split-up.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "ezzlit intro.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "gpuClass": "standard",
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q2XYqaSMRU9G",
        "outputId": "2b344309-5efa-4490-ab23-b17d0f638faf"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Wed Jun 29 09:04:34 2022       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 460.32.03    Driver Version: 460.32.03    CUDA Version: 11.2     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla P100-PCIE...  Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   34C    P0    25W / 250W |      0MiB / 16280MiB |      0%      Default |\n",
            "|                               |                      |                  N/A |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ]
        }
      ],
      "source": [
        "!nvidia-smi"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!nvidia-smi -L"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qrhGgVLxR-yt",
        "outputId": "950433ce-950c-4d02-e51f-f954f6087926"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "GPU 0: Tesla P100-PCIE-16GB (UUID: GPU-4f91e494-b31e-1fc3-0206-19f33bdeb100)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://github.com/Franck-Dernoncourt/pubmed-rct.git"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uo_EWIYuSqoV",
        "outputId": "b820cb56-50ab-4385-83f4-7046cec25afb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'pubmed-rct'...\n",
            "remote: Enumerating objects: 33, done.\u001b[K\n",
            "remote: Counting objects: 100% (8/8), done.\u001b[K\n",
            "remote: Compressing objects: 100% (3/3), done.\u001b[K\n",
            "remote: Total 33 (delta 5), reused 5 (delta 5), pack-reused 25\u001b[K\n",
            "Unpacking objects: 100% (33/33), done.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!ls pubmed-rct"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-QwcPbb7TFmx",
        "outputId": "6d846fee-8fcc-477f-b10d-b9bbc5c47e3c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "PubMed_200k_RCT\n",
            "PubMed_200k_RCT_numbers_replaced_with_at_sign\n",
            "PubMed_20k_RCT\n",
            "PubMed_20k_RCT_numbers_replaced_with_at_sign\n",
            "README.md\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!ls pubmed-rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/"
      ],
      "metadata": {
        "id": "fq6PaXXcTiIq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ec16c9f7-b615-4b99-f8e0-dc2fd3f8a51c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "dev.txt  test.txt  train.txt\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_dir = \"pubmed-rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/\""
      ],
      "metadata": {
        "id": "cbz0QG5kgkGv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def get_lines(filename):\n",
        "  with open(filename, \"r\") as f:\n",
        "    return f.readlines()"
      ],
      "metadata": {
        "id": "Eg2zR6wahDck"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_lines = get_lines(data_dir+\"train.txt\")\n",
        "for i in train_lines[:20]:\n",
        "  print(i, end=\"\\n\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9O3OkGOthV-V",
        "outputId": "a8ada3fb-8229-4bbc-bf2c-b32ddfe66215"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "###24293578\n",
            "\n",
            "OBJECTIVE\tTo investigate the efficacy of @ weeks of daily low-dose oral prednisolone in improving pain , mobility , and systemic low-grade inflammation in the short term and whether the effect would be sustained at @ weeks in older adults with moderate to severe knee osteoarthritis ( OA ) .\n",
            "\n",
            "METHODS\tA total of @ patients with primary knee OA were randomized @:@ ; @ received @ mg/day of prednisolone and @ received placebo for @ weeks .\n",
            "\n",
            "METHODS\tOutcome measures included pain reduction and improvement in function scores and systemic inflammation markers .\n",
            "\n",
            "METHODS\tPain was assessed using the visual analog pain scale ( @-@ mm ) .\n",
            "\n",
            "METHODS\tSecondary outcome measures included the Western Ontario and McMaster Universities Osteoarthritis Index scores , patient global assessment ( PGA ) of the severity of knee OA , and @-min walk distance ( @MWD ) .\n",
            "\n",
            "METHODS\tSerum levels of interleukin @ ( IL-@ ) , IL-@ , tumor necrosis factor ( TNF ) - , and high-sensitivity C-reactive protein ( hsCRP ) were measured .\n",
            "\n",
            "RESULTS\tThere was a clinically relevant reduction in the intervention group compared to the placebo group for knee pain , physical function , PGA , and @MWD at @ weeks .\n",
            "\n",
            "RESULTS\tThe mean difference between treatment arms ( @ % CI ) was @ ( @-@ @ ) , p < @ ; @ ( @-@ @ ) , p < @ ; @ ( @-@ @ ) , p < @ ; and @ ( @-@ @ ) , p < @ , respectively .\n",
            "\n",
            "RESULTS\tFurther , there was a clinically relevant reduction in the serum levels of IL-@ , IL-@ , TNF - , and hsCRP at @ weeks in the intervention group when compared to the placebo group .\n",
            "\n",
            "RESULTS\tThese differences remained significant at @ weeks .\n",
            "\n",
            "RESULTS\tThe Outcome Measures in Rheumatology Clinical Trials-Osteoarthritis Research Society International responder rate was @ % in the intervention group and @ % in the placebo group ( p < @ ) .\n",
            "\n",
            "CONCLUSIONS\tLow-dose oral prednisolone had both a short-term and a longer sustained effect resulting in less knee pain , better physical function , and attenuation of systemic inflammation in older patients with knee OA ( ClinicalTrials.gov identifier NCT@ ) .\n",
            "\n",
            "\n",
            "\n",
            "###24854809\n",
            "\n",
            "BACKGROUND\tEmotional eating is associated with overeating and the development of obesity .\n",
            "\n",
            "BACKGROUND\tYet , empirical evidence for individual ( trait ) differences in emotional eating and cognitive mechanisms that contribute to eating during sad mood remain equivocal .\n",
            "\n",
            "OBJECTIVE\tThe aim of this study was to test if attention bias for food moderates the effect of self-reported emotional eating during sad mood ( vs neutral mood ) on actual food intake .\n",
            "\n",
