step - loss: 0.4717 - f1_score: 0.4157 - val_loss: 0.4714 - val_f1_score: 0.4157\n",
            "Epoch 96/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.4698 - f1_score: 0.4157 - val_loss: 0.4696 - val_f1_score: 0.4157\n",
            "Epoch 97/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4680 - f1_score: 0.4157 - val_loss: 0.4678 - val_f1_score: 0.4157\n",
            "Epoch 98/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4662 - f1_score: 0.4157 - val_loss: 0.4660 - val_f1_score: 0.4157\n",
            "Epoch 99/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4644 - f1_score: 0.4157 - val_loss: 0.4642 - val_f1_score: 0.4157\n",
            "Epoch 100/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4626 - f1_score: 0.4157 - val_loss: 0.4625 - val_f1_score: 0.4157\n",
            "Epoch 101/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4608 - f1_score: 0.4157 - val_loss: 0.4608 - val_f1_score: 0.4157\n",
            "Epoch 102/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4591 - f1_score: 0.4157 - val_loss: 0.4591 - val_f1_score: 0.4157\n",
            "Epoch 103/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4572 - f1_score: 0.4157 - val_loss: 0.4575 - val_f1_score: 0.4157\n",
            "Epoch 104/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4555 - f1_score: 0.4157 - val_loss: 0.4558 - val_f1_score: 0.4157\n",
            "Epoch 105/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4538 - f1_score: 0.4157 - val_loss: 0.4541 - val_f1_score: 0.4157\n",
            "Epoch 106/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4520 - f1_score: 0.4157 - val_loss: 0.4523 - val_f1_score: 0.4157\n",
            "Epoch 107/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4502 - f1_score: 0.4157 - val_loss: 0.4507 - val_f1_score: 0.4157\n",
            "Epoch 108/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4486 - f1_score: 0.4157 - val_loss: 0.4491 - val_f1_score: 0.4157\n",
            "Epoch 109/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4470 - f1_score: 0.4157 - val_loss: 0.4475 - val_f1_score: 0.4157\n",
            "Epoch 110/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4454 - f1_score: 0.4157 - val_loss: 0.4460 - val_f1_score: 0.4157\n",
            "Epoch 111/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4439 - f1_score: 0.4157 - val_loss: 0.4445 - val_f1_score: 0.4157\n",
            "Epoch 112/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4422 - f1_score: 0.4157 - val_loss: 0.4427 - val_f1_score: 0.4157\n",
            "Epoch 113/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4405 - f1_score: 0.4157 - val_loss: 0.4411 - val_f1_score: 0.4157\n",
            "Epoch 114/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4389 - f1_score: 0.4157 - val_loss: 0.4395 - val_f1_score: 0.4157\n",
            "Epoch 115/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4373 - f1_score: 0.4157 - val_loss: 0.4380 - val_f1_score: 0.4157\n",
            "Epoch 116/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4356 - f1_score: 0.4157 - val_loss: 0.4363 - val_f1_score: 0.4157\n",
            "Epoch 117/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4338 - f1_score: 0.4157 - val_loss: 0.4345 - val_f1_score: 0.4157\n",
            "Epoch 118/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4321 - f1_score: 0.4157 - val_loss: 0.4330 - val_f1_score: 0.4157\n",
            "Epoch 119/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4303 - f1_score: 0.4157 - val_loss: 0.4312 - val_f1_score: 0.4157\n",
            "Epoch 120/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4285 - f1_score: 0.4157 - val_loss: 0.4295 - val_f1_score: 0.4157\n",
            "Epoch 121/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4266 - f1_score: 0.4157 - val_loss: 0.4275 - val_f1_score: 0.4157\n",
            "Epoch 122/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4244 - f1_score: 0.4157 - val_loss: 0.4255 - val_f1_score: 0.4157\n",
            "Epoch 123/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4220 - f1_score: 0.4157 - val_loss: 0.4230 - val_f1_score: 0.4157\n",
            "Epoch 124/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4199 - f1_score: 0.4239 - val_loss: 0.4215 - val_f1_score: 0.5301\n",
            "Epoch 125/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4188 - f1_score: 0.4710 - val_loss: 0.4205 - val_f1_score: 0.6109\n",
            "Epoch 126/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4178 - f1_score: 0.5473 - val_loss: 0.4196 - val_f1_score: 0.6469\n",
            "Epoch 127/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4169 - f1_score: 0.6006 - val_loss: 0.4187 - val_f1_score: 0.6668\n",
            "Epoch 128/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4160 - f1_score: 0.6199 - val_loss: 0.4178 - val_f1_score: 0.6878\n",
            "Epoch 129/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4151 - f1_score: 0.6304 - val_loss: 0.4169 - val_f1_score: 0.7030\n",
            "Epoch 130/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4142 - f1_score: 0.6387 - val_loss: 0.4160 - val_f1_score: 0.7172\n",
            "Epoch 131/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4133 - f1_score: 0.6462 - val_loss: 0.4151 - val_f1_score: 0.7299\n",
            "Epoch 132/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4124 - f1_score: 0.6509 - val_loss: 0.4142 - val_f1_score: 0.7435\n",
            "Epoch 133/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4116 - f1_score: 0.6556 - val_loss: 0.4133 - val_f1_score: 0.7624\n",
            "Epoch 134/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4107 - f1_score: 0.6612 - val_loss: 0.4124 - val_f1_score: 0.7737\n",
            "Epoch 135/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4098 - f1_score: 0.6659 - val_loss: 0.4115 - val_f1_score: 0.7828\n",
            "Epoch 136/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4090 - f1_score: 0.6716 - val_loss: 0.4107 - val_f1_score: 0.7910\n",
            "Epoch 137/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4082 - f1_score: 0.6782 - val_loss: 0.4098 - val_f1_score: 0.7953\n",
            "Epoch 138/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4073 - f1_score: 0.6863 - val_loss: 0.4089 - val_f1_score: 0.8005\n",
            "Epoch 139/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4065 - f1_score: 0.6939 - val_loss: 0.4080 - val_f1_score: 0.8043\n",
            "Epoch 140/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4056 - f1_score: 0.7000 - val_loss: 0.4071 - val_f1_score: 0.8104\n",
            "Epoch 141/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4048 - f1_score: 0.7072 - val_loss: 0.4061 - val_f1_score: 0.8132\n",
            "Epoch 142/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4039 - f1_score: 0.7140 - val_loss: 0.4052 - val_f1_score: 0.8129\n",
            "Epoch 143/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4031 - f1_score: 0.7202 - val_loss: 0.4042 - val_f1_score: 0.8089\n",
            "Epoch 144/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4022 - f1_score: 0.7276 - val_loss: 0.4033 - val_f1_score: 0.8032\n",
            "Epoch 145/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.4012 - f1_score: 0.7349 - val_loss: 0.4021 - val_f1_score: 0.7929\n",
            "Epoch 146/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.4000 - f1_score: 0.7325 - val_loss: 0.4000 - val_f1_score: 0.7696\n",
            "Epoch 147/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3985 - f1_score: 0.7235 - val_loss: 0.3987 - val_f1_score: 0.7513\n",
            "Epoch 148/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3972 - f1_score: 0.7244 - val_loss: 0.3975 - val_f1_score: 0.7416\n",
            "Epoch 149/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3958 - f1_score: 0.7214 - val_loss: 0.3962 - val_f1_score: 0.7461\n",
            "Epoch 150/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3943 - f1_score: 0.7253 - val_loss: 0.3949 - val_f1_score: 0.7366\n",
            "Epoch 151/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3929 - f1_score: 0.7340 - val_loss: 0.3937 - val_f1_score: 0.7283\n",
            "Epoch 152/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3917 - f1_score: 0.7410 - val_loss: 0.3926 - val_f1_score: 0.7113\n",
            "Epoch 153/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3906 - f1_score: 0.7487 - val_loss: 0.3917 - val_f1_score: 0.6982\n",
            "Epoch 154/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3895 - f1_score: 0.7541 - val_loss: 0.3908 - val_f1_score: 0.7002\n",
            "Epoch 155/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3885 - f1_score: 0.7597 - val_loss: 0.3900 - val_f1_score: 0.6989\n",
            "Epoch 156/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3876 - f1_score: 0.7640 - val_loss: 0.3892 - val_f1_score: 0.6974\n",
            "Epoch 157/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3867 - f1_score: 0.7669 - val_loss: 0.3884 - val_f1_score: 0.6967\n",
            "Epoch 158/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3858 - f1_score: 0.7702 - val_loss: 0.3876 - val_f1_score: 0.6964\n",
            "Epoch 159/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3849 - f1_score: 0.7735 - val_loss: 0.3868 - val_f1_score: 0.6962\n",
            "Epoch 160/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3840 - f1_score: 0.7772 - val_loss: 0.3860 - val_f1_score: 0.6978\n",
            "Epoch 161/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3831 - f1_score: 0.7802 - val_loss: 0.3853 - val_f1_score: 0.6985\n",
            "Epoch 162/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3823 - f1_score: 0.7830 - val_loss: 0.3845 - val_f1_score: 0.7089\n",
            "Epoch 163/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3814 - f1_score: 0.7854 - val_loss: 0.3837 - val_f1_score: 0.7135\n",
            "Epoch 164/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3805 - f1_score: 0.7875 - val_loss: 0.3829 - val_f1_score: 0.7163\n",
            "Epoch 165/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3797 - f1_score: 0.7892 - val_loss: 0.3821 - val_f1_score: 0.7182\n",
            "Epoch 166/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3788 - f1_score: 0.7907 - val_loss: 0.3813 - val_f1_score: 0.7201\n",
            "Epoch 167/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3780 - f1_score: 0.7928 - val_loss: 0.3805 - val_f1_score: 0.7229\n",
            "Epoch 168/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3772 - f1_score: 0.7949 - val_loss: 0.3797 - val_f1_score: 0.7247\n",
            "Epoch 169/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3763 - f1_score: 0.7964 - val_loss: 0.3789 - val_f1_score: 0.7275\n",
            "Epoch 170/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3755 - f1_score: 0.7975 - val_loss: 0.3780 - val_f1_score: 0.7301\n",
            "Epoch 171/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3747 - f1_score: 0.7989 - val_loss: 0.3772 - val_f1_score: 0.7333\n",
            "Epoch 172/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3738 - f1_score: 0.7998 - val_loss: 0.3763 - val_f1_score: 0.7356\n",
            "Epoch 173/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3730 - f1_score: 0.8008 - val_loss: 0.3754 - val_f1_score: 0.7393\n",
            "Epoch 174/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3722 - f1_score: 0.8019 - val_loss: 0.3745 - val_f1_score: 0.7423\n",
            "Epoch 175/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3713 - f1_score: 0.8030 - val_loss: 0.3736 - val_f1_score: 0.7452\n",
            "Epoch 176/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3705 - f1_score: 0.8042 - val_loss: 0.3726 - val_f1_score: 0.7486\n",
            "Epoch 177/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3697 - f1_score: 0.8050 - val_loss: 0.3717 - val_f1_score: 0.7532\n",
            "Epoch 178/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3689 - f1_score: 0.8057 - val_loss: 0.3707 - val_f1_score: 0.7585\n",
            "Epoch 179/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3681 - f1_score: 0.8067 - val_loss: 0.3697 - val_f1_score: 0.7631\n",
            "Epoch 180/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3672 - f1_score: 0.8075 - val_loss: 0.3688 - val_f1_score: 0.7668\n",
            "Epoch 181/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3664 - f1_score: 0.8087 - val_loss: 0.3678 - val_f1_score: 0.7718\n",
            "Epoch 182/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3656 - f1_score: 0.8094 - val_loss: 0.3668 - val_f1_score: 0.7763\n",
            "Epoch 183/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3648 - f1_score: 0.8104 - val_loss: 0.3658 - val_f1_score: 0.7810\n",
            "Epoch 184/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3640 - f1_score: 0.8115 - val_loss: 0.3649 - val_f1_score: 0.7853\n",
            "Epoch 185/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3632 - f1_score: 0.8123 - val_loss: 0.3639 - val_f1_score: 0.7885\n",
            "Epoch 186/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3624 - f1_score: 0.8132 - val_loss: 0.3629 - val_f1_score: 0.7898\n",
            "Epoch 187/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3615 - f1_score: 0.8136 - val_loss: 0.3620 - val_f1_score: 0.7924\n",
            "Epoch 188/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3607 - f1_score: 0.8141 - val_loss: 0.3610 - val_f1_score: 0.7940\n",
            "Epoch 189/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3599 - f1_score: 0.8150 - val_loss: 0.3601 - val_f1_score: 0.7957\n",
            "Epoch 190/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3591 - f1_score: 0.8156 - val_loss: 0.3591 - val_f1_score: 0.7981\n",
            "Epoch 191/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3583 - f1_score: 0.8159 - val_loss: 0.3582 - val_f1_score: 0.8007\n",
            "Epoch 192/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3575 - f1_score: 0.8165 - val_loss: 0.3572 - val_f1_score: 0.8029\n",
            "Epoch 193/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3567 - f1_score: 0.8170 - val_loss: 0.3563 - val_f1_score: 0.8050\n",
            "Epoch 194/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3559 - f1_score: 0.8173 - val_loss: 0.3554 - val_f1_score: 0.8071\n",
            "Epoch 195/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3551 - f1_score: 0.8177 - val_loss: 0.3545 - val_f1_score: 0.8086\n",
            "Epoch 196/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3543 - f1_score: 0.8187 - val_loss: 0.3536 - val_f1_score: 0.8109\n",
            "Epoch 197/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3535 - f1_score: 0.8196 - val_loss: 0.3527 - val_f1_score: 0.8140\n",
            "Epoch 198/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3527 - f1_score: 0.8200 - val_loss: 0.3518 - val_f1_score: 0.8173\n",
            "Epoch 199/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3519 - f1_score: 0.8204 - val_loss: 0.3509 - val_f1_score: 0.8202\n",
            "Epoch 200/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3511 - f1_score: 0.8206 - val_loss: 0.3500 - val_f1_score: 0.8235\n",
            "Epoch 201/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3503 - f1_score: 0.8212 - val_loss: 0.3492 - val_f1_score: 0.8269\n",
            "Epoch 202/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3495 - f1_score: 0.8221 - val_loss: 0.3483 - val_f1_score: 0.8309\n",
            "Epoch 203/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3487 - f1_score: 0.8227 - val_loss: 0.3475 - val_f1_score: 0.8345\n",
            "Epoch 204/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3480 - f1_score: 0.8235 - val_loss: 0.3466 - val_f1_score: 0.8373\n",
            "Epoch 205/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3472 - f1_score: 0.8241 - val_loss: 0.3458 - val_f1_score: 0.8411\n",
            "Epoch 206/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3464 - f1_score: 0.8249 - val_loss: 0.3450 - val_f1_score: 0.8454\n",
            "Epoch 207/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3456 - f1_score: 0.8256 - val_loss: 0.3442 - val_f1_score: 0.8507\n",
            "Epoch 208/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3448 - f1_score: 0.8265 - val_loss: 0.3434 - val_f1_score: 0.8582\n",
            "Epoch 209/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3440 - f1_score: 0.8277 - val_loss: 0.3426 - val_f1_score: 0.8657\n",
            "Epoch 210/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3433 - f1_score: 0.8277 - val_loss: 0.3419 - val_f1_score: 0.8738\n",
            "Epoch 211/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3425 - f1_score: 0.8281 - val_loss: 0.3411 - val_f1_score: 0.8808\n",
            "Epoch 212/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3417 - f1_score: 0.8286 - val_loss: 0.3404 - val_f1_score: 0.8804\n",
            "Epoch 213/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3409 - f1_score: 0.8287 - val_loss: 0.3397 - val_f1_score: 0.8795\n",
            "Epoch 214/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3402 - f1_score: 0.8293 - val_loss: 0.3390 - val_f1_score: 0.8788\n",
            "Epoch 215/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3394 - f1_score: 0.8297 - val_loss: 0.3383 - val_f1_score: 0.8778\n",
            "Epoch 216/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3386 - f1_score: 0.8299 - val_loss: 0.3376 - val_f1_score: 0.8762\n",
            "Epoch 217/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3379 - f1_score: 0.8305 - val_loss: 0.3369 - val_f1_score: 0.8763\n",
            "Epoch 218/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3371 - f1_score: 0.8310 - val_loss: 0.3363 - val_f1_score: 0.8763\n",
            "Epoch 219/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3364 - f1_score: 0.8312 - val_loss: 0.3356 - val_f1_score: 0.8762\n",
            "Epoch 220/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3356 - f1_score: 0.8316 - val_loss: 0.3350 - val_f1_score: 0.8763\n",
            "Epoch 221/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3349 - f1_score: 0.8318 - val_loss: 0.3343 - val_f1_score: 0.8764\n",
            "Epoch 222/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3341 - f1_score: 0.8321 - val_loss: 0.3337 - val_f1_score: 0.8765\n",
            "Epoch 223/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3334 - f1_score: 0.8326 - val_loss: 0.3330 - val_f1_score: 0.8764\n",
            "Epoch 224/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3326 - f1_score: 0.8327 - val_loss: 0.3324 - val_f1_score: 0.8764\n",
            "Epoch 225/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3319 - f1_score: 0.8327 - val_loss: 0.3318 - val_f1_score: 0.8773\n",
            "Epoch 226/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3312 - f1_score: 0.8330 - val_loss: 0.3311 - val_f1_score: 0.8793\n",
            "Epoch 227/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3304 - f1_score: 0.8333 - val_loss: 0.3305 - val_f1_score: 0.8803\n",
            "Epoch 228/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3297 - f1_score: 0.8335 - val_loss: 0.3299 - val_f1_score: 0.8814\n",
            "Epoch 229/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3290 - f1_score: 0.8336 - val_loss: 0.3292 - val_f1_score: 0.8825\n",
            "Epoch 230/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3282 - f1_score: 0.8338 - val_loss: 0.3286 - val_f1_score: 0.8828\n",
            "Epoch 231/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3275 - f1_score: 0.8344 - val_loss: 0.3279 - val_f1_score: 0.8835\n",
            "Epoch 232/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3268 - f1_score: 0.8343 - val_loss: 0.3273 - val_f1_score: 0.8839\n",
            "Epoch 233/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3261 - f1_score: 0.8342 - val_loss: 0.3266 - val_f1_score: 0.8844\n",
            "Epoch 234/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3254 - f1_score: 0.8343 - val_loss: 0.3260 - val_f1_score: 0.8849\n",
            "Epoch 235/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3247 - f1_score: 0.8343 - val_loss: 0.3253 - val_f1_score: 0.8831\n",
            "Epoch 236/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3240 - f1_score: 0.8343 - val_loss: 0.3246 - val_f1_score: 0.8820\n",
            "Epoch 237/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3233 - f1_score: 0.8344 - val_loss: 0.3240 - val_f1_score: 0.8811\n",
            "Epoch 238/2000\n",
            "322/322 [==============================] - 2s 6ms/step - loss: 0.3226 - f1_score: 0.8345 - val_loss: 0.3233 - val_f1_score: 0.8805\n",
            "Epoch 239/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3219 - f1_score: 0.8344 - val_loss: 0.3226 - val_f1_score: 0.8808\n",
            "Epoch 240/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3212 - f1_score: 0.8342 - val_loss: 0.3219 - val_f1_score: 0.8808\n",
            "Epoch 241/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3205 - f1_score: 0.8344 - val_loss: 0.3213 - val_f1_score: 0.8815\n",
            "Epoch 242/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3198 - f1_score: 0.8345 - val_loss: 0.3206 - val_f1_score: 0.8823\n",
            "Epoch 243/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3192 - f1_score: 0.8346 - val_loss: 0.3199 - val_f1_score: 0.8833\n",
            "Epoch 244/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3185 - f1_score: 0.8347 - val_loss: 0.3192 - val_f1_score: 0.8848\n",
            "Epoch 245/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3178 - f1_score: 0.8346 - val_loss: 0.3185 - val_f1_score: 0.8860\n",
            "Epoch 246/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3171 - f1_score: 0.8343 - val_loss: 0.3178 - val_f1_score: 0.8875\n",
            "Epoch 247/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3165 - f1_score: 0.8343 - val_loss: 0.3171 - val_f1_score: 0.8894\n",
            "Epoch 248/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3158 - f1_score: 0.8344 - val_loss: 0.3164 - val_f1_score: 0.8899\n",
            "Epoch 249/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3151 - f1_score: 0.8347 - val_loss: 0.3157 - val_f1_score: 0.8901\n",
            "Epoch 250/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3145 - f1_score: 0.8348 - val_loss: 0.3150 - val_f1_score: 0.8902\n",
            "Epoch 251/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3138 - f1_score: 0.8350 - val_loss: 0.3143 - val_f1_score: 0.8903\n",
            "Epoch 252/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3132 - f1_score: 0.8348 - val_loss: 0.3136 - val_f1_score: 0.8904\n",
            "Epoch 253/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3125 - f1_score: 0.8348 - val_loss: 0.3129 - val_f1_score: 0.8906\n",
            "Epoch 254/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3119 - f1_score: 0.8348 - val_loss: 0.3122 - val_f1_score: 0.8907\n",
            "Epoch 255/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3113 - f1_score: 0.8346 - val_loss: 0.3115 - val_f1_score: 0.8907\n",
            "Epoch 256/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3106 - f1_score: 0.8346 - val_loss: 0.3108 - val_f1_score: 0.8907\n",
            "Epoch 257/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3100 - f1_score: 0.8342 - val_loss: 0.3101 - val_f1_score: 0.8906\n",
            "Epoch 258/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3094 - f1_score: 0.8343 - val_loss: 0.3094 - val_f1_score: 0.8906\n",
            "Epoch 259/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3087 - f1_score: 0.8341 - val_loss: 0.3087 - val_f1_score: 0.8907\n",
            "Epoch 260/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3081 - f1_score: 0.8337 - val_loss: 0.3080 - val_f1_score: 0.8905\n",
            "Epoch 261/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3075 - f1_score: 0.8337 - val_loss: 0.3073 - val_f1_score: 0.8905\n",
            "Epoch 262/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3069 - f1_score: 0.8337 - val_loss: 0.3066 - val_f1_score: 0.8905\n",
            "Epoch 263/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3063 - f1_score: 0.8335 - val_loss: 0.3059 - val_f1_score: 0.8905\n",
            "Epoch 264/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3057 - f1_score: 0.8334 - val_loss: 0.3052 - val_f1_score: 0.8905\n",
            "Epoch 265/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3051 - f1_score: 0.8331 - val_loss: 0.3046 - val_f1_score: 0.8904\n",
            "Epoch 266/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3045 - f1_score: 0.8327 - val_loss: 0.3039 - val_f1_score: 0.8904\n",
            "Epoch 267/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3039 - f1_score: 0.8322 - val_loss: 0.3032 - val_f1_score: 0.8904\n",
            "Epoch 268/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3033 - f1_score: 0.8320 - val_loss: 0.3025 - val_f1_score: 0.8904\n",
            "Epoch 269/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3027 - f1_score: 0.8321 - val_loss: 0.3019 - val_f1_score: 0.8904\n",
            "Epoch 270/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3021 - f1_score: 0.8322 - val_loss: 0.3012 - val_f1_score: 0.8903\n",
            "Epoch 271/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3015 - f1_score: 0.8322 - val_loss: 0.3005 - val_f1_score: 0.8904\n",
            "Epoch 272/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3009 - f1_score: 0.8319 - val_loss: 0.2999 - val_f1_score: 0.8905\n",
            "Epoch 273/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.3004 - f1_score: 0.8321 - val_loss: 0.2992 - val_f1_score: 0.8905\n",
            "Epoch 274/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2998 - f1_score: 0.8320 - val_loss: 0.2986 - val_f1_score: 0.8893\n",
            "Epoch 275/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2992 - f1_score: 0.8315 - val_loss: 0.2980 - val_f1_score: 0.8886\n",
            "Epoch 276/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2987 - f1_score: 0.8315 - val_loss: 0.2973 - val_f1_score: 0.8882\n",
            "Epoch 277/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2981 - f1_score: 0.8311 - val_loss: 0.2967 - val_f1_score: 0.8876\n",
            "Epoch 278/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2976 - f1_score: 0.8310 - val_loss: 0.2961 - val_f1_score: 0.8872\n",
            "Epoch 279/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2970 - f1_score: 0.8309 - val_loss: 0.2955 - val_f1_score: 0.8865\n",
            "Epoch 280/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2965 - f1_score: 0.8308 - val_loss: 0.2949 - val_f1_score: 0.8858\n",
            "Epoch 281/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2959 - f1_score: 0.8307 - val_loss: 0.2943 - val_f1_score: 0.8851\n",
            "Epoch 282/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2954 - f1_score: 0.8306 - val_loss: 0.2937 - val_f1_score: 0.8848\n",
            "Epoch 283/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2949 - f1_score: 0.8303 - val_loss: 0.2932 - val_f1_score: 0.8845\n",
            "Epoch 284/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2943 - f1_score: 0.8305 - val_loss: 0.2926 - val_f1_score: 0.8841\n",
            "Epoch 285/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2938 - f1_score: 0.8306 - val_loss: 0.2920 - val_f1_score: 0.8839\n",
            "Epoch 286/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2933 - f1_score: 0.8305 - val_loss: 0.2915 - val_f1_score: 0.8836\n",
            "Epoch 287/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2928 - f1_score: 0.8304 - val_loss: 0.2909 - val_f1_score: 0.8833\n",
            "Epoch 288/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2923 - f1_score: 0.8304 - val_loss: 0.2904 - val_f1_score: 0.8828\n",
            "Epoch 289/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2918 - f1_score: 0.8304 - val_loss: 0.2898 - val_f1_score: 0.8820\n",
            "Epoch 290/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2913 - f1_score: 0.8304 - val_loss: 0.2893 - val_f1_score: 0.8818\n",
            "Epoch 291/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2908 - f1_score: 0.8302 - val_loss: 0.2888 - val_f1_score: 0.8818\n",
            "Epoch 292/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2903 - f1_score: 0.8301 - val_loss: 0.2882 - val_f1_score: 0.8814\n",
            "Epoch 293/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2898 - f1_score: 0.8301 - val_loss: 0.2877 - val_f1_score: 0.8811\n",
            "Epoch 294/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2893 - f1_score: 0.8302 - val_loss: 0.2872 - val_f1_score: 0.8810\n",
            "Epoch 295/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2888 - f1_score: 0.8301 - val_loss: 0.2867 - val_f1_score: 0.8807\n",
            "Epoch 296/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2883 - f1_score: 0.8300 - val_loss: 0.2862 - val_f1_score: 0.8807\n",
            "Epoch 297/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2878 - f1_score: 0.8300 - val_loss: 0.2857 - val_f1_score: 0.8804\n",
            "Epoch 298/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2874 - f1_score: 0.8301 - val_loss: 0.2852 - val_f1_score: 0.8802\n",
            "Epoch 299/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2869 - f1_score: 0.8303 - val_loss: 0.2847 - val_f1_score: 0.8799\n",
            "Epoch 300/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2864 - f1_score: 0.8304 - val_loss: 0.2842 - val_f1_score: 0.8798\n",
            "Epoch 301/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2859 - f1_score: 0.8304 - val_loss: 0.2838 - val_f1_score: 0.8795\n",
            "Epoch 302/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2855 - f1_score: 0.8301 - val_loss: 0.2833 - val_f1_score: 0.8797\n",
            "Epoch 303/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2850 - f1_score: 0.8303 - val_loss: 0.2829 - val_f1_score: 0.8793\n",
            "Epoch 304/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2846 - f1_score: 0.8303 - val_loss: 0.2824 - val_f1_score: 0.8794\n",
            "Epoch 305/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2841 - f1_score: 0.8304 - val_loss: 0.2820 - val_f1_score: 0.8790\n",
            "Epoch 306/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2837 - f1_score: 0.8306 - val_loss: 0.2815 - val_f1_score: 0.8792\n",
            "Epoch 307/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2833 - f1_score: 0.8305 - val_loss: 0.2811 - val_f1_score: 0.8792\n",
            "Epoch 308/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2828 - f1_score: 0.8307 - val_loss: 0.2807 - val_f1_score: 0.8792\n",
            "Epoch 309/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2824 - f1_score: 0.8310 - val_loss: 0.2802 - val_f1_score: 0.8792\n",
            "Epoch 310/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2820 - f1_score: 0.8312 - val_loss: 0.2798 - val_f1_score: 0.8792\n",
            "Epoch 311/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2815 - f1_score: 0.8313 - val_loss: 0.2794 - val_f1_score: 0.8793\n",
            "Epoch 312/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2811 - f1_score: 0.8316 - val_loss: 0.2790 - val_f1_score: 0.8797\n",
            "Epoch 313/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.2807 - f1_score: 0.8316 - val_loss: 0.2786 - val_f1_score: 0.8801\n",
            "Epoch 314/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2803 - f1_score: 0.8317 - val_loss: 0.2782 - val_f1_score: 0.8802\n",
            "Epoch 315/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2799 - f1_score: 0.8318 - val_loss: 0.2778 - val_f1_score: 0.8807\n",
            "Epoch 316/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2795 - f1_score: 0.8319 - val_loss: 0.2774 - val_f1_score: 0.8809\n",
            "Epoch 317/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2791 - f1_score: 0.8321 - val_loss: 0.2770 - val_f1_score: 0.8812\n",
            "Epoch 318/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2787 - f1_score: 0.8321 - val_loss: 0.2766 - val_f1_score: 0.8813\n",
            "Epoch 319/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2783 - f1_score: 0.8322 - val_loss: 0.2762 - val_f1_score: 0.8815\n",
            "Epoch 320/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2779 - f1_score: 0.8323 - val_loss: 0.2759 - val_f1_score: 0.8816\n",
            "Epoch 321/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2775 - f1_score: 0.8325 - val_loss: 0.2755 - val_f1_score: 0.8816\n",
            "Epoch 322/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2771 - f1_score: 0.8326 - val_loss: 0.2751 - val_f1_score: 0.8817\n",
            "Epoch 323/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2767 - f1_score: 0.8328 - val_loss: 0.2748 - val_f1_score: 0.8819\n",
            "Epoch 324/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2764 - f1_score: 0.8328 - val_loss: 0.2744 - val_f1_score: 0.8820\n",
            "Epoch 325/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2760 - f1_score: 0.8329 - val_loss: 0.2741 - val_f1_score: 0.8825\n",
            "Epoch 326/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2756 - f1_score: 0.8330 - val_loss: 0.2737 - val_f1_score: 0.8825\n",
            "Epoch 327/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2752 - f1_score: 0.8329 - val_loss: 0.2734 - val_f1_score: 0.8827\n",
            "Epoch 328/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2749 - f1_score: 0.8329 - val_loss: 0.2730 - val_f1_score: 0.8827\n",
            "Epoch 329/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2745 - f1_score: 0.8329 - val_loss: 0.2727 - val_f1_score: 0.8828\n",
            "Epoch 330/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2741 - f1_score: 0.8330 - val_loss: 0.2724 - val_f1_score: 0.8830\n",
            "Epoch 331/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2738 - f1_score: 0.8332 - val_loss: 0.2720 - val_f1_score: 0.8831\n",
            "Epoch 332/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2734 - f1_score: 0.8334 - val_loss: 0.2717 - val_f1_score: 0.8834\n",
            "Epoch 333/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2731 - f1_score: 0.8334 - val_loss: 0.2714 - val_f1_score: 0.8837\n",
            "Epoch 334/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2727 - f1_score: 0.8336 - val_loss: 0.2711 - val_f1_score: 0.8838\n",
            "Epoch 335/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2724 - f1_score: 0.8338 - val_loss: 0.2708 - val_f1_score: 0.8843\n",
            "Epoch 336/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2721 - f1_score: 0.8339 - val_loss: 0.2705 - val_f1_score: 0.8844\n",
            "Epoch 337/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2717 - f1_score: 0.8339 - val_loss: 0.2701 - val_f1_score: 0.8844\n",
            "Epoch 338/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2714 - f1_score: 0.8341 - val_loss: 0.2698 - val_f1_score: 0.8844\n",
            "Epoch 339/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2710 - f1_score: 0.8342 - val_loss: 0.2695 - val_f1_score: 0.8846\n",
            "Epoch 340/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2707 - f1_score: 0.8344 - val_loss: 0.2692 - val_f1_score: 0.8848\n",
            "Epoch 341/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2704 - f1_score: 0.8344 - val_loss: 0.2689 - val_f1_score: 0.8851\n",
            "Epoch 342/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2701 - f1_score: 0.8346 - val_loss: 0.2686 - val_f1_score: 0.8853\n",
            "Epoch 343/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2697 - f1_score: 0.8347 - val_loss: 0.2684 - val_f1_score: 0.8857\n",
            "Epoch 344/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2694 - f1_score: 0.8349 - val_loss: 0.2681 - val_f1_score: 0.8859\n",
            "Epoch 345/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2691 - f1_score: 0.8348 - val_loss: 0.2678 - val_f1_score: 0.8859\n",
            "Epoch 346/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2688 - f1_score: 0.8350 - val_loss: 0.2675 - val_f1_score: 0.8860\n",
            "Epoch 347/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2685 - f1_score: 0.8351 - val_loss: 0.2672 - val_f1_score: 0.8863\n",
            "Epoch 348/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2682 - f1_score: 0.8352 - val_loss: 0.2669 - val_f1_score: 0.8868\n",
            "Epoch 349/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2678 - f1_score: 0.8353 - val_loss: 0.2667 - val_f1_score: 0.8871\n",
            "Epoch 350/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2675 - f1_score: 0.8356 - val_loss: 0.2664 - val_f1_score: 0.8873\n",
            "Epoch 351/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2672 - f1_score: 0.8358 - val_loss: 0.2661 - val_f1_score: 0.8875\n",
            "Epoch 352/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2669 - f1_score: 0.8360 - val_loss: 0.2659 - val_f1_score: 0.8876\n",
            "Epoch 353/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2666 - f1_score: 0.8363 - val_loss: 0.2656 - val_f1_score: 0.8879\n",
            "Epoch 354/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2664 - f1_score: 0.8365 - val_loss: 0.2654 - val_f1_score: 0.8881\n",
            "Epoch 355/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2661 - f1_score: 0.8366 - val_loss: 0.2651 - val_f1_score: 0.8881\n",
            "Epoch 356/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2658 - f1_score: 0.8368 - val_loss: 0.2649 - val_f1_score: 0.8884\n",
            "Epoch 357/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2655 - f1_score: 0.8370 - val_loss: 0.2646 - val_f1_score: 0.8886\n",
            "Epoch 358/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2652 - f1_score: 0.8372 - val_loss: 0.2644 - val_f1_score: 0.8889\n",
            "Epoch 359/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2649 - f1_score: 0.8373 - val_loss: 0.2641 - val_f1_score: 0.8890\n",
            "Epoch 360/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2646 - f1_score: 0.8373 - val_loss: 0.2639 - val_f1_score: 0.8892\n",
            "Epoch 361/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2644 - f1_score: 0.8373 - val_loss: 0.2637 - val_f1_score: 0.8891\n",
            "Epoch 362/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2641 - f1_score: 0.8375 - val_loss: 0.2634 - val_f1_score: 0.8893\n",
            "Epoch 363/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2638 - f1_score: 0.8377 - val_loss: 0.2632 - val_f1_score: 0.8896\n",
            "Epoch 364/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2635 - f1_score: 0.8378 - val_loss: 0.2630 - val_f1_score: 0.8898\n",
            "Epoch 365/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2633 - f1_score: 0.8379 - val_loss: 0.2628 - val_f1_score: 0.8898\n",
            "Epoch 366/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2630 - f1_score: 0.8380 - val_loss: 0.2625 - val_f1_score: 0.8901\n",
            "Epoch 367/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2627 - f1_score: 0.8383 - val_loss: 0.2623 - val_f1_score: 0.8901\n",
            "Epoch 368/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2625 - f1_score: 0.8404 - val_loss: 0.2621 - val_f1_score: 0.8933\n",
            "Epoch 369/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2622 - f1_score: 0.8408 - val_loss: 0.2619 - val_f1_score: 0.8936\n",
            "Epoch 370/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2620 - f1_score: 0.8408 - val_loss: 0.2617 - val_f1_score: 0.8939\n",
            "Epoch 371/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2617 - f1_score: 0.8410 - val_loss: 0.2615 - val_f1_score: 0.8941\n",
            "Epoch 372/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2615 - f1_score: 0.8411 - val_loss: 0.2613 - val_f1_score: 0.8943\n",
            "Epoch 373/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2612 - f1_score: 0.8414 - val_loss: 0.2610 - val_f1_score: 0.8947\n",
            "Epoch 374/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2609 - f1_score: 0.8417 - val_loss: 0.2608 - val_f1_score: 0.8948\n",
            "Epoch 375/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2607 - f1_score: 0.8418 - val_loss: 0.2606 - val_f1_score: 0.8950\n",
            "Epoch 376/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2605 - f1_score: 0.8419 - val_loss: 0.2604 - val_f1_score: 0.8951\n",
            "Epoch 377/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2602 - f1_score: 0.8421 - val_loss: 0.2602 - val_f1_score: 0.8953\n",
            "Epoch 378/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2600 - f1_score: 0.8424 - val_loss: 0.2600 - val_f1_score: 0.8955\n",
            "Epoch 379/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2597 - f1_score: 0.8426 - val_loss: 0.2598 - val_f1_score: 0.8954\n",
            "Epoch 380/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2595 - f1_score: 0.8426 - val_loss: 0.2596 - val_f1_score: 0.8955\n",
            "Epoch 381/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2592 - f1_score: 0.8428 - val_loss: 0.2595 - val_f1_score: 0.8956\n",
            "Epoch 382/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2590 - f1_score: 0.8429 - val_loss: 0.2593 - val_f1_score: 0.8955\n",
            "Epoch 383/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2588 - f1_score: 0.8431 - val_loss: 0.2591 - val_f1_score: 0.8957\n",
            "Epoch 384/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2585 - f1_score: 0.8433 - val_loss: 0.2589 - val_f1_score: 0.8960\n",
            "Epoch 385/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2583 - f1_score: 0.8435 - val_loss: 0.2587 - val_f1_score: 0.8957\n",
            "Epoch 386/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2581 - f1_score: 0.8436 - val_loss: 0.2585 - val_f1_score: 0.8956\n",
            "Epoch 387/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2578 - f1_score: 0.8439 - val_loss: 0.2583 - val_f1_score: 0.8951\n",
            "Epoch 388/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2576 - f1_score: 0.8442 - val_loss: 0.2581 - val_f1_score: 0.8947\n",
            "Epoch 389/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2574 - f1_score: 0.8444 - val_loss: 0.2580 - val_f1_score: 0.8943\n",
            "Epoch 390/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2572 - f1_score: 0.8447 - val_loss: 0.2578 - val_f1_score: 0.8941\n",
            "Epoch 391/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2570 - f1_score: 0.8447 - val_loss: 0.2576 - val_f1_score: 0.8938\n",
            "Epoch 392/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2567 - f1_score: 0.8449 - val_loss: 0.2574 - val_f1_score: 0.8934\n",
            "Epoch 393/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2565 - f1_score: 0.8451 - val_loss: 0.2572 - val_f1_score: 0.8933\n",
            "Epoch 394/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2563 - f1_score: 0.8452 - val_loss: 0.2571 - val_f1_score: 0.8928\n",
            "Epoch 395/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2561 - f1_score: 0.8454 - val_loss: 0.2569 - val_f1_score: 0.8922\n",
            "Epoch 396/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2559 - f1_score: 0.8456 - val_loss: 0.2567 - val_f1_score: 0.8921\n",
            "Epoch 397/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2557 - f1_score: 0.8458 - val_loss: 0.2565 - val_f1_score: 0.8916\n",
            "Epoch 398/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2554 - f1_score: 0.8460 - val_loss: 0.2564 - val_f1_score: 0.8912\n",
            "Epoch 399/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2552 - f1_score: 0.8461 - val_loss: 0.2562 - val_f1_score: 0.8911\n",
            "Epoch 400/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2550 - f1_score: 0.8463 - val_loss: 0.2560 - val_f1_score: 0.8910\n",
            "Epoch 401/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2548 - f1_score: 0.8465 - val_loss: 0.2559 - val_f1_score: 0.8911\n",
            "Epoch 402/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2546 - f1_score: 0.8466 - val_loss: 0.2557 - val_f1_score: 0.8908\n",
            "Epoch 403/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2544 - f1_score: 0.8468 - val_loss: 0.2555 - val_f1_score: 0.8909\n",
            "Epoch 404/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2542 - f1_score: 0.8470 - val_loss: 0.2554 - val_f1_score: 0.8905\n",
            "Epoch 405/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2540 - f1_score: 0.8473 - val_loss: 0.2552 - val_f1_score: 0.8907\n",
            "Epoch 406/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2538 - f1_score: 0.8474 - val_loss: 0.2550 - val_f1_score: 0.8905\n",
            "Epoch 407/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2536 - f1_score: 0.8478 - val_loss: 0.2549 - val_f1_score: 0.8907\n",
            "Epoch 408/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2534 - f1_score: 0.8480 - val_loss: 0.2547 - val_f1_score: 0.8904\n",
            "Epoch 409/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2532 - f1_score: 0.8482 - val_loss: 0.2545 - val_f1_score: 0.8902\n",
            "Epoch 410/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2530 - f1_score: 0.8484 - val_loss: 0.2544 - val_f1_score: 0.8900\n",
            "Epoch 411/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2528 - f1_score: 0.8487 - val_loss: 0.2542 - val_f1_score: 0.8899\n",
            "Epoch 412/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2526 - f1_score: 0.8490 - val_loss: 0.2540 - val_f1_score: 0.8897\n",
            "Epoch 413/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2524 - f1_score: 0.8493 - val_loss: 0.2539 - val_f1_score: 0.8895\n",
            "Epoch 414/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2522 - f1_score: 0.8496 - val_loss: 0.2537 - val_f1_score: 0.8895\n",
            "Epoch 415/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2520 - f1_score: 0.8498 - val_loss: 0.2536 - val_f1_score: 0.8893\n",
            "Epoch 416/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2518 - f1_score: 0.8500 - val_loss: 0.2534 - val_f1_score: 0.8891\n",
            "Epoch 417/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2516 - f1_score: 0.8503 - val_loss: 0.2532 - val_f1_score: 0.8889\n",
            "Epoch 418/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2514 - f1_score: 0.8505 - val_loss: 0.2531 - val_f1_score: 0.8888\n",
            "Epoch 419/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2512 - f1_score: 0.8507 - val_loss: 0.2529 - val_f1_score: 0.8888\n",
            "Epoch 420/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2511 - f1_score: 0.8510 - val_loss: 0.2528 - val_f1_score: 0.8886\n",
            "Epoch 421/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2509 - f1_score: 0.8513 - val_loss: 0.2526 - val_f1_score: 0.8885\n",
            "Epoch 422/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2507 - f1_score: 0.8514 - val_loss: 0.2525 - val_f1_score: 0.8884\n",
            "Epoch 423/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2505 - f1_score: 0.8515 - val_loss: 0.2523 - val_f1_score: 0.8883\n",
            "Epoch 424/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2503 - f1_score: 0.8518 - val_loss: 0.2522 - val_f1_score: 0.8882\n",
            "Epoch 425/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2501 - f1_score: 0.8522 - val_loss: 0.2520 - val_f1_score: 0.8882\n",
            "Epoch 426/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2499 - f1_score: 0.8525 - val_loss: 0.2519 - val_f1_score: 0.8880\n",
            "Epoch 427/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2498 - f1_score: 0.8528 - val_loss: 0.2517 - val_f1_score: 0.8878\n",
            "Epoch 428/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2496 - f1_score: 0.8530 - val_loss: 0.2515 - val_f1_score: 0.8876\n",
            "Epoch 429/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2494 - f1_score: 0.8532 - val_loss: 0.2514 - val_f1_score: 0.8877\n",
            "Epoch 430/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2492 - f1_score: 0.8535 - val_loss: 0.2512 - val_f1_score: 0.8877\n",
            "Epoch 431/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2490 - f1_score: 0.8538 - val_loss: 0.2511 - val_f1_score: 0.8877\n",
            "Epoch 432/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2488 - f1_score: 0.8540 - val_loss: 0.2509 - val_f1_score: 0.8879\n",
            "Epoch 433/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2487 - f1_score: 0.8541 - val_loss: 0.2508 - val_f1_score: 0.8876\n",
            "Epoch 434/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2485 - f1_score: 0.8543 - val_loss: 0.2507 - val_f1_score: 0.8875\n",
            "Epoch 435/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2483 - f1_score: 0.8545 - val_loss: 0.2505 - val_f1_score: 0.8875\n",
            "Epoch 436/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2481 - f1_score: 0.8548 - val_loss: 0.2504 - val_f1_score: 0.8876\n",
            "Epoch 437/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2480 - f1_score: 0.8550 - val_loss: 0.2502 - val_f1_score: 0.8876\n",
            "Epoch 438/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2478 - f1_score: 0.8551 - val_loss: 0.2501 - val_f1_score: 0.8875\n",
            "Epoch 439/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2476 - f1_score: 0.8553 - val_loss: 0.2499 - val_f1_score: 0.8875\n",
            "Epoch 440/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2475 - f1_score: 0.8556 - val_loss: 0.2498 - val_f1_score: 0.8874\n",
            "Epoch 441/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2473 - f1_score: 0.8558 - val_loss: 0.2497 - val_f1_score: 0.8873\n",
            "Epoch 442/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2471 - f1_score: 0.8559 - val_loss: 0.2495 - val_f1_score: 0.8872\n",
            "Epoch 443/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2470 - f1_score: 0.8561 - val_loss: 0.2494 - val_f1_score: 0.8870\n",
            "Epoch 444/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2468 - f1_score: 0.8564 - val_loss: 0.2492 - val_f1_score: 0.8867\n",
            "Epoch 445/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2466 - f1_score: 0.8565 - val_loss: 0.2491 - val_f1_score: 0.8868\n",
            "Epoch 446/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2465 - f1_score: 0.8568 - val_loss: 0.2490 - val_f1_score: 0.8867\n",
            "Epoch 447/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2463 - f1_score: 0.8570 - val_loss: 0.2488 - val_f1_score: 0.8867\n",
            "Epoch 448/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2462 - f1_score: 0.8572 - val_loss: 0.2487 - val_f1_score: 0.8871\n",
            "Epoch 449/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2460 - f1_score: 0.8574 - val_loss: 0.2486 - val_f1_score: 0.8869\n",
            "Epoch 450/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2458 - f1_score: 0.8577 - val_loss: 0.2484 - val_f1_score: 0.8868\n",
            "Epoch 451/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2457 - f1_score: 0.8578 - val_loss: 0.2483 - val_f1_score: 0.8868\n",
            "Epoch 452/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2455 - f1_score: 0.8580 - val_loss: 0.2482 - val_f1_score: 0.8867\n",
            "Epoch 453/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2454 - f1_score: 0.8582 - val_loss: 0.2480 - val_f1_score: 0.8868\n",
            "Epoch 454/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2452 - f1_score: 0.8584 - val_loss: 0.2479 - val_f1_score: 0.8868\n",
            "Epoch 455/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2450 - f1_score: 0.8586 - val_loss: 0.2478 - val_f1_score: 0.8869\n",
            "Epoch 456/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2449 - f1_score: 0.8589 - val_loss: 0.2476 - val_f1_score: 0.8869\n",
            "Epoch 457/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2447 - f1_score: 0.8591 - val_loss: 0.2475 - val_f1_score: 0.8871\n",
            "Epoch 458/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2446 - f1_score: 0.8594 - val_loss: 0.2474 - val_f1_score: 0.8871\n",
            "Epoch 459/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2444 - f1_score: 0.8596 - val_loss: 0.2473 - val_f1_score: 0.8871\n",
            "Epoch 460/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2443 - f1_score: 0.8598 - val_loss: 0.2471 - val_f1_score: 0.8871\n",
            "Epoch 461/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2441 - f1_score: 0.8599 - val_loss: 0.2470 - val_f1_score: 0.8869\n",
            "Epoch 462/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2440 - f1_score: 0.8600 - val_loss: 0.2469 - val_f1_score: 0.8869\n",
            "Epoch 463/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2438 - f1_score: 0.8603 - val_loss: 0.2468 - val_f1_score: 0.8870\n",
            "Epoch 464/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2437 - f1_score: 0.8605 - val_loss: 0.2466 - val_f1_score: 0.8872\n",
            "Epoch 465/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2435 - f1_score: 0.8607 - val_loss: 0.2465 - val_f1_score: 0.8872\n",
            "Epoch 466/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2434 - f1_score: 0.8609 - val_loss: 0.2464 - val_f1_score: 0.8872\n",
            "Epoch 467/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2432 - f1_score: 0.8611 - val_loss: 0.2463 - val_f1_score: 0.8873\n",
            "Epoch 468/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2431 - f1_score: 0.8613 - val_loss: 0.2461 - val_f1_score: 0.8870\n",
            "Epoch 469/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2429 - f1_score: 0.8615 - val_loss: 0.2460 - val_f1_score: 0.8869\n",
            "Epoch 470/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2428 - f1_score: 0.8617 - val_loss: 0.2459 - val_f1_score: 0.8868\n",
            "Epoch 471/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2426 - f1_score: 0.8619 - val_loss: 0.2458 - val_f1_score: 0.8868\n",
            "Epoch 472/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2425 - f1_score: 0.8621 - val_loss: 0.2456 - val_f1_score: 0.8870\n",
            "Epoch 473/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2423 - f1_score: 0.8624 - val_loss: 0.2455 - val_f1_score: 0.8869\n",
            "Epoch 474/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2422 - f1_score: 0.8626 - val_loss: 0.2454 - val_f1_score: 0.8871\n",
            "Epoch 475/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2421 - f1_score: 0.8627 - val_loss: 0.2453 - val_f1_score: 0.8872\n",
            "Epoch 476/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2419 - f1_score: 0.8630 - val_loss: 0.2452 - val_f1_score: 0.8872\n",
            "Epoch 477/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2418 - f1_score: 0.8633 - val_loss: 0.2451 - val_f1_score: 0.8873\n",
            "Epoch 478/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2416 - f1_score: 0.8635 - val_loss: 0.2449 - val_f1_score: 0.8872\n",
            "Epoch 479/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2415 - f1_score: 0.8637 - val_loss: 0.2448 - val_f1_score: 0.8872\n",
            "Epoch 480/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2413 - f1_score: 0.8639 - val_loss: 0.2447 - val_f1_score: 0.8873\n",
            "Epoch 481/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2412 - f1_score: 0.8640 - val_loss: 0.2446 - val_f1_score: 0.8872\n",
            "Epoch 482/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2410 - f1_score: 0.8642 - val_loss: 0.2445 - val_f1_score: 0.8875\n",
            "Epoch 483/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2409 - f1_score: 0.8644 - val_loss: 0.2444 - val_f1_score: 0.8876\n",
            "Epoch 484/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2408 - f1_score: 0.8646 - val_loss: 0.2442 - val_f1_score: 0.8877\n",
            "Epoch 485/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2406 - f1_score: 0.8647 - val_loss: 0.2441 - val_f1_score: 0.8877\n",
            "Epoch 486/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2405 - f1_score: 0.8648 - val_loss: 0.2440 - val_f1_score: 0.8878\n",
            "Epoch 487/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2403 - f1_score: 0.8651 - val_loss: 0.2439 - val_f1_score: 0.8877\n",
            "Epoch 488/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2402 - f1_score: 0.8654 - val_loss: 0.2438 - val_f1_score: 0.8878\n",
            "Epoch 489/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2401 - f1_score: 0.8656 - val_loss: 0.2437 - val_f1_score: 0.8879\n",
            "Epoch 490/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2399 - f1_score: 0.8659 - val_loss: 0.2436 - val_f1_score: 0.8880\n",
            "Epoch 491/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2398 - f1_score: 0.8660 - val_loss: 0.2435 - val_f1_score: 0.8879\n",
            "Epoch 492/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2397 - f1_score: 0.8662 - val_loss: 0.2433 - val_f1_score: 0.8880\n",
            "Epoch 493/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2395 - f1_score: 0.8664 - val_loss: 0.2432 - val_f1_score: 0.8882\n",
            "Epoch 494/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2394 - f1_score: 0.8668 - val_loss: 0.2431 - val_f1_score: 0.8882\n",
            "Epoch 495/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2392 - f1_score: 0.8670 - val_loss: 0.2430 - val_f1_score: 0.8885\n",
            "Epoch 496/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2391 - f1_score: 0.8672 - val_loss: 0.2429 - val_f1_score: 0.8885\n",
            "Epoch 497/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2390 - f1_score: 0.8674 - val_loss: 0.2428 - val_f1_score: 0.8886\n",
            "Epoch 498/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2388 - f1_score: 0.8676 - val_loss: 0.2427 - val_f1_score: 0.8888\n",
            "Epoch 499/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2387 - f1_score: 0.8678 - val_loss: 0.2426 - val_f1_score: 0.8888\n",
            "Epoch 500/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2386 - f1_score: 0.8680 - val_loss: 0.2425 - val_f1_score: 0.8889\n",
            "Epoch 501/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2384 - f1_score: 0.8683 - val_loss: 0.2424 - val_f1_score: 0.8890\n",
            "Epoch 502/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2383 - f1_score: 0.8685 - val_loss: 0.2423 - val_f1_score: 0.8894\n",
            "Epoch 503/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2382 - f1_score: 0.8687 - val_loss: 0.2422 - val_f1_score: 0.8894\n",
            "Epoch 504/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2380 - f1_score: 0.8690 - val_loss: 0.2421 - val_f1_score: 0.8895\n",
            "Epoch 505/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2379 - f1_score: 0.8692 - val_loss: 0.2419 - val_f1_score: 0.8896\n",
            "Epoch 506/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2378 - f1_score: 0.8695 - val_loss: 0.2418 - val_f1_score: 0.8899\n",
            "Epoch 507/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2376 - f1_score: 0.8697 - val_loss: 0.2417 - val_f1_score: 0.8898\n",
            "Epoch 508/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2375 - f1_score: 0.8699 - val_loss: 0.2416 - val_f1_score: 0.8897\n",
            "Epoch 509/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2374 - f1_score: 0.8701 - val_loss: 0.2415 - val_f1_score: 0.8898\n",
            "Epoch 510/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2372 - f1_score: 0.8704 - val_loss: 0.2414 - val_f1_score: 0.8899\n",
            "Epoch 511/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2371 - f1_score: 0.8707 - val_loss: 0.2413 - val_f1_score: 0.8901\n",
            "Epoch 512/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2370 - f1_score: 0.8710 - val_loss: 0.2412 - val_f1_score: 0.8903\n",
            "Epoch 513/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2368 - f1_score: 0.8712 - val_loss: 0.2411 - val_f1_score: 0.8906\n",
            "Epoch 514/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2367 - f1_score: 0.8715 - val_loss: 0.2410 - val_f1_score: 0.8907\n",
            "Epoch 515/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2366 - f1_score: 0.8717 - val_loss: 0.2409 - val_f1_score: 0.8908\n",
            "Epoch 516/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2364 - f1_score: 0.8719 - val_loss: 0.2408 - val_f1_score: 0.8909\n",
            "Epoch 517/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2363 - f1_score: 0.8721 - val_loss: 0.2407 - val_f1_score: 0.8910\n",
            "Epoch 518/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2362 - f1_score: 0.8722 - val_loss: 0.2406 - val_f1_score: 0.8912\n",
            "Epoch 519/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2361 - f1_score: 0.8726 - val_loss: 0.2405 - val_f1_score: 0.8914\n",
            "Epoch 520/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2359 - f1_score: 0.8727 - val_loss: 0.2404 - val_f1_score: 0.8915\n",
            "Epoch 521/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2358 - f1_score: 0.8729 - val_loss: 0.2403 - val_f1_score: 0.8917\n",
            "Epoch 522/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2357 - f1_score: 0.8732 - val_loss: 0.2402 - val_f1_score: 0.8919\n",
            "Epoch 523/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2356 - f1_score: 0.8734 - val_loss: 0.2401 - val_f1_score: 0.8921\n",
            "Epoch 524/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2354 - f1_score: 0.8736 - val_loss: 0.2400 - val_f1_score: 0.8925\n",
            "Epoch 525/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2353 - f1_score: 0.8739 - val_loss: 0.2399 - val_f1_score: 0.8928\n",
            "Epoch 526/2000\n",
            "322/322 [==============================] - 1s 3ms/step - loss: 0.2352 - f1_score: 0.8742 - val_loss: 0.2398 - val_f1_score: 0.8930\n",
            "Epoch 527/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2350 - f1_score: 0.8745 - val_loss: 0.2397 - val_f1_score: 0.8930\n",
            "Epoch 528/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2349 - f1_score: 0.8746 - val_loss: 0.2397 - val_f1_score: 0.8930\n",
            "Epoch 529/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2348 - f1_score: 0.8748 - val_loss: 0.2396 - val_f1_score: 0.8931\n",
            "Epoch 530/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2347 - f1_score: 0.8752 - val_loss: 0.2395 - val_f1_score: 0.8931\n",
            "Epoch 531/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2345 - f1_score: 0.8755 - val_loss: 0.2394 - val_f1_score: 0.8933\n",
            "Epoch 532/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2344 - f1_score: 0.8758 - val_loss: 0.2393 - val_f1_score: 0.8935\n",
            "Epoch 533/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2343 - f1_score: 0.8760 - val_loss: 0.2392 - val_f1_score: 0.8936\n",
            "Epoch 534/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2342 - f1_score: 0.8762 - val_loss: 0.2391 - val_f1_score: 0.8936\n",
            "Epoch 535/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2340 - f1_score: 0.8766 - val_loss: 0.2390 - val_f1_score: 0.8937\n",
            "Epoch 536/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2339 - f1_score: 0.8769 - val_loss: 0.2389 - val_f1_score: 0.8938\n",
            "Epoch 537/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2338 - f1_score: 0.8771 - val_loss: 0.2388 - val_f1_score: 0.8937\n",
            "Epoch 538/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2337 - f1_score: 0.8775 - val_loss: 0.2387 - val_f1_score: 0.8937\n",
            "Epoch 539/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2336 - f1_score: 0.8778 - val_loss: 0.2386 - val_f1_score: 0.8938\n",
            "Epoch 540/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2334 - f1_score: 0.8781 - val_loss: 0.2385 - val_f1_score: 0.8940\n",
            "Epoch 541/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2333 - f1_score: 0.8783 - val_loss: 0.2384 - val_f1_score: 0.8939\n",
            "Epoch 542/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2332 - f1_score: 0.8786 - val_loss: 0.2383 - val_f1_score: 0.8940\n",
            "Epoch 543/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2331 - f1_score: 0.8790 - val_loss: 0.2383 - val_f1_score: 0.8942\n",
            "Epoch 544/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2329 - f1_score: 0.8794 - val_loss: 0.2382 - val_f1_score: 0.8943\n",
            "Epoch 545/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2328 - f1_score: 0.8796 - val_loss: 0.2381 - val_f1_score: 0.8943\n",
            "Epoch 546/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2327 - f1_score: 0.8798 - val_loss: 0.2380 - val_f1_score: 0.8944\n",
            "Epoch 547/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2326 - f1_score: 0.8801 - val_loss: 0.2379 - val_f1_score: 0.8943\n",
            "Epoch 548/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2325 - f1_score: 0.8803 - val_loss: 0.2378 - val_f1_score: 0.8943\n",
            "Epoch 549/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2323 - f1_score: 0.8806 - val_loss: 0.2377 - val_f1_score: 0.8944\n",
            "Epoch 550/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2322 - f1_score: 0.8809 - val_loss: 0.2376 - val_f1_score: 0.8944\n",
            "Epoch 551/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2321 - f1_score: 0.8812 - val_loss: 0.2375 - val_f1_score: 0.8945\n",
            "Epoch 552/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2320 - f1_score: 0.8816 - val_loss: 0.2374 - val_f1_score: 0.8946\n",
            "Epoch 553/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2319 - f1_score: 0.8819 - val_loss: 0.2374 - val_f1_score: 0.8947\n",
            "Epoch 554/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2318 - f1_score: 0.8823 - val_loss: 0.2373 - val_f1_score: 0.8944\n",
            "Epoch 555/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2316 - f1_score: 0.8825 - val_loss: 0.2372 - val_f1_score: 0.8944\n",
            "Epoch 556/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2315 - f1_score: 0.8828 - val_loss: 0.2371 - val_f1_score: 0.8945\n",
            "Epoch 557/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2314 - f1_score: 0.8831 - val_loss: 0.2370 - val_f1_score: 0.8945\n",
            "Epoch 558/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2313 - f1_score: 0.8833 - val_loss: 0.2369 - val_f1_score: 0.8944\n",
            "Epoch 559/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2312 - f1_score: 0.8835 - val_loss: 0.2368 - val_f1_score: 0.8944\n",
            "Epoch 560/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2310 - f1_score: 0.8837 - val_loss: 0.2367 - val_f1_score: 0.8944\n",
            "Epoch 561/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2309 - f1_score: 0.8839 - val_loss: 0.2366 - val_f1_score: 0.8944\n",
            "Epoch 562/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2308 - f1_score: 0.8840 - val_loss: 0.2366 - val_f1_score: 0.8944\n",
            "Epoch 563/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2307 - f1_score: 0.8843 - val_loss: 0.2365 - val_f1_score: 0.8944\n",
            "Epoch 564/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2306 - f1_score: 0.8845 - val_loss: 0.2364 - val_f1_score: 0.8944\n",
            "Epoch 565/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2305 - f1_score: 0.8846 - val_loss: 0.2363 - val_f1_score: 0.8944\n",
            "Epoch 566/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2304 - f1_score: 0.8848 - val_loss: 0.2362 - val_f1_score: 0.8945\n",
            "Epoch 567/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2302 - f1_score: 0.8852 - val_loss: 0.2361 - val_f1_score: 0.8946\n",
            "Epoch 568/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2301 - f1_score: 0.8853 - val_loss: 0.2360 - val_f1_score: 0.8948\n",
            "Epoch 569/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2300 - f1_score: 0.8856 - val_loss: 0.2360 - val_f1_score: 0.8949\n",
            "Epoch 570/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2299 - f1_score: 0.8858 - val_loss: 0.2359 - val_f1_score: 0.8948\n",
            "Epoch 571/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2298 - f1_score: 0.8861 - val_loss: 0.2358 - val_f1_score: 0.8948\n",
            "Epoch 572/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2297 - f1_score: 0.8862 - val_loss: 0.2357 - val_f1_score: 0.8948\n",
            "Epoch 573/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2296 - f1_score: 0.8864 - val_loss: 0.2356 - val_f1_score: 0.8948\n",
            "Epoch 574/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2294 - f1_score: 0.8866 - val_loss: 0.2355 - val_f1_score: 0.8950\n",