            "OBJECTIVE\tIt was expected that emotional eating is predictive of elevated attention for food and higher food intake after an experimentally induced sad mood and that attentional maintenance on food predicts food intake during a sad versus a neutral mood .\n",
            "\n",
            "METHODS\tParticipants ( N = @ ) were randomly assigned to one of the two experimental mood induction conditions ( sad/neutral ) .\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_lines[:20]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "F_BkP1W_noCU",
        "outputId": "d657d106-7379-4456-f34f-0918e599a701"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['###24293578\\n',\n",
              " 'OBJECTIVE\\tTo investigate the efficacy of @ weeks of daily low-dose oral prednisolone in improving pain , mobility , and systemic low-grade inflammation in the short term and whether the effect would be sustained at @ weeks in older adults with moderate to severe knee osteoarthritis ( OA ) .\\n',\n",
              " 'METHODS\\tA total of @ patients with primary knee OA were randomized @:@ ; @ received @ mg/day of prednisolone and @ received placebo for @ weeks .\\n',\n",
              " 'METHODS\\tOutcome measures included pain reduction and improvement in function scores and systemic inflammation markers .\\n',\n",
              " 'METHODS\\tPain was assessed using the visual analog pain scale ( @-@ mm ) .\\n',\n",
              " 'METHODS\\tSecondary outcome measures included the Western Ontario and McMaster Universities Osteoarthritis Index scores , patient global assessment ( PGA ) of the severity of knee OA , and @-min walk distance ( @MWD ) .\\n',\n",
              " 'METHODS\\tSerum levels of interleukin @ ( IL-@ ) , IL-@ , tumor necrosis factor ( TNF ) - , and high-sensitivity C-reactive protein ( hsCRP ) were measured .\\n',\n",
              " 'RESULTS\\tThere was a clinically relevant reduction in the intervention group compared to the placebo group for knee pain , physical function , PGA , and @MWD at @ weeks .\\n',\n",
              " 'RESULTS\\tThe mean difference between treatment arms ( @ % CI ) was @ ( @-@ @ ) , p < @ ; @ ( @-@ @ ) , p < @ ; @ ( @-@ @ ) , p < @ ; and @ ( @-@ @ ) , p < @ , respectively .\\n',\n",
              " 'RESULTS\\tFurther , there was a clinically relevant reduction in the serum levels of IL-@ , IL-@ , TNF - , and hsCRP at @ weeks in the intervention group when compared to the placebo group .\\n',\n",
              " 'RESULTS\\tThese differences remained significant at @ weeks .\\n',\n",
              " 'RESULTS\\tThe Outcome Measures in Rheumatology Clinical Trials-Osteoarthritis Research Society International responder rate was @ % in the intervention group and @ % in the placebo group ( p < @ ) .\\n',\n",
              " 'CONCLUSIONS\\tLow-dose oral prednisolone had both a short-term and a longer sustained effect resulting in less knee pain , better physical function , and attenuation of systemic inflammation in older patients with knee OA ( ClinicalTrials.gov identifier NCT@ ) .\\n',\n",
              " '\\n',\n",
              " '###24854809\\n',\n",
              " 'BACKGROUND\\tEmotional eating is associated with overeating and the development of obesity .\\n',\n",
              " 'BACKGROUND\\tYet , empirical evidence for individual ( trait ) differences in emotional eating and cognitive mechanisms that contribute to eating during sad mood remain equivocal .\\n',\n",
              " 'OBJECTIVE\\tThe aim of this study was to test if attention bias for food moderates the effect of self-reported emotional eating during sad mood ( vs neutral mood ) on actual food intake .\\n',\n",
              " 'OBJECTIVE\\tIt was expected that emotional eating is predictive of elevated attention for food and higher food intake after an experimentally induced sad mood and that attentional maintenance on food predicts food intake during a sad versus a neutral mood .\\n',\n",
              " 'METHODS\\tParticipants ( N = @ ) were randomly assigned to one of the two experimental mood induction conditions ( sad/neutral ) .\\n']"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def preprocess_text_with_line_numbers(filename):\n",
        "  input_lines = get_lines(filename)\n",
        "  abstract_lines = \"\"\n",
        "  abstract_samples = []\n",
        "\n",
        "  for line in input_lines:\n",
        "    if line.startswith(\"###\"):\n",
        "      abstract_id = line\n",
        "      abstract_lines = \"\"\n",
        "    elif line.isspace():\n",
        "      abstract_line_split = abstract_lines.splitlines()\n",
        "\n",
        "      for abstract_line_number, abstract_line in enumerate(abstract_line_split):\n",
        "        line_data = {}\n",
        "        target_text_split = abstract_line.split(\"\\t\")\n",
        "        line_data[\"target\"] = target_text_split[0]\n",
        "        line_data[\"text\"] = target_text_split[1].lower()\n",
        "        line_data[\"line_number\"] = abstract_line_number\n",
        "        line_data[\"total_lines\"] = len(abstract_line_split) - 1\n",
        "        abstract_samples.append(line_data)\n",
        "    else:\n",
        "      abstract_lines+=line\n",
        "  return abstract_samples\n"
      ],
      "metadata": {
        "id": "BRdxCBm3hdym"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_samples = preprocess_text_with_line_numbers(data_dir + \"train.txt\")\n",
        "val_samples = preprocess_text_with_line_numbers(data_dir + \"dev.txt\") \n",
        "test_samples = preprocess_text_with_line_numbers(data_dir + \"test.txt\")\n",
        "len(train_samples), len(val_samples), len(test_samples)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yup491N7pghp",
        "outputId": "f21b3fc3-bcf3-441a-b179-f27985b87e8a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(180040, 30212, 30135)"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_samples[:12]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NwbBuR80pr8h",
        "outputId": "812fdf81-3d24-47d2-d0df-56b7598ef1b0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[{'line_number': 0,\n",
              "  'target': 'OBJECTIVE',\n",
              "  'text': 'to investigate the efficacy of @ weeks of daily low-dose oral prednisolone in improving pain , mobility , and systemic low-grade inflammation in the short term and whether the effect would be sustained at @ weeks in older adults with moderate to severe knee osteoarthritis ( oa ) .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 1,\n",
              "  'target': 'METHODS',\n",
              "  'text': 'a total of @ patients with primary knee oa were randomized @:@ ; @ received @ mg/day of prednisolone and @ received placebo for @ weeks .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 2,\n",
              "  'target': 'METHODS',\n",
              "  'text': 'outcome measures included pain reduction and improvement in function scores and systemic inflammation markers .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 3,\n",
              "  'target': 'METHODS',\n",
              "  'text': 'pain was assessed using the visual analog pain scale ( @-@ mm ) .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 4,\n",
              "  'target': 'METHODS',\n",
              "  'text': 'secondary outcome measures included the western ontario and mcmaster universities osteoarthritis index scores , patient global assessment ( pga ) of the severity of knee oa , and @-min walk distance ( @mwd ) .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 5,\n",
              "  'target': 'METHODS',\n",
              "  'text': 'serum levels of interleukin @ ( il-@ ) , il-@ , tumor necrosis factor ( tnf ) - , and high-sensitivity c-reactive protein ( hscrp ) were measured .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 6,\n",
              "  'target': 'RESULTS',\n",
              "  'text': 'there was a clinically relevant reduction in the intervention group compared to the placebo group for knee pain , physical function , pga , and @mwd at @ weeks .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 7,\n",
              "  'target': 'RESULTS',\n",
              "  'text': 'the mean difference between treatment arms ( @ % ci ) was @ ( @-@ @ ) , p < @ ; @ ( @-@ @ ) , p < @ ; @ ( @-@ @ ) , p < @ ; and @ ( @-@ @ ) , p < @ , respectively .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 8,\n",
              "  'target': 'RESULTS',\n",
              "  'text': 'further , there was a clinically relevant reduction in the serum levels of il-@ , il-@ , tnf - , and hscrp at @ weeks in the intervention group when compared to the placebo group .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 9,\n",
              "  'target': 'RESULTS',\n",
              "  'text': 'these differences remained significant at @ weeks .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 10,\n",
              "  'target': 'RESULTS',\n",
              "  'text': 'the outcome measures in rheumatology clinical trials-osteoarthritis research society international responder rate was @ % in the intervention group and @ % in the placebo group ( p < @ ) .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 11,\n",
              "  'target': 'CONCLUSIONS',\n",
              "  'text': 'low-dose oral prednisolone had both a short-term and a longer sustained effect resulting in less knee pain , better physical function , and attenuation of systemic inflammation in older patients with knee oa ( clinicaltrials.gov identifier nct@ ) .',\n",
              "  'total_lines': 11}]"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "train_df = pd.DataFrame(train_samples)\n",
        "val_df = pd.DataFrame(val_samples)\n",
        "test_df = pd.DataFrame(test_samples)\n",
        "train_df.head(15)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 824
        },
        "id": "jkDogmRMp3oP",
        "outputId": "b4fdc3ed-dba6-451c-a0a8-4f70414b14c3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         target                                               text  \\\n",
              "0     OBJECTIVE  to investigate the efficacy of @ weeks of dail...   \n",
              "1       METHODS  a total of @ patients with primary knee oa wer...   \n",
              "2       METHODS  outcome measures included pain reduction and i...   \n",
              "3       METHODS  pain was assessed using the visual analog pain...   \n",
              "4       METHODS  secondary outcome measures included the wester...   \n",
              "5       METHODS  serum levels of interleukin @ ( il-@ ) , il-@ ...   \n",
              "6       RESULTS  there was a clinically relevant reduction in t...   \n",
              "7       RESULTS  the mean difference between treatment arms ( @...   \n",
              "8       RESULTS  further , there was a clinically relevant redu...   \n",
              "9       RESULTS  these differences remained significant at @ we...   \n",
              "10      RESULTS  the outcome measures in rheumatology clinical ...   \n",
              "11  CONCLUSIONS  low-dose oral prednisolone had both a short-te...   \n",
              "12   BACKGROUND  emotional eating is associated with overeating...   \n",
              "13   BACKGROUND  yet , empirical evidence for individual ( trai...   \n",
              "14    OBJECTIVE  the aim of this study was to test if attention...   \n",
              "\n",
              "    line_number  total_lines  \n",
              "0             0           11  \n",
              "1             1           11  \n",
              "2             2           11  \n",
              "3             3           11  \n",
              "4             4           11  \n",
              "5             5           11  \n",
              "6             6           11  \n",
              "7             7           11  \n",
              "8             8           11  \n",
              "9             9           11  \n",
              "10           10           11  \n",
              "11           11           11  \n",
              "12            0           10  \n",
              "13            1           10  \n",
              "14            2           10  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-34d6210a-acd8-4263-8014-c29237f31c38\">\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>target</th>\n",
              "      <th>text</th>\n",
              "      <th>line_number</th>\n",
              "      <th>total_lines</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>OBJECTIVE</td>\n",
              "      <td>to investigate the efficacy of @ weeks of dail...</td>\n",
              "      <td>0</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>a total of @ patients with primary knee oa wer...</td>\n",
              "      <td>1</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>outcome measures included pain reduction and i...</td>\n",
              "      <td>2</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>pain was assessed using the visual analog pain...</td>\n",
              "      <td>3</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>secondary outcome measures included the wester...</td>\n",
              "      <td>4</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>serum levels of interleukin @ ( il-@ ) , il-@ ...</td>\n",