            "Epoch 575/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2293 - f1_score: 0.8869 - val_loss: 0.2354 - val_f1_score: 0.8952\n",
            "Epoch 576/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2292 - f1_score: 0.8870 - val_loss: 0.2354 - val_f1_score: 0.8953\n",
            "Epoch 577/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2291 - f1_score: 0.8872 - val_loss: 0.2353 - val_f1_score: 0.8953\n",
            "Epoch 578/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2290 - f1_score: 0.8874 - val_loss: 0.2352 - val_f1_score: 0.8953\n",
            "Epoch 579/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2289 - f1_score: 0.8876 - val_loss: 0.2351 - val_f1_score: 0.8954\n",
            "Epoch 580/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2288 - f1_score: 0.8878 - val_loss: 0.2350 - val_f1_score: 0.8955\n",
            "Epoch 581/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2287 - f1_score: 0.8881 - val_loss: 0.2349 - val_f1_score: 0.8956\n",
            "Epoch 582/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2285 - f1_score: 0.8883 - val_loss: 0.2348 - val_f1_score: 0.8958\n",
            "Epoch 583/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2284 - f1_score: 0.8884 - val_loss: 0.2348 - val_f1_score: 0.8956\n",
            "Epoch 584/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2283 - f1_score: 0.8886 - val_loss: 0.2347 - val_f1_score: 0.8957\n",
            "Epoch 585/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2282 - f1_score: 0.8887 - val_loss: 0.2346 - val_f1_score: 0.8957\n",
            "Epoch 586/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2281 - f1_score: 0.8888 - val_loss: 0.2345 - val_f1_score: 0.8957\n",
            "Epoch 587/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2280 - f1_score: 0.8890 - val_loss: 0.2344 - val_f1_score: 0.8955\n",
            "Epoch 588/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2279 - f1_score: 0.8890 - val_loss: 0.2343 - val_f1_score: 0.8955\n",
            "Epoch 589/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2278 - f1_score: 0.8891 - val_loss: 0.2342 - val_f1_score: 0.8956\n",
            "Epoch 590/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2277 - f1_score: 0.8891 - val_loss: 0.2342 - val_f1_score: 0.8957\n",
            "Epoch 591/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2276 - f1_score: 0.8893 - val_loss: 0.2341 - val_f1_score: 0.8957\n",
            "Epoch 592/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2274 - f1_score: 0.8895 - val_loss: 0.2340 - val_f1_score: 0.8957\n",
            "Epoch 593/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2273 - f1_score: 0.8896 - val_loss: 0.2339 - val_f1_score: 0.8957\n",
            "Epoch 594/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2272 - f1_score: 0.8897 - val_loss: 0.2338 - val_f1_score: 0.8957\n",
            "Epoch 595/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2271 - f1_score: 0.8899 - val_loss: 0.2337 - val_f1_score: 0.8959\n",
            "Epoch 596/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2270 - f1_score: 0.8900 - val_loss: 0.2337 - val_f1_score: 0.8960\n",
            "Epoch 597/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2269 - f1_score: 0.8900 - val_loss: 0.2336 - val_f1_score: 0.8961\n",
            "Epoch 598/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2268 - f1_score: 0.8902 - val_loss: 0.2335 - val_f1_score: 0.8963\n",
            "Epoch 599/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2267 - f1_score: 0.8903 - val_loss: 0.2334 - val_f1_score: 0.8963\n",
            "Epoch 600/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2266 - f1_score: 0.8904 - val_loss: 0.2333 - val_f1_score: 0.8964\n",
            "Epoch 601/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2265 - f1_score: 0.8906 - val_loss: 0.2332 - val_f1_score: 0.8964\n",
            "Epoch 602/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2264 - f1_score: 0.8908 - val_loss: 0.2332 - val_f1_score: 0.8964\n",
            "Epoch 603/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2263 - f1_score: 0.8909 - val_loss: 0.2331 - val_f1_score: 0.8965\n",
            "Epoch 604/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2262 - f1_score: 0.8910 - val_loss: 0.2330 - val_f1_score: 0.8966\n",
            "Epoch 605/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2261 - f1_score: 0.8911 - val_loss: 0.2329 - val_f1_score: 0.8966\n",
            "Epoch 606/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2259 - f1_score: 0.8912 - val_loss: 0.2328 - val_f1_score: 0.8967\n",
            "Epoch 607/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2258 - f1_score: 0.8913 - val_loss: 0.2327 - val_f1_score: 0.8969\n",
            "Epoch 608/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2257 - f1_score: 0.8914 - val_loss: 0.2326 - val_f1_score: 0.8969\n",
            "Epoch 609/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2256 - f1_score: 0.8915 - val_loss: 0.2326 - val_f1_score: 0.8969\n",
            "Epoch 610/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2255 - f1_score: 0.8917 - val_loss: 0.2325 - val_f1_score: 0.8969\n",
            "Epoch 611/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2254 - f1_score: 0.8919 - val_loss: 0.2324 - val_f1_score: 0.8969\n",
            "Epoch 612/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2253 - f1_score: 0.8921 - val_loss: 0.2323 - val_f1_score: 0.8969\n",
            "Epoch 613/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2252 - f1_score: 0.8923 - val_loss: 0.2322 - val_f1_score: 0.8969\n",
            "Epoch 614/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2251 - f1_score: 0.8924 - val_loss: 0.2321 - val_f1_score: 0.8971\n",
            "Epoch 615/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2250 - f1_score: 0.8925 - val_loss: 0.2321 - val_f1_score: 0.8972\n",
            "Epoch 616/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2249 - f1_score: 0.8926 - val_loss: 0.2320 - val_f1_score: 0.8972\n",
            "Epoch 617/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2248 - f1_score: 0.8927 - val_loss: 0.2319 - val_f1_score: 0.8972\n",
            "Epoch 618/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2247 - f1_score: 0.8929 - val_loss: 0.2318 - val_f1_score: 0.8972\n",
            "Epoch 619/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2246 - f1_score: 0.8929 - val_loss: 0.2317 - val_f1_score: 0.8972\n",
            "Epoch 620/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2245 - f1_score: 0.8931 - val_loss: 0.2316 - val_f1_score: 0.8972\n",
            "Epoch 621/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2244 - f1_score: 0.8932 - val_loss: 0.2315 - val_f1_score: 0.8973\n",
            "Epoch 622/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2243 - f1_score: 0.8933 - val_loss: 0.2315 - val_f1_score: 0.8973\n",
            "Epoch 623/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2242 - f1_score: 0.8934 - val_loss: 0.2314 - val_f1_score: 0.8973\n",
            "Epoch 624/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2241 - f1_score: 0.8936 - val_loss: 0.2313 - val_f1_score: 0.8974\n",
            "Epoch 625/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2240 - f1_score: 0.8938 - val_loss: 0.2312 - val_f1_score: 0.8975\n",
            "Epoch 626/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2239 - f1_score: 0.8940 - val_loss: 0.2311 - val_f1_score: 0.8975\n",
            "Epoch 627/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2238 - f1_score: 0.8941 - val_loss: 0.2310 - val_f1_score: 0.8975\n",
            "Epoch 628/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2237 - f1_score: 0.8942 - val_loss: 0.2310 - val_f1_score: 0.8976\n",
            "Epoch 629/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2236 - f1_score: 0.8943 - val_loss: 0.2309 - val_f1_score: 0.8976\n",
            "Epoch 630/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2235 - f1_score: 0.8944 - val_loss: 0.2308 - val_f1_score: 0.8978\n",
            "Epoch 631/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2234 - f1_score: 0.8945 - val_loss: 0.2307 - val_f1_score: 0.8979\n",
            "Epoch 632/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2233 - f1_score: 0.8945 - val_loss: 0.2306 - val_f1_score: 0.8981\n",
            "Epoch 633/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2232 - f1_score: 0.8946 - val_loss: 0.2305 - val_f1_score: 0.8981\n",
            "Epoch 634/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2231 - f1_score: 0.8948 - val_loss: 0.2304 - val_f1_score: 0.8982\n",
            "Epoch 635/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2230 - f1_score: 0.8948 - val_loss: 0.2304 - val_f1_score: 0.8982\n",
            "Epoch 636/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2229 - f1_score: 0.8950 - val_loss: 0.2303 - val_f1_score: 0.8985\n",
            "Epoch 637/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2228 - f1_score: 0.8951 - val_loss: 0.2302 - val_f1_score: 0.8986\n",
            "Epoch 638/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2227 - f1_score: 0.8953 - val_loss: 0.2301 - val_f1_score: 0.8986\n",
            "Epoch 639/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2226 - f1_score: 0.8953 - val_loss: 0.2300 - val_f1_score: 0.8986\n",
            "Epoch 640/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2225 - f1_score: 0.8955 - val_loss: 0.2299 - val_f1_score: 0.8988\n",
            "Epoch 641/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2224 - f1_score: 0.8956 - val_loss: 0.2298 - val_f1_score: 0.8987\n",
            "Epoch 642/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2223 - f1_score: 0.8957 - val_loss: 0.2298 - val_f1_score: 0.8987\n",
            "Epoch 643/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2222 - f1_score: 0.8958 - val_loss: 0.2297 - val_f1_score: 0.8988\n",
            "Epoch 644/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2221 - f1_score: 0.8958 - val_loss: 0.2296 - val_f1_score: 0.8990\n",
            "Epoch 645/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2220 - f1_score: 0.8959 - val_loss: 0.2295 - val_f1_score: 0.8990\n",
            "Epoch 646/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2219 - f1_score: 0.8960 - val_loss: 0.2294 - val_f1_score: 0.8990\n",
            "Epoch 647/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2218 - f1_score: 0.8961 - val_loss: 0.2293 - val_f1_score: 0.8991\n",
            "Epoch 648/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2217 - f1_score: 0.8961 - val_loss: 0.2292 - val_f1_score: 0.8994\n",
            "Epoch 649/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2216 - f1_score: 0.8962 - val_loss: 0.2292 - val_f1_score: 0.8995\n",
            "Epoch 650/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2215 - f1_score: 0.8963 - val_loss: 0.2291 - val_f1_score: 0.8996\n",
            "Epoch 651/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2214 - f1_score: 0.8964 - val_loss: 0.2290 - val_f1_score: 0.8996\n",
            "Epoch 652/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2213 - f1_score: 0.8965 - val_loss: 0.2289 - val_f1_score: 0.8997\n",
            "Epoch 653/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2212 - f1_score: 0.8966 - val_loss: 0.2288 - val_f1_score: 0.8998\n",
            "Epoch 654/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2211 - f1_score: 0.8967 - val_loss: 0.2287 - val_f1_score: 0.8998\n",
            "Epoch 655/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2210 - f1_score: 0.8967 - val_loss: 0.2286 - val_f1_score: 0.8998\n",
            "Epoch 656/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2209 - f1_score: 0.8968 - val_loss: 0.2286 - val_f1_score: 0.8998\n",
            "Epoch 657/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2208 - f1_score: 0.8969 - val_loss: 0.2285 - val_f1_score: 0.8998\n",
            "Epoch 658/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2207 - f1_score: 0.8970 - val_loss: 0.2284 - val_f1_score: 0.9001\n",
            "Epoch 659/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2207 - f1_score: 0.8971 - val_loss: 0.2283 - val_f1_score: 0.9002\n",
            "Epoch 660/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2206 - f1_score: 0.8972 - val_loss: 0.2282 - val_f1_score: 0.9004\n",
            "Epoch 661/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2205 - f1_score: 0.8973 - val_loss: 0.2281 - val_f1_score: 0.9005\n",
            "Epoch 662/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2204 - f1_score: 0.8974 - val_loss: 0.2280 - val_f1_score: 0.9005\n",
            "Epoch 663/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2203 - f1_score: 0.8974 - val_loss: 0.2280 - val_f1_score: 0.9007\n",
            "Epoch 664/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2202 - f1_score: 0.8975 - val_loss: 0.2279 - val_f1_score: 0.9008\n",
            "Epoch 665/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2201 - f1_score: 0.8976 - val_loss: 0.2278 - val_f1_score: 0.9008\n",
            "Epoch 666/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2200 - f1_score: 0.8976 - val_loss: 0.2277 - val_f1_score: 0.9008\n",
            "Epoch 667/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2199 - f1_score: 0.8977 - val_loss: 0.2276 - val_f1_score: 0.9009\n",
            "Epoch 668/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2198 - f1_score: 0.8977 - val_loss: 0.2275 - val_f1_score: 0.9010\n",
            "Epoch 669/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2197 - f1_score: 0.8978 - val_loss: 0.2274 - val_f1_score: 0.9012\n",
            "Epoch 670/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2196 - f1_score: 0.8978 - val_loss: 0.2273 - val_f1_score: 0.9015\n",
            "Epoch 671/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2195 - f1_score: 0.8979 - val_loss: 0.2273 - val_f1_score: 0.9016\n",
            "Epoch 672/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2194 - f1_score: 0.8979 - val_loss: 0.2272 - val_f1_score: 0.9016\n",
            "Epoch 673/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2193 - f1_score: 0.8980 - val_loss: 0.2271 - val_f1_score: 0.9018\n",
            "Epoch 674/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2192 - f1_score: 0.8981 - val_loss: 0.2270 - val_f1_score: 0.9018\n",
            "Epoch 675/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2192 - f1_score: 0.8982 - val_loss: 0.2269 - val_f1_score: 0.9018\n",
            "Epoch 676/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2191 - f1_score: 0.8983 - val_loss: 0.2268 - val_f1_score: 0.9019\n",
            "Epoch 677/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2190 - f1_score: 0.8984 - val_loss: 0.2267 - val_f1_score: 0.9021\n",
            "Epoch 678/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2189 - f1_score: 0.8984 - val_loss: 0.2267 - val_f1_score: 0.9022\n",
            "Epoch 679/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2188 - f1_score: 0.8985 - val_loss: 0.2266 - val_f1_score: 0.9022\n",
            "Epoch 680/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2187 - f1_score: 0.8985 - val_loss: 0.2265 - val_f1_score: 0.9023\n",
            "Epoch 681/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2186 - f1_score: 0.8985 - val_loss: 0.2264 - val_f1_score: 0.9023\n",
            "Epoch 682/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2185 - f1_score: 0.8986 - val_loss: 0.2263 - val_f1_score: 0.9023\n",
            "Epoch 683/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2184 - f1_score: 0.8986 - val_loss: 0.2262 - val_f1_score: 0.9024\n",
            "Epoch 684/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2183 - f1_score: 0.8987 - val_loss: 0.2261 - val_f1_score: 0.9024\n",
            "Epoch 685/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2182 - f1_score: 0.8988 - val_loss: 0.2260 - val_f1_score: 0.9024\n",
            "Epoch 686/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2182 - f1_score: 0.8989 - val_loss: 0.2260 - val_f1_score: 0.9024\n",
            "Epoch 687/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2181 - f1_score: 0.8989 - val_loss: 0.2259 - val_f1_score: 0.9025\n",
            "Epoch 688/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2180 - f1_score: 0.8990 - val_loss: 0.2258 - val_f1_score: 0.9028\n",
            "Epoch 689/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2179 - f1_score: 0.8991 - val_loss: 0.2257 - val_f1_score: 0.9029\n",
            "Epoch 690/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2178 - f1_score: 0.8991 - val_loss: 0.2256 - val_f1_score: 0.9029\n",
            "Epoch 691/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2177 - f1_score: 0.8991 - val_loss: 0.2255 - val_f1_score: 0.9029\n",
            "Epoch 692/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2176 - f1_score: 0.8992 - val_loss: 0.2254 - val_f1_score: 0.9031\n",
            "Epoch 693/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2175 - f1_score: 0.8992 - val_loss: 0.2254 - val_f1_score: 0.9033\n",
            "Epoch 694/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2174 - f1_score: 0.8992 - val_loss: 0.2253 - val_f1_score: 0.9034\n",
            "Epoch 695/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2174 - f1_score: 0.8993 - val_loss: 0.2252 - val_f1_score: 0.9036\n",
            "Epoch 696/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2173 - f1_score: 0.8993 - val_loss: 0.2251 - val_f1_score: 0.9036\n",
            "Epoch 697/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2172 - f1_score: 0.8994 - val_loss: 0.2250 - val_f1_score: 0.9037\n",
            "Epoch 698/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2171 - f1_score: 0.8994 - val_loss: 0.2249 - val_f1_score: 0.9037\n",
            "Epoch 699/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2170 - f1_score: 0.8995 - val_loss: 0.2248 - val_f1_score: 0.9037\n",
            "Epoch 700/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2169 - f1_score: 0.8995 - val_loss: 0.2247 - val_f1_score: 0.9037\n",
            "Epoch 701/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2168 - f1_score: 0.8996 - val_loss: 0.2247 - val_f1_score: 0.9038\n",
            "Epoch 702/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2167 - f1_score: 0.8997 - val_loss: 0.2246 - val_f1_score: 0.9040\n",
            "Epoch 703/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2167 - f1_score: 0.8997 - val_loss: 0.2245 - val_f1_score: 0.9042\n",
            "Epoch 704/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2166 - f1_score: 0.8998 - val_loss: 0.2244 - val_f1_score: 0.9043\n",
            "Epoch 705/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2165 - f1_score: 0.8999 - val_loss: 0.2243 - val_f1_score: 0.9044\n",
            "Epoch 706/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2164 - f1_score: 0.9000 - val_loss: 0.2242 - val_f1_score: 0.9046\n",
            "Epoch 707/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2163 - f1_score: 0.9000 - val_loss: 0.2241 - val_f1_score: 0.9048\n",
            "Epoch 708/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2162 - f1_score: 0.9002 - val_loss: 0.2240 - val_f1_score: 0.9048\n",
            "Epoch 709/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2161 - f1_score: 0.9002 - val_loss: 0.2240 - val_f1_score: 0.9049\n",
            "Epoch 710/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2160 - f1_score: 0.9002 - val_loss: 0.2239 - val_f1_score: 0.9049\n",
            "Epoch 711/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2160 - f1_score: 0.9002 - val_loss: 0.2238 - val_f1_score: 0.9052\n",
            "Epoch 712/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2159 - f1_score: 0.9003 - val_loss: 0.2237 - val_f1_score: 0.9052\n",
            "Epoch 713/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2158 - f1_score: 0.9003 - val_loss: 0.2236 - val_f1_score: 0.9053\n",
            "Epoch 714/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2157 - f1_score: 0.9004 - val_loss: 0.2235 - val_f1_score: 0.9054\n",
            "Epoch 715/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2156 - f1_score: 0.9004 - val_loss: 0.2234 - val_f1_score: 0.9054\n",
            "Epoch 716/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2155 - f1_score: 0.9005 - val_loss: 0.2233 - val_f1_score: 0.9055\n",
            "Epoch 717/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2154 - f1_score: 0.9006 - val_loss: 0.2233 - val_f1_score: 0.9057\n",
            "Epoch 718/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2154 - f1_score: 0.9006 - val_loss: 0.2232 - val_f1_score: 0.9058\n",
            "Epoch 719/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2153 - f1_score: 0.9006 - val_loss: 0.2231 - val_f1_score: 0.9059\n",
            "Epoch 720/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2152 - f1_score: 0.9007 - val_loss: 0.2230 - val_f1_score: 0.9060\n",
            "Epoch 721/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2151 - f1_score: 0.9008 - val_loss: 0.2229 - val_f1_score: 0.9061\n",
            "Epoch 722/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2150 - f1_score: 0.9008 - val_loss: 0.2228 - val_f1_score: 0.9062\n",
            "Epoch 723/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2149 - f1_score: 0.9009 - val_loss: 0.2227 - val_f1_score: 0.9062\n",
            "Epoch 724/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2149 - f1_score: 0.9009 - val_loss: 0.2227 - val_f1_score: 0.9062\n",
            "Epoch 725/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2148 - f1_score: 0.9010 - val_loss: 0.2226 - val_f1_score: 0.9063\n",
            "Epoch 726/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2147 - f1_score: 0.9010 - val_loss: 0.2225 - val_f1_score: 0.9064\n",
            "Epoch 727/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2146 - f1_score: 0.9010 - val_loss: 0.2224 - val_f1_score: 0.9065\n",
            "Epoch 728/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2145 - f1_score: 0.9011 - val_loss: 0.2223 - val_f1_score: 0.9066\n",
            "Epoch 729/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2144 - f1_score: 0.9011 - val_loss: 0.2222 - val_f1_score: 0.9067\n",
            "Epoch 730/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2144 - f1_score: 0.9012 - val_loss: 0.2221 - val_f1_score: 0.9068\n",
            "Epoch 731/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2143 - f1_score: 0.9013 - val_loss: 0.2220 - val_f1_score: 0.9069\n",
            "Epoch 732/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2142 - f1_score: 0.9013 - val_loss: 0.2220 - val_f1_score: 0.9070\n",
            "Epoch 733/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2141 - f1_score: 0.9014 - val_loss: 0.2219 - val_f1_score: 0.9070\n",
            "Epoch 734/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2140 - f1_score: 0.9015 - val_loss: 0.2218 - val_f1_score: 0.9071\n",
            "Epoch 735/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2139 - f1_score: 0.9015 - val_loss: 0.2217 - val_f1_score: 0.9071\n",
            "Epoch 736/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2139 - f1_score: 0.9016 - val_loss: 0.2216 - val_f1_score: 0.9072\n",
            "Epoch 737/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2138 - f1_score: 0.9016 - val_loss: 0.2215 - val_f1_score: 0.9074\n",
            "Epoch 738/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2137 - f1_score: 0.9017 - val_loss: 0.2214 - val_f1_score: 0.9075\n",
            "Epoch 739/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2136 - f1_score: 0.9018 - val_loss: 0.2213 - val_f1_score: 0.9077\n",
            "Epoch 740/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2135 - f1_score: 0.9018 - val_loss: 0.2213 - val_f1_score: 0.9079\n",
            "Epoch 741/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2135 - f1_score: 0.9018 - val_loss: 0.2212 - val_f1_score: 0.9080\n",
            "Epoch 742/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2134 - f1_score: 0.9019 - val_loss: 0.2211 - val_f1_score: 0.9080\n",
            "Epoch 743/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2133 - f1_score: 0.9020 - val_loss: 0.2210 - val_f1_score: 0.9080\n",
            "Epoch 744/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2132 - f1_score: 0.9020 - val_loss: 0.2209 - val_f1_score: 0.9079\n",
            "Epoch 745/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2131 - f1_score: 0.9020 - val_loss: 0.2208 - val_f1_score: 0.9080\n",
            "Epoch 746/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2130 - f1_score: 0.9021 - val_loss: 0.2207 - val_f1_score: 0.9081\n",
            "Epoch 747/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2130 - f1_score: 0.9021 - val_loss: 0.2207 - val_f1_score: 0.9082\n",
            "Epoch 748/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2129 - f1_score: 0.9022 - val_loss: 0.2206 - val_f1_score: 0.9084\n",
            "Epoch 749/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2128 - f1_score: 0.9022 - val_loss: 0.2205 - val_f1_score: 0.9086\n",
            "Epoch 750/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2127 - f1_score: 0.9022 - val_loss: 0.2204 - val_f1_score: 0.9087\n",
            "Epoch 751/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2126 - f1_score: 0.9022 - val_loss: 0.2203 - val_f1_score: 0.9087\n",
            "Epoch 752/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2126 - f1_score: 0.9023 - val_loss: 0.2202 - val_f1_score: 0.9088\n",
            "Epoch 753/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2125 - f1_score: 0.9024 - val_loss: 0.2201 - val_f1_score: 0.9088\n",
            "Epoch 754/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2124 - f1_score: 0.9025 - val_loss: 0.2200 - val_f1_score: 0.9090\n",
            "Epoch 755/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2123 - f1_score: 0.9026 - val_loss: 0.2200 - val_f1_score: 0.9090\n",
            "Epoch 756/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2122 - f1_score: 0.9026 - val_loss: 0.2199 - val_f1_score: 0.9091\n",
            "Epoch 757/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2122 - f1_score: 0.9026 - val_loss: 0.2198 - val_f1_score: 0.9093\n",
            "Epoch 758/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2121 - f1_score: 0.9027 - val_loss: 0.2197 - val_f1_score: 0.9093\n",
            "Epoch 759/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2120 - f1_score: 0.9027 - val_loss: 0.2196 - val_f1_score: 0.9095\n",
            "Epoch 760/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2119 - f1_score: 0.9028 - val_loss: 0.2195 - val_f1_score: 0.9096\n",
            "Epoch 761/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2118 - f1_score: 0.9029 - val_loss: 0.2194 - val_f1_score: 0.9097\n",
            "Epoch 762/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2118 - f1_score: 0.9029 - val_loss: 0.2194 - val_f1_score: 0.9098\n",
            "Epoch 763/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2117 - f1_score: 0.9030 - val_loss: 0.2193 - val_f1_score: 0.9098\n",
            "Epoch 764/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2116 - f1_score: 0.9030 - val_loss: 0.2192 - val_f1_score: 0.9099\n",
            "Epoch 765/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2115 - f1_score: 0.9030 - val_loss: 0.2191 - val_f1_score: 0.9099\n",
            "Epoch 766/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2115 - f1_score: 0.9030 - val_loss: 0.2190 - val_f1_score: 0.9100\n",
            "Epoch 767/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2114 - f1_score: 0.9031 - val_loss: 0.2189 - val_f1_score: 0.9102\n",
            "Epoch 768/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2113 - f1_score: 0.9032 - val_loss: 0.2188 - val_f1_score: 0.9103\n",
            "Epoch 769/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2112 - f1_score: 0.9032 - val_loss: 0.2187 - val_f1_score: 0.9105\n",
            "Epoch 770/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2111 - f1_score: 0.9033 - val_loss: 0.2187 - val_f1_score: 0.9106\n",
            "Epoch 771/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2111 - f1_score: 0.9033 - val_loss: 0.2186 - val_f1_score: 0.9107\n",
            "Epoch 772/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2110 - f1_score: 0.9034 - val_loss: 0.2185 - val_f1_score: 0.9108\n",
            "Epoch 773/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2109 - f1_score: 0.9035 - val_loss: 0.2184 - val_f1_score: 0.9110\n",
            "Epoch 774/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2108 - f1_score: 0.9035 - val_loss: 0.2183 - val_f1_score: 0.9110\n",
            "Epoch 775/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2108 - f1_score: 0.9035 - val_loss: 0.2182 - val_f1_score: 0.9112\n",
            "Epoch 776/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2107 - f1_score: 0.9035 - val_loss: 0.2181 - val_f1_score: 0.9113\n",
            "Epoch 777/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2106 - f1_score: 0.9036 - val_loss: 0.2181 - val_f1_score: 0.9113\n",
            "Epoch 778/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2105 - f1_score: 0.9036 - val_loss: 0.2180 - val_f1_score: 0.9114\n",
            "Epoch 779/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2104 - f1_score: 0.9036 - val_loss: 0.2179 - val_f1_score: 0.9116\n",
            "Epoch 780/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2104 - f1_score: 0.9037 - val_loss: 0.2178 - val_f1_score: 0.9116\n",
            "Epoch 781/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2103 - f1_score: 0.9037 - val_loss: 0.2177 - val_f1_score: 0.9118\n",
            "Epoch 782/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2102 - f1_score: 0.9038 - val_loss: 0.2176 - val_f1_score: 0.9119\n",
            "Epoch 783/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2101 - f1_score: 0.9038 - val_loss: 0.2175 - val_f1_score: 0.9119\n",
            "Epoch 784/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2101 - f1_score: 0.9038 - val_loss: 0.2175 - val_f1_score: 0.9119\n",
            "Epoch 785/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2100 - f1_score: 0.9039 - val_loss: 0.2174 - val_f1_score: 0.9120\n",
            "Epoch 786/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2099 - f1_score: 0.9040 - val_loss: 0.2173 - val_f1_score: 0.9122\n",
            "Epoch 787/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2098 - f1_score: 0.9041 - val_loss: 0.2172 - val_f1_score: 0.9122\n",
            "Epoch 788/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2098 - f1_score: 0.9041 - val_loss: 0.2171 - val_f1_score: 0.9123\n",
            "Epoch 789/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2097 - f1_score: 0.9042 - val_loss: 0.2170 - val_f1_score: 0.9124\n",
            "Epoch 790/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2096 - f1_score: 0.9042 - val_loss: 0.2169 - val_f1_score: 0.9124\n",
            "Epoch 791/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2095 - f1_score: 0.9043 - val_loss: 0.2169 - val_f1_score: 0.9127\n",
            "Epoch 792/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2095 - f1_score: 0.9043 - val_loss: 0.2168 - val_f1_score: 0.9129\n",
            "Epoch 793/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2094 - f1_score: 0.9043 - val_loss: 0.2167 - val_f1_score: 0.9131\n",
            "Epoch 794/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2093 - f1_score: 0.9044 - val_loss: 0.2166 - val_f1_score: 0.9132\n",
            "Epoch 795/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2092 - f1_score: 0.9044 - val_loss: 0.2165 - val_f1_score: 0.9133\n",
            "Epoch 796/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2092 - f1_score: 0.9045 - val_loss: 0.2164 - val_f1_score: 0.9134\n",
            "Epoch 797/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2091 - f1_score: 0.9046 - val_loss: 0.2163 - val_f1_score: 0.9137\n",
            "Epoch 798/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2090 - f1_score: 0.9046 - val_loss: 0.2163 - val_f1_score: 0.9139\n",
            "Epoch 799/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2089 - f1_score: 0.9047 - val_loss: 0.2162 - val_f1_score: 0.9141\n",
            "Epoch 800/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2089 - f1_score: 0.9047 - val_loss: 0.2161 - val_f1_score: 0.9141\n",
            "Epoch 801/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2088 - f1_score: 0.9047 - val_loss: 0.2160 - val_f1_score: 0.9142\n",
            "Epoch 802/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2087 - f1_score: 0.9048 - val_loss: 0.2159 - val_f1_score: 0.9142\n",
            "Epoch 803/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2086 - f1_score: 0.9048 - val_loss: 0.2158 - val_f1_score: 0.9143\n",
            "Epoch 804/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2086 - f1_score: 0.9049 - val_loss: 0.2157 - val_f1_score: 0.9145\n",
            "Epoch 805/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2085 - f1_score: 0.9050 - val_loss: 0.2157 - val_f1_score: 0.9145\n",
            "Epoch 806/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2084 - f1_score: 0.9050 - val_loss: 0.2156 - val_f1_score: 0.9146\n",
            "Epoch 807/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2083 - f1_score: 0.9050 - val_loss: 0.2155 - val_f1_score: 0.9146\n",
            "Epoch 808/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2083 - f1_score: 0.9051 - val_loss: 0.2154 - val_f1_score: 0.9146\n",
            "Epoch 809/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2082 - f1_score: 0.9052 - val_loss: 0.2153 - val_f1_score: 0.9147\n",
            "Epoch 810/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2081 - f1_score: 0.9053 - val_loss: 0.2152 - val_f1_score: 0.9148\n",
            "Epoch 811/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2080 - f1_score: 0.9053 - val_loss: 0.2152 - val_f1_score: 0.9148\n",
            "Epoch 812/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2080 - f1_score: 0.9054 - val_loss: 0.2151 - val_f1_score: 0.9150\n",
            "Epoch 813/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2079 - f1_score: 0.9054 - val_loss: 0.2150 - val_f1_score: 0.9150\n",
            "Epoch 814/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2078 - f1_score: 0.9055 - val_loss: 0.2149 - val_f1_score: 0.9151\n",
            "Epoch 815/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2078 - f1_score: 0.9056 - val_loss: 0.2148 - val_f1_score: 0.9151\n",
            "Epoch 816/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2077 - f1_score: 0.9056 - val_loss: 0.2147 - val_f1_score: 0.9149\n",
            "Epoch 817/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2076 - f1_score: 0.9057 - val_loss: 0.2146 - val_f1_score: 0.9150\n",
            "Epoch 818/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2075 - f1_score: 0.9057 - val_loss: 0.2146 - val_f1_score: 0.9152\n",
            "Epoch 819/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2075 - f1_score: 0.9056 - val_loss: 0.2145 - val_f1_score: 0.9153\n",
            "Epoch 820/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2074 - f1_score: 0.9056 - val_loss: 0.2144 - val_f1_score: 0.9155\n",
            "Epoch 821/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2073 - f1_score: 0.9057 - val_loss: 0.2143 - val_f1_score: 0.9158\n",
            "Epoch 822/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2072 - f1_score: 0.9057 - val_loss: 0.2142 - val_f1_score: 0.9158\n",
            "Epoch 823/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2072 - f1_score: 0.9058 - val_loss: 0.2141 - val_f1_score: 0.9159\n",
            "Epoch 824/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2071 - f1_score: 0.9058 - val_loss: 0.2141 - val_f1_score: 0.9158\n",
            "Epoch 825/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2070 - f1_score: 0.9059 - val_loss: 0.2140 - val_f1_score: 0.9159\n",
            "Epoch 826/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2070 - f1_score: 0.9059 - val_loss: 0.2139 - val_f1_score: 0.9161\n",
            "Epoch 827/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2069 - f1_score: 0.9060 - val_loss: 0.2138 - val_f1_score: 0.9163\n",
            "Epoch 828/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2068 - f1_score: 0.9060 - val_loss: 0.2137 - val_f1_score: 0.9165\n",
            "Epoch 829/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2067 - f1_score: 0.9060 - val_loss: 0.2136 - val_f1_score: 0.9165\n",
            "Epoch 830/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2067 - f1_score: 0.9061 - val_loss: 0.2135 - val_f1_score: 0.9166\n",
            "Epoch 831/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2066 - f1_score: 0.9061 - val_loss: 0.2135 - val_f1_score: 0.9169\n",
            "Epoch 832/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2065 - f1_score: 0.9062 - val_loss: 0.2134 - val_f1_score: 0.9170\n",
            "Epoch 833/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2065 - f1_score: 0.9062 - val_loss: 0.2133 - val_f1_score: 0.9171\n",
            "Epoch 834/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2064 - f1_score: 0.9062 - val_loss: 0.2132 - val_f1_score: 0.9173\n",
            "Epoch 835/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2063 - f1_score: 0.9063 - val_loss: 0.2131 - val_f1_score: 0.9173\n",
            "Epoch 836/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2062 - f1_score: 0.9063 - val_loss: 0.2130 - val_f1_score: 0.9174\n",
            "Epoch 837/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2062 - f1_score: 0.9064 - val_loss: 0.2130 - val_f1_score: 0.9175\n",
            "Epoch 838/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2061 - f1_score: 0.9064 - val_loss: 0.2129 - val_f1_score: 0.9178\n",
            "Epoch 839/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2060 - f1_score: 0.9064 - val_loss: 0.2128 - val_f1_score: 0.9178\n",
            "Epoch 840/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2060 - f1_score: 0.9065 - val_loss: 0.2127 - val_f1_score: 0.9178\n",
            "Epoch 841/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2059 - f1_score: 0.9065 - val_loss: 0.2126 - val_f1_score: 0.9179\n",
            "Epoch 842/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2058 - f1_score: 0.9066 - val_loss: 0.2125 - val_f1_score: 0.9179\n",
            "Epoch 843/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2057 - f1_score: 0.9066 - val_loss: 0.2125 - val_f1_score: 0.9181\n",
            "Epoch 844/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2057 - f1_score: 0.9066 - val_loss: 0.2124 - val_f1_score: 0.9183\n",
            "Epoch 845/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2056 - f1_score: 0.9067 - val_loss: 0.2123 - val_f1_score: 0.9184\n",
            "Epoch 846/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2055 - f1_score: 0.9068 - val_loss: 0.2122 - val_f1_score: 0.9184\n",
            "Epoch 847/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2055 - f1_score: 0.9068 - val_loss: 0.2121 - val_f1_score: 0.9185\n",
            "Epoch 848/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2054 - f1_score: 0.9068 - val_loss: 0.2120 - val_f1_score: 0.9185\n",
            "Epoch 849/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2053 - f1_score: 0.9068 - val_loss: 0.2120 - val_f1_score: 0.9187\n",
            "Epoch 850/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2053 - f1_score: 0.9069 - val_loss: 0.2119 - val_f1_score: 0.9189\n",
            "Epoch 851/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2052 - f1_score: 0.9069 - val_loss: 0.2118 - val_f1_score: 0.9189\n",
            "Epoch 852/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2051 - f1_score: 0.9069 - val_loss: 0.2117 - val_f1_score: 0.9190\n",
            "Epoch 853/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2050 - f1_score: 0.9070 - val_loss: 0.2116 - val_f1_score: 0.9191\n",
            "Epoch 854/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2050 - f1_score: 0.9070 - val_loss: 0.2116 - val_f1_score: 0.9192\n",
            "Epoch 855/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2049 - f1_score: 0.9071 - val_loss: 0.2115 - val_f1_score: 0.9193\n",
            "Epoch 856/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2048 - f1_score: 0.9071 - val_loss: 0.2114 - val_f1_score: 0.9194\n",
            "Epoch 857/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2048 - f1_score: 0.9071 - val_loss: 0.2113 - val_f1_score: 0.9195\n",
            "Epoch 858/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2047 - f1_score: 0.9072 - val_loss: 0.2112 - val_f1_score: 0.9198\n",
            "Epoch 859/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2046 - f1_score: 0.9072 - val_loss: 0.2111 - val_f1_score: 0.9199\n",
            "Epoch 860/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2046 - f1_score: 0.9072 - val_loss: 0.2111 - val_f1_score: 0.9200\n",
            "Epoch 861/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2045 - f1_score: 0.9073 - val_loss: 0.2110 - val_f1_score: 0.9203\n",
            "Epoch 862/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2044 - f1_score: 0.9073 - val_loss: 0.2109 - val_f1_score: 0.9203\n",
            "Epoch 863/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2044 - f1_score: 0.9074 - val_loss: 0.2108 - val_f1_score: 0.9203\n",
            "Epoch 864/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2043 - f1_score: 0.9074 - val_loss: 0.2107 - val_f1_score: 0.9204\n",
            "Epoch 865/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2042 - f1_score: 0.9074 - val_loss: 0.2106 - val_f1_score: 0.9205\n",
            "Epoch 866/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2041 - f1_score: 0.9075 - val_loss: 0.2106 - val_f1_score: 0.9208\n",
            "Epoch 867/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2041 - f1_score: 0.9075 - val_loss: 0.2105 - val_f1_score: 0.9208\n",
            "Epoch 868/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2040 - f1_score: 0.9076 - val_loss: 0.2104 - val_f1_score: 0.9209\n",
            "Epoch 869/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2039 - f1_score: 0.9076 - val_loss: 0.2103 - val_f1_score: 0.9210\n",
            "Epoch 870/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2039 - f1_score: 0.9077 - val_loss: 0.2102 - val_f1_score: 0.9210\n",
            "Epoch 871/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2038 - f1_score: 0.9077 - val_loss: 0.2102 - val_f1_score: 0.9212\n",
            "Epoch 872/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2037 - f1_score: 0.9077 - val_loss: 0.2101 - val_f1_score: 0.9212\n",
            "Epoch 873/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2037 - f1_score: 0.9077 - val_loss: 0.2100 - val_f1_score: 0.9213\n",
            "Epoch 874/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2036 - f1_score: 0.9078 - val_loss: 0.2099 - val_f1_score: 0.9214\n",
            "Epoch 875/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2035 - f1_score: 0.9078 - val_loss: 0.2098 - val_f1_score: 0.9214\n",
            "Epoch 876/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2035 - f1_score: 0.9078 - val_loss: 0.2098 - val_f1_score: 0.9215\n",
            "Epoch 877/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2034 - f1_score: 0.9078 - val_loss: 0.2097 - val_f1_score: 0.9215\n",
            "Epoch 878/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2033 - f1_score: 0.9079 - val_loss: 0.2096 - val_f1_score: 0.9216\n",
            "Epoch 879/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2033 - f1_score: 0.9079 - val_loss: 0.2095 - val_f1_score: 0.9218\n",
            "Epoch 880/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2032 - f1_score: 0.9079 - val_loss: 0.2094 - val_f1_score: 0.9219\n",
            "Epoch 881/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2031 - f1_score: 0.9080 - val_loss: 0.2093 - val_f1_score: 0.9220\n",
            "Epoch 882/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2031 - f1_score: 0.9080 - val_loss: 0.2093 - val_f1_score: 0.9222\n",
            "Epoch 883/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2030 - f1_score: 0.9081 - val_loss: 0.2092 - val_f1_score: 0.9224\n",
            "Epoch 884/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2029 - f1_score: 0.9081 - val_loss: 0.2091 - val_f1_score: 0.9225\n",
            "Epoch 885/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2029 - f1_score: 0.9081 - val_loss: 0.2090 - val_f1_score: 0.9228\n",
            "Epoch 886/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2028 - f1_score: 0.9082 - val_loss: 0.2089 - val_f1_score: 0.9228\n",
            "Epoch 887/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2027 - f1_score: 0.9082 - val_loss: 0.2089 - val_f1_score: 0.9229\n",
            "Epoch 888/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2027 - f1_score: 0.9083 - val_loss: 0.2088 - val_f1_score: 0.9229\n",
            "Epoch 889/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2026 - f1_score: 0.9083 - val_loss: 0.2087 - val_f1_score: 0.9231\n",
            "Epoch 890/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2025 - f1_score: 0.9084 - val_loss: 0.2086 - val_f1_score: 0.9232\n",
            "Epoch 891/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2025 - f1_score: 0.9084 - val_loss: 0.2085 - val_f1_score: 0.9232\n",
            "Epoch 892/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2024 - f1_score: 0.9085 - val_loss: 0.2085 - val_f1_score: 0.9234\n",
            "Epoch 893/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2023 - f1_score: 0.9085 - val_loss: 0.2084 - val_f1_score: 0.9236\n",
            "Epoch 894/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2023 - f1_score: 0.9086 - val_loss: 0.2083 - val_f1_score: 0.9236\n",
            "Epoch 895/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2022 - f1_score: 0.9086 - val_loss: 0.2082 - val_f1_score: 0.9236\n",
            "Epoch 896/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2021 - f1_score: 0.9087 - val_loss: 0.2081 - val_f1_score: 0.9237\n",
            "Epoch 897/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2021 - f1_score: 0.9087 - val_loss: 0.2081 - val_f1_score: 0.9240\n",
            "Epoch 898/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2020 - f1_score: 0.9087 - val_loss: 0.2080 - val_f1_score: 0.9240\n",
            "Epoch 899/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2019 - f1_score: 0.9088 - val_loss: 0.2079 - val_f1_score: 0.9241\n",
            "Epoch 900/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2019 - f1_score: 0.9088 - val_loss: 0.2078 - val_f1_score: 0.9242\n",
            "Epoch 901/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2018 - f1_score: 0.9089 - val_loss: 0.2077 - val_f1_score: 0.9242\n",
            "Epoch 902/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2017 - f1_score: 0.9090 - val_loss: 0.2077 - val_f1_score: 0.9244\n",
            "Epoch 903/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2017 - f1_score: 0.9090 - val_loss: 0.2076 - val_f1_score: 0.9244\n",
            "Epoch 904/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2016 - f1_score: 0.9090 - val_loss: 0.2075 - val_f1_score: 0.9244\n",
            "Epoch 905/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2015 - f1_score: 0.9091 - val_loss: 0.2074 - val_f1_score: 0.9245\n",
            "Epoch 906/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2015 - f1_score: 0.9091 - val_loss: 0.2074 - val_f1_score: 0.9248\n",
            "Epoch 907/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2014 - f1_score: 0.9091 - val_loss: 0.2073 - val_f1_score: 0.9249\n",
            "Epoch 908/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2013 - f1_score: 0.9092 - val_loss: 0.2072 - val_f1_score: 0.9249\n",
            "Epoch 909/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2013 - f1_score: 0.9093 - val_loss: 0.2071 - val_f1_score: 0.9250\n",
            "Epoch 910/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2012 - f1_score: 0.9093 - val_loss: 0.2070 - val_f1_score: 0.9249\n",
            "Epoch 911/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2012 - f1_score: 0.9093 - val_loss: 0.2070 - val_f1_score: 0.9251\n",
            "Epoch 912/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2011 - f1_score: 0.9093 - val_loss: 0.2069 - val_f1_score: 0.9251\n",
            "Epoch 913/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2010 - f1_score: 0.9094 - val_loss: 0.2068 - val_f1_score: 0.9251\n",
            "Epoch 914/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2010 - f1_score: 0.9094 - val_loss: 0.2067 - val_f1_score: 0.9253\n",
            "Epoch 915/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2009 - f1_score: 0.9094 - val_loss: 0.2066 - val_f1_score: 0.9255\n",
            "Epoch 916/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2008 - f1_score: 0.9094 - val_loss: 0.2066 - val_f1_score: 0.9255\n",
            "Epoch 917/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2008 - f1_score: 0.9095 - val_loss: 0.2065 - val_f1_score: 0.9255\n",
            "Epoch 918/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2007 - f1_score: 0.9095 - val_loss: 0.2064 - val_f1_score: 0.9257\n",
            "Epoch 919/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2006 - f1_score: 0.9096 - val_loss: 0.2063 - val_f1_score: 0.9257\n",
            "Epoch 920/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2006 - f1_score: 0.9096 - val_loss: 0.2063 - val_f1_score: 0.9259\n",
            "Epoch 921/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2005 - f1_score: 0.9096 - val_loss: 0.2062 - val_f1_score: 0.9260\n",
            "Epoch 922/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2004 - f1_score: 0.9096 - val_loss: 0.2061 - val_f1_score: 0.9262\n",
            "Epoch 923/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2004 - f1_score: 0.9096 - val_loss: 0.2060 - val_f1_score: 0.9264\n",
            "Epoch 924/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2003 - f1_score: 0.9097 - val_loss: 0.2059 - val_f1_score: 0.9265\n",
            "Epoch 925/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2002 - f1_score: 0.9097 - val_loss: 0.2059 - val_f1_score: 0.9265\n",
            "Epoch 926/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2002 - f1_score: 0.9098 - val_loss: 0.2058 - val_f1_score: 0.9267\n",
            "Epoch 927/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2001 - f1_score: 0.9098 - val_loss: 0.2057 - val_f1_score: 0.9267\n",
            "Epoch 928/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2001 - f1_score: 0.9099 - val_loss: 0.2056 - val_f1_score: 0.9269\n",
            "Epoch 929/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.2000 - f1_score: 0.9099 - val_loss: 0.2056 - val_f1_score: 0.9270\n",
            "Epoch 930/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1999 - f1_score: 0.9100 - val_loss: 0.2055 - val_f1_score: 0.9271\n",
            "Epoch 931/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1999 - f1_score: 0.9100 - val_loss: 0.2054 - val_f1_score: 0.9271\n",
            "Epoch 932/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1998 - f1_score: 0.9101 - val_loss: 0.2053 - val_f1_score: 0.9273\n",
            "Epoch 933/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1997 - f1_score: 0.9101 - val_loss: 0.2052 - val_f1_score: 0.9274\n",
            "Epoch 934/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1997 - f1_score: 0.9102 - val_loss: 0.2052 - val_f1_score: 0.9274\n",
            "Epoch 935/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1996 - f1_score: 0.9102 - val_loss: 0.2051 - val_f1_score: 0.9277\n",
            "Epoch 936/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1995 - f1_score: 0.9102 - val_loss: 0.2050 - val_f1_score: 0.9279\n",
            "Epoch 937/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1995 - f1_score: 0.9102 - val_loss: 0.2049 - val_f1_score: 0.9280\n",
            "Epoch 938/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1994 - f1_score: 0.9103 - val_loss: 0.2049 - val_f1_score: 0.9280\n",
            "Epoch 939/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1994 - f1_score: 0.9103 - val_loss: 0.2048 - val_f1_score: 0.9281\n",
            "Epoch 940/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1993 - f1_score: 0.9103 - val_loss: 0.2047 - val_f1_score: 0.9283\n",
            "Epoch 941/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1992 - f1_score: 0.9104 - val_loss: 0.2046 - val_f1_score: 0.9284\n",
            "Epoch 942/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1992 - f1_score: 0.9104 - val_loss: 0.2046 - val_f1_score: 0.9287\n",
            "Epoch 943/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1991 - f1_score: 0.9104 - val_loss: 0.2045 - val_f1_score: 0.9288\n",
            "Epoch 944/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1990 - f1_score: 0.9105 - val_loss: 0.2044 - val_f1_score: 0.9289\n",
            "Epoch 945/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1990 - f1_score: 0.9105 - val_loss: 0.2043 - val_f1_score: 0.9289\n",
            "Epoch 946/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1989 - f1_score: 0.9105 - val_loss: 0.2042 - val_f1_score: 0.9291\n",
            "Epoch 947/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1989 - f1_score: 0.9105 - val_loss: 0.2042 - val_f1_score: 0.9291\n",
            "Epoch 948/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1988 - f1_score: 0.9105 - val_loss: 0.2041 - val_f1_score: 0.9293\n",
            "Epoch 949/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1987 - f1_score: 0.9106 - val_loss: 0.2040 - val_f1_score: 0.9293\n",
            "Epoch 950/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1987 - f1_score: 0.9106 - val_loss: 0.2039 - val_f1_score: 0.9294\n",
            "Epoch 951/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1986 - f1_score: 0.9106 - val_loss: 0.2039 - val_f1_score: 0.9296\n",
            "Epoch 952/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1985 - f1_score: 0.9106 - val_loss: 0.2038 - val_f1_score: 0.9298\n",
            "Epoch 953/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1985 - f1_score: 0.9106 - val_loss: 0.2037 - val_f1_score: 0.9298\n",
            "Epoch 954/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1984 - f1_score: 0.9106 - val_loss: 0.2036 - val_f1_score: 0.9299\n",
            "Epoch 955/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1984 - f1_score: 0.9106 - val_loss: 0.2036 - val_f1_score: 0.9300\n",
            "Epoch 956/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1983 - f1_score: 0.9107 - val_loss: 0.2035 - val_f1_score: 0.9302\n",
            "Epoch 957/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1982 - f1_score: 0.9107 - val_loss: 0.2034 - val_f1_score: 0.9304\n",
            "Epoch 958/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1982 - f1_score: 0.9108 - val_loss: 0.2033 - val_f1_score: 0.9305\n",
            "Epoch 959/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1981 - f1_score: 0.9108 - val_loss: 0.2033 - val_f1_score: 0.9307\n",
            "Epoch 960/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1980 - f1_score: 0.9108 - val_loss: 0.2032 - val_f1_score: 0.9307\n",
            "Epoch 961/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1980 - f1_score: 0.9108 - val_loss: 0.2031 - val_f1_score: 0.9308\n",
            "Epoch 962/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1979 - f1_score: 0.9109 - val_loss: 0.2030 - val_f1_score: 0.9308\n",
            "Epoch 963/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1979 - f1_score: 0.9109 - val_loss: 0.2030 - val_f1_score: 0.9308\n",
            "Epoch 964/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1978 - f1_score: 0.9110 - val_loss: 0.2029 - val_f1_score: 0.9311\n",
            "Epoch 965/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1977 - f1_score: 0.9110 - val_loss: 0.2028 - val_f1_score: 0.9313\n",
            "Epoch 966/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1977 - f1_score: 0.9111 - val_loss: 0.2027 - val_f1_score: 0.9313\n",
            "Epoch 967/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1976 - f1_score: 0.9111 - val_loss: 0.2027 - val_f1_score: 0.9315\n",
            "Epoch 968/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1976 - f1_score: 0.9111 - val_loss: 0.2026 - val_f1_score: 0.9316\n",
            "Epoch 969/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1975 - f1_score: 0.9112 - val_loss: 0.2025 - val_f1_score: 0.9316\n",
            "Epoch 970/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1974 - f1_score: 0.9112 - val_loss: 0.2024 - val_f1_score: 0.9320\n",
            "Epoch 971/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1974 - f1_score: 0.9113 - val_loss: 0.2024 - val_f1_score: 0.9320\n",
            "Epoch 972/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1973 - f1_score: 0.9113 - val_loss: 0.2023 - val_f1_score: 0.9322\n",
            "Epoch 973/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1972 - f1_score: 0.9113 - val_loss: 0.2022 - val_f1_score: 0.9323\n",
            "Epoch 974/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1972 - f1_score: 0.9113 - val_loss: 0.2021 - val_f1_score: 0.9325\n",
            "Epoch 975/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1971 - f1_score: 0.9113 - val_loss: 0.2021 - val_f1_score: 0.9326\n",
            "Epoch 976/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1971 - f1_score: 0.9114 - val_loss: 0.2020 - val_f1_score: 0.9327\n",
            "Epoch 977/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1970 - f1_score: 0.9114 - val_loss: 0.2019 - val_f1_score: 0.9329\n",
            "Epoch 978/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1969 - f1_score: 0.9114 - val_loss: 0.2018 - val_f1_score: 0.9329\n",
            "Epoch 979/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1969 - f1_score: 0.9114 - val_loss: 0.2018 - val_f1_score: 0.9331\n",
            "Epoch 980/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1968 - f1_score: 0.9115 - val_loss: 0.2017 - val_f1_score: 0.9332\n",
            "Epoch 981/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1968 - f1_score: 0.9115 - val_loss: 0.2016 - val_f1_score: 0.9332\n",
            "Epoch 982/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1967 - f1_score: 0.9115 - val_loss: 0.2015 - val_f1_score: 0.9333\n",
            "Epoch 983/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1966 - f1_score: 0.9116 - val_loss: 0.2015 - val_f1_score: 0.9333\n",
            "Epoch 984/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1966 - f1_score: 0.9116 - val_loss: 0.2014 - val_f1_score: 0.9334\n",
            "Epoch 985/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1965 - f1_score: 0.9117 - val_loss: 0.2013 - val_f1_score: 0.9336\n",
            "Epoch 986/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1965 - f1_score: 0.9117 - val_loss: 0.2012 - val_f1_score: 0.9338\n",
            "Epoch 987/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1964 - f1_score: 0.9118 - val_loss: 0.2012 - val_f1_score: 0.9338\n",
            "Epoch 988/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1963 - f1_score: 0.9118 - val_loss: 0.2011 - val_f1_score: 0.9338\n",
            "Epoch 989/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1963 - f1_score: 0.9119 - val_loss: 0.2010 - val_f1_score: 0.9338\n",
            "Epoch 990/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1962 - f1_score: 0.9120 - val_loss: 0.2009 - val_f1_score: 0.9338\n",
            "Epoch 991/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1962 - f1_score: 0.9120 - val_loss: 0.2009 - val_f1_score: 0.9339\n",
            "Epoch 992/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1961 - f1_score: 0.9121 - val_loss: 0.2008 - val_f1_score: 0.9339\n",
            "Epoch 993/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1960 - f1_score: 0.9121 - val_loss: 0.2007 - val_f1_score: 0.9340\n",
            "Epoch 994/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1960 - f1_score: 0.9122 - val_loss: 0.2006 - val_f1_score: 0.9340\n",
            "Epoch 995/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1959 - f1_score: 0.9122 - val_loss: 0.2006 - val_f1_score: 0.9340\n",
            "Epoch 996/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1959 - f1_score: 0.9122 - val_loss: 0.2005 - val_f1_score: 0.9341\n",
            "Epoch 997/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1958 - f1_score: 0.9122 - val_loss: 0.2004 - val_f1_score: 0.9341\n",
            "Epoch 998/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1957 - f1_score: 0.9122 - val_loss: 0.2003 - val_f1_score: 0.9341\n",
            "Epoch 999/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1957 - f1_score: 0.9122 - val_loss: 0.2003 - val_f1_score: 0.9342\n",
            "Epoch 1000/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1956 - f1_score: 0.9122 - val_loss: 0.2002 - val_f1_score: 0.9342\n",
            "Epoch 1001/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1956 - f1_score: 0.9123 - val_loss: 0.2001 - val_f1_score: 0.9342\n",
            "Epoch 1002/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1955 - f1_score: 0.9123 - val_loss: 0.2000 - val_f1_score: 0.9342\n",
            "Epoch 1003/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1954 - f1_score: 0.9124 - val_loss: 0.2000 - val_f1_score: 0.9342\n",
            "Epoch 1004/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1954 - f1_score: 0.9124 - val_loss: 0.1999 - val_f1_score: 0.9342\n",
            "Epoch 1005/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1953 - f1_score: 0.9124 - val_loss: 0.1998 - val_f1_score: 0.9342\n",
            "Epoch 1006/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1953 - f1_score: 0.9125 - val_loss: 0.1998 - val_f1_score: 0.9342\n",
            "Epoch 1007/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1952 - f1_score: 0.9125 - val_loss: 0.1997 - val_f1_score: 0.9342\n",
            "Epoch 1008/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1951 - f1_score: 0.9125 - val_loss: 0.1996 - val_f1_score: 0.9342\n",
            "Epoch 1009/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1951 - f1_score: 0.9126 - val_loss: 0.1995 - val_f1_score: 0.9342\n",
            "Epoch 1010/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1950 - f1_score: 0.9126 - val_loss: 0.1995 - val_f1_score: 0.9342\n",
            "Epoch 1011/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1950 - f1_score: 0.9126 - val_loss: 0.1994 - val_f1_score: 0.9343\n",
            "Epoch 1012/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1949 - f1_score: 0.9127 - val_loss: 0.1993 - val_f1_score: 0.9343\n",
            "Epoch 1013/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1948 - f1_score: 0.9127 - val_loss: 0.1992 - val_f1_score: 0.9344\n",
            "Epoch 1014/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1948 - f1_score: 0.9128 - val_loss: 0.1992 - val_f1_score: 0.9344\n",
            "Epoch 1015/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1947 - f1_score: 0.9129 - val_loss: 0.1991 - val_f1_score: 0.9344\n",
            "Epoch 1016/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1947 - f1_score: 0.9130 - val_loss: 0.1990 - val_f1_score: 0.9344\n",
            "Epoch 1017/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1946 - f1_score: 0.9130 - val_loss: 0.1989 - val_f1_score: 0.9344\n",
            "Epoch 1018/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1945 - f1_score: 0.9130 - val_loss: 0.1989 - val_f1_score: 0.9344\n",
            "Epoch 1019/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1945 - f1_score: 0.9131 - val_loss: 0.1988 - val_f1_score: 0.9344\n",
            "Epoch 1020/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1944 - f1_score: 0.9131 - val_loss: 0.1987 - val_f1_score: 0.9344\n",
            "Epoch 1021/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1944 - f1_score: 0.9132 - val_loss: 0.1986 - val_f1_score: 0.9344\n",
            "Epoch 1022/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1943 - f1_score: 0.9132 - val_loss: 0.1986 - val_f1_score: 0.9344\n",
            "Epoch 1023/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1943 - f1_score: 0.9133 - val_loss: 0.1985 - val_f1_score: 0.9345\n",
            "Epoch 1024/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1942 - f1_score: 0.9133 - val_loss: 0.1984 - val_f1_score: 0.9345\n",
            "Epoch 1025/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1941 - f1_score: 0.9133 - val_loss: 0.1984 - val_f1_score: 0.9345\n",
            "Epoch 1026/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1941 - f1_score: 0.9133 - val_loss: 0.1983 - val_f1_score: 0.9346\n",
            "Epoch 1027/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1940 - f1_score: 0.9134 - val_loss: 0.1982 - val_f1_score: 0.9346\n",
            "Epoch 1028/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1940 - f1_score: 0.9134 - val_loss: 0.1981 - val_f1_score: 0.9347\n",
            "Epoch 1029/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1939 - f1_score: 0.9135 - val_loss: 0.1981 - val_f1_score: 0.9347\n",
            "Epoch 1030/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1938 - f1_score: 0.9135 - val_loss: 0.1980 - val_f1_score: 0.9347\n",
            "Epoch 1031/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1938 - f1_score: 0.9135 - val_loss: 0.1979 - val_f1_score: 0.9347\n",
            "Epoch 1032/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1937 - f1_score: 0.9136 - val_loss: 0.1978 - val_f1_score: 0.9348\n",
            "Epoch 1033/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1937 - f1_score: 0.9136 - val_loss: 0.1978 - val_f1_score: 0.9348\n",
            "Epoch 1034/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1936 - f1_score: 0.9136 - val_loss: 0.1977 - val_f1_score: 0.9348\n",
            "Epoch 1035/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1936 - f1_score: 0.9137 - val_loss: 0.1976 - val_f1_score: 0.9348\n",
            "Epoch 1036/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1935 - f1_score: 0.9138 - val_loss: 0.1976 - val_f1_score: 0.9349\n",
            "Epoch 1037/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1934 - f1_score: 0.9138 - val_loss: 0.1975 - val_f1_score: 0.9349\n",
            "Epoch 1038/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1934 - f1_score: 0.9138 - val_loss: 0.1974 - val_f1_score: 0.9349\n",
            "Epoch 1039/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1933 - f1_score: 0.9139 - val_loss: 0.1973 - val_f1_score: 0.9349\n",
            "Epoch 1040/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1933 - f1_score: 0.9139 - val_loss: 0.1973 - val_f1_score: 0.9349\n",
            "Epoch 1041/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1932 - f1_score: 0.9139 - val_loss: 0.1972 - val_f1_score: 0.9349\n",
            "Epoch 1042/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1932 - f1_score: 0.9140 - val_loss: 0.1971 - val_f1_score: 0.9349\n",
            "Epoch 1043/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1931 - f1_score: 0.9140 - val_loss: 0.1970 - val_f1_score: 0.9349\n",
            "Epoch 1044/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1930 - f1_score: 0.9140 - val_loss: 0.1970 - val_f1_score: 0.9349\n",
            "Epoch 1045/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1930 - f1_score: 0.9141 - val_loss: 0.1969 - val_f1_score: 0.9349\n",
            "Epoch 1046/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1929 - f1_score: 0.9141 - val_loss: 0.1968 - val_f1_score: 0.9349\n",
            "Epoch 1047/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1929 - f1_score: 0.9141 - val_loss: 0.1968 - val_f1_score: 0.9349\n",
            "Epoch 1048/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1928 - f1_score: 0.9142 - val_loss: 0.1967 - val_f1_score: 0.9349\n",
            "Epoch 1049/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1927 - f1_score: 0.9142 - val_loss: 0.1966 - val_f1_score: 0.9350\n",
            "Epoch 1050/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1927 - f1_score: 0.9143 - val_loss: 0.1965 - val_f1_score: 0.9350\n",
            "Epoch 1051/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1926 - f1_score: 0.9143 - val_loss: 0.1965 - val_f1_score: 0.9350\n",
            "Epoch 1052/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1926 - f1_score: 0.9144 - val_loss: 0.1964 - val_f1_score: 0.9351\n",
            "Epoch 1053/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1925 - f1_score: 0.9145 - val_loss: 0.1963 - val_f1_score: 0.9351\n",
            "Epoch 1054/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1925 - f1_score: 0.9145 - val_loss: 0.1962 - val_f1_score: 0.9351\n",