              "      <td>5</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>RESULTS</td>\n",
              "      <td>there was a clinically relevant reduction in t...</td>\n",
              "      <td>6</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>RESULTS</td>\n",
              "      <td>the mean difference between treatment arms ( @...</td>\n",
              "      <td>7</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>RESULTS</td>\n",
              "      <td>further , there was a clinically relevant redu...</td>\n",
              "      <td>8</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>RESULTS</td>\n",
              "      <td>these differences remained significant at @ we...</td>\n",
              "      <td>9</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>RESULTS</td>\n",
              "      <td>the outcome measures in rheumatology clinical ...</td>\n",
              "      <td>10</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>CONCLUSIONS</td>\n",
              "      <td>low-dose oral prednisolone had both a short-te...</td>\n",
              "      <td>11</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>BACKGROUND</td>\n",
              "      <td>emotional eating is associated with overeating...</td>\n",
              "      <td>0</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>BACKGROUND</td>\n",
              "      <td>yet , empirical evidence for individual ( trai...</td>\n",
              "      <td>1</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>OBJECTIVE</td>\n",
              "      <td>the aim of this study was to test if attention...</td>\n",
              "      <td>2</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-34d6210a-acd8-4263-8014-c29237f31c38')\"\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-34d6210a-acd8-4263-8014-c29237f31c38 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-34d6210a-acd8-4263-8014-c29237f31c38');\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",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.target.value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fz1N-6P4rbMf",
        "outputId": "52cc1e62-b1d4-4898-9419-6d532cd2fd6a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "METHODS        59353\n",
              "RESULTS        57953\n",
              "CONCLUSIONS    27168\n",
              "BACKGROUND     21727\n",
              "OBJECTIVE      13839\n",
              "Name: target, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.total_lines.plot.hist(figsize=(5,3.5))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 257
        },
        "id": "VqstvUCqsJcQ",
        "outputId": "7f8c0e50-2485-4741-a592-6cf391d368ed"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f2ffd803dd0>"
            ]
          },
          "metadata": {},
          "execution_count": 15
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 360x252 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVkAAADfCAYAAABcbKCqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAWrElEQVR4nO3df5Bd5X3f8ffHAgImxpLCRqUSRCTR2CHExmIN8sRNCYyFgMQirUugcbRlGOQZ5I49bqcIT6YiEDpypzGxEoeGGBXJsS3L+AdqkKOsFTtpZiLQCmTEDzNaY1GtLCTFAmSMAwV/+sd5Fm7E3dWVtM/u3eXzmrlzz/me55zzvYfRl7PPec45sk1ERNTxpolOICJiKkuRjYioKEU2IqKiFNmIiIpSZCMiKkqRjYioqFqRlfQ2SdtbPockfVTSTEn9knaW7xmlvSStkjQo6WFJ81u21Vfa75TU1xI/X9KOss4qSar1eyIijkW1Imv7Cdvn2T4POB94AfgqsBzYbHsesLnMA1wGzCufpcAdAJJmAiuAC4ELgBXDhbm0ub5lvUW1fk9ExLE4YZz2cwnwXdtPSVoMXFTia4BvATcCi4G1bu6O2CJpuqQzStt+2wcBJPUDiyR9CzjN9pYSXwtcCXx9tEROP/10z507d0x/XETEtm3b/tF2z+Hx8SqyVwNfKNOzbO8t008Ds8r0bGB3yzpDJTZafKhNfFRz585lYGDgaPOPiBiVpKfaxatf+JJ0EvB+4EuHLytnrdXv65W0VNKApIEDBw7U3l1ExKvGY3TBZcCDtveV+X2lG4Dyvb/E9wBntqw3p8RGi89pE38d23fa7rXd29PzurP5iIhqxqPIXsNrXQUAG4DhEQJ9wL0t8SVllMEC4LnSrbAJWChpRrngtRDYVJYdkrSgjCpY0rKtiIiuULVPVtKpwPuAD7WEVwLrJV0HPAVcVeIbgcuBQZqRCNcC2D4o6VZga2l3y/BFMOAG4G7gFJoLXqNe9IqIGG96oz3qsLe317nwFRFjTdI2272Hx3PHV0RERSmyEREVjdc42egCc5ffV30fu1ZeUX0fEZNJzmQjIipKkY2IqChFNiKiohTZiIiKUmQjIipKkY2IqChFNiKiohTZiIiKUmQjIipKkY2IqChFNiKiohTZiIiKUmQjIipKkY2IqChFNiKioqpFVtJ0SfdI+o6kxyW9R9JMSf2SdpbvGaWtJK2SNCjpYUnzW7bTV9rvlNTXEj9f0o6yzqryQsWIiK5R+0z2U8Bf2X478E7gcWA5sNn2PGBzmYfm1eHzymcpcAeApJnACuBC4AJgxXBhLm2ub1lvUeXfExFxVKoVWUlvBX4NuAvA9ku2nwUWA2tKszXAlWV6MbDWjS3AdElnAJcC/bYP2n4G6AcWlWWn2d7i5m2Qa1u2FRHRFWqeyZ4NHAD+l6SHJH2mvCJ8lu29pc3TwKwyPRvY3bL+UImNFh9qE4+I6Bo1i+wJwHzgDtvvAn7Ea10DAJQz0OrvJJe0VNKApIEDBw7U3l1ExKtqFtkhYMj2/WX+Hpqiu6/8qU/53l+W7wHObFl/TomNFp/TJv46tu+03Wu7t6en57h+VETE0ahWZG0/DeyW9LYSugR4DNgADI8Q6APuLdMbgCVllMEC4LnSrbAJWChpRrngtRDYVJYdkrSgjCpY0rKtiIiuUPuV4P8R+Jykk4AngWtpCvt6SdcBTwFXlbYbgcuBQeCF0hbbByXdCmwt7W6xfbBM3wDcDZwCfL18IiK6RtUia3s70Ntm0SVt2hpYNsJ2VgOr28QHgHOPM82IiGpyx1dEREUpshERFaXIRkRUlCIbEVFRimxEREUpshERFaXIRkRUlCIbEVFRimxEREUpshERFaXIRkRUlCIbEVFRimxEREUpshERFaXIRkRUlCIbEVFRimxEREUpshERFVUtspJ2SdohabukgRKbKalf0s7yPaPEJWmVpEFJD0ua37KdvtJ+p6S+lvj5ZfuDZV3V/D0REUdrPM5kf932ebaH3/W1HNhsex6wucwDXAbMK5+lwB3QFGVgBXAhcAGwYrgwlzbXt6y3qP7PiYjo3ER0FywG1pTpNcCVLfG1bmwBpks6A7gU6Ld90PYzQD+wqCw7zfaW8hLGtS3biojoCrWLrIG/lrRN0tISm2V7b5l+GphVpmcDu1vWHSqx0eJDbeIREV2j6ivBgffa3iPpZ4F+Sd9pXWjbklw5B0qBXwpw1lln1d5dRMSrqp7J2t5TvvcDX6XpU91X/tSnfO8vzfcAZ7asPqfERovPaRNvl8edtntt9/b09Bzvz4qI6Fi1IivpVElvGZ4GFgKPABuA4RECfcC9ZXoDsKSMMlgAPFe6FTYBCyXNKBe8FgKbyrJDkhaUUQVLWrYVEdEVanYXzAK+WkZVnQB83vZfSdoKrJd0HfAUcFVpvxG4HBgEXgCuBbB9UNKtwNbS7hbbB8v0DcDdwCnA18snIqJrVCuytp8E3tkm/gPgkjZxA8tG2NZqYHWb+ABw7nEnGxFRSe74ioioKEU2IqKiFNmIiIpSZCMiKkqRjYioKEU2IqKiFNmIiIpSZCMiKkqRjYioKEU2IqKijoqspF+pnUhExFTU6Znsn0p6QNINkt5aNaOIiCmkoyJr+18Bv0PzXNdtkj4v6X1VM4uImAI67pO1vRP4PeBG4F8DqyR9R9K/qZVcRMRk12mf7Dsk3Q48DlwM/KbtXyrTt1fMLyJiUuv0ebJ/DHwG+LjtHw8HbX9f0u9VySwiYgrotMheAfzY9isAkt4EnGz7BdufrZZdRMQk12mf7DdoXvEy7M0lFhERo+i0yJ5s+/nhmTL95k5WlDRN0kOS/rLMny3pfkmDkr4o6aQS/6kyP1iWz23Zxk0l/oSkS1vii0psUNLyDn9LRMS46bTI/kjS/OEZSecDPx6lfauP0FwwG/YJ4Hbbvwg8A1xX4tcBz5T47aUdks4BrgZ+GVhEM2Z3mqRpwKeBy4BzgGtK24iIrtFpn+xHgS9J+j4g4F8Av32klSTNoenPvQ34WHl198XAvy9N1gA3A3cAi8s0wD3An5T2i4F1tl8EvidpELigtBssL2xE0rrS9rEOf1NUMHf5feOyn10rrxiX/UQcr46KrO2tkt4OvK2EnrD9/zpY9Y+A/wK8pcz/DPCs7ZfL/BAwu0zPBnaX/b0s6bnSfjawpWWbrevsPix+YbskJC0FlgKcddZZHaQdETE2juYBMe8G3gHMp/nTfMlojSX9BrDf9rbjyG9M2L7Tdq/t3p6enolOJyLeQDo6k5X0WeAXgO3AKyVsYO0oq/0q8H5JlwMnA6cBnwKmSzqhnM3OAfaU9ntobtsdknQC8FbgBy3xYa3rjBSPiOgKnfbJ9gLn2HanG7Z9E3ATgKSLgP9s+3ckfQn4ALAO6APuLatsKPP/UJb/jW1L2gB8XtIngX8JzAMeoOkbnifpbJriejWv9fVGRHSFTovsIzQXu/aOwT5vBNZJ+gPgIeCuEr8L+Gy5sHWQpmhi+1FJ62kuaL0MLGu5KeLDwCZgGrDa9qNjkF9ExJjptMieDjwm6QHgxeGg7fd3srLtbwHfKtNP8trogNY2/wT8uxHWv41mhMLh8Y3Axk5yiIiYCJ0W2ZtrJhERMVV1OoTrbyX9HDDP9jckvZnmT/SIiBhFp486vJ7mBoE/K6HZwNdqJRURMVV0Ok52Gc2QrEPw6gO8f7ZWUhERU