            "Epoch 1055/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1924 - f1_score: 0.9145 - val_loss: 0.1962 - val_f1_score: 0.9351\n",
            "Epoch 1056/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1923 - f1_score: 0.9145 - val_loss: 0.1961 - val_f1_score: 0.9351\n",
            "Epoch 1057/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1923 - f1_score: 0.9146 - val_loss: 0.1960 - val_f1_score: 0.9351\n",
            "Epoch 1058/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1922 - f1_score: 0.9146 - val_loss: 0.1960 - val_f1_score: 0.9351\n",
            "Epoch 1059/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1922 - f1_score: 0.9146 - val_loss: 0.1959 - val_f1_score: 0.9351\n",
            "Epoch 1060/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1921 - f1_score: 0.9146 - val_loss: 0.1958 - val_f1_score: 0.9351\n",
            "Epoch 1061/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1921 - f1_score: 0.9147 - val_loss: 0.1957 - val_f1_score: 0.9351\n",
            "Epoch 1062/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1920 - f1_score: 0.9147 - val_loss: 0.1957 - val_f1_score: 0.9352\n",
            "Epoch 1063/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1920 - f1_score: 0.9148 - val_loss: 0.1956 - val_f1_score: 0.9352\n",
            "Epoch 1064/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1919 - f1_score: 0.9148 - val_loss: 0.1955 - val_f1_score: 0.9352\n",
            "Epoch 1065/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1918 - f1_score: 0.9149 - val_loss: 0.1955 - val_f1_score: 0.9352\n",
            "Epoch 1066/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1918 - f1_score: 0.9149 - val_loss: 0.1954 - val_f1_score: 0.9352\n",
            "Epoch 1067/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1917 - f1_score: 0.9150 - val_loss: 0.1953 - val_f1_score: 0.9352\n",
            "Epoch 1068/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1917 - f1_score: 0.9150 - val_loss: 0.1952 - val_f1_score: 0.9352\n",
            "Epoch 1069/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1916 - f1_score: 0.9151 - val_loss: 0.1952 - val_f1_score: 0.9352\n",
            "Epoch 1070/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1916 - f1_score: 0.9151 - val_loss: 0.1951 - val_f1_score: 0.9352\n",
            "Epoch 1071/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1915 - f1_score: 0.9152 - val_loss: 0.1950 - val_f1_score: 0.9353\n",
            "Epoch 1072/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1914 - f1_score: 0.9152 - val_loss: 0.1950 - val_f1_score: 0.9353\n",
            "Epoch 1073/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1914 - f1_score: 0.9153 - val_loss: 0.1949 - val_f1_score: 0.9353\n",
            "Epoch 1074/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1913 - f1_score: 0.9153 - val_loss: 0.1948 - val_f1_score: 0.9353\n",
            "Epoch 1075/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1913 - f1_score: 0.9154 - val_loss: 0.1947 - val_f1_score: 0.9354\n",
            "Epoch 1076/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1912 - f1_score: 0.9154 - val_loss: 0.1947 - val_f1_score: 0.9356\n",
            "Epoch 1077/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1912 - f1_score: 0.9154 - val_loss: 0.1946 - val_f1_score: 0.9356\n",
            "Epoch 1078/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1911 - f1_score: 0.9155 - val_loss: 0.1945 - val_f1_score: 0.9356\n",
            "Epoch 1079/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1911 - f1_score: 0.9155 - val_loss: 0.1945 - val_f1_score: 0.9356\n",
            "Epoch 1080/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1910 - f1_score: 0.9156 - val_loss: 0.1944 - val_f1_score: 0.9356\n",
            "Epoch 1081/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1909 - f1_score: 0.9156 - val_loss: 0.1943 - val_f1_score: 0.9356\n",
            "Epoch 1082/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1909 - f1_score: 0.9156 - val_loss: 0.1942 - val_f1_score: 0.9356\n",
            "Epoch 1083/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1908 - f1_score: 0.9157 - val_loss: 0.1942 - val_f1_score: 0.9357\n",
            "Epoch 1084/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1908 - f1_score: 0.9157 - val_loss: 0.1941 - val_f1_score: 0.9357\n",
            "Epoch 1085/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1907 - f1_score: 0.9157 - val_loss: 0.1940 - val_f1_score: 0.9357\n",
            "Epoch 1086/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1907 - f1_score: 0.9157 - val_loss: 0.1940 - val_f1_score: 0.9357\n",
            "Epoch 1087/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1906 - f1_score: 0.9157 - val_loss: 0.1939 - val_f1_score: 0.9357\n",
            "Epoch 1088/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1906 - f1_score: 0.9158 - val_loss: 0.1938 - val_f1_score: 0.9357\n",
            "Epoch 1089/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1905 - f1_score: 0.9158 - val_loss: 0.1937 - val_f1_score: 0.9357\n",
            "Epoch 1090/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1904 - f1_score: 0.9158 - val_loss: 0.1937 - val_f1_score: 0.9359\n",
            "Epoch 1091/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1904 - f1_score: 0.9159 - val_loss: 0.1936 - val_f1_score: 0.9359\n",
            "Epoch 1092/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1903 - f1_score: 0.9159 - val_loss: 0.1935 - val_f1_score: 0.9359\n",
            "Epoch 1093/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1903 - f1_score: 0.9159 - val_loss: 0.1935 - val_f1_score: 0.9359\n",
            "Epoch 1094/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1902 - f1_score: 0.9160 - val_loss: 0.1934 - val_f1_score: 0.9359\n",
            "Epoch 1095/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1902 - f1_score: 0.9160 - val_loss: 0.1933 - val_f1_score: 0.9359\n",
            "Epoch 1096/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1901 - f1_score: 0.9161 - val_loss: 0.1932 - val_f1_score: 0.9360\n",
            "Epoch 1097/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1901 - f1_score: 0.9161 - val_loss: 0.1932 - val_f1_score: 0.9361\n",
            "Epoch 1098/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1900 - f1_score: 0.9161 - val_loss: 0.1931 - val_f1_score: 0.9361\n",
            "Epoch 1099/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1899 - f1_score: 0.9162 - val_loss: 0.1930 - val_f1_score: 0.9361\n",
            "Epoch 1100/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1899 - f1_score: 0.9163 - val_loss: 0.1930 - val_f1_score: 0.9362\n",
            "Epoch 1101/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1898 - f1_score: 0.9163 - val_loss: 0.1929 - val_f1_score: 0.9362\n",
            "Epoch 1102/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1898 - f1_score: 0.9163 - val_loss: 0.1928 - val_f1_score: 0.9363\n",
            "Epoch 1103/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1897 - f1_score: 0.9164 - val_loss: 0.1927 - val_f1_score: 0.9363\n",
            "Epoch 1104/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1897 - f1_score: 0.9164 - val_loss: 0.1927 - val_f1_score: 0.9363\n",
            "Epoch 1105/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1896 - f1_score: 0.9165 - val_loss: 0.1926 - val_f1_score: 0.9363\n",
            "Epoch 1106/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1896 - f1_score: 0.9165 - val_loss: 0.1925 - val_f1_score: 0.9363\n",
            "Epoch 1107/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1895 - f1_score: 0.9166 - val_loss: 0.1925 - val_f1_score: 0.9363\n",
            "Epoch 1108/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1895 - f1_score: 0.9167 - val_loss: 0.1924 - val_f1_score: 0.9363\n",
            "Epoch 1109/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1894 - f1_score: 0.9167 - val_loss: 0.1923 - val_f1_score: 0.9364\n",
            "Epoch 1110/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1893 - f1_score: 0.9168 - val_loss: 0.1922 - val_f1_score: 0.9364\n",
            "Epoch 1111/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1893 - f1_score: 0.9168 - val_loss: 0.1922 - val_f1_score: 0.9364\n",
            "Epoch 1112/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1892 - f1_score: 0.9169 - val_loss: 0.1921 - val_f1_score: 0.9364\n",
            "Epoch 1113/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1892 - f1_score: 0.9169 - val_loss: 0.1920 - val_f1_score: 0.9365\n",
            "Epoch 1114/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1891 - f1_score: 0.9170 - val_loss: 0.1920 - val_f1_score: 0.9365\n",
            "Epoch 1115/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1891 - f1_score: 0.9170 - val_loss: 0.1919 - val_f1_score: 0.9365\n",
            "Epoch 1116/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1890 - f1_score: 0.9170 - val_loss: 0.1918 - val_f1_score: 0.9365\n",
            "Epoch 1117/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1890 - f1_score: 0.9171 - val_loss: 0.1917 - val_f1_score: 0.9365\n",
            "Epoch 1118/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1889 - f1_score: 0.9171 - val_loss: 0.1917 - val_f1_score: 0.9365\n",
            "Epoch 1119/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1889 - f1_score: 0.9171 - val_loss: 0.1916 - val_f1_score: 0.9365\n",
            "Epoch 1120/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1888 - f1_score: 0.9171 - val_loss: 0.1915 - val_f1_score: 0.9365\n",
            "Epoch 1121/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1888 - f1_score: 0.9171 - val_loss: 0.1915 - val_f1_score: 0.9365\n",
            "Epoch 1122/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1887 - f1_score: 0.9171 - val_loss: 0.1914 - val_f1_score: 0.9366\n",
            "Epoch 1123/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1886 - f1_score: 0.9172 - val_loss: 0.1913 - val_f1_score: 0.9367\n",
            "Epoch 1124/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1886 - f1_score: 0.9173 - val_loss: 0.1913 - val_f1_score: 0.9367\n",
            "Epoch 1125/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1885 - f1_score: 0.9173 - val_loss: 0.1912 - val_f1_score: 0.9367\n",
            "Epoch 1126/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1885 - f1_score: 0.9173 - val_loss: 0.1911 - val_f1_score: 0.9368\n",
            "Epoch 1127/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1884 - f1_score: 0.9174 - val_loss: 0.1910 - val_f1_score: 0.9369\n",
            "Epoch 1128/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1884 - f1_score: 0.9175 - val_loss: 0.1910 - val_f1_score: 0.9369\n",
            "Epoch 1129/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1883 - f1_score: 0.9175 - val_loss: 0.1909 - val_f1_score: 0.9369\n",
            "Epoch 1130/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1883 - f1_score: 0.9176 - val_loss: 0.1908 - val_f1_score: 0.9369\n",
            "Epoch 1131/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1882 - f1_score: 0.9177 - val_loss: 0.1908 - val_f1_score: 0.9370\n",
            "Epoch 1132/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1882 - f1_score: 0.9177 - val_loss: 0.1907 - val_f1_score: 0.9370\n",
            "Epoch 1133/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1881 - f1_score: 0.9178 - val_loss: 0.1906 - val_f1_score: 0.9370\n",
            "Epoch 1134/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1881 - f1_score: 0.9178 - val_loss: 0.1905 - val_f1_score: 0.9370\n",
            "Epoch 1135/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1880 - f1_score: 0.9179 - val_loss: 0.1905 - val_f1_score: 0.9370\n",
            "Epoch 1136/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1880 - f1_score: 0.9180 - val_loss: 0.1904 - val_f1_score: 0.9370\n",
            "Epoch 1137/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1879 - f1_score: 0.9180 - val_loss: 0.1903 - val_f1_score: 0.9370\n",
            "Epoch 1138/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1878 - f1_score: 0.9181 - val_loss: 0.1903 - val_f1_score: 0.9370\n",
            "Epoch 1139/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1878 - f1_score: 0.9181 - val_loss: 0.1902 - val_f1_score: 0.9370\n",
            "Epoch 1140/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1877 - f1_score: 0.9181 - val_loss: 0.1901 - val_f1_score: 0.9371\n",
            "Epoch 1141/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1877 - f1_score: 0.9182 - val_loss: 0.1901 - val_f1_score: 0.9371\n",
            "Epoch 1142/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1876 - f1_score: 0.9183 - val_loss: 0.1900 - val_f1_score: 0.9371\n",
            "Epoch 1143/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1876 - f1_score: 0.9183 - val_loss: 0.1899 - val_f1_score: 0.9371\n",
            "Epoch 1144/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1875 - f1_score: 0.9184 - val_loss: 0.1898 - val_f1_score: 0.9371\n",
            "Epoch 1145/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1875 - f1_score: 0.9184 - val_loss: 0.1898 - val_f1_score: 0.9372\n",
            "Epoch 1146/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1874 - f1_score: 0.9184 - val_loss: 0.1897 - val_f1_score: 0.9372\n",
            "Epoch 1147/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1874 - f1_score: 0.9184 - val_loss: 0.1896 - val_f1_score: 0.9372\n",
            "Epoch 1148/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1873 - f1_score: 0.9185 - val_loss: 0.1896 - val_f1_score: 0.9373\n",
            "Epoch 1149/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1873 - f1_score: 0.9186 - val_loss: 0.1895 - val_f1_score: 0.9373\n",
            "Epoch 1150/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1872 - f1_score: 0.9187 - val_loss: 0.1894 - val_f1_score: 0.9373\n",
            "Epoch 1151/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1872 - f1_score: 0.9187 - val_loss: 0.1894 - val_f1_score: 0.9373\n",
            "Epoch 1152/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1871 - f1_score: 0.9188 - val_loss: 0.1893 - val_f1_score: 0.9373\n",
            "Epoch 1153/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1871 - f1_score: 0.9189 - val_loss: 0.1892 - val_f1_score: 0.9373\n",
            "Epoch 1154/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1870 - f1_score: 0.9190 - val_loss: 0.1892 - val_f1_score: 0.9373\n",
            "Epoch 1155/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1870 - f1_score: 0.9190 - val_loss: 0.1891 - val_f1_score: 0.9374\n",
            "Epoch 1156/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1869 - f1_score: 0.9191 - val_loss: 0.1890 - val_f1_score: 0.9375\n",
            "Epoch 1157/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1869 - f1_score: 0.9191 - val_loss: 0.1889 - val_f1_score: 0.9375\n",
            "Epoch 1158/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1868 - f1_score: 0.9192 - val_loss: 0.1889 - val_f1_score: 0.9375\n",
            "Epoch 1159/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1867 - f1_score: 0.9192 - val_loss: 0.1888 - val_f1_score: 0.9375\n",
            "Epoch 1160/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1867 - f1_score: 0.9193 - val_loss: 0.1887 - val_f1_score: 0.9375\n",
            "Epoch 1161/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1866 - f1_score: 0.9194 - val_loss: 0.1887 - val_f1_score: 0.9375\n",
            "Epoch 1162/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1866 - f1_score: 0.9195 - val_loss: 0.1886 - val_f1_score: 0.9375\n",
            "Epoch 1163/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1865 - f1_score: 0.9196 - val_loss: 0.1885 - val_f1_score: 0.9376\n",
            "Epoch 1164/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1865 - f1_score: 0.9197 - val_loss: 0.1885 - val_f1_score: 0.9376\n",
            "Epoch 1165/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1864 - f1_score: 0.9197 - val_loss: 0.1884 - val_f1_score: 0.9376\n",
            "Epoch 1166/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1864 - f1_score: 0.9198 - val_loss: 0.1883 - val_f1_score: 0.9376\n",
            "Epoch 1167/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1863 - f1_score: 0.9198 - val_loss: 0.1882 - val_f1_score: 0.9377\n",
            "Epoch 1168/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1863 - f1_score: 0.9198 - val_loss: 0.1882 - val_f1_score: 0.9377\n",
            "Epoch 1169/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1862 - f1_score: 0.9199 - val_loss: 0.1881 - val_f1_score: 0.9377\n",
            "Epoch 1170/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1862 - f1_score: 0.9199 - val_loss: 0.1880 - val_f1_score: 0.9378\n",
            "Epoch 1171/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1861 - f1_score: 0.9199 - val_loss: 0.1880 - val_f1_score: 0.9380\n",
            "Epoch 1172/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1861 - f1_score: 0.9200 - val_loss: 0.1879 - val_f1_score: 0.9380\n",
            "Epoch 1173/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1860 - f1_score: 0.9201 - val_loss: 0.1878 - val_f1_score: 0.9380\n",
            "Epoch 1174/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1860 - f1_score: 0.9201 - val_loss: 0.1878 - val_f1_score: 0.9380\n",
            "Epoch 1175/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1859 - f1_score: 0.9202 - val_loss: 0.1877 - val_f1_score: 0.9380\n",
            "Epoch 1176/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1859 - f1_score: 0.9203 - val_loss: 0.1876 - val_f1_score: 0.9381\n",
            "Epoch 1177/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1858 - f1_score: 0.9203 - val_loss: 0.1876 - val_f1_score: 0.9381\n",
            "Epoch 1178/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1858 - f1_score: 0.9203 - val_loss: 0.1875 - val_f1_score: 0.9382\n",
            "Epoch 1179/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1857 - f1_score: 0.9204 - val_loss: 0.1874 - val_f1_score: 0.9383\n",
            "Epoch 1180/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1857 - f1_score: 0.9205 - val_loss: 0.1874 - val_f1_score: 0.9383\n",
            "Epoch 1181/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1856 - f1_score: 0.9205 - val_loss: 0.1873 - val_f1_score: 0.9383\n",
            "Epoch 1182/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1856 - f1_score: 0.9206 - val_loss: 0.1872 - val_f1_score: 0.9383\n",
            "Epoch 1183/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1855 - f1_score: 0.9206 - val_loss: 0.1871 - val_f1_score: 0.9384\n",
            "Epoch 1184/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1855 - f1_score: 0.9207 - val_loss: 0.1871 - val_f1_score: 0.9384\n",
            "Epoch 1185/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1854 - f1_score: 0.9208 - val_loss: 0.1870 - val_f1_score: 0.9385\n",
            "Epoch 1186/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1854 - f1_score: 0.9208 - val_loss: 0.1869 - val_f1_score: 0.9385\n",
            "Epoch 1187/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1853 - f1_score: 0.9209 - val_loss: 0.1869 - val_f1_score: 0.9385\n",
            "Epoch 1188/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1853 - f1_score: 0.9208 - val_loss: 0.1868 - val_f1_score: 0.9385\n",
            "Epoch 1189/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1852 - f1_score: 0.9209 - val_loss: 0.1867 - val_f1_score: 0.9385\n",
            "Epoch 1190/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1852 - f1_score: 0.9210 - val_loss: 0.1867 - val_f1_score: 0.9386\n",
            "Epoch 1191/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1851 - f1_score: 0.9210 - val_loss: 0.1866 - val_f1_score: 0.9386\n",
            "Epoch 1192/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1851 - f1_score: 0.9211 - val_loss: 0.1865 - val_f1_score: 0.9386\n",
            "Epoch 1193/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1850 - f1_score: 0.9212 - val_loss: 0.1865 - val_f1_score: 0.9386\n",
            "Epoch 1194/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1850 - f1_score: 0.9212 - val_loss: 0.1864 - val_f1_score: 0.9386\n",
            "Epoch 1195/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1849 - f1_score: 0.9212 - val_loss: 0.1863 - val_f1_score: 0.9386\n",
            "Epoch 1196/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1849 - f1_score: 0.9213 - val_loss: 0.1863 - val_f1_score: 0.9388\n",
            "Epoch 1197/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1848 - f1_score: 0.9214 - val_loss: 0.1862 - val_f1_score: 0.9388\n",
            "Epoch 1198/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1848 - f1_score: 0.9214 - val_loss: 0.1861 - val_f1_score: 0.9388\n",
            "Epoch 1199/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1847 - f1_score: 0.9215 - val_loss: 0.1861 - val_f1_score: 0.9389\n",
            "Epoch 1200/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1847 - f1_score: 0.9215 - val_loss: 0.1860 - val_f1_score: 0.9389\n",
            "Epoch 1201/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1846 - f1_score: 0.9216 - val_loss: 0.1859 - val_f1_score: 0.9389\n",
            "Epoch 1202/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1846 - f1_score: 0.9216 - val_loss: 0.1858 - val_f1_score: 0.9390\n",
            "Epoch 1203/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1845 - f1_score: 0.9217 - val_loss: 0.1858 - val_f1_score: 0.9390\n",
            "Epoch 1204/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1845 - f1_score: 0.9217 - val_loss: 0.1857 - val_f1_score: 0.9390\n",
            "Epoch 1205/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1844 - f1_score: 0.9218 - val_loss: 0.1856 - val_f1_score: 0.9390\n",
            "Epoch 1206/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1844 - f1_score: 0.9218 - val_loss: 0.1856 - val_f1_score: 0.9391\n",
            "Epoch 1207/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1843 - f1_score: 0.9219 - val_loss: 0.1855 - val_f1_score: 0.9391\n",
            "Epoch 1208/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1843 - f1_score: 0.9220 - val_loss: 0.1854 - val_f1_score: 0.9391\n",
            "Epoch 1209/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1842 - f1_score: 0.9220 - val_loss: 0.1854 - val_f1_score: 0.9391\n",
            "Epoch 1210/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1842 - f1_score: 0.9221 - val_loss: 0.1853 - val_f1_score: 0.9391\n",
            "Epoch 1211/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1841 - f1_score: 0.9222 - val_loss: 0.1852 - val_f1_score: 0.9391\n",
            "Epoch 1212/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1841 - f1_score: 0.9222 - val_loss: 0.1852 - val_f1_score: 0.9391\n",
            "Epoch 1213/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1840 - f1_score: 0.9223 - val_loss: 0.1851 - val_f1_score: 0.9391\n",
            "Epoch 1214/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1840 - f1_score: 0.9224 - val_loss: 0.1850 - val_f1_score: 0.9391\n",
            "Epoch 1215/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1839 - f1_score: 0.9225 - val_loss: 0.1850 - val_f1_score: 0.9391\n",
            "Epoch 1216/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1839 - f1_score: 0.9225 - val_loss: 0.1849 - val_f1_score: 0.9391\n",
            "Epoch 1217/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1838 - f1_score: 0.9226 - val_loss: 0.1848 - val_f1_score: 0.9392\n",
            "Epoch 1218/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1838 - f1_score: 0.9227 - val_loss: 0.1848 - val_f1_score: 0.9392\n",
            "Epoch 1219/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1837 - f1_score: 0.9227 - val_loss: 0.1847 - val_f1_score: 0.9392\n",
            "Epoch 1220/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1837 - f1_score: 0.9227 - val_loss: 0.1846 - val_f1_score: 0.9392\n",
            "Epoch 1221/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1836 - f1_score: 0.9227 - val_loss: 0.1846 - val_f1_score: 0.9393\n",
            "Epoch 1222/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1836 - f1_score: 0.9228 - val_loss: 0.1845 - val_f1_score: 0.9393\n",
            "Epoch 1223/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1835 - f1_score: 0.9228 - val_loss: 0.1844 - val_f1_score: 0.9393\n",
            "Epoch 1224/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1835 - f1_score: 0.9229 - val_loss: 0.1844 - val_f1_score: 0.9393\n",
            "Epoch 1225/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1834 - f1_score: 0.9229 - val_loss: 0.1843 - val_f1_score: 0.9393\n",
            "Epoch 1226/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1834 - f1_score: 0.9230 - val_loss: 0.1842 - val_f1_score: 0.9393\n",
            "Epoch 1227/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1833 - f1_score: 0.9230 - val_loss: 0.1842 - val_f1_score: 0.9393\n",
            "Epoch 1228/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1833 - f1_score: 0.9230 - val_loss: 0.1841 - val_f1_score: 0.9393\n",
            "Epoch 1229/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1833 - f1_score: 0.9231 - val_loss: 0.1840 - val_f1_score: 0.9393\n",
            "Epoch 1230/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1832 - f1_score: 0.9231 - val_loss: 0.1840 - val_f1_score: 0.9394\n",
            "Epoch 1231/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1832 - f1_score: 0.9231 - val_loss: 0.1839 - val_f1_score: 0.9394\n",
            "Epoch 1232/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1831 - f1_score: 0.9232 - val_loss: 0.1838 - val_f1_score: 0.9394\n",
            "Epoch 1233/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1831 - f1_score: 0.9232 - val_loss: 0.1838 - val_f1_score: 0.9394\n",
            "Epoch 1234/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1830 - f1_score: 0.9233 - val_loss: 0.1837 - val_f1_score: 0.9396\n",
            "Epoch 1235/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1830 - f1_score: 0.9233 - val_loss: 0.1836 - val_f1_score: 0.9396\n",
            "Epoch 1236/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1829 - f1_score: 0.9233 - val_loss: 0.1836 - val_f1_score: 0.9396\n",
            "Epoch 1237/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1829 - f1_score: 0.9234 - val_loss: 0.1835 - val_f1_score: 0.9396\n",
            "Epoch 1238/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1828 - f1_score: 0.9235 - val_loss: 0.1834 - val_f1_score: 0.9396\n",
            "Epoch 1239/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1828 - f1_score: 0.9235 - val_loss: 0.1834 - val_f1_score: 0.9396\n",
            "Epoch 1240/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1827 - f1_score: 0.9236 - val_loss: 0.1833 - val_f1_score: 0.9396\n",
            "Epoch 1241/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1827 - f1_score: 0.9237 - val_loss: 0.1832 - val_f1_score: 0.9396\n",
            "Epoch 1242/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1826 - f1_score: 0.9238 - val_loss: 0.1832 - val_f1_score: 0.9396\n",
            "Epoch 1243/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1826 - f1_score: 0.9239 - val_loss: 0.1831 - val_f1_score: 0.9397\n",
            "Epoch 1244/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1825 - f1_score: 0.9239 - val_loss: 0.1830 - val_f1_score: 0.9397\n",
            "Epoch 1245/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1825 - f1_score: 0.9240 - val_loss: 0.1830 - val_f1_score: 0.9397\n",
            "Epoch 1246/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1824 - f1_score: 0.9241 - val_loss: 0.1829 - val_f1_score: 0.9397\n",
            "Epoch 1247/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1824 - f1_score: 0.9241 - val_loss: 0.1828 - val_f1_score: 0.9398\n",
            "Epoch 1248/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1823 - f1_score: 0.9242 - val_loss: 0.1828 - val_f1_score: 0.9398\n",
            "Epoch 1249/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1823 - f1_score: 0.9243 - val_loss: 0.1827 - val_f1_score: 0.9398\n",
            "Epoch 1250/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1823 - f1_score: 0.9243 - val_loss: 0.1826 - val_f1_score: 0.9398\n",
            "Epoch 1251/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1822 - f1_score: 0.9244 - val_loss: 0.1826 - val_f1_score: 0.9398\n",
            "Epoch 1252/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1822 - f1_score: 0.9245 - val_loss: 0.1825 - val_f1_score: 0.9399\n",
            "Epoch 1253/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1821 - f1_score: 0.9246 - val_loss: 0.1824 - val_f1_score: 0.9399\n",
            "Epoch 1254/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1821 - f1_score: 0.9246 - val_loss: 0.1824 - val_f1_score: 0.9399\n",
            "Epoch 1255/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1820 - f1_score: 0.9247 - val_loss: 0.1823 - val_f1_score: 0.9399\n",
            "Epoch 1256/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1820 - f1_score: 0.9248 - val_loss: 0.1822 - val_f1_score: 0.9400\n",
            "Epoch 1257/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1819 - f1_score: 0.9249 - val_loss: 0.1822 - val_f1_score: 0.9400\n",
            "Epoch 1258/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1819 - f1_score: 0.9249 - val_loss: 0.1821 - val_f1_score: 0.9400\n",
            "Epoch 1259/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1818 - f1_score: 0.9250 - val_loss: 0.1820 - val_f1_score: 0.9400\n",
            "Epoch 1260/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1818 - f1_score: 0.9250 - val_loss: 0.1820 - val_f1_score: 0.9400\n",
            "Epoch 1261/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1817 - f1_score: 0.9250 - val_loss: 0.1819 - val_f1_score: 0.9402\n",
            "Epoch 1262/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1817 - f1_score: 0.9251 - val_loss: 0.1819 - val_f1_score: 0.9402\n",
            "Epoch 1263/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1816 - f1_score: 0.9251 - val_loss: 0.1818 - val_f1_score: 0.9402\n",
            "Epoch 1264/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1816 - f1_score: 0.9252 - val_loss: 0.1817 - val_f1_score: 0.9402\n",
            "Epoch 1265/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1816 - f1_score: 0.9252 - val_loss: 0.1817 - val_f1_score: 0.9402\n",
            "Epoch 1266/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1815 - f1_score: 0.9253 - val_loss: 0.1816 - val_f1_score: 0.9403\n",
            "Epoch 1267/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1815 - f1_score: 0.9254 - val_loss: 0.1815 - val_f1_score: 0.9403\n",
            "Epoch 1268/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1814 - f1_score: 0.9255 - val_loss: 0.1815 - val_f1_score: 0.9404\n",
            "Epoch 1269/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1814 - f1_score: 0.9255 - val_loss: 0.1814 - val_f1_score: 0.9404\n",
            "Epoch 1270/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1813 - f1_score: 0.9256 - val_loss: 0.1813 - val_f1_score: 0.9404\n",
            "Epoch 1271/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1813 - f1_score: 0.9257 - val_loss: 0.1813 - val_f1_score: 0.9405\n",
            "Epoch 1272/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1812 - f1_score: 0.9257 - val_loss: 0.1812 - val_f1_score: 0.9405\n",
            "Epoch 1273/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1812 - f1_score: 0.9258 - val_loss: 0.1811 - val_f1_score: 0.9405\n",
            "Epoch 1274/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1811 - f1_score: 0.9259 - val_loss: 0.1811 - val_f1_score: 0.9405\n",
            "Epoch 1275/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1811 - f1_score: 0.9259 - val_loss: 0.1810 - val_f1_score: 0.9405\n",
            "Epoch 1276/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1811 - f1_score: 0.9260 - val_loss: 0.1809 - val_f1_score: 0.9406\n",
            "Epoch 1277/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1810 - f1_score: 0.9260 - val_loss: 0.1809 - val_f1_score: 0.9406\n",
            "Epoch 1278/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1810 - f1_score: 0.9261 - val_loss: 0.1808 - val_f1_score: 0.9407\n",
            "Epoch 1279/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1809 - f1_score: 0.9262 - val_loss: 0.1808 - val_f1_score: 0.9408\n",
            "Epoch 1280/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1809 - f1_score: 0.9262 - val_loss: 0.1807 - val_f1_score: 0.9408\n",
            "Epoch 1281/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1808 - f1_score: 0.9263 - val_loss: 0.1806 - val_f1_score: 0.9408\n",
            "Epoch 1282/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1808 - f1_score: 0.9263 - val_loss: 0.1806 - val_f1_score: 0.9408\n",
            "Epoch 1283/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1807 - f1_score: 0.9264 - val_loss: 0.1805 - val_f1_score: 0.9408\n",
            "Epoch 1284/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1807 - f1_score: 0.9265 - val_loss: 0.1804 - val_f1_score: 0.9408\n",
            "Epoch 1285/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1806 - f1_score: 0.9266 - val_loss: 0.1804 - val_f1_score: 0.9408\n",
            "Epoch 1286/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1806 - f1_score: 0.9267 - val_loss: 0.1803 - val_f1_score: 0.9408\n",
            "Epoch 1287/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1806 - f1_score: 0.9267 - val_loss: 0.1802 - val_f1_score: 0.9410\n",
            "Epoch 1288/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1805 - f1_score: 0.9269 - val_loss: 0.1802 - val_f1_score: 0.9410\n",
            "Epoch 1289/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1805 - f1_score: 0.9270 - val_loss: 0.1801 - val_f1_score: 0.9410\n",
            "Epoch 1290/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1804 - f1_score: 0.9270 - val_loss: 0.1801 - val_f1_score: 0.9410\n",
            "Epoch 1291/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1804 - f1_score: 0.9271 - val_loss: 0.1800 - val_f1_score: 0.9410\n",
            "Epoch 1292/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1803 - f1_score: 0.9271 - val_loss: 0.1799 - val_f1_score: 0.9410\n",
            "Epoch 1293/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1803 - f1_score: 0.9273 - val_loss: 0.1799 - val_f1_score: 0.9410\n",
            "Epoch 1294/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1802 - f1_score: 0.9274 - val_loss: 0.1798 - val_f1_score: 0.9411\n",
            "Epoch 1295/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1802 - f1_score: 0.9274 - val_loss: 0.1797 - val_f1_score: 0.9411\n",
            "Epoch 1296/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1802 - f1_score: 0.9275 - val_loss: 0.1797 - val_f1_score: 0.9411\n",
            "Epoch 1297/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1801 - f1_score: 0.9276 - val_loss: 0.1796 - val_f1_score: 0.9411\n",
            "Epoch 1298/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1801 - f1_score: 0.9277 - val_loss: 0.1795 - val_f1_score: 0.9411\n",
            "Epoch 1299/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1800 - f1_score: 0.9278 - val_loss: 0.1795 - val_f1_score: 0.9411\n",
            "Epoch 1300/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1800 - f1_score: 0.9279 - val_loss: 0.1794 - val_f1_score: 0.9411\n",
            "Epoch 1301/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1799 - f1_score: 0.9279 - val_loss: 0.1794 - val_f1_score: 0.9412\n",
            "Epoch 1302/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1799 - f1_score: 0.9280 - val_loss: 0.1793 - val_f1_score: 0.9412\n",
            "Epoch 1303/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1798 - f1_score: 0.9280 - val_loss: 0.1792 - val_f1_score: 0.9412\n",
            "Epoch 1304/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1798 - f1_score: 0.9281 - val_loss: 0.1792 - val_f1_score: 0.9412\n",
            "Epoch 1305/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1798 - f1_score: 0.9282 - val_loss: 0.1791 - val_f1_score: 0.9411\n",
            "Epoch 1306/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1797 - f1_score: 0.9282 - val_loss: 0.1790 - val_f1_score: 0.9411\n",
            "Epoch 1307/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1797 - f1_score: 0.9283 - val_loss: 0.1790 - val_f1_score: 0.9411\n",
            "Epoch 1308/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1796 - f1_score: 0.9284 - val_loss: 0.1789 - val_f1_score: 0.9411\n",
            "Epoch 1309/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1796 - f1_score: 0.9285 - val_loss: 0.1789 - val_f1_score: 0.9411\n",
            "Epoch 1310/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1795 - f1_score: 0.9286 - val_loss: 0.1788 - val_f1_score: 0.9411\n",
            "Epoch 1311/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1795 - f1_score: 0.9287 - val_loss: 0.1787 - val_f1_score: 0.9412\n",
            "Epoch 1312/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1795 - f1_score: 0.9287 - val_loss: 0.1787 - val_f1_score: 0.9412\n",
            "Epoch 1313/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1794 - f1_score: 0.9287 - val_loss: 0.1786 - val_f1_score: 0.9412\n",
            "Epoch 1314/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1794 - f1_score: 0.9287 - val_loss: 0.1785 - val_f1_score: 0.9413\n",
            "Epoch 1315/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1793 - f1_score: 0.9288 - val_loss: 0.1785 - val_f1_score: 0.9413\n",
            "Epoch 1316/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1793 - f1_score: 0.9288 - val_loss: 0.1784 - val_f1_score: 0.9413\n",
            "Epoch 1317/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1792 - f1_score: 0.9289 - val_loss: 0.1784 - val_f1_score: 0.9413\n",
            "Epoch 1318/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1792 - f1_score: 0.9289 - val_loss: 0.1783 - val_f1_score: 0.9413\n",
            "Epoch 1319/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1792 - f1_score: 0.9290 - val_loss: 0.1782 - val_f1_score: 0.9413\n",
            "Epoch 1320/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1791 - f1_score: 0.9291 - val_loss: 0.1782 - val_f1_score: 0.9413\n",
            "Epoch 1321/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1791 - f1_score: 0.9293 - val_loss: 0.1781 - val_f1_score: 0.9413\n",
            "Epoch 1322/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1790 - f1_score: 0.9293 - val_loss: 0.1781 - val_f1_score: 0.9414\n",
            "Epoch 1323/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1790 - f1_score: 0.9294 - val_loss: 0.1780 - val_f1_score: 0.9414\n",
            "Epoch 1324/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1789 - f1_score: 0.9294 - val_loss: 0.1779 - val_f1_score: 0.9414\n",
            "Epoch 1325/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1789 - f1_score: 0.9295 - val_loss: 0.1779 - val_f1_score: 0.9414\n",
            "Epoch 1326/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1789 - f1_score: 0.9296 - val_loss: 0.1778 - val_f1_score: 0.9414\n",
            "Epoch 1327/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1788 - f1_score: 0.9296 - val_loss: 0.1777 - val_f1_score: 0.9414\n",
            "Epoch 1328/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1788 - f1_score: 0.9297 - val_loss: 0.1777 - val_f1_score: 0.9416\n",
            "Epoch 1329/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1787 - f1_score: 0.9298 - val_loss: 0.1776 - val_f1_score: 0.9416\n",
            "Epoch 1330/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1787 - f1_score: 0.9299 - val_loss: 0.1776 - val_f1_score: 0.9416\n",
            "Epoch 1331/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1786 - f1_score: 0.9299 - val_loss: 0.1775 - val_f1_score: 0.9416\n",
            "Epoch 1332/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1786 - f1_score: 0.9299 - val_loss: 0.1774 - val_f1_score: 0.9416\n",
            "Epoch 1333/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1786 - f1_score: 0.9300 - val_loss: 0.1774 - val_f1_score: 0.9416\n",
            "Epoch 1334/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1785 - f1_score: 0.9302 - val_loss: 0.1773 - val_f1_score: 0.9416\n",
            "Epoch 1335/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1785 - f1_score: 0.9302 - val_loss: 0.1773 - val_f1_score: 0.9416\n",
            "Epoch 1336/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1784 - f1_score: 0.9302 - val_loss: 0.1772 - val_f1_score: 0.9416\n",
            "Epoch 1337/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1784 - f1_score: 0.9302 - val_loss: 0.1771 - val_f1_score: 0.9417\n",
            "Epoch 1338/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1783 - f1_score: 0.9303 - val_loss: 0.1771 - val_f1_score: 0.9417\n",
            "Epoch 1339/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1783 - f1_score: 0.9304 - val_loss: 0.1770 - val_f1_score: 0.9417\n",
            "Epoch 1340/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1783 - f1_score: 0.9305 - val_loss: 0.1770 - val_f1_score: 0.9417\n",
            "Epoch 1341/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1782 - f1_score: 0.9305 - val_loss: 0.1769 - val_f1_score: 0.9417\n",
            "Epoch 1342/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1782 - f1_score: 0.9307 - val_loss: 0.1768 - val_f1_score: 0.9417\n",
            "Epoch 1343/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1781 - f1_score: 0.9308 - val_loss: 0.1768 - val_f1_score: 0.9417\n",
            "Epoch 1344/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1781 - f1_score: 0.9309 - val_loss: 0.1767 - val_f1_score: 0.9417\n",
            "Epoch 1345/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1781 - f1_score: 0.9309 - val_loss: 0.1767 - val_f1_score: 0.9417\n",
            "Epoch 1346/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1780 - f1_score: 0.9310 - val_loss: 0.1766 - val_f1_score: 0.9417\n",
            "Epoch 1347/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1780 - f1_score: 0.9311 - val_loss: 0.1765 - val_f1_score: 0.9417\n",
            "Epoch 1348/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1779 - f1_score: 0.9311 - val_loss: 0.1765 - val_f1_score: 0.9417\n",
            "Epoch 1349/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1779 - f1_score: 0.9312 - val_loss: 0.1764 - val_f1_score: 0.9417\n",
            "Epoch 1350/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1779 - f1_score: 0.9313 - val_loss: 0.1764 - val_f1_score: 0.9417\n",
            "Epoch 1351/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1778 - f1_score: 0.9313 - val_loss: 0.1763 - val_f1_score: 0.9418\n",
            "Epoch 1352/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1778 - f1_score: 0.9314 - val_loss: 0.1762 - val_f1_score: 0.9418\n",
            "Epoch 1353/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1777 - f1_score: 0.9315 - val_loss: 0.1762 - val_f1_score: 0.9420\n",
            "Epoch 1354/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1777 - f1_score: 0.9316 - val_loss: 0.1761 - val_f1_score: 0.9420\n",
            "Epoch 1355/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1776 - f1_score: 0.9317 - val_loss: 0.1761 - val_f1_score: 0.9420\n",
            "Epoch 1356/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1776 - f1_score: 0.9317 - val_loss: 0.1760 - val_f1_score: 0.9420\n",
            "Epoch 1357/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1776 - f1_score: 0.9318 - val_loss: 0.1759 - val_f1_score: 0.9420\n",
            "Epoch 1358/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1775 - f1_score: 0.9319 - val_loss: 0.1759 - val_f1_score: 0.9420\n",
            "Epoch 1359/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1775 - f1_score: 0.9320 - val_loss: 0.1758 - val_f1_score: 0.9420\n",
            "Epoch 1360/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1774 - f1_score: 0.9321 - val_loss: 0.1758 - val_f1_score: 0.9420\n",
            "Epoch 1361/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1774 - f1_score: 0.9322 - val_loss: 0.1757 - val_f1_score: 0.9420\n",
            "Epoch 1362/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1774 - f1_score: 0.9323 - val_loss: 0.1756 - val_f1_score: 0.9421\n",
            "Epoch 1363/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1773 - f1_score: 0.9324 - val_loss: 0.1756 - val_f1_score: 0.9422\n",
            "Epoch 1364/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1773 - f1_score: 0.9324 - val_loss: 0.1755 - val_f1_score: 0.9422\n",
            "Epoch 1365/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1772 - f1_score: 0.9325 - val_loss: 0.1755 - val_f1_score: 0.9422\n",
            "Epoch 1366/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1772 - f1_score: 0.9326 - val_loss: 0.1754 - val_f1_score: 0.9422\n",
            "Epoch 1367/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1772 - f1_score: 0.9327 - val_loss: 0.1754 - val_f1_score: 0.9422\n",
            "Epoch 1368/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1771 - f1_score: 0.9328 - val_loss: 0.1753 - val_f1_score: 0.9422\n",
            "Epoch 1369/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1771 - f1_score: 0.9329 - val_loss: 0.1752 - val_f1_score: 0.9421\n",
            "Epoch 1370/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1770 - f1_score: 0.9330 - val_loss: 0.1752 - val_f1_score: 0.9421\n",
            "Epoch 1371/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1770 - f1_score: 0.9331 - val_loss: 0.1751 - val_f1_score: 0.9421\n",
            "Epoch 1372/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1770 - f1_score: 0.9331 - val_loss: 0.1751 - val_f1_score: 0.9422\n",
            "Epoch 1373/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1769 - f1_score: 0.9332 - val_loss: 0.1750 - val_f1_score: 0.9422\n",
            "Epoch 1374/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1769 - f1_score: 0.9333 - val_loss: 0.1749 - val_f1_score: 0.9422\n",
            "Epoch 1375/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1768 - f1_score: 0.9334 - val_loss: 0.1749 - val_f1_score: 0.9422\n",
            "Epoch 1376/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1768 - f1_score: 0.9334 - val_loss: 0.1748 - val_f1_score: 0.9422\n",
            "Epoch 1377/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1768 - f1_score: 0.9335 - val_loss: 0.1748 - val_f1_score: 0.9422\n",
            "Epoch 1378/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1767 - f1_score: 0.9336 - val_loss: 0.1747 - val_f1_score: 0.9424\n",
            "Epoch 1379/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1767 - f1_score: 0.9338 - val_loss: 0.1747 - val_f1_score: 0.9424\n",
            "Epoch 1380/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1766 - f1_score: 0.9338 - val_loss: 0.1746 - val_f1_score: 0.9424\n",
            "Epoch 1381/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1766 - f1_score: 0.9339 - val_loss: 0.1745 - val_f1_score: 0.9424\n",
            "Epoch 1382/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1766 - f1_score: 0.9340 - val_loss: 0.1745 - val_f1_score: 0.9424\n",
            "Epoch 1383/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1765 - f1_score: 0.9341 - val_loss: 0.1744 - val_f1_score: 0.9424\n",
            "Epoch 1384/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1765 - f1_score: 0.9341 - val_loss: 0.1744 - val_f1_score: 0.9424\n",
            "Epoch 1385/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1764 - f1_score: 0.9341 - val_loss: 0.1743 - val_f1_score: 0.9424\n",
            "Epoch 1386/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1764 - f1_score: 0.9342 - val_loss: 0.1743 - val_f1_score: 0.9425\n",
            "Epoch 1387/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1764 - f1_score: 0.9343 - val_loss: 0.1742 - val_f1_score: 0.9424\n",
            "Epoch 1388/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1763 - f1_score: 0.9344 - val_loss: 0.1741 - val_f1_score: 0.9424\n",
            "Epoch 1389/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1763 - f1_score: 0.9345 - val_loss: 0.1741 - val_f1_score: 0.9424\n",
            "Epoch 1390/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1762 - f1_score: 0.9346 - val_loss: 0.1740 - val_f1_score: 0.9424\n",
            "Epoch 1391/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1762 - f1_score: 0.9347 - val_loss: 0.1740 - val_f1_score: 0.9424\n",
            "Epoch 1392/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1762 - f1_score: 0.9348 - val_loss: 0.1739 - val_f1_score: 0.9424\n",
            "Epoch 1393/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1761 - f1_score: 0.9348 - val_loss: 0.1739 - val_f1_score: 0.9424\n",
            "Epoch 1394/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1761 - f1_score: 0.9350 - val_loss: 0.1738 - val_f1_score: 0.9425\n",
            "Epoch 1395/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1761 - f1_score: 0.9350 - val_loss: 0.1738 - val_f1_score: 0.9425\n",
            "Epoch 1396/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1760 - f1_score: 0.9351 - val_loss: 0.1737 - val_f1_score: 0.9425\n",
            "Epoch 1397/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1760 - f1_score: 0.9351 - val_loss: 0.1736 - val_f1_score: 0.9425\n",
            "Epoch 1398/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1759 - f1_score: 0.9352 - val_loss: 0.1736 - val_f1_score: 0.9425\n",
            "Epoch 1399/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1759 - f1_score: 0.9352 - val_loss: 0.1735 - val_f1_score: 0.9426\n",
            "Epoch 1400/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1759 - f1_score: 0.9353 - val_loss: 0.1735 - val_f1_score: 0.9426\n",
            "Epoch 1401/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1758 - f1_score: 0.9353 - val_loss: 0.1734 - val_f1_score: 0.9426\n",
            "Epoch 1402/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1758 - f1_score: 0.9354 - val_loss: 0.1734 - val_f1_score: 0.9426\n",
            "Epoch 1403/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1757 - f1_score: 0.9355 - val_loss: 0.1733 - val_f1_score: 0.9426\n",
            "Epoch 1404/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1757 - f1_score: 0.9356 - val_loss: 0.1732 - val_f1_score: 0.9426\n",
            "Epoch 1405/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1757 - f1_score: 0.9356 - val_loss: 0.1732 - val_f1_score: 0.9426\n",
            "Epoch 1406/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1756 - f1_score: 0.9357 - val_loss: 0.1731 - val_f1_score: 0.9426\n",
            "Epoch 1407/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1756 - f1_score: 0.9358 - val_loss: 0.1731 - val_f1_score: 0.9426\n",
            "Epoch 1408/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1756 - f1_score: 0.9358 - val_loss: 0.1730 - val_f1_score: 0.9426\n",
            "Epoch 1409/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1755 - f1_score: 0.9359 - val_loss: 0.1730 - val_f1_score: 0.9426\n",
            "Epoch 1410/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1755 - f1_score: 0.9360 - val_loss: 0.1729 - val_f1_score: 0.9426\n",
            "Epoch 1411/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1754 - f1_score: 0.9360 - val_loss: 0.1729 - val_f1_score: 0.9426\n",
            "Epoch 1412/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1754 - f1_score: 0.9361 - val_loss: 0.1728 - val_f1_score: 0.9426\n",
            "Epoch 1413/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1754 - f1_score: 0.9362 - val_loss: 0.1728 - val_f1_score: 0.9426\n",
            "Epoch 1414/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1753 - f1_score: 0.9362 - val_loss: 0.1727 - val_f1_score: 0.9426\n",
            "Epoch 1415/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1753 - f1_score: 0.9363 - val_loss: 0.1726 - val_f1_score: 0.9426\n",
            "Epoch 1416/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1753 - f1_score: 0.9363 - val_loss: 0.1726 - val_f1_score: 0.9426\n",
            "Epoch 1417/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1752 - f1_score: 0.9363 - val_loss: 0.1725 - val_f1_score: 0.9427\n",
            "Epoch 1418/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1752 - f1_score: 0.9364 - val_loss: 0.1725 - val_f1_score: 0.9427\n",
            "Epoch 1419/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1751 - f1_score: 0.9364 - val_loss: 0.1724 - val_f1_score: 0.9427\n",
            "Epoch 1420/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1751 - f1_score: 0.9365 - val_loss: 0.1724 - val_f1_score: 0.9427\n",
            "Epoch 1421/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1751 - f1_score: 0.9364 - val_loss: 0.1723 - val_f1_score: 0.9427\n",
            "Epoch 1422/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1750 - f1_score: 0.9365 - val_loss: 0.1723 - val_f1_score: 0.9427\n",
            "Epoch 1423/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1750 - f1_score: 0.9365 - val_loss: 0.1722 - val_f1_score: 0.9427\n",
            "Epoch 1424/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1750 - f1_score: 0.9366 - val_loss: 0.1722 - val_f1_score: 0.9427\n",
            "Epoch 1425/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1749 - f1_score: 0.9367 - val_loss: 0.1721 - val_f1_score: 0.9427\n",
            "Epoch 1426/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1749 - f1_score: 0.9367 - val_loss: 0.1720 - val_f1_score: 0.9427\n",
            "Epoch 1427/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1749 - f1_score: 0.9368 - val_loss: 0.1720 - val_f1_score: 0.9427\n",
            "Epoch 1428/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1748 - f1_score: 0.9368 - val_loss: 0.1719 - val_f1_score: 0.9427\n",
            "Epoch 1429/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1748 - f1_score: 0.9368 - val_loss: 0.1719 - val_f1_score: 0.9427\n",
            "Epoch 1430/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1747 - f1_score: 0.9368 - val_loss: 0.1718 - val_f1_score: 0.9427\n",
            "Epoch 1431/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1747 - f1_score: 0.9368 - val_loss: 0.1718 - val_f1_score: 0.9427\n",
            "Epoch 1432/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1747 - f1_score: 0.9369 - val_loss: 0.1717 - val_f1_score: 0.9427\n",
            "Epoch 1433/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1746 - f1_score: 0.9369 - val_loss: 0.1717 - val_f1_score: 0.9427\n",
            "Epoch 1434/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1746 - f1_score: 0.9370 - val_loss: 0.1716 - val_f1_score: 0.9427\n",
            "Epoch 1435/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1746 - f1_score: 0.9370 - val_loss: 0.1716 - val_f1_score: 0.9427\n",
            "Epoch 1436/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1745 - f1_score: 0.9370 - val_loss: 0.1715 - val_f1_score: 0.9427\n",
            "Epoch 1437/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1745 - f1_score: 0.9370 - val_loss: 0.1715 - val_f1_score: 0.9427\n",
            "Epoch 1438/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1744 - f1_score: 0.9371 - val_loss: 0.1714 - val_f1_score: 0.9427\n",
            "Epoch 1439/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1744 - f1_score: 0.9371 - val_loss: 0.1714 - val_f1_score: 0.9427\n",
            "Epoch 1440/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1744 - f1_score: 0.9371 - val_loss: 0.1713 - val_f1_score: 0.9427\n",
            "Epoch 1441/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1743 - f1_score: 0.9372 - val_loss: 0.1712 - val_f1_score: 0.9427\n",
            "Epoch 1442/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1743 - f1_score: 0.9372 - val_loss: 0.1712 - val_f1_score: 0.9427\n",
            "Epoch 1443/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1743 - f1_score: 0.9373 - val_loss: 0.1711 - val_f1_score: 0.9427\n",
            "Epoch 1444/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1742 - f1_score: 0.9373 - val_loss: 0.1711 - val_f1_score: 0.9427\n",
            "Epoch 1445/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1742 - f1_score: 0.9373 - val_loss: 0.1710 - val_f1_score: 0.9427\n",
            "Epoch 1446/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1742 - f1_score: 0.9374 - val_loss: 0.1710 - val_f1_score: 0.9427\n",
            "Epoch 1447/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1741 - f1_score: 0.9374 - val_loss: 0.1709 - val_f1_score: 0.9427\n",
            "Epoch 1448/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1741 - f1_score: 0.9374 - val_loss: 0.1709 - val_f1_score: 0.9427\n",
            "Epoch 1449/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1741 - f1_score: 0.9374 - val_loss: 0.1708 - val_f1_score: 0.9427\n",
            "Epoch 1450/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1740 - f1_score: 0.9374 - val_loss: 0.1708 - val_f1_score: 0.9427\n",
            "Epoch 1451/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1740 - f1_score: 0.9374 - val_loss: 0.1707 - val_f1_score: 0.9427\n",
            "Epoch 1452/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1739 - f1_score: 0.9374 - val_loss: 0.1707 - val_f1_score: 0.9427\n",
            "Epoch 1453/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1739 - f1_score: 0.9374 - val_loss: 0.1706 - val_f1_score: 0.9427\n",
            "Epoch 1454/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1739 - f1_score: 0.9374 - val_loss: 0.1706 - val_f1_score: 0.9427\n",
            "Epoch 1455/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1738 - f1_score: 0.9375 - val_loss: 0.1705 - val_f1_score: 0.9427\n",
            "Epoch 1456/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1738 - f1_score: 0.9375 - val_loss: 0.1705 - val_f1_score: 0.9427\n",
            "Epoch 1457/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1738 - f1_score: 0.9376 - val_loss: 0.1704 - val_f1_score: 0.9427\n",
            "Epoch 1458/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1737 - f1_score: 0.9376 - val_loss: 0.1704 - val_f1_score: 0.9427\n",
            "Epoch 1459/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1737 - f1_score: 0.9376 - val_loss: 0.1703 - val_f1_score: 0.9427\n",
            "Epoch 1460/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1737 - f1_score: 0.9376 - val_loss: 0.1703 - val_f1_score: 0.9427\n",
            "Epoch 1461/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1736 - f1_score: 0.9376 - val_loss: 0.1702 - val_f1_score: 0.9427\n",
            "Epoch 1462/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1736 - f1_score: 0.9377 - val_loss: 0.1702 - val_f1_score: 0.9427\n",
            "Epoch 1463/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1736 - f1_score: 0.9377 - val_loss: 0.1701 - val_f1_score: 0.9427\n",
            "Epoch 1464/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1735 - f1_score: 0.9377 - val_loss: 0.1701 - val_f1_score: 0.9427\n",
            "Epoch 1465/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1735 - f1_score: 0.9378 - val_loss: 0.1700 - val_f1_score: 0.9427\n",
            "Epoch 1466/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1735 - f1_score: 0.9378 - val_loss: 0.1700 - val_f1_score: 0.9427\n",
            "Epoch 1467/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1734 - f1_score: 0.9378 - val_loss: 0.1699 - val_f1_score: 0.9427\n",
            "Epoch 1468/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1734 - f1_score: 0.9378 - val_loss: 0.1699 - val_f1_score: 0.9427\n",
            "Epoch 1469/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1734 - f1_score: 0.9379 - val_loss: 0.1698 - val_f1_score: 0.9427\n",
            "Epoch 1470/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1733 - f1_score: 0.9379 - val_loss: 0.1698 - val_f1_score: 0.9427\n",
            "Epoch 1471/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1733 - f1_score: 0.9379 - val_loss: 0.1697 - val_f1_score: 0.9427\n",
            "Epoch 1472/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1733 - f1_score: 0.9379 - val_loss: 0.1697 - val_f1_score: 0.9427\n",
            "Epoch 1473/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1732 - f1_score: 0.9379 - val_loss: 0.1696 - val_f1_score: 0.9427\n",
            "Epoch 1474/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1732 - f1_score: 0.9380 - val_loss: 0.1696 - val_f1_score: 0.9427\n",
            "Epoch 1475/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1731 - f1_score: 0.9380 - val_loss: 0.1695 - val_f1_score: 0.9427\n",
            "Epoch 1476/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1731 - f1_score: 0.9381 - val_loss: 0.1695 - val_f1_score: 0.9427\n",
            "Epoch 1477/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1731 - f1_score: 0.9381 - val_loss: 0.1694 - val_f1_score: 0.9428\n",
            "Epoch 1478/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1730 - f1_score: 0.9381 - val_loss: 0.1694 - val_f1_score: 0.9428\n",
            "Epoch 1479/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1730 - f1_score: 0.9381 - val_loss: 0.1693 - val_f1_score: 0.9428\n",
            "Epoch 1480/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1730 - f1_score: 0.9381 - val_loss: 0.1693 - val_f1_score: 0.9428\n",
            "Epoch 1481/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1729 - f1_score: 0.9381 - val_loss: 0.1692 - val_f1_score: 0.9428\n",
            "Epoch 1482/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1729 - f1_score: 0.9381 - val_loss: 0.1692 - val_f1_score: 0.9428\n",
            "Epoch 1483/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1729 - f1_score: 0.9382 - val_loss: 0.1691 - val_f1_score: 0.9428\n",
            "Epoch 1484/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1728 - f1_score: 0.9382 - val_loss: 0.1691 - val_f1_score: 0.9428\n",
            "Epoch 1485/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1728 - f1_score: 0.9382 - val_loss: 0.1690 - val_f1_score: 0.9428\n",
            "Epoch 1486/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1728 - f1_score: 0.9382 - val_loss: 0.1690 - val_f1_score: 0.9428\n",
            "Epoch 1487/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1727 - f1_score: 0.9382 - val_loss: 0.1689 - val_f1_score: 0.9428\n",
            "Epoch 1488/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1727 - f1_score: 0.9382 - val_loss: 0.1689 - val_f1_score: 0.9428\n",
            "Epoch 1489/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1727 - f1_score: 0.9382 - val_loss: 0.1688 - val_f1_score: 0.9428\n",
            "Epoch 1490/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1726 - f1_score: 0.9382 - val_loss: 0.1688 - val_f1_score: 0.9429\n",
            "Epoch 1491/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1726 - f1_score: 0.9382 - val_loss: 0.1687 - val_f1_score: 0.9429\n",
            "Epoch 1492/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1726 - f1_score: 0.9382 - val_loss: 0.1687 - val_f1_score: 0.9429\n",
            "Epoch 1493/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1725 - f1_score: 0.9383 - val_loss: 0.1686 - val_f1_score: 0.9429\n",
            "Epoch 1494/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1725 - f1_score: 0.9383 - val_loss: 0.1686 - val_f1_score: 0.9429\n",
            "Epoch 1495/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1725 - f1_score: 0.9383 - val_loss: 0.1685 - val_f1_score: 0.9429\n",
            "Epoch 1496/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1724 - f1_score: 0.9383 - val_loss: 0.1685 - val_f1_score: 0.9429\n",
            "Epoch 1497/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1724 - f1_score: 0.9383 - val_loss: 0.1684 - val_f1_score: 0.9429\n",
            "Epoch 1498/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1724 - f1_score: 0.9383 - val_loss: 0.1684 - val_f1_score: 0.9429\n",
            "Epoch 1499/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1723 - f1_score: 0.9383 - val_loss: 0.1683 - val_f1_score: 0.9429\n",
            "Epoch 1500/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1723 - f1_score: 0.9383 - val_loss: 0.1683 - val_f1_score: 0.9429\n",
            "Epoch 1501/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1723 - f1_score: 0.9384 - val_loss: 0.1682 - val_f1_score: 0.9429\n",
            "Epoch 1502/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1722 - f1_score: 0.9384 - val_loss: 0.1682 - val_f1_score: 0.9429\n",
            "Epoch 1503/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1722 - f1_score: 0.9384 - val_loss: 0.1681 - val_f1_score: 0.9429\n",
            "Epoch 1504/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1722 - f1_score: 0.9384 - val_loss: 0.1681 - val_f1_score: 0.9429\n",
            "Epoch 1505/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1721 - f1_score: 0.9384 - val_loss: 0.1680 - val_f1_score: 0.9429\n",
            "Epoch 1506/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1721 - f1_score: 0.9384 - val_loss: 0.1680 - val_f1_score: 0.9429\n",
            "Epoch 1507/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1721 - f1_score: 0.9384 - val_loss: 0.1679 - val_f1_score: 0.9430\n",
            "Epoch 1508/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1721 - f1_score: 0.9384 - val_loss: 0.1679 - val_f1_score: 0.9430\n",
            "Epoch 1509/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1720 - f1_score: 0.9384 - val_loss: 0.1679 - val_f1_score: 0.9430\n",
            "Epoch 1510/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1720 - f1_score: 0.9384 - val_loss: 0.1678 - val_f1_score: 0.9430\n",
            "Epoch 1511/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1720 - f1_score: 0.9384 - val_loss: 0.1678 - val_f1_score: 0.9430\n",
            "Epoch 1512/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1719 - f1_score: 0.9384 - val_loss: 0.1677 - val_f1_score: 0.9430\n",
            "Epoch 1513/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1719 - f1_score: 0.9384 - val_loss: 0.1677 - val_f1_score: 0.9430\n",
            "Epoch 1514/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1719 - f1_score: 0.9384 - val_loss: 0.1676 - val_f1_score: 0.9430\n",
            "Epoch 1515/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1718 - f1_score: 0.9384 - val_loss: 0.1676 - val_f1_score: 0.9430\n",
            "Epoch 1516/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1718 - f1_score: 0.9385 - val_loss: 0.1675 - val_f1_score: 0.9431\n",
            "Epoch 1517/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1718 - f1_score: 0.9385 - val_loss: 0.1675 - val_f1_score: 0.9431\n",
            "Epoch 1518/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1717 - f1_score: 0.9385 - val_loss: 0.1674 - val_f1_score: 0.9431\n",
            "Epoch 1519/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1717 - f1_score: 0.9385 - val_loss: 0.1674 - val_f1_score: 0.9431\n",
            "Epoch 1520/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1717 - f1_score: 0.9385 - val_loss: 0.1673 - val_f1_score: 0.9431\n",
            "Epoch 1521/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1716 - f1_score: 0.9385 - val_loss: 0.1673 - val_f1_score: 0.9431\n",
            "Epoch 1522/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1716 - f1_score: 0.9385 - val_loss: 0.1672 - val_f1_score: 0.9431\n",
            "Epoch 1523/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1716 - f1_score: 0.9385 - val_loss: 0.1672 - val_f1_score: 0.9431\n",
            "Epoch 1524/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1715 - f1_score: 0.9385 - val_loss: 0.1672 - val_f1_score: 0.9431\n",
            "Epoch 1525/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1715 - f1_score: 0.9386 - val_loss: 0.1671 - val_f1_score: 0.9431\n",
            "Epoch 1526/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1715 - f1_score: 0.9386 - val_loss: 0.1671 - val_f1_score: 0.9431\n",
            "Epoch 1527/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1714 - f1_score: 0.9386 - val_loss: 0.1670 - val_f1_score: 0.9431\n",
            "Epoch 1528/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1714 - f1_score: 0.9386 - val_loss: 0.1670 - val_f1_score: 0.9431\n",
            "Epoch 1529/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1714 - f1_score: 0.9386 - val_loss: 0.1669 - val_f1_score: 0.9431\n",
            "Epoch 1530/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1714 - f1_score: 0.9386 - val_loss: 0.1669 - val_f1_score: 0.9431\n",
            "Epoch 1531/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1713 - f1_score: 0.9386 - val_loss: 0.1668 - val_f1_score: 0.9431\n",