0WnRfZF2y8Nz5RxrB0P54qIeKPqtMj+raSPA6eUd3t9Cfjf9dKKiJgaOi2yy4EDwA7gQzTDpvJGhIiII+h0dMFPgD8vn4iI6FCnzy74Hm36YG3//JhnFBExhRzNswuGnUxzZ9bMsU8nImJq6ahP1vYPWj57bP8RzcO4IyJiFJ12F8xvmX0TzZltp2fBERFvWJ0Wyj9smX4Z2AVcNebZRERMMZ2OLvj12olERExFnXYXfGy05bY/OTbpRHRmPF7YmJc1xlg4mtEF76Z5ewHAb9K8nWBnjaQiIqaKTovsHGC+7R8CSLoZuM/2B2slFhExFXR6W+0s4KWW+ZdKLCIiRtFpkV0LPCDp5nIWez+wZrQVJJ0s6QFJ35b0qKTfL/GzJd0vaVDSFyWdVOI/VeYHy/K5Ldu6qcSfkHRpS3xRiQ1KWn5UvzwiYhx0ejPCbcC1wDPlc63t/3aE1V4ELrb9TuA8YJGkBcAngNtt/2LZ1nWl/XXAMyV+e2mHpHNoXqr4y8Ai4E8lTZM0Dfg0cBlwDnBNaRsR0TU6PZMFeDNwyPangKHyKu4RufF8mT2xfAxcTPOWBWjOhq8s04t57ez4HuASSSrxdbZftP09YJDmRYwXAIO2nyzPul1X2kZEdI1OXz+zguZV3jeV0InAX3Sw3jRJ24H9QD/wXeBZ2y+XJkM0r7KhfO8GKMufA36mNX7YOiPF2+WxVNKApIEDBw4cKe2IiDHT6ZnsbwHvB34EYPv7wFuOtJLtV2yfRzM64QLg7ceY53GxfaftXtu9PT09E5FCRLxBdVpkX7JtyuMOJZ16NDux/SzwTeA9wPTy+hpoiu+eMr0HOLNs/wTgrcAPWuOHrTNSPCKia3RaZNdL+jOaAnk98A2O8ABvST2SppfpU4D3AY/TFNsPlGZ9wL1lekOZpyz/m1LYNwBXl9EHZwPzaG6E2ArMK6MVTqK5ODZ8s0RERFc44s0I5eLTF2n+1D8EvA34r7b7j7DqGcCaMgrgTcB6238p6TFgnaQ/AB4C7irt7wI+K2kQOEhTNLH9qKT1wGM0D6dZZvuVktuHgU3ANGC17Uc7/+kREfUdscjatqSNtn+F5uJVR2w/DLyrTfxJmv7Zw+P/RPMw8Hbbug24rU18I837xiIiulKn3QUPSnp31UwiIqagTp9dcCHwQUm7aEYYiOYk9x21EouImApGLbKSzrL9f4FLR2sXERHtHelM9ms0T996StKXbf/b8UgqImKqOFKfrFqm8/rviIijdKQi6xGmIyKiA0fqLninpEM0Z7SnlGl47cLXaVWzi4iY5EYtsranjVciERFT0dE86jAiIo5SimxEREUpshERFaXIRkRUlCIbEVFRp88uiMrmLr9volOIiApyJhsRUVGKbERERSmyEREVVSuyks6U9E1Jj0l6VNJHSnympH5JO8v3jBKXpFWSBiU9LGl+y7b6Svudkvpa4udL2lHWWVVelRMR0TVqnsm+DPwn2+cAC4Blks4BlgObbc8DNpd5gMtoXpI4D1gK3AFNUQZW0Dw4/AJgxXBhLm2ub1lvUcXfExFx1KoVWdt7bT9Ypn9I86ba2cBiYE1ptga4skwvBta6sYXmzbhn0DwwvN/2QdvP0LxnbFFZdprtLeWttmtbthUR0RXGpU9W0lyalyreD8yyvbcsehqYVaZnA7tbVhsqsdHiQ23iERFdo3qRlfTTwJeBj9o+1LqsnIFWf06tpKWSBiQNHDhwoPbuIiJeVbXISjqRpsB+zvZXSnhf+VOf8r2/xPcAZ7asPqfERovPaRN/Hdt32u613dvT03N8Pyoi4ijUHF0g4C7gcdufbFm0ARgeIdAH3NsSX1JGGSwAnivdCpuAhZJmlAteC4FNZdkhSQvKvpa0bCsioivUvK32V4HfBXZI2l5iHwdWAuslXQc8BVxVlm0ELgcGgReAawFsH5R0K7C1tLvF9sEyfQNwN3AK8PXyiYjoGtWKrO2/55+/iLHVJW3aG1g2wrZWA6vbxAeAc48jzYiIqnLHV0RERSmyEREVpchGRFSUIhsRUVGKbERERSmyEREVpchGRFSUIhsRUVGKbERERSmyEREVpchGRFSUIhsRUVGKbERERSmyEREVpchGRFSUIhsRUVGKbERERSmyEREV1XyR4mpJ+yU90hKbKalf0s7yPaPEJWmVpEFJD0ua37JOX2m/U1JfS/x8STvKOqvKyxQjIrpKzRcp3g38CbC2JbYc2Gx7paTlZf5G4DJgXvlcCNwBXChpJrAC6AUMbJO0wfYzpc31wP00L2FcRF6kGGNo7vL7qu9j18orqu8jJla1M1nbfwccPCy8GFhTptcAV7bE17qxBZgu6QzgUqDf9sFSWPuBRWXZaba3lBcwrm3ZVkRE1xjvPtlZtveW6aeBWWV6NrC7pd1QiY0WH2oTj4joKhN24aucgXo89iVpqaQBSQMHDhwYj11GRADjX2T3lT/1Kd/7S3wPcGZLuzklNlp8Tpt4W7bvtN1ru7enp+e4f0RERKfGu8huAIZHCPQB97bEl5RRBguA50q3wiZgoaQZZSTCQmBTWXZI0oIyqmBJy7YiIrpGtdEFkr4AXAScLmmIZpTASmC9pOuAp4CrSvONwOXAIPACcC2A7YOSbgW2lna32B6+mHYDzQiGU2hGFWRkQUR0nWpF1vY1Iyy6pE1bA8tG2M5qYHWb+ABw7vHkGBFRW+74ioioKEU2IqKiFNmIiIpSZCMiKkqRjYioKEU2IqKiFNmIiIpSZCMiKkqRjYioKEU2IqKiFNmIiIpSZCMiKkqRjYioKEU2IqKiFNmIiIpSZCMiKqr20O6IOLK5y+8bl/3sWnnFuOwnXi9nshERFU36M1lJi4BPAdOAz9heOZbbH68zjYiYmiZ1kZU0Dfg08D5gCNgqaYPtxyY2s4juMh4nC+mSaG+ydxdcAAzaftL2S8A6YPEE5xQR8apJfSYLzAZ2t8wPARdOUC4Rb2i5iNfeZC+yHZG0FFhaZp+X9MQEpXI68I8TtO/jldwnRnI/jD4x1lts61hy/7l2wcleZPcAZ7bMzymxf8b2ncCd45XUSCQN2O6d6DyORXKfGMl9Yoxl7pO9T3YrME/S2ZJOAq4GNkxwThERr5rUZ7K2X5b0YWATzRCu1bYfneC0IiJeNamLLIDtjcDGic6jQxPeZXEckvvESO4TY8xyl+2x2lZERBxmsvfJRkR0tRTZcSJpl6QdkrZLGpjofEYjabWk/ZIeaYnNlNQvaWf5njGROY5khNxvlrSnHPvtki6fyBzbkXSmpG9KekzSo5I+UuJdf9xHyX0yHPeTJT0g6dsl998v8bMl3S9pUNIXy4X1Y9tHugvGh6RdQK/trh/zKOnXgOeBtbbPLbH/Dhy0vVLScmCG7RsnMs92Rsj9ZuB52/9jInMbjaQzgDNsPyjpLcA24ErgP9Dlx32U3K+i+4+7gFNtPy/pRODvgY8AHwO+YnudpP8JfNv2Hceyj5zJxuvY/jvg4GHhxcCaMr2G5h9R1xkh965ne6/tB8v0D4HHae5o7PrjPkruXc+N58vsieVj4GLgnhI/ruOeIjt+DPy1pG3lDrTJZpbtvWX6aWDWRCZzDD4s6eHSndB1f3K3kjQXeBdwP5PsuB+WO0yC4y5pmqTtwH6gH/gu8Kztl0uTIY7jfxopsuPnvbbnA5cBy8qftZOSmz6mydTPdAfwC8B5wF7gDyc2nZFJ+mngy8BHbR9qXdbtx71N7pPiuNt+xfZ5NHeMXgC8fSy3nyI7TmzvKd/7ga/S/MecTPaVvrfhPrj9E5xPx2zvK/+QfgL8OV167Euf4JeBz9n+SglPiuPeLvfJctyH2X4W+CbwHmC6pOH7CNrert+pFNlxIOnUckEASacCC4FHRl+r62wA+sp0H3DvBOZyVIaLVPFbdOGxLxdg7gIet/3JlkVdf9xHyn2SHPceSdPL9Ck0z6Z+nKbYfqA0O67jntEF40DSz9OcvUJzl93nbd82gSmNStIXgItonkS0D1gBfA1YD5wFPAVcZbvrLjCNkPtFNH+yGtgFfKiln7MrSHov8H+AHcBPSvjjNH2bXX3cR8n9Grr/uL+D5sLWNJqTzvW2byn/ZtcBM4GHgA/afvGY9pEiGxFRT7oLIiIqSpGNiKgoRTYioqIU2YiIilJkIyIqSpGNiKgoRTYioqIU2YiIiv4/6vB/jg99XfwAAAAASUVORK5CYII=\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_sentences = train_df[\"text\"].tolist()\n",
        "val_sentences = val_df[\"text\"].tolist()\n",
        "test_sentences = test_df[\"text\"].tolist()\n",
        "len(train_sentences), len(val_sentences), len(test_sentences)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VdIG43VIsQPu",
        "outputId": "f481af37-59cf-4add-96fc-4029c1ca27b8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(180040, 30212, 30135)"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_sentences[:10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "erMwdXRXsqza",
        "outputId": "a530d479-1bc6-46c1-f6f9-1d8ef5924d95"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['to investigate the efficacy of @ weeks of daily low-dose oral prednisolone in improving pain , mobility , and systemic low-grade inflammation in the short term and whether the effect would be sustained at @ weeks in older adults with moderate to severe knee osteoarthritis ( oa ) .',\n",
              " 'a total of @ patients with primary knee oa were randomized @:@ ; @ received @ mg/day of prednisolone and @ received placebo for @ weeks .',\n",
              " 'outcome measures included pain reduction and improvement in function scores and systemic inflammation markers .',\n",
              " 'pain was assessed using the visual analog pain scale ( @-@ mm ) .',\n",
              " 'secondary outcome measures included the western ontario and mcmaster universities osteoarthritis index scores , patient global assessment ( pga ) of the severity of knee oa , and @-min walk distance ( @mwd ) .',\n",
              " 'serum levels of interleukin @ ( il-@ ) , il-@ , tumor necrosis factor ( tnf ) - , and high-sensitivity c-reactive protein ( hscrp ) were measured .',\n",
              " 'there was a clinically relevant reduction in the intervention group compared to the placebo group for knee pain , physical function , pga , and @mwd at @ weeks .',\n",
              " 'the mean difference between treatment arms ( @ % ci ) was @ ( @-@ @ ) , p < @ ; @ ( @-@ @ ) , p < @ ; @ ( @-@ @ ) , p < @ ; and @ ( @-@ @ ) , p < @ , respectively .',\n",
              " 'further , there was a clinically relevant reduction in the serum levels of il-@ , il-@ , tnf - , and hscrp at @ weeks in the intervention group when compared to the placebo group .',\n",
              " 'these differences remained significant at @ weeks .']"