            "Epoch 1532/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1713 - f1_score: 0.9386 - val_loss: 0.1668 - val_f1_score: 0.9431\n",
            "Epoch 1533/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1713 - f1_score: 0.9386 - val_loss: 0.1667 - val_f1_score: 0.9432\n",
            "Epoch 1534/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1712 - f1_score: 0.9386 - val_loss: 0.1667 - val_f1_score: 0.9432\n",
            "Epoch 1535/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1712 - f1_score: 0.9387 - val_loss: 0.1667 - val_f1_score: 0.9432\n",
            "Epoch 1536/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1712 - f1_score: 0.9387 - val_loss: 0.1666 - val_f1_score: 0.9432\n",
            "Epoch 1537/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1711 - f1_score: 0.9387 - val_loss: 0.1666 - val_f1_score: 0.9432\n",
            "Epoch 1538/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1711 - f1_score: 0.9387 - val_loss: 0.1665 - val_f1_score: 0.9432\n",
            "Epoch 1539/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1711 - f1_score: 0.9387 - val_loss: 0.1665 - val_f1_score: 0.9432\n",
            "Epoch 1540/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1710 - f1_score: 0.9387 - val_loss: 0.1664 - val_f1_score: 0.9432\n",
            "Epoch 1541/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1710 - f1_score: 0.9388 - val_loss: 0.1664 - val_f1_score: 0.9432\n",
            "Epoch 1542/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1710 - f1_score: 0.9388 - val_loss: 0.1663 - val_f1_score: 0.9432\n",
            "Epoch 1543/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1710 - f1_score: 0.9388 - val_loss: 0.1663 - val_f1_score: 0.9432\n",
            "Epoch 1544/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1709 - f1_score: 0.9388 - val_loss: 0.1663 - val_f1_score: 0.9432\n",
            "Epoch 1545/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1709 - f1_score: 0.9388 - val_loss: 0.1662 - val_f1_score: 0.9432\n",
            "Epoch 1546/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1709 - f1_score: 0.9388 - val_loss: 0.1662 - val_f1_score: 0.9432\n",
            "Epoch 1547/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1708 - f1_score: 0.9388 - val_loss: 0.1661 - val_f1_score: 0.9432\n",
            "Epoch 1548/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1708 - f1_score: 0.9388 - val_loss: 0.1661 - val_f1_score: 0.9432\n",
            "Epoch 1549/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1708 - f1_score: 0.9388 - val_loss: 0.1660 - val_f1_score: 0.9432\n",
            "Epoch 1550/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1707 - f1_score: 0.9388 - val_loss: 0.1660 - val_f1_score: 0.9432\n",
            "Epoch 1551/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1707 - f1_score: 0.9388 - val_loss: 0.1659 - val_f1_score: 0.9432\n",
            "Epoch 1552/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1707 - f1_score: 0.9388 - val_loss: 0.1659 - val_f1_score: 0.9432\n",
            "Epoch 1553/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1706 - f1_score: 0.9388 - val_loss: 0.1659 - val_f1_score: 0.9432\n",
            "Epoch 1554/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1706 - f1_score: 0.9388 - val_loss: 0.1658 - val_f1_score: 0.9432\n",
            "Epoch 1555/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1706 - f1_score: 0.9388 - val_loss: 0.1658 - val_f1_score: 0.9432\n",
            "Epoch 1556/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1706 - f1_score: 0.9389 - val_loss: 0.1657 - val_f1_score: 0.9432\n",
            "Epoch 1557/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1705 - f1_score: 0.9389 - val_loss: 0.1657 - val_f1_score: 0.9432\n",
            "Epoch 1558/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1705 - f1_score: 0.9389 - val_loss: 0.1656 - val_f1_score: 0.9432\n",
            "Epoch 1559/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1705 - f1_score: 0.9389 - val_loss: 0.1656 - val_f1_score: 0.9432\n",
            "Epoch 1560/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1704 - f1_score: 0.9389 - val_loss: 0.1656 - val_f1_score: 0.9432\n",
            "Epoch 1561/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1704 - f1_score: 0.9389 - val_loss: 0.1655 - val_f1_score: 0.9432\n",
            "Epoch 1562/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1704 - f1_score: 0.9389 - val_loss: 0.1655 - val_f1_score: 0.9432\n",
            "Epoch 1563/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1703 - f1_score: 0.9390 - val_loss: 0.1654 - val_f1_score: 0.9432\n",
            "Epoch 1564/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1703 - f1_score: 0.9390 - val_loss: 0.1654 - val_f1_score: 0.9432\n",
            "Epoch 1565/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1703 - f1_score: 0.9390 - val_loss: 0.1653 - val_f1_score: 0.9432\n",
            "Epoch 1566/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1703 - f1_score: 0.9390 - val_loss: 0.1653 - val_f1_score: 0.9432\n",
            "Epoch 1567/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1702 - f1_score: 0.9390 - val_loss: 0.1653 - val_f1_score: 0.9432\n",
            "Epoch 1568/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1702 - f1_score: 0.9390 - val_loss: 0.1652 - val_f1_score: 0.9432\n",
            "Epoch 1569/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1702 - f1_score: 0.9390 - val_loss: 0.1652 - val_f1_score: 0.9432\n",
            "Epoch 1570/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1701 - f1_score: 0.9390 - val_loss: 0.1651 - val_f1_score: 0.9432\n",
            "Epoch 1571/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1701 - f1_score: 0.9390 - val_loss: 0.1651 - val_f1_score: 0.9432\n",
            "Epoch 1572/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1701 - f1_score: 0.9391 - val_loss: 0.1650 - val_f1_score: 0.9432\n",
            "Epoch 1573/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1701 - f1_score: 0.9391 - val_loss: 0.1650 - val_f1_score: 0.9432\n",
            "Epoch 1574/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1700 - f1_score: 0.9391 - val_loss: 0.1650 - val_f1_score: 0.9432\n",
            "Epoch 1575/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1700 - f1_score: 0.9392 - val_loss: 0.1649 - val_f1_score: 0.9432\n",
            "Epoch 1576/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1700 - f1_score: 0.9392 - val_loss: 0.1649 - val_f1_score: 0.9432\n",
            "Epoch 1577/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1699 - f1_score: 0.9392 - val_loss: 0.1648 - val_f1_score: 0.9432\n",
            "Epoch 1578/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1699 - f1_score: 0.9392 - val_loss: 0.1648 - val_f1_score: 0.9432\n",
            "Epoch 1579/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1699 - f1_score: 0.9392 - val_loss: 0.1647 - val_f1_score: 0.9432\n",
            "Epoch 1580/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1699 - f1_score: 0.9392 - val_loss: 0.1647 - val_f1_score: 0.9433\n",
            "Epoch 1581/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1698 - f1_score: 0.9392 - val_loss: 0.1647 - val_f1_score: 0.9433\n",
            "Epoch 1582/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1698 - f1_score: 0.9393 - val_loss: 0.1646 - val_f1_score: 0.9433\n",
            "Epoch 1583/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1698 - f1_score: 0.9393 - val_loss: 0.1646 - val_f1_score: 0.9434\n",
            "Epoch 1584/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1697 - f1_score: 0.9393 - val_loss: 0.1645 - val_f1_score: 0.9434\n",
            "Epoch 1585/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1697 - f1_score: 0.9393 - val_loss: 0.1645 - val_f1_score: 0.9434\n",
            "Epoch 1586/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1697 - f1_score: 0.9393 - val_loss: 0.1645 - val_f1_score: 0.9434\n",
            "Epoch 1587/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1697 - f1_score: 0.9393 - val_loss: 0.1644 - val_f1_score: 0.9434\n",
            "Epoch 1588/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1696 - f1_score: 0.9393 - val_loss: 0.1644 - val_f1_score: 0.9434\n",
            "Epoch 1589/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1696 - f1_score: 0.9393 - val_loss: 0.1643 - val_f1_score: 0.9434\n",
            "Epoch 1590/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1696 - f1_score: 0.9393 - val_loss: 0.1643 - val_f1_score: 0.9434\n",
            "Epoch 1591/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1695 - f1_score: 0.9393 - val_loss: 0.1643 - val_f1_score: 0.9434\n",
            "Epoch 1592/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1695 - f1_score: 0.9393 - val_loss: 0.1642 - val_f1_score: 0.9434\n",
            "Epoch 1593/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1695 - f1_score: 0.9394 - val_loss: 0.1642 - val_f1_score: 0.9434\n",
            "Epoch 1594/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1695 - f1_score: 0.9394 - val_loss: 0.1641 - val_f1_score: 0.9434\n",
            "Epoch 1595/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1694 - f1_score: 0.9394 - val_loss: 0.1641 - val_f1_score: 0.9434\n",
            "Epoch 1596/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1694 - f1_score: 0.9394 - val_loss: 0.1641 - val_f1_score: 0.9434\n",
            "Epoch 1597/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1694 - f1_score: 0.9394 - val_loss: 0.1640 - val_f1_score: 0.9434\n",
            "Epoch 1598/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1693 - f1_score: 0.9394 - val_loss: 0.1640 - val_f1_score: 0.9434\n",
            "Epoch 1599/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1693 - f1_score: 0.9394 - val_loss: 0.1639 - val_f1_score: 0.9434\n",
            "Epoch 1600/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1693 - f1_score: 0.9395 - val_loss: 0.1639 - val_f1_score: 0.9434\n",
            "Epoch 1601/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1693 - f1_score: 0.9395 - val_loss: 0.1638 - val_f1_score: 0.9435\n",
            "Epoch 1602/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1692 - f1_score: 0.9395 - val_loss: 0.1638 - val_f1_score: 0.9435\n",
            "Epoch 1603/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1692 - f1_score: 0.9395 - val_loss: 0.1638 - val_f1_score: 0.9435\n",
            "Epoch 1604/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1692 - f1_score: 0.9395 - val_loss: 0.1637 - val_f1_score: 0.9435\n",
            "Epoch 1605/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1691 - f1_score: 0.9395 - val_loss: 0.1637 - val_f1_score: 0.9435\n",
            "Epoch 1606/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1691 - f1_score: 0.9395 - val_loss: 0.1637 - val_f1_score: 0.9435\n",
            "Epoch 1607/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1691 - f1_score: 0.9395 - val_loss: 0.1636 - val_f1_score: 0.9435\n",
            "Epoch 1608/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1691 - f1_score: 0.9395 - val_loss: 0.1636 - val_f1_score: 0.9435\n",
            "Epoch 1609/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1690 - f1_score: 0.9395 - val_loss: 0.1635 - val_f1_score: 0.9435\n",
            "Epoch 1610/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1690 - f1_score: 0.9395 - val_loss: 0.1635 - val_f1_score: 0.9434\n",
            "Epoch 1611/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1690 - f1_score: 0.9395 - val_loss: 0.1635 - val_f1_score: 0.9434\n",
            "Epoch 1612/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1690 - f1_score: 0.9395 - val_loss: 0.1634 - val_f1_score: 0.9434\n",
            "Epoch 1613/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1689 - f1_score: 0.9395 - val_loss: 0.1634 - val_f1_score: 0.9434\n",
            "Epoch 1614/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1689 - f1_score: 0.9395 - val_loss: 0.1633 - val_f1_score: 0.9434\n",
            "Epoch 1615/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1689 - f1_score: 0.9395 - val_loss: 0.1633 - val_f1_score: 0.9434\n",
            "Epoch 1616/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1688 - f1_score: 0.9395 - val_loss: 0.1633 - val_f1_score: 0.9434\n",
            "Epoch 1617/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1688 - f1_score: 0.9395 - val_loss: 0.1632 - val_f1_score: 0.9434\n",
            "Epoch 1618/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1688 - f1_score: 0.9395 - val_loss: 0.1632 - val_f1_score: 0.9434\n",
            "Epoch 1619/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1688 - f1_score: 0.9395 - val_loss: 0.1631 - val_f1_score: 0.9434\n",
            "Epoch 1620/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1687 - f1_score: 0.9395 - val_loss: 0.1631 - val_f1_score: 0.9434\n",
            "Epoch 1621/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1687 - f1_score: 0.9395 - val_loss: 0.1631 - val_f1_score: 0.9434\n",
            "Epoch 1622/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1687 - f1_score: 0.9395 - val_loss: 0.1630 - val_f1_score: 0.9434\n",
            "Epoch 1623/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1687 - f1_score: 0.9395 - val_loss: 0.1630 - val_f1_score: 0.9435\n",
            "Epoch 1624/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1686 - f1_score: 0.9395 - val_loss: 0.1630 - val_f1_score: 0.9435\n",
            "Epoch 1625/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1686 - f1_score: 0.9395 - val_loss: 0.1629 - val_f1_score: 0.9435\n",
            "Epoch 1626/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1686 - f1_score: 0.9395 - val_loss: 0.1629 - val_f1_score: 0.9435\n",
            "Epoch 1627/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1686 - f1_score: 0.9395 - val_loss: 0.1628 - val_f1_score: 0.9435\n",
            "Epoch 1628/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1685 - f1_score: 0.9395 - val_loss: 0.1628 - val_f1_score: 0.9435\n",
            "Epoch 1629/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1685 - f1_score: 0.9395 - val_loss: 0.1628 - val_f1_score: 0.9435\n",
            "Epoch 1630/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1685 - f1_score: 0.9395 - val_loss: 0.1627 - val_f1_score: 0.9435\n",
            "Epoch 1631/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1684 - f1_score: 0.9395 - val_loss: 0.1627 - val_f1_score: 0.9435\n",
            "Epoch 1632/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1684 - f1_score: 0.9395 - val_loss: 0.1626 - val_f1_score: 0.9435\n",
            "Epoch 1633/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1684 - f1_score: 0.9395 - val_loss: 0.1626 - val_f1_score: 0.9435\n",
            "Epoch 1634/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1684 - f1_score: 0.9396 - val_loss: 0.1626 - val_f1_score: 0.9435\n",
            "Epoch 1635/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1683 - f1_score: 0.9396 - val_loss: 0.1625 - val_f1_score: 0.9435\n",
            "Epoch 1636/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1683 - f1_score: 0.9396 - val_loss: 0.1625 - val_f1_score: 0.9435\n",
            "Epoch 1637/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1683 - f1_score: 0.9396 - val_loss: 0.1625 - val_f1_score: 0.9435\n",
            "Epoch 1638/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1683 - f1_score: 0.9396 - val_loss: 0.1624 - val_f1_score: 0.9435\n",
            "Epoch 1639/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1682 - f1_score: 0.9396 - val_loss: 0.1624 - val_f1_score: 0.9435\n",
            "Epoch 1640/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1682 - f1_score: 0.9396 - val_loss: 0.1624 - val_f1_score: 0.9435\n",
            "Epoch 1641/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1682 - f1_score: 0.9396 - val_loss: 0.1623 - val_f1_score: 0.9435\n",
            "Epoch 1642/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1682 - f1_score: 0.9395 - val_loss: 0.1623 - val_f1_score: 0.9435\n",
            "Epoch 1643/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1681 - f1_score: 0.9395 - val_loss: 0.1622 - val_f1_score: 0.9435\n",
            "Epoch 1644/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1681 - f1_score: 0.9395 - val_loss: 0.1622 - val_f1_score: 0.9435\n",
            "Epoch 1645/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1681 - f1_score: 0.9395 - val_loss: 0.1622 - val_f1_score: 0.9435\n",
            "Epoch 1646/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1681 - f1_score: 0.9395 - val_loss: 0.1621 - val_f1_score: 0.9435\n",
            "Epoch 1647/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1680 - f1_score: 0.9395 - val_loss: 0.1621 - val_f1_score: 0.9435\n",
            "Epoch 1648/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1680 - f1_score: 0.9395 - val_loss: 0.1621 - val_f1_score: 0.9435\n",
            "Epoch 1649/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1680 - f1_score: 0.9395 - val_loss: 0.1620 - val_f1_score: 0.9435\n",
            "Epoch 1650/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1680 - f1_score: 0.9396 - val_loss: 0.1620 - val_f1_score: 0.9435\n",
            "Epoch 1651/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1679 - f1_score: 0.9396 - val_loss: 0.1619 - val_f1_score: 0.9435\n",
            "Epoch 1652/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1679 - f1_score: 0.9396 - val_loss: 0.1619 - val_f1_score: 0.9435\n",
            "Epoch 1653/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1679 - f1_score: 0.9396 - val_loss: 0.1619 - val_f1_score: 0.9435\n",
            "Epoch 1654/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1679 - f1_score: 0.9396 - val_loss: 0.1618 - val_f1_score: 0.9435\n",
            "Epoch 1655/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1678 - f1_score: 0.9396 - val_loss: 0.1618 - val_f1_score: 0.9435\n",
            "Epoch 1656/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1678 - f1_score: 0.9396 - val_loss: 0.1618 - val_f1_score: 0.9435\n",
            "Epoch 1657/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1678 - f1_score: 0.9396 - val_loss: 0.1617 - val_f1_score: 0.9435\n",
            "Epoch 1658/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1678 - f1_score: 0.9396 - val_loss: 0.1617 - val_f1_score: 0.9435\n",
            "Epoch 1659/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1677 - f1_score: 0.9396 - val_loss: 0.1617 - val_f1_score: 0.9436\n",
            "Epoch 1660/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1677 - f1_score: 0.9396 - val_loss: 0.1616 - val_f1_score: 0.9436\n",
            "Epoch 1661/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1677 - f1_score: 0.9396 - val_loss: 0.1616 - val_f1_score: 0.9436\n",
            "Epoch 1662/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1677 - f1_score: 0.9396 - val_loss: 0.1616 - val_f1_score: 0.9436\n",
            "Epoch 1663/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1676 - f1_score: 0.9396 - val_loss: 0.1615 - val_f1_score: 0.9436\n",
            "Epoch 1664/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1676 - f1_score: 0.9396 - val_loss: 0.1615 - val_f1_score: 0.9436\n",
            "Epoch 1665/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1676 - f1_score: 0.9396 - val_loss: 0.1615 - val_f1_score: 0.9436\n",
            "Epoch 1666/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1676 - f1_score: 0.9396 - val_loss: 0.1614 - val_f1_score: 0.9436\n",
            "Epoch 1667/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1675 - f1_score: 0.9396 - val_loss: 0.1614 - val_f1_score: 0.9436\n",
            "Epoch 1668/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1675 - f1_score: 0.9396 - val_loss: 0.1613 - val_f1_score: 0.9436\n",
            "Epoch 1669/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1675 - f1_score: 0.9396 - val_loss: 0.1613 - val_f1_score: 0.9436\n",
            "Epoch 1670/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1675 - f1_score: 0.9396 - val_loss: 0.1613 - val_f1_score: 0.9436\n",
            "Epoch 1671/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1674 - f1_score: 0.9396 - val_loss: 0.1612 - val_f1_score: 0.9436\n",
            "Epoch 1672/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1674 - f1_score: 0.9396 - val_loss: 0.1612 - val_f1_score: 0.9436\n",
            "Epoch 1673/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1674 - f1_score: 0.9396 - val_loss: 0.1612 - val_f1_score: 0.9436\n",
            "Epoch 1674/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1674 - f1_score: 0.9396 - val_loss: 0.1611 - val_f1_score: 0.9436\n",
            "Epoch 1675/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1673 - f1_score: 0.9396 - val_loss: 0.1611 - val_f1_score: 0.9436\n",
            "Epoch 1676/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1673 - f1_score: 0.9396 - val_loss: 0.1611 - val_f1_score: 0.9436\n",
            "Epoch 1677/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1673 - f1_score: 0.9396 - val_loss: 0.1610 - val_f1_score: 0.9436\n",
            "Epoch 1678/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1673 - f1_score: 0.9396 - val_loss: 0.1610 - val_f1_score: 0.9436\n",
            "Epoch 1679/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1672 - f1_score: 0.9397 - val_loss: 0.1610 - val_f1_score: 0.9436\n",
            "Epoch 1680/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1672 - f1_score: 0.9397 - val_loss: 0.1609 - val_f1_score: 0.9436\n",
            "Epoch 1681/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1672 - f1_score: 0.9397 - val_loss: 0.1609 - val_f1_score: 0.9436\n",
            "Epoch 1682/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1672 - f1_score: 0.9397 - val_loss: 0.1609 - val_f1_score: 0.9436\n",
            "Epoch 1683/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1671 - f1_score: 0.9397 - val_loss: 0.1608 - val_f1_score: 0.9436\n",
            "Epoch 1684/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1671 - f1_score: 0.9397 - val_loss: 0.1608 - val_f1_score: 0.9436\n",
            "Epoch 1685/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1671 - f1_score: 0.9397 - val_loss: 0.1608 - val_f1_score: 0.9436\n",
            "Epoch 1686/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1671 - f1_score: 0.9397 - val_loss: 0.1607 - val_f1_score: 0.9436\n",
            "Epoch 1687/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1670 - f1_score: 0.9397 - val_loss: 0.1607 - val_f1_score: 0.9436\n",
            "Epoch 1688/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1670 - f1_score: 0.9397 - val_loss: 0.1607 - val_f1_score: 0.9436\n",
            "Epoch 1689/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1670 - f1_score: 0.9397 - val_loss: 0.1606 - val_f1_score: 0.9436\n",
            "Epoch 1690/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1670 - f1_score: 0.9397 - val_loss: 0.1606 - val_f1_score: 0.9436\n",
            "Epoch 1691/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1669 - f1_score: 0.9397 - val_loss: 0.1606 - val_f1_score: 0.9436\n",
            "Epoch 1692/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1669 - f1_score: 0.9397 - val_loss: 0.1605 - val_f1_score: 0.9436\n",
            "Epoch 1693/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1669 - f1_score: 0.9397 - val_loss: 0.1605 - val_f1_score: 0.9436\n",
            "Epoch 1694/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1669 - f1_score: 0.9397 - val_loss: 0.1605 - val_f1_score: 0.9436\n",
            "Epoch 1695/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1668 - f1_score: 0.9397 - val_loss: 0.1604 - val_f1_score: 0.9436\n",
            "Epoch 1696/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1668 - f1_score: 0.9397 - val_loss: 0.1604 - val_f1_score: 0.9436\n",
            "Epoch 1697/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1668 - f1_score: 0.9397 - val_loss: 0.1604 - val_f1_score: 0.9436\n",
            "Epoch 1698/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1668 - f1_score: 0.9397 - val_loss: 0.1603 - val_f1_score: 0.9436\n",
            "Epoch 1699/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1668 - f1_score: 0.9397 - val_loss: 0.1603 - val_f1_score: 0.9436\n",
            "Epoch 1700/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1667 - f1_score: 0.9397 - val_loss: 0.1603 - val_f1_score: 0.9436\n",
            "Epoch 1701/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1667 - f1_score: 0.9397 - val_loss: 0.1602 - val_f1_score: 0.9436\n",
            "Epoch 1702/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1667 - f1_score: 0.9397 - val_loss: 0.1602 - val_f1_score: 0.9436\n",
            "Epoch 1703/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1667 - f1_score: 0.9397 - val_loss: 0.1602 - val_f1_score: 0.9436\n",
            "Epoch 1704/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1666 - f1_score: 0.9397 - val_loss: 0.1601 - val_f1_score: 0.9436\n",
            "Epoch 1705/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1666 - f1_score: 0.9397 - val_loss: 0.1601 - val_f1_score: 0.9436\n",
            "Epoch 1706/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1666 - f1_score: 0.9397 - val_loss: 0.1601 - val_f1_score: 0.9436\n",
            "Epoch 1707/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1666 - f1_score: 0.9397 - val_loss: 0.1600 - val_f1_score: 0.9436\n",
            "Epoch 1708/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1665 - f1_score: 0.9397 - val_loss: 0.1600 - val_f1_score: 0.9436\n",
            "Epoch 1709/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1665 - f1_score: 0.9397 - val_loss: 0.1600 - val_f1_score: 0.9436\n",
            "Epoch 1710/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1665 - f1_score: 0.9397 - val_loss: 0.1599 - val_f1_score: 0.9436\n",
            "Epoch 1711/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1665 - f1_score: 0.9397 - val_loss: 0.1599 - val_f1_score: 0.9436\n",
            "Epoch 1712/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1665 - f1_score: 0.9397 - val_loss: 0.1599 - val_f1_score: 0.9436\n",
            "Epoch 1713/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1664 - f1_score: 0.9397 - val_loss: 0.1598 - val_f1_score: 0.9436\n",
            "Epoch 1714/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1664 - f1_score: 0.9397 - val_loss: 0.1598 - val_f1_score: 0.9436\n",
            "Epoch 1715/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1664 - f1_score: 0.9397 - val_loss: 0.1598 - val_f1_score: 0.9436\n",
            "Epoch 1716/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1664 - f1_score: 0.9397 - val_loss: 0.1597 - val_f1_score: 0.9436\n",
            "Epoch 1717/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1663 - f1_score: 0.9397 - val_loss: 0.1597 - val_f1_score: 0.9436\n",
            "Epoch 1718/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1663 - f1_score: 0.9397 - val_loss: 0.1597 - val_f1_score: 0.9436\n",
            "Epoch 1719/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1663 - f1_score: 0.9397 - val_loss: 0.1597 - val_f1_score: 0.9436\n",
            "Epoch 1720/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1663 - f1_score: 0.9397 - val_loss: 0.1596 - val_f1_score: 0.9436\n",
            "Epoch 1721/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1662 - f1_score: 0.9397 - val_loss: 0.1596 - val_f1_score: 0.9436\n",
            "Epoch 1722/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1662 - f1_score: 0.9397 - val_loss: 0.1596 - val_f1_score: 0.9436\n",
            "Epoch 1723/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1662 - f1_score: 0.9397 - val_loss: 0.1595 - val_f1_score: 0.9436\n",
            "Epoch 1724/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1662 - f1_score: 0.9397 - val_loss: 0.1595 - val_f1_score: 0.9436\n",
            "Epoch 1725/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1662 - f1_score: 0.9397 - val_loss: 0.1595 - val_f1_score: 0.9436\n",
            "Epoch 1726/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1661 - f1_score: 0.9397 - val_loss: 0.1594 - val_f1_score: 0.9436\n",
            "Epoch 1727/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1661 - f1_score: 0.9397 - val_loss: 0.1594 - val_f1_score: 0.9436\n",
            "Epoch 1728/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1661 - f1_score: 0.9397 - val_loss: 0.1594 - val_f1_score: 0.9436\n",
            "Epoch 1729/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1661 - f1_score: 0.9397 - val_loss: 0.1593 - val_f1_score: 0.9436\n",
            "Epoch 1730/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1660 - f1_score: 0.9397 - val_loss: 0.1593 - val_f1_score: 0.9436\n",
            "Epoch 1731/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1660 - f1_score: 0.9397 - val_loss: 0.1593 - val_f1_score: 0.9436\n",
            "Epoch 1732/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1660 - f1_score: 0.9397 - val_loss: 0.1593 - val_f1_score: 0.9436\n",
            "Epoch 1733/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1660 - f1_score: 0.9397 - val_loss: 0.1592 - val_f1_score: 0.9436\n",
            "Epoch 1734/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1660 - f1_score: 0.9397 - val_loss: 0.1592 - val_f1_score: 0.9436\n",
            "Epoch 1735/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1659 - f1_score: 0.9397 - val_loss: 0.1592 - val_f1_score: 0.9436\n",
            "Epoch 1736/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1659 - f1_score: 0.9397 - val_loss: 0.1591 - val_f1_score: 0.9436\n",
            "Epoch 1737/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1659 - f1_score: 0.9397 - val_loss: 0.1591 - val_f1_score: 0.9436\n",
            "Epoch 1738/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1659 - f1_score: 0.9397 - val_loss: 0.1591 - val_f1_score: 0.9436\n",
            "Epoch 1739/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1658 - f1_score: 0.9397 - val_loss: 0.1590 - val_f1_score: 0.9436\n",
            "Epoch 1740/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1658 - f1_score: 0.9397 - val_loss: 0.1590 - val_f1_score: 0.9436\n",
            "Epoch 1741/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1658 - f1_score: 0.9397 - val_loss: 0.1590 - val_f1_score: 0.9436\n",
            "Epoch 1742/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1658 - f1_score: 0.9397 - val_loss: 0.1590 - val_f1_score: 0.9436\n",
            "Epoch 1743/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1658 - f1_score: 0.9397 - val_loss: 0.1589 - val_f1_score: 0.9436\n",
            "Epoch 1744/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1657 - f1_score: 0.9397 - val_loss: 0.1589 - val_f1_score: 0.9436\n",
            "Epoch 1745/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1657 - f1_score: 0.9397 - val_loss: 0.1589 - val_f1_score: 0.9436\n",
            "Epoch 1746/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1657 - f1_score: 0.9397 - val_loss: 0.1588 - val_f1_score: 0.9436\n",
            "Epoch 1747/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1657 - f1_score: 0.9397 - val_loss: 0.1588 - val_f1_score: 0.9437\n",
            "Epoch 1748/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1657 - f1_score: 0.9397 - val_loss: 0.1588 - val_f1_score: 0.9437\n",
            "Epoch 1749/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1656 - f1_score: 0.9397 - val_loss: 0.1587 - val_f1_score: 0.9437\n",
            "Epoch 1750/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1656 - f1_score: 0.9397 - val_loss: 0.1587 - val_f1_score: 0.9437\n",
            "Epoch 1751/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1656 - f1_score: 0.9397 - val_loss: 0.1587 - val_f1_score: 0.9437\n",
            "Epoch 1752/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1656 - f1_score: 0.9397 - val_loss: 0.1587 - val_f1_score: 0.9437\n",
            "Epoch 1753/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1655 - f1_score: 0.9397 - val_loss: 0.1586 - val_f1_score: 0.9437\n",
            "Epoch 1754/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1655 - f1_score: 0.9397 - val_loss: 0.1586 - val_f1_score: 0.9437\n",
            "Epoch 1755/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1655 - f1_score: 0.9397 - val_loss: 0.1586 - val_f1_score: 0.9437\n",
            "Epoch 1756/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1655 - f1_score: 0.9397 - val_loss: 0.1585 - val_f1_score: 0.9437\n",
            "Epoch 1757/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1655 - f1_score: 0.9397 - val_loss: 0.1585 - val_f1_score: 0.9437\n",
            "Epoch 1758/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1654 - f1_score: 0.9398 - val_loss: 0.1585 - val_f1_score: 0.9437\n",
            "Epoch 1759/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1654 - f1_score: 0.9398 - val_loss: 0.1585 - val_f1_score: 0.9437\n",
            "Epoch 1760/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1654 - f1_score: 0.9398 - val_loss: 0.1584 - val_f1_score: 0.9437\n",
            "Epoch 1761/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1654 - f1_score: 0.9398 - val_loss: 0.1584 - val_f1_score: 0.9437\n",
            "Epoch 1762/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1654 - f1_score: 0.9398 - val_loss: 0.1584 - val_f1_score: 0.9437\n",
            "Epoch 1763/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1653 - f1_score: 0.9398 - val_loss: 0.1583 - val_f1_score: 0.9437\n",
            "Epoch 1764/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1653 - f1_score: 0.9398 - val_loss: 0.1583 - val_f1_score: 0.9437\n",
            "Epoch 1765/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1653 - f1_score: 0.9398 - val_loss: 0.1583 - val_f1_score: 0.9437\n",
            "Epoch 1766/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1653 - f1_score: 0.9398 - val_loss: 0.1583 - val_f1_score: 0.9437\n",
            "Epoch 1767/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1653 - f1_score: 0.9398 - val_loss: 0.1582 - val_f1_score: 0.9437\n",
            "Epoch 1768/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1652 - f1_score: 0.9398 - val_loss: 0.1582 - val_f1_score: 0.9437\n",
            "Epoch 1769/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1652 - f1_score: 0.9398 - val_loss: 0.1582 - val_f1_score: 0.9437\n",
            "Epoch 1770/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1652 - f1_score: 0.9398 - val_loss: 0.1581 - val_f1_score: 0.9437\n",