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import OneHotEncoder\n",
        "one_hot_encoder = OneHotEncoder(sparse = False)\n",
        "train_labels_one_hot = one_hot_encoder.fit_transform(train_df[\"target\"].to_numpy().reshape(-1,1))"
      ],
      "metadata": {
        "id": "TbXqt9oBstU9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_labels_one_hot"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3GG12tjXteaA",
        "outputId": "b4f3033d-2285-4de2-da6a-95be1ef3dd87"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0., 0., 0., 1., 0.],\n",
              "       [0., 0., 1., 0., 0.],\n",
              "       [0., 0., 1., 0., 0.],\n",
              "       ...,\n",
              "       [0., 0., 0., 0., 1.],\n",
              "       [0., 1., 0., 0., 0.],\n",
              "       [0., 1., 0., 0., 0.]])"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "val_labels_one_hot = one_hot_encoder.transform(val_df[\"target\"].to_numpy().reshape(-1, 1))\n",
        "test_labels_one_hot = one_hot_encoder.transform(test_df[\"target\"].to_numpy().reshape(-1, 1))\n"
      ],
      "metadata": {
        "id": "A0HRolyttgUE"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "val_labels_one_hot"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G9_AA7obt20Z",
        "outputId": "04612cf2-4328-4873-b5d3-c772460065d0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1., 0., 0., 0., 0.],\n",
              "       [1., 0., 0., 0., 0.],\n",
              "       [0., 0., 0., 1., 0.],\n",
              "       ...,\n",
              "       [0., 0., 0., 0., 1.],\n",
              "       [0., 1., 0., 0., 0.],\n",
              "       [0., 1., 0., 0., 0.]])"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_labels_one_hot"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XUgFx2Ptt6IT",
        "outputId": "4b6e2d4d-6388-436c-8cd0-12c6130dda84"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1., 0., 0., 0., 0.],\n",
              "       [0., 0., 0., 0., 1.],\n",
              "       [0., 0., 0., 0., 1.],\n",
              "       ...,\n",
              "       [0., 0., 0., 0., 1.],\n",
              "       [0., 0., 0., 0., 1.],\n",
              "       [0., 1., 0., 0., 0.]])"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.preprocessing import LabelEncoder\n",
        "label_encoder = LabelEncoder()\n",
        "train_labels_encoded = label_encoder.fit_transform(train_df[\"target\"].to_numpy())\n",
        "val_labels_encoded = label_encoder.transform(val_df[\"target\"].to_numpy())\n",
        "test_labels_encoded = label_encoder.transform(test_df[\"target\"].to_numpy())\n",
        "train_labels_encoded\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "17C_2kdRt7mF",
        "outputId": "6951058a-748b-40ec-8d2e-c0fcd7f2bb4d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([3, 2, 2, ..., 4, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(train_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4UV3IW5hvaPH",
        "outputId": "2890ae54-bc97-48d4-9819-24ab3d9e6040"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "180040"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "num_classes = len(label_encoder.classes_)\n",
        "class_names = label_encoder.classes_\n",
        "num_classes, class_names"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ic4v9GAJvgYk",
        "outputId": "c89c4646-0a27-4b19-d42b-b2195725a3ba"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(5, array(['BACKGROUND', 'CONCLUSIONS', 'METHODS', 'OBJECTIVE', 'RESULTS'],\n",
              "       dtype=object))"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Baseline - Tfidf + Naive Bayes"
      ],
      "metadata": {
        "id": "lw_Qi-yev06f"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.naive_bayes import MultinomialNB\n",
        "from sklearn.pipeline import Pipeline\n",
        "\n",
        "model_0 = Pipeline([\n",
        "                    (\"tf-idf\", TfidfVectorizer()),\n",
        "                    (\"clf\", MultinomialNB())\n",
        "])\n",
        "\n",
        "model_0.fit(X=train_sentences, y=train_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "eibwMo1dwmpo",
        "outputId": "5a08f090-01b4-4882-cad5-922817f76c95"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Pipeline(steps=[('tf-idf', TfidfVectorizer()), ('clf', MultinomialNB())])"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_0.score(X=val_sentences, y = val_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rBqI95D2xedJ",
        "outputId": "848adffa-cf5b-4e4c-c8f5-6c7c0416e222"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7218323844829869"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "baseline_preds = model_0.predict(val_sentences)\n",
        "baseline_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YeEkrmbSxrdp",
        "outputId": "ded67f12-f258-4167-8dd2-89516a4a70fd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([4, 1, 3, ..., 4, 4, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import accuracy_score, precision_recall_fscore_support\n",
        "\n",
        "def calculate_results(y_true, y_pred):\n",
        "\n",
        "  model_accuracy = accuracy_score(y_true, y_pred) * 100\n",
        "\n",
        "  model_precision, model_recall, model_f1, _ = precision_recall_fscore_support(y_true, y_pred, average=\"weighted\")\n",
        "  model_results = {\"accuracy\": model_accuracy,\n",
        "                  \"precision\": model_precision,\n",
        "                  \"recall\": model_recall,\n",
        "                  \"f1\": model_f1}\n",
        "  return model_results"
      ],
      "metadata": {
        "id": "4E7SZn2Rzg3Z"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "baseline_results = calculate_results(y_true=val_labels_encoded,\n",
        "                                     y_pred=baseline_preds)\n",
        "baseline_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dw7aBiSs0Jj8",
        "outputId": "f57a0884-36e8-4e9b-ec89-77e74a81da58"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 72.1832384482987,\n",
              " 'f1': 0.6989250353450294,\n",
              " 'precision': 0.7186466952323352,\n",
              " 'recall': 0.7218323844829869}"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "###Building Sequence Models"
      ],
      "metadata": {
        "id": "Qmn_LhI90ZPF"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras import layers"
      ],
      "metadata": {
        "id": "kRjIdX4V0fpk"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sent_lens = [len(sentence.split()) for sentence in train_sentences]\n",
        "avg_sent_len = np.mean(sent_lens)\n",
        "avg_sent_len"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wLPWN_5O0o5M",
        "outputId": "3d2f9115-80ec-4968-8c3a-02c5b92eeaeb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "26.338269273494777"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "plt.hist(sent_lens, bins=7);\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "Z7lWJYUS3qE2",
        "outputId": "2330c738-fceb-4502-b338-a5797757b903"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "output_seq_len = int(np.percentile(sent_lens, 95))\n",
        "output_seq_len"
      ],
      "metadata": {
        "id": "YAAFhhBF31wr",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f008dcd2-720c-46b4-e4db-e41e6b073a13"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "55"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        " max(sent_lens)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D5Bk9_0X_ona",
        "outputId": "8a30102e-f393-4170-c179-fd8dea9921d5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "296"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "max_tokens = 68000 #Given in the paper."
      ],
      "metadata": {
        "id": "4uZbSyJhAr2V"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.layers.experimental.preprocessing import TextVectorization\n",
        "\n",
        "text_vectorizer = TextVectorization(max_tokens = max_tokens,\n",
        "                                    output_sequence_length = 55)"
      ],
      "metadata": {
        "id": "t_ZpDXaXBHE-"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "text_vectorizer.adapt(train_sentences)"
      ],
      "metadata": {
        "id": "gTn67tn4Bj3x"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import random\n",
        "target_sentence = random.choice(train_sentences)\n",
        "print(f\"Text:\\n{target_sentence}\")\n",
        "print(f\"\\nLength of text: {len(target_sentence.split())}\")\n",
        "print(f\"\\nVectorized text:\\n{text_vectorizer([target_sentence])}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dJ3GhaDcByJc",
        "outputId": "72088423-e956-45c4-f01e-24043b4e10a4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Text:\n",
            "the clinical utility of genotype-guided ( pharmacogenetically based ) dosing of warfarin has been tested only in small clinical trials or observational studies , with equivocal results .\n",
            "\n",
            "Length of text: 28\n",
            "\n",
            "Vectorized text:\n",
            "[[    2    47  2049     4 10743 49129   261   915     4  2163   139   167\n",
            "    567   192     5   521    47   233    16  1636   202     7 16036   117\n",