            "Epoch 1771/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1652 - f1_score: 0.9398 - val_loss: 0.1581 - val_f1_score: 0.9437\n",
            "Epoch 1772/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1651 - f1_score: 0.9398 - val_loss: 0.1581 - val_f1_score: 0.9437\n",
            "Epoch 1773/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1651 - f1_score: 0.9398 - val_loss: 0.1581 - val_f1_score: 0.9437\n",
            "Epoch 1774/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1651 - f1_score: 0.9398 - val_loss: 0.1580 - val_f1_score: 0.9437\n",
            "Epoch 1775/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1651 - f1_score: 0.9398 - val_loss: 0.1580 - val_f1_score: 0.9437\n",
            "Epoch 1776/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1651 - f1_score: 0.9398 - val_loss: 0.1580 - val_f1_score: 0.9438\n",
            "Epoch 1777/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1650 - f1_score: 0.9398 - val_loss: 0.1579 - val_f1_score: 0.9438\n",
            "Epoch 1778/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1650 - f1_score: 0.9398 - val_loss: 0.1579 - val_f1_score: 0.9438\n",
            "Epoch 1779/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1650 - f1_score: 0.9398 - val_loss: 0.1579 - val_f1_score: 0.9438\n",
            "Epoch 1780/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1650 - f1_score: 0.9398 - val_loss: 0.1579 - val_f1_score: 0.9438\n",
            "Epoch 1781/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1650 - f1_score: 0.9398 - val_loss: 0.1578 - val_f1_score: 0.9438\n",
            "Epoch 1782/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1649 - f1_score: 0.9398 - val_loss: 0.1578 - val_f1_score: 0.9438\n",
            "Epoch 1783/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1649 - f1_score: 0.9398 - val_loss: 0.1578 - val_f1_score: 0.9438\n",
            "Epoch 1784/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1649 - f1_score: 0.9398 - val_loss: 0.1578 - val_f1_score: 0.9438\n",
            "Epoch 1785/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1649 - f1_score: 0.9398 - val_loss: 0.1577 - val_f1_score: 0.9438\n",
            "Epoch 1786/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1649 - f1_score: 0.9398 - val_loss: 0.1577 - val_f1_score: 0.9438\n",
            "Epoch 1787/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1648 - f1_score: 0.9398 - val_loss: 0.1577 - val_f1_score: 0.9438\n",
            "Epoch 1788/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1648 - f1_score: 0.9398 - val_loss: 0.1576 - val_f1_score: 0.9438\n",
            "Epoch 1789/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1648 - f1_score: 0.9399 - val_loss: 0.1576 - val_f1_score: 0.9438\n",
            "Epoch 1790/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1648 - f1_score: 0.9399 - val_loss: 0.1576 - val_f1_score: 0.9438\n",
            "Epoch 1791/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1648 - f1_score: 0.9398 - val_loss: 0.1576 - val_f1_score: 0.9438\n",
            "Epoch 1792/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1647 - f1_score: 0.9398 - val_loss: 0.1575 - val_f1_score: 0.9438\n",
            "Epoch 1793/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1647 - f1_score: 0.9398 - val_loss: 0.1575 - val_f1_score: 0.9438\n",
            "Epoch 1794/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1647 - f1_score: 0.9398 - val_loss: 0.1575 - val_f1_score: 0.9438\n",
            "Epoch 1795/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1647 - f1_score: 0.9398 - val_loss: 0.1575 - val_f1_score: 0.9438\n",
            "Epoch 1796/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1647 - f1_score: 0.9398 - val_loss: 0.1574 - val_f1_score: 0.9438\n",
            "Epoch 1797/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1646 - f1_score: 0.9398 - val_loss: 0.1574 - val_f1_score: 0.9438\n",
            "Epoch 1798/2000\n",
            "322/322 [==============================] - 2s 7ms/step - loss: 0.1646 - f1_score: 0.9398 - val_loss: 0.1574 - val_f1_score: 0.9438\n",
            "Epoch 1799/2000\n",
            "322/322 [==============================] - 3s 10ms/step - loss: 0.1646 - f1_score: 0.9398 - val_loss: 0.1574 - val_f1_score: 0.9438\n",
            "Epoch 1800/2000\n",
            "322/322 [==============================] - 3s 9ms/step - loss: 0.1646 - f1_score: 0.9398 - val_loss: 0.1573 - val_f1_score: 0.9438\n",
            "Epoch 1801/2000\n",
            "322/322 [==============================] - 3s 9ms/step - loss: 0.1646 - f1_score: 0.9398 - val_loss: 0.1573 - val_f1_score: 0.9438\n",
            "Epoch 1802/2000\n",
            "322/322 [==============================] - 2s 6ms/step - loss: 0.1645 - f1_score: 0.9398 - val_loss: 0.1573 - val_f1_score: 0.9438\n",
            "Epoch 1803/2000\n",
            "322/322 [==============================] - 2s 7ms/step - loss: 0.1645 - f1_score: 0.9398 - val_loss: 0.1572 - val_f1_score: 0.9438\n",
            "Epoch 1804/2000\n",
            "322/322 [==============================] - 3s 9ms/step - loss: 0.1645 - f1_score: 0.9398 - val_loss: 0.1572 - val_f1_score: 0.9438\n",
            "Epoch 1805/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1645 - f1_score: 0.9398 - val_loss: 0.1572 - val_f1_score: 0.9438\n",
            "Epoch 1806/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1645 - f1_score: 0.9398 - val_loss: 0.1572 - val_f1_score: 0.9438\n",
            "Epoch 1807/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1644 - f1_score: 0.9398 - val_loss: 0.1571 - val_f1_score: 0.9438\n",
            "Epoch 1808/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1644 - f1_score: 0.9399 - val_loss: 0.1571 - val_f1_score: 0.9438\n",
            "Epoch 1809/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1644 - f1_score: 0.9398 - val_loss: 0.1571 - val_f1_score: 0.9439\n",
            "Epoch 1810/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1644 - f1_score: 0.9398 - val_loss: 0.1571 - val_f1_score: 0.9439\n",
            "Epoch 1811/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1644 - f1_score: 0.9399 - val_loss: 0.1570 - val_f1_score: 0.9439\n",
            "Epoch 1812/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1644 - f1_score: 0.9399 - val_loss: 0.1570 - val_f1_score: 0.9439\n",
            "Epoch 1813/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1643 - f1_score: 0.9399 - val_loss: 0.1570 - val_f1_score: 0.9439\n",
            "Epoch 1814/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1643 - f1_score: 0.9399 - val_loss: 0.1570 - val_f1_score: 0.9439\n",
            "Epoch 1815/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1643 - f1_score: 0.9399 - val_loss: 0.1569 - val_f1_score: 0.9439\n",
            "Epoch 1816/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1643 - f1_score: 0.9399 - val_loss: 0.1569 - val_f1_score: 0.9439\n",
            "Epoch 1817/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1643 - f1_score: 0.9399 - val_loss: 0.1569 - val_f1_score: 0.9439\n",
            "Epoch 1818/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1642 - f1_score: 0.9399 - val_loss: 0.1569 - val_f1_score: 0.9439\n",
            "Epoch 1819/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1642 - f1_score: 0.9399 - val_loss: 0.1568 - val_f1_score: 0.9439\n",
            "Epoch 1820/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1642 - f1_score: 0.9399 - val_loss: 0.1568 - val_f1_score: 0.9439\n",
            "Epoch 1821/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1642 - f1_score: 0.9399 - val_loss: 0.1568 - val_f1_score: 0.9439\n",
            "Epoch 1822/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1641 - f1_score: 0.9399 - val_loss: 0.1568 - val_f1_score: 0.9439\n",
            "Epoch 1823/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1641 - f1_score: 0.9399 - val_loss: 0.1567 - val_f1_score: 0.9439\n",
            "Epoch 1824/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1641 - f1_score: 0.9399 - val_loss: 0.1567 - val_f1_score: 0.9439\n",
            "Epoch 1825/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1641 - f1_score: 0.9399 - val_loss: 0.1567 - val_f1_score: 0.9439\n",
            "Epoch 1826/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1640 - f1_score: 0.9398 - val_loss: 0.1567 - val_f1_score: 0.9439\n",
            "Epoch 1827/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1640 - f1_score: 0.9398 - val_loss: 0.1566 - val_f1_score: 0.9439\n",
            "Epoch 1828/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1640 - f1_score: 0.9398 - val_loss: 0.1566 - val_f1_score: 0.9439\n",
            "Epoch 1829/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1640 - f1_score: 0.9398 - val_loss: 0.1566 - val_f1_score: 0.9439\n",
            "Epoch 1830/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1639 - f1_score: 0.9398 - val_loss: 0.1566 - val_f1_score: 0.9439\n",
            "Epoch 1831/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1639 - f1_score: 0.9398 - val_loss: 0.1565 - val_f1_score: 0.9439\n",
            "Epoch 1832/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1639 - f1_score: 0.9398 - val_loss: 0.1565 - val_f1_score: 0.9439\n",
            "Epoch 1833/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1639 - f1_score: 0.9399 - val_loss: 0.1565 - val_f1_score: 0.9439\n",
            "Epoch 1834/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1638 - f1_score: 0.9399 - val_loss: 0.1565 - val_f1_score: 0.9439\n",
            "Epoch 1835/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1638 - f1_score: 0.9399 - val_loss: 0.1564 - val_f1_score: 0.9439\n",
            "Epoch 1836/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1638 - f1_score: 0.9399 - val_loss: 0.1564 - val_f1_score: 0.9439\n",
            "Epoch 1837/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1637 - f1_score: 0.9399 - val_loss: 0.1564 - val_f1_score: 0.9438\n",
            "Epoch 1838/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1637 - f1_score: 0.9399 - val_loss: 0.1564 - val_f1_score: 0.9438\n",
            "Epoch 1839/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1637 - f1_score: 0.9399 - val_loss: 0.1563 - val_f1_score: 0.9438\n",
            "Epoch 1840/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1636 - f1_score: 0.9399 - val_loss: 0.1563 - val_f1_score: 0.9438\n",
            "Epoch 1841/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1636 - f1_score: 0.9399 - val_loss: 0.1563 - val_f1_score: 0.9437\n",
            "Epoch 1842/2000\n",
            "322/322 [==============================] - 2s 8ms/step - loss: 0.1636 - f1_score: 0.9399 - val_loss: 0.1562 - val_f1_score: 0.9437\n",
            "Epoch 1843/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1635 - f1_score: 0.9399 - val_loss: 0.1562 - val_f1_score: 0.9436\n",
            "Epoch 1844/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1635 - f1_score: 0.9399 - val_loss: 0.1562 - val_f1_score: 0.9436\n",
            "Epoch 1845/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1635 - f1_score: 0.9399 - val_loss: 0.1562 - val_f1_score: 0.9436\n",
            "Epoch 1846/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1634 - f1_score: 0.9399 - val_loss: 0.1561 - val_f1_score: 0.9436\n",
            "Epoch 1847/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1634 - f1_score: 0.9399 - val_loss: 0.1561 - val_f1_score: 0.9436\n",
            "Epoch 1848/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1634 - f1_score: 0.9399 - val_loss: 0.1561 - val_f1_score: 0.9436\n",
            "Epoch 1849/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1633 - f1_score: 0.9399 - val_loss: 0.1560 - val_f1_score: 0.9436\n",
            "Epoch 1850/2000\n",
            "322/322 [==============================] - 3s 8ms/step - loss: 0.1633 - f1_score: 0.9399 - val_loss: 0.1560 - val_f1_score: 0.9435\n",
            "Epoch 1851/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1633 - f1_score: 0.9399 - val_loss: 0.1560 - val_f1_score: 0.9435\n",
            "Epoch 1852/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1632 - f1_score: 0.9399 - val_loss: 0.1560 - val_f1_score: 0.9435\n",
            "Epoch 1853/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1632 - f1_score: 0.9399 - val_loss: 0.1559 - val_f1_score: 0.9435\n",
            "Epoch 1854/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1632 - f1_score: 0.9399 - val_loss: 0.1559 - val_f1_score: 0.9435\n",
            "Epoch 1855/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1631 - f1_score: 0.9399 - val_loss: 0.1559 - val_f1_score: 0.9435\n",
            "Epoch 1856/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1631 - f1_score: 0.9399 - val_loss: 0.1558 - val_f1_score: 0.9435\n",
            "Epoch 1857/2000\n",
            "322/322 [==============================] - 2s 8ms/step - loss: 0.1631 - f1_score: 0.9399 - val_loss: 0.1558 - val_f1_score: 0.9435\n",
            "Epoch 1858/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1630 - f1_score: 0.9399 - val_loss: 0.1558 - val_f1_score: 0.9435\n",
            "Epoch 1859/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1630 - f1_score: 0.9399 - val_loss: 0.1557 - val_f1_score: 0.9434\n",
            "Epoch 1860/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1630 - f1_score: 0.9399 - val_loss: 0.1557 - val_f1_score: 0.9434\n",
            "Epoch 1861/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1629 - f1_score: 0.9399 - val_loss: 0.1557 - val_f1_score: 0.9434\n",
            "Epoch 1862/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1629 - f1_score: 0.9399 - val_loss: 0.1557 - val_f1_score: 0.9434\n",
            "Epoch 1863/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1629 - f1_score: 0.9399 - val_loss: 0.1556 - val_f1_score: 0.9434\n",
            "Epoch 1864/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1628 - f1_score: 0.9399 - val_loss: 0.1556 - val_f1_score: 0.9434\n",
            "Epoch 1865/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1628 - f1_score: 0.9399 - val_loss: 0.1556 - val_f1_score: 0.9434\n",
            "Epoch 1866/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1628 - f1_score: 0.9399 - val_loss: 0.1555 - val_f1_score: 0.9434\n",
            "Epoch 1867/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1627 - f1_score: 0.9399 - val_loss: 0.1555 - val_f1_score: 0.9434\n",
            "Epoch 1868/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1627 - f1_score: 0.9399 - val_loss: 0.1555 - val_f1_score: 0.9434\n",
            "Epoch 1869/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1627 - f1_score: 0.9399 - val_loss: 0.1554 - val_f1_score: 0.9434\n",
            "Epoch 1870/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1626 - f1_score: 0.9399 - val_loss: 0.1554 - val_f1_score: 0.9434\n",
            "Epoch 1871/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1626 - f1_score: 0.9399 - val_loss: 0.1554 - val_f1_score: 0.9434\n",
            "Epoch 1872/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1625 - f1_score: 0.9399 - val_loss: 0.1554 - val_f1_score: 0.9434\n",
            "Epoch 1873/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1625 - f1_score: 0.9399 - val_loss: 0.1553 - val_f1_score: 0.9434\n",
            "Epoch 1874/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1625 - f1_score: 0.9399 - val_loss: 0.1553 - val_f1_score: 0.9434\n",
            "Epoch 1875/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1624 - f1_score: 0.9399 - val_loss: 0.1553 - val_f1_score: 0.9434\n",
            "Epoch 1876/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1624 - f1_score: 0.9399 - val_loss: 0.1552 - val_f1_score: 0.9434\n",
            "Epoch 1877/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1624 - f1_score: 0.9399 - val_loss: 0.1552 - val_f1_score: 0.9434\n",
            "Epoch 1878/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1623 - f1_score: 0.9399 - val_loss: 0.1552 - val_f1_score: 0.9434\n",
            "Epoch 1879/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1623 - f1_score: 0.9399 - val_loss: 0.1552 - val_f1_score: 0.9434\n",
            "Epoch 1880/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1623 - f1_score: 0.9399 - val_loss: 0.1551 - val_f1_score: 0.9434\n",
            "Epoch 1881/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1622 - f1_score: 0.9399 - val_loss: 0.1551 - val_f1_score: 0.9434\n",
            "Epoch 1882/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1622 - f1_score: 0.9399 - val_loss: 0.1551 - val_f1_score: 0.9434\n",
            "Epoch 1883/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1622 - f1_score: 0.9399 - val_loss: 0.1550 - val_f1_score: 0.9434\n",
            "Epoch 1884/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1621 - f1_score: 0.9399 - val_loss: 0.1550 - val_f1_score: 0.9434\n",
            "Epoch 1885/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1621 - f1_score: 0.9399 - val_loss: 0.1550 - val_f1_score: 0.9434\n",
            "Epoch 1886/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1621 - f1_score: 0.9399 - val_loss: 0.1550 - val_f1_score: 0.9434\n",
            "Epoch 1887/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1620 - f1_score: 0.9399 - val_loss: 0.1549 - val_f1_score: 0.9434\n",
            "Epoch 1888/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1620 - f1_score: 0.9399 - val_loss: 0.1549 - val_f1_score: 0.9434\n",
            "Epoch 1889/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1620 - f1_score: 0.9398 - val_loss: 0.1549 - val_f1_score: 0.9434\n",
            "Epoch 1890/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1619 - f1_score: 0.9398 - val_loss: 0.1548 - val_f1_score: 0.9434\n",
            "Epoch 1891/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1619 - f1_score: 0.9398 - val_loss: 0.1548 - val_f1_score: 0.9434\n",
            "Epoch 1892/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1619 - f1_score: 0.9398 - val_loss: 0.1548 - val_f1_score: 0.9434\n",
            "Epoch 1893/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1618 - f1_score: 0.9398 - val_loss: 0.1547 - val_f1_score: 0.9433\n",
            "Epoch 1894/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1618 - f1_score: 0.9398 - val_loss: 0.1547 - val_f1_score: 0.9432\n",
            "Epoch 1895/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1618 - f1_score: 0.9398 - val_loss: 0.1547 - val_f1_score: 0.9432\n",
            "Epoch 1896/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1617 - f1_score: 0.9398 - val_loss: 0.1547 - val_f1_score: 0.9432\n",
            "Epoch 1897/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1617 - f1_score: 0.9398 - val_loss: 0.1546 - val_f1_score: 0.9432\n",
            "Epoch 1898/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1617 - f1_score: 0.9398 - val_loss: 0.1546 - val_f1_score: 0.9432\n",
            "Epoch 1899/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1616 - f1_score: 0.9398 - val_loss: 0.1546 - val_f1_score: 0.9432\n",
            "Epoch 1900/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1616 - f1_score: 0.9398 - val_loss: 0.1546 - val_f1_score: 0.9432\n",
            "Epoch 1901/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1616 - f1_score: 0.9398 - val_loss: 0.1545 - val_f1_score: 0.9432\n",
            "Epoch 1902/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1616 - f1_score: 0.9398 - val_loss: 0.1545 - val_f1_score: 0.9432\n",
            "Epoch 1903/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1615 - f1_score: 0.9398 - val_loss: 0.1545 - val_f1_score: 0.9432\n",
            "Epoch 1904/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1615 - f1_score: 0.9398 - val_loss: 0.1544 - val_f1_score: 0.9432\n",
            "Epoch 1905/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1615 - f1_score: 0.9398 - val_loss: 0.1544 - val_f1_score: 0.9432\n",
            "Epoch 1906/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1614 - f1_score: 0.9398 - val_loss: 0.1544 - val_f1_score: 0.9432\n",
            "Epoch 1907/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1614 - f1_score: 0.9398 - val_loss: 0.1544 - val_f1_score: 0.9432\n",
            "Epoch 1908/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1614 - f1_score: 0.9398 - val_loss: 0.1543 - val_f1_score: 0.9432\n",
            "Epoch 1909/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1613 - f1_score: 0.9398 - val_loss: 0.1543 - val_f1_score: 0.9432\n",
            "Epoch 1910/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1613 - f1_score: 0.9398 - val_loss: 0.1543 - val_f1_score: 0.9432\n",
            "Epoch 1911/2000\n",
            "322/322 [==============================] - 3s 8ms/step - loss: 0.1613 - f1_score: 0.9398 - val_loss: 0.1542 - val_f1_score: 0.9432\n",
            "Epoch 1912/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1612 - f1_score: 0.9398 - val_loss: 0.1542 - val_f1_score: 0.9432\n",
            "Epoch 1913/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1612 - f1_score: 0.9398 - val_loss: 0.1542 - val_f1_score: 0.9431\n",
            "Epoch 1914/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1612 - f1_score: 0.9398 - val_loss: 0.1542 - val_f1_score: 0.9431\n",
            "Epoch 1915/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1611 - f1_score: 0.9398 - val_loss: 0.1541 - val_f1_score: 0.9431\n",
            "Epoch 1916/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1611 - f1_score: 0.9398 - val_loss: 0.1541 - val_f1_score: 0.9431\n",
            "Epoch 1917/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1611 - f1_score: 0.9398 - val_loss: 0.1541 - val_f1_score: 0.9431\n",
            "Epoch 1918/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1610 - f1_score: 0.9398 - val_loss: 0.1541 - val_f1_score: 0.9431\n",
            "Epoch 1919/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1610 - f1_score: 0.9398 - val_loss: 0.1540 - val_f1_score: 0.9431\n",
            "Epoch 1920/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1610 - f1_score: 0.9398 - val_loss: 0.1540 - val_f1_score: 0.9431\n",
            "Epoch 1921/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1609 - f1_score: 0.9398 - val_loss: 0.1540 - val_f1_score: 0.9430\n",
            "Epoch 1922/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1609 - f1_score: 0.9398 - val_loss: 0.1539 - val_f1_score: 0.9430\n",
            "Epoch 1923/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1609 - f1_score: 0.9398 - val_loss: 0.1539 - val_f1_score: 0.9430\n",
            "Epoch 1924/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1608 - f1_score: 0.9398 - val_loss: 0.1539 - val_f1_score: 0.9430\n",
            "Epoch 1925/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1608 - f1_score: 0.9398 - val_loss: 0.1539 - val_f1_score: 0.9430\n",
            "Epoch 1926/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1608 - f1_score: 0.9398 - val_loss: 0.1538 - val_f1_score: 0.9431\n",
            "Epoch 1927/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1607 - f1_score: 0.9398 - val_loss: 0.1538 - val_f1_score: 0.9431\n",
            "Epoch 1928/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1607 - f1_score: 0.9397 - val_loss: 0.1538 - val_f1_score: 0.9431\n",
            "Epoch 1929/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1607 - f1_score: 0.9397 - val_loss: 0.1538 - val_f1_score: 0.9431\n",
            "Epoch 1930/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1607 - f1_score: 0.9397 - val_loss: 0.1537 - val_f1_score: 0.9430\n",
            "Epoch 1931/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1606 - f1_score: 0.9397 - val_loss: 0.1537 - val_f1_score: 0.9430\n",
            "Epoch 1932/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1606 - f1_score: 0.9397 - val_loss: 0.1537 - val_f1_score: 0.9430\n",
            "Epoch 1933/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1606 - f1_score: 0.9397 - val_loss: 0.1536 - val_f1_score: 0.9430\n",
            "Epoch 1934/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1605 - f1_score: 0.9397 - val_loss: 0.1536 - val_f1_score: 0.9430\n",
            "Epoch 1935/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1605 - f1_score: 0.9397 - val_loss: 0.1536 - val_f1_score: 0.9430\n",
            "Epoch 1936/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1605 - f1_score: 0.9397 - val_loss: 0.1536 - val_f1_score: 0.9430\n",
            "Epoch 1937/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1604 - f1_score: 0.9397 - val_loss: 0.1535 - val_f1_score: 0.9430\n",
            "Epoch 1938/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1604 - f1_score: 0.9397 - val_loss: 0.1535 - val_f1_score: 0.9430\n",
            "Epoch 1939/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1604 - f1_score: 0.9397 - val_loss: 0.1535 - val_f1_score: 0.9430\n",
            "Epoch 1940/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1603 - f1_score: 0.9397 - val_loss: 0.1535 - val_f1_score: 0.9430\n",
            "Epoch 1941/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1603 - f1_score: 0.9397 - val_loss: 0.1534 - val_f1_score: 0.9430\n",
            "Epoch 1942/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1603 - f1_score: 0.9397 - val_loss: 0.1534 - val_f1_score: 0.9430\n",
            "Epoch 1943/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1602 - f1_score: 0.9397 - val_loss: 0.1534 - val_f1_score: 0.9430\n",
            "Epoch 1944/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1602 - f1_score: 0.9397 - val_loss: 0.1534 - val_f1_score: 0.9430\n",
            "Epoch 1945/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1602 - f1_score: 0.9397 - val_loss: 0.1533 - val_f1_score: 0.9430\n",
            "Epoch 1946/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1602 - f1_score: 0.9397 - val_loss: 0.1533 - val_f1_score: 0.9430\n",
            "Epoch 1947/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1601 - f1_score: 0.9397 - val_loss: 0.1533 - val_f1_score: 0.9430\n",
            "Epoch 1948/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1601 - f1_score: 0.9397 - val_loss: 0.1533 - val_f1_score: 0.9430\n",
            "Epoch 1949/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1601 - f1_score: 0.9397 - val_loss: 0.1532 - val_f1_score: 0.9430\n",
            "Epoch 1950/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1600 - f1_score: 0.9397 - val_loss: 0.1532 - val_f1_score: 0.9430\n",
            "Epoch 1951/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1600 - f1_score: 0.9397 - val_loss: 0.1532 - val_f1_score: 0.9430\n",
            "Epoch 1952/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1600 - f1_score: 0.9397 - val_loss: 0.1531 - val_f1_score: 0.9430\n",
            "Epoch 1953/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1599 - f1_score: 0.9397 - val_loss: 0.1531 - val_f1_score: 0.9430\n",
            "Epoch 1954/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1599 - f1_score: 0.9397 - val_loss: 0.1531 - val_f1_score: 0.9430\n",
            "Epoch 1955/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1599 - f1_score: 0.9397 - val_loss: 0.1531 - val_f1_score: 0.9430\n",
            "Epoch 1956/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1599 - f1_score: 0.9397 - val_loss: 0.1530 - val_f1_score: 0.9430\n",
            "Epoch 1957/2000\n",
            "322/322 [==============================] - 2s 6ms/step - loss: 0.1598 - f1_score: 0.9397 - val_loss: 0.1530 - val_f1_score: 0.9430\n",
            "Epoch 1958/2000\n",
            "322/322 [==============================] - 2s 7ms/step - loss: 0.1598 - f1_score: 0.9397 - val_loss: 0.1530 - val_f1_score: 0.9430\n",
            "Epoch 1959/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1598 - f1_score: 0.9397 - val_loss: 0.1530 - val_f1_score: 0.9430\n",
            "Epoch 1960/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1597 - f1_score: 0.9397 - val_loss: 0.1529 - val_f1_score: 0.9430\n",
            "Epoch 1961/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1597 - f1_score: 0.9397 - val_loss: 0.1529 - val_f1_score: 0.9430\n",
            "Epoch 1962/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1597 - f1_score: 0.9397 - val_loss: 0.1529 - val_f1_score: 0.9430\n",
            "Epoch 1963/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1596 - f1_score: 0.9397 - val_loss: 0.1529 - val_f1_score: 0.9430\n",
            "Epoch 1964/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1596 - f1_score: 0.9397 - val_loss: 0.1528 - val_f1_score: 0.9430\n",
            "Epoch 1965/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1596 - f1_score: 0.9397 - val_loss: 0.1528 - val_f1_score: 0.9430\n",
            "Epoch 1966/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1596 - f1_score: 0.9397 - val_loss: 0.1528 - val_f1_score: 0.9430\n",
            "Epoch 1967/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1595 - f1_score: 0.9397 - val_loss: 0.1528 - val_f1_score: 0.9430\n",
            "Epoch 1968/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1595 - f1_score: 0.9397 - val_loss: 0.1527 - val_f1_score: 0.9429\n",
            "Epoch 1969/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1595 - f1_score: 0.9397 - val_loss: 0.1527 - val_f1_score: 0.9429\n",
            "Epoch 1970/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1594 - f1_score: 0.9397 - val_loss: 0.1527 - val_f1_score: 0.9429\n",
            "Epoch 1971/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1594 - f1_score: 0.9397 - val_loss: 0.1527 - val_f1_score: 0.9429\n",
            "Epoch 1972/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1594 - f1_score: 0.9397 - val_loss: 0.1526 - val_f1_score: 0.9429\n",
            "Epoch 1973/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1593 - f1_score: 0.9397 - val_loss: 0.1526 - val_f1_score: 0.9429\n",
            "Epoch 1974/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1593 - f1_score: 0.9397 - val_loss: 0.1526 - val_f1_score: 0.9429\n",
            "Epoch 1975/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1593 - f1_score: 0.9397 - val_loss: 0.1525 - val_f1_score: 0.9429\n",
            "Epoch 1976/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1593 - f1_score: 0.9397 - val_loss: 0.1525 - val_f1_score: 0.9429\n",
            "Epoch 1977/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1592 - f1_score: 0.9397 - val_loss: 0.1525 - val_f1_score: 0.9429\n",
            "Epoch 1978/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1592 - f1_score: 0.9397 - val_loss: 0.1525 - val_f1_score: 0.9429\n",
            "Epoch 1979/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1592 - f1_score: 0.9397 - val_loss: 0.1524 - val_f1_score: 0.9429\n",
            "Epoch 1980/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1591 - f1_score: 0.9397 - val_loss: 0.1524 - val_f1_score: 0.9428\n",
            "Epoch 1981/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1591 - f1_score: 0.9397 - val_loss: 0.1524 - val_f1_score: 0.9428\n",
            "Epoch 1982/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1591 - f1_score: 0.9397 - val_loss: 0.1524 - val_f1_score: 0.9428\n",
            "Epoch 1983/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1591 - f1_score: 0.9397 - val_loss: 0.1523 - val_f1_score: 0.9428\n",
            "Epoch 1984/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1590 - f1_score: 0.9397 - val_loss: 0.1523 - val_f1_score: 0.9428\n",
            "Epoch 1985/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1590 - f1_score: 0.9397 - val_loss: 0.1523 - val_f1_score: 0.9428\n",
            "Epoch 1986/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1590 - f1_score: 0.9397 - val_loss: 0.1523 - val_f1_score: 0.9428\n",
            "Epoch 1987/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1589 - f1_score: 0.9397 - val_loss: 0.1522 - val_f1_score: 0.9428\n",
            "Epoch 1988/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1589 - f1_score: 0.9397 - val_loss: 0.1522 - val_f1_score: 0.9428\n",
            "Epoch 1989/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1589 - f1_score: 0.9397 - val_loss: 0.1522 - val_f1_score: 0.9428\n",
            "Epoch 1990/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1589 - f1_score: 0.9397 - val_loss: 0.1522 - val_f1_score: 0.9428\n",
            "Epoch 1991/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1588 - f1_score: 0.9397 - val_loss: 0.1521 - val_f1_score: 0.9428\n",
            "Epoch 1992/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1588 - f1_score: 0.9397 - val_loss: 0.1521 - val_f1_score: 0.9428\n",
            "Epoch 1993/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1588 - f1_score: 0.9397 - val_loss: 0.1521 - val_f1_score: 0.9428\n",
            "Epoch 1994/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1587 - f1_score: 0.9397 - val_loss: 0.1521 - val_f1_score: 0.9428\n",
            "Epoch 1995/2000\n",
            "322/322 [==============================] - 1s 4ms/step - loss: 0.1587 - f1_score: 0.9397 - val_loss: 0.1520 - val_f1_score: 0.9427\n",
            "Epoch 1996/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1587 - f1_score: 0.9397 - val_loss: 0.1520 - val_f1_score: 0.9427\n",
            "Epoch 1997/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1587 - f1_score: 0.9397 - val_loss: 0.1520 - val_f1_score: 0.9427\n",
            "Epoch 1998/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1586 - f1_score: 0.9397 - val_loss: 0.1519 - val_f1_score: 0.9427\n",
            "Epoch 1999/2000\n",
            "322/322 [==============================] - 1s 5ms/step - loss: 0.1586 - f1_score: 0.9397 - val_loss: 0.1519 - val_f1_score: 0.9427\n",
            "Epoch 2000/2000\n",
            "322/322 [==============================] - 2s 5ms/step - loss: 0.1586 - f1_score: 0.9397 - val_loss: 0.1519 - val_f1_score: 0.9427\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "mlp_pred = tf.argmax(model.predict(x_test), axis=-1)\n",
        "print(f1_score(y_test, mlp_pred))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d8VuIJEcfwvT",
        "outputId": "88bf1400-d1f1-4a5b-9551-7d0a0c9f72f2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.9116929367539406\n"
          ]
        }
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.0"
    },
    "colab": {
      "name": "Models.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "dCP33m3FXc3i",
        "TLFFSEHrXc3r",
        "pthifNLqXc3t",
        "qVYZPpPMXc33",
        "fCxLDLc3Xc39"
      ]
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}