            "      0     0     0     0     0     0     0     0     0     0     0     0\n",
            "      0     0     0     0     0     0     0     0     0     0     0     0\n",
            "      0     0     0     0     0     0     0]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "rct_20k_text_vocab = text_vectorizer.get_vocabulary()\n",
        "print(f\"Number of words in vocabulary: {len(rct_20k_text_vocab)}\"), \n",
        "print(f\"Most common words in the vocabulary: {rct_20k_text_vocab[:5]}\")\n",
        "print(f\"Least common words in the vocabulary: {rct_20k_text_vocab[-5:]}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tnHhCc0SC_PJ",
        "outputId": "22009f8d-e50e-4892-ce40-81e99efe9375"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of words in vocabulary: 64841\n",
            "Most common words in the vocabulary: ['', '[UNK]', 'the', 'and', 'of']\n",
            "Least common words in the vocabulary: ['aainduced', 'aaigroup', 'aachener', 'aachen', 'aaacp']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "text_vectorizer.get_config()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xODrgKMcEyZF",
        "outputId": "d176974a-3b76-48eb-84c3-106cd2cc244d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'batch_input_shape': (None,),\n",
              " 'dtype': 'string',\n",
              " 'idf_weights': None,\n",
              " 'max_tokens': 68000,\n",
              " 'name': 'text_vectorization',\n",
              " 'ngrams': None,\n",
              " 'output_mode': 'int',\n",
              " 'output_sequence_length': 55,\n",
              " 'pad_to_max_tokens': False,\n",
              " 'ragged': False,\n",
              " 'sparse': False,\n",
              " 'split': 'whitespace',\n",
              " 'standardize': 'lower_and_strip_punctuation',\n",
              " 'trainable': True,\n",
              " 'vocabulary': None}"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "token_embed = layers.Embedding(input_dim = len(rct_20k_text_vocab),\n",
        "                               output_dim = 128,\n",
        "                               mask_zero = True,\n",
        "                               name = 'token_embedding')"
      ],
      "metadata": {
        "id": "Em_NIRQhE_06"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"Sentence before vectorization:\\n{target_sentence}\\n\")\n",
        "vectorized_sentence = text_vectorizer([target_sentence])\n",
        "print(f\"Sentence after vectorization (before embedding):\\n{vectorized_sentence}\\n\")\n",
        "embedded_sentence = token_embed(vectorized_sentence)\n",
        "print(f\"Sentence after embedding:\\n{embedded_sentence}\\n\")\n",
        "print(f\"Embedded sentence shape: {embedded_sentence.shape}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xedayLKAvD-g",
        "outputId": "1bfed0e0-9c7c-454b-8ffb-5c12670ac917"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Sentence before vectorization:\n",
            "the clinical utility of genotype-guided ( pharmacogenetically based ) dosing of warfarin has been tested only in small clinical trials or observational studies , with equivocal results .\n",
            "\n",
            "Sentence after vectorization (before embedding):\n",
            "[[    2    47  2049     4 10743 49129   261   915     4  2163   139   167\n",
            "    567   192     5   521    47   233    16  1636   202     7 16036   117\n",
            "      0     0     0     0     0     0     0     0     0     0     0     0\n",
            "      0     0     0     0     0     0     0     0     0     0     0     0\n",
            "      0     0     0     0     0     0     0]]\n",
            "\n",
            "Sentence after embedding:\n",
            "[[[-0.00688001 -0.02390089  0.01608186 ...  0.02081187 -0.03330153\n",
            "    0.04096079]\n",
            "  [-0.000638   -0.01897868 -0.04263452 ...  0.04853748 -0.02753742\n",
            "   -0.03579951]\n",
            "  [ 0.04595437  0.02021721 -0.02634802 ... -0.0152688  -0.04516705\n",
            "    0.00199192]\n",
            "  ...\n",
            "  [ 0.0366185   0.02505573 -0.01422445 ...  0.02800094  0.02159094\n",
            "   -0.02358586]\n",
            "  [ 0.0366185   0.02505573 -0.01422445 ...  0.02800094  0.02159094\n",
            "   -0.02358586]\n",
            "  [ 0.0366185   0.02505573 -0.01422445 ...  0.02800094  0.02159094\n",
            "   -0.02358586]]]\n",
            "\n",
            "Embedded sentence shape: (1, 55, 128)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_dataset = tf.data.Dataset.from_tensor_slices((train_sentences, train_labels_one_hot))\n",
        "valid_dataset = tf.data.Dataset.from_tensor_slices((val_sentences, val_labels_one_hot))\n",
        "test_dataset = tf.data.Dataset.from_tensor_slices((test_sentences, test_labels_one_hot))\n",
        "\n",
        "train_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q7rkkk29vETL",
        "outputId": "05fa3e7a-1a5d-4a92-b192-c9ed8793c986"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<TensorSliceDataset element_spec=(TensorSpec(shape=(), dtype=tf.string, name=None), TensorSpec(shape=(5,), dtype=tf.float64, name=None))>"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_dataset = train_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "test_dataset = test_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "valid_dataset = valid_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "\n",
        "train_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "R8abS6Mkvr5_",
        "outputId": "ec76f35f-7ddb-431a-8b06-3fdc94ece650"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<PrefetchDataset element_spec=(TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>"
            ]
          },
          "metadata": {},
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 1D Convolutional Model"
      ],
      "metadata": {
        "id": "lcWNnzvhw0_W"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "inputs = layers.Input(shape=(1,), dtype = tf.string)\n",
        "text_vectors = text_vectorizer(inputs)\n",
        "token_embeddings = token_embed(text_vectors)\n",
        "x = layers.Conv1D(64, kernel_size = 5, padding='same', activation = 'relu')(token_embeddings)\n",
        "x = layers.GlobalAveragePooling1D()(x)\n",
        "outputs = layers.Dense(num_classes, activation = \"softmax\")(x)\n",
        "model_1 = tf.keras.Model(inputs, outputs)\n",
        "\n",
        "model_1.compile(loss = \"categorical_crossentropy\",\n",
        "                optimizer = tf.keras.optimizers.Adam(),\n",
        "                metrics = ['accuracy'])"
      ],
      "metadata": {
        "id": "EvnbNuf4-iwm"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_1.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "V3yizt56ABAL",
        "outputId": "82b9be29-edd7-4bb9-febd-ae489f1db21f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_1 (InputLayer)        [(None, 1)]               0         \n",
            "                                                                 \n",
            " text_vectorization (TextVec  (None, 55)               0         \n",
            " torization)                                                     \n",
            "                                                                 \n",
            " token_embedding (Embedding)  (None, 55, 128)          8299648   \n",
            "                                                                 \n",
            " conv1d (Conv1D)             (None, 55, 64)            41024     \n",
            "                                                                 \n",
            " global_average_pooling1d (G  (None, 64)               0         \n",
            " lobalAveragePooling1D)                                          \n",
            "                                                                 \n",
            " dense (Dense)               (None, 5)                 325       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 8,340,997\n",
            "Trainable params: 8,340,997\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1_history = model_1.fit(train_dataset,\n",
        "                              steps_per_epoch = int(0.1 * len(train_dataset)),\n",
        "                              epochs = 3,\n",
        "                              validation_data = valid_dataset,\n",
        "                              validation_steps = int(0.1 * len(valid_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nh9Qbu0sALwt",
        "outputId": "244c30c1-99e4-4777-bf5d-2ff0965f9194"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/3\n",
            "562/562 [==============================] - 15s 6ms/step - loss: 0.9121 - accuracy: 0.6388 - val_loss: 0.6782 - val_accuracy: 0.7417\n",
            "Epoch 2/3\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.6510 - accuracy: 0.7571 - val_loss: 0.6250 - val_accuracy: 0.7763\n",
            "Epoch 3/3\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.6123 - accuracy: 0.7766 - val_loss: 0.5944 - val_accuracy: 0.7872\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1_history = model_1.fit(train_dataset,\n",
        "                              steps_per_epoch = int(0.1 * len(train_dataset)),\n",
        "                              epochs = 10,\n",
        "                              validation_data = valid_dataset,\n",
        "                              validation_steps = int(0.1 * len(valid_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mGMZgGcxBH-q",
        "outputId": "d6fe4c35-5fa2-4503-b30e-7fcf75910cc9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.5124 - accuracy: 0.8221 - val_loss: 0.5866 - val_accuracy: 0.7892\n",
            "Epoch 2/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.4487 - accuracy: 0.8416 - val_loss: 0.5963 - val_accuracy: 0.7852\n",
            "Epoch 3/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.4366 - accuracy: 0.8463 - val_loss: 0.6068 - val_accuracy: 0.7789\n",
            "Epoch 4/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.5861 - accuracy: 0.7900 - val_loss: 0.5726 - val_accuracy: 0.7916\n",
            "Epoch 5/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.5834 - accuracy: 0.7930 - val_loss: 0.5527 - val_accuracy: 0.8005\n",
            "Epoch 6/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.5708 - accuracy: 0.7938 - val_loss: 0.5478 - val_accuracy: 0.8009\n",
            "Epoch 7/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.5494 - accuracy: 0.8027 - val_loss: 0.5367 - val_accuracy: 0.8098\n",
            "Epoch 8/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.5310 - accuracy: 0.8141 - val_loss: 0.5211 - val_accuracy: 0.8168\n",
            "Epoch 9/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.5342 - accuracy: 0.8111 - val_loss: 0.5446 - val_accuracy: 0.7999\n",
            "Epoch 10/10\n",
            "562/562 [==============================] - 3s 6ms/step - loss: 0.5351 - accuracy: 0.8081 - val_loss: 0.5216 - val_accuracy: 0.8092\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1.evaluate(valid_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BbbNZMvEB47c",
        "outputId": "aea8af02-7c3c-436c-d396-e8921cd53621"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 3s 3ms/step - loss: 0.5183 - accuracy: 0.8142\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.5182808637619019, 0.8142459988594055]"
            ]
          },
          "metadata": {},
          "execution_count": 50
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1_pred_probs = model_1.predict(val_sentences)\n"
      ],
      "metadata": {
        "id": "PUUdXjxMCVGX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_1_pred_probs"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "d7A1DrpcDAKY",
        "outputId": "559d2eea-004c-4576-cfc7-b25eea0299b5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[5.8314604e-01, 9.8013006e-02, 1.1343796e-01, 1.5973660e-01,\n",
              "        4.5666382e-02],\n",
              "       [4.8367018e-01, 2.6245981e-01, 1.4903126e-02, 2.0440808e-01,\n",
              "        3.4558848e-02],\n",
              "       [1.7438562e-01, 5.0688926e-03, 1.5458276e-03, 8.1891930e-01,\n",
              "        8.0333106e-05],\n",
              "       ...,\n",
              "       [4.2336123e-06, 1.2230979e-04, 2.5369814e-03, 4.5328329e-06,\n",
              "        9.9733198e-01],\n",
              "       [7.8159176e-02, 5.2544206e-01, 1.3685213e-01, 4.5528833e-02,\n",
              "        2.1401779e-01],\n",
              "       [1.8394575e-02, 9.6430278e-01, 1.3088840e-02, 9.3090831e-04,\n",
              "        3.2828168e-03]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1_preds = tf.argmax(model_1_pred_probs, axis = 1)\n",
        "model_1_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x6h5jZrgDD0N",
        "outputId": "598001f1-87ff-4e25-8f7b-b4004c2a6adb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(30212,), dtype=int64, numpy=array([0, 0, 3, ..., 4, 1, 1])>"
            ]
          },
          "metadata": {},
          "execution_count": 53
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1_results = calculate_results(y_true = val_labels_encoded, y_pred = model_1_preds)\n",
        "model_1_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4A3ahQX5fkOC",
        "outputId": "6f473473-822d-4cbc-c595-47a849b0e2ae"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 81.42459949688865,\n",
              " 'f1': 0.8109578148437292,\n",
              " 'precision': 0.8103560542117628,\n",
              " 'recall': 0.8142459949688865}"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow_hub as hub\n",
        "tf_hub_embedding_layer = hub.KerasLayer(\"https://tfhub.dev/google/universal-sentence-encoder/4\",\n",
        "                                        trainable = False,\n",
        "                                        name = \"universal_sentence_encoder\")\n"
      ],
      "metadata": {
        "id": "q8Zdmdykf49v"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "random_training_sentence = random.choice(train_sentences)\n",
        "print(f\"Random training sentence:\\n{random_training_sentence}\\n\")\n",
        "use_embedded_sentence = tf_hub_embedding_layer([random_training_sentence])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "f3IIJowv31qQ",
        "outputId": "81183e41-d456-429e-f92d-e918385f8bfa"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Random training sentence:\n",
            "metatarsus adductus is a common pediatric foot deformity related to intrauterine molding .\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "use_embedded_sentence.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jw5jRiy64R2m",
        "outputId": "27d4239c-9e3a-4a3a-aef2-5c86983c3399"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "TensorShape([1, 512])"
            ]
          },
          "metadata": {},
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"Sentence after embedding:\\n{use_embedded_sentence}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HHgeDQi94eno",
        "outputId": "8694ec51-7508-49d3-8a2b-abc858833134"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Sentence after embedding:\n",
            "[[ 0.01957086  0.08954732 -0.01761952  0.04193586 -0.05153193 -0.04189713\n",
            "   0.0116812   0.01661068  0.02446553  0.0102042   0.06820686  0.08281638\n",
            "  -0.03082899  0.0632665   0.08031683 -0.05631074 -0.09380297 -0.04544327\n",
            "  -0.07376179 -0.04736659 -0.08101864  0.00674424  0.059961   -0.03385213\n",
            "   0.05389145 -0.01615187  0.01146602  0.01931871 -0.0062017   0.06874909\n",
            "  -0.03956238  0.08879916 -0.02741722  0.07104471 -0.0278207  -0.05533249\n",
            "  -0.06330281  0.09008925 -0.06594521 -0.01992238  0.01554852 -0.02986084\n",
            "   0.05221952  0.0546467   0.00924512 -0.03150042 -0.02100791 -0.00481115\n",
            "   0.01355759  0.03569749  0.03637778 -0.02997683  0.02317179 -0.06750373\n",
            "   0.07688599 -0.08273236  0.05529003 -0.05196289  0.07793091  0.01980902\n",
            "  -0.03748023 -0.04052385 -0.02550196  0.04486858 -0.06900846 -0.07005762\n",
            "   0.0534706  -0.02571663  0.002371    0.00863319 -0.0264296   0.0277202\n",
            "   0.04342524 -0.053192   -0.06285149 -0.0048966  -0.04324923  0.03811843\n",
            "   0.04876234 -0.00976113 -0.06034793  0.00359192  0.06706546 -0.06601188\n",
            "   0.02299598  0.00907571  0.0417807  -0.02088893  0.02038957 -0.03230781\n",
            "  -0.06769302  0.03774251  0.04085069 -0.01584129 -0.04929853 -0.04724627\n",
            "  -0.02335434 -0.02835358  0.02985167  0.0375288  -0.07904751  0.04529027\n",
            "   0.06068949  0.01419103  0.01174389 -0.00158152 -0.0064821  -0.02463797\n",
            "  -0.06484875  0.01564053  0.03122854 -0.06428005 -0.01846337  0.01149792\n",
            "   0.05975178 -0.0272798   0.0007897   0.05180314  0.03971013  0.02439643\n",
            "  -0.04273842 -0.05247685  0.05350896 -0.00602085 -0.00933527  0.05649856\n",
            "  -0.0519825   0.00257149 -0.00179869 -0.05220347 -0.01985253  0.09409639\n",
            "   0.06110804  0.05506196  0.0192419  -0.0016512   0.03424878  0.01468859\n",
            "  -0.02381545  0.02044787 -0.06625874  0.08662589 -0.03515561 -0.07373956\n",
            "  -0.02000428 -0.07603349  0.03123674 -0.04199476  0.00901881  0.07099955\n",
            "  -0.06299911 -0.05941433 -0.04519234  0.00585754 -0.0050576   0.01738177\n",
            "   0.01399237 -0.01200368 -0.06170848 -0.07690553 -0.00276054  0.00444904\n",
            "   0.01184624 -0.05765938  0.01338599  0.02152771 -0.07092874  0.03729984\n",
            "  -0.00577889 -0.01596756 -0.03891257 -0.04392525  0.04075004  0.04858528\n",
            "   0.00964988  0.0425039   0.0367907   0.04243207  0.07856364  0.03867161\n",
            "  -0.02188081  0.04834295 -0.00217526 -0.02083032 -0.07698565 -0.05241542\n",
            "  -0.05062084  0.01924795 -0.00902898 -0.02416566 -0.07641876  0.028318\n",
            "  -0.00721746 -0.02929205  0.03435311  0.02534853 -0.06583491 -0.08342755\n",
            "  -0.01701012  0.05412903  0.04238635 -0.03118632 -0.02447527 -0.01009285\n",
            "   0.03148796  0.00619647  0.02722032  0.07792858 -0.00589021  0.01632634\n",
            "   0.05418113  0.02674591  0.00579423  0.02915842 -0.04646397  0.0545577\n",
            "   0.02381732  0.0573324  -0.03909324  0.09003461 -0.08422428  0.00766455\n",
            "  -0.00266284 -0.04151268  0.05874179 -0.00803572  0.0329501   0.0609503\n",
            "  -0.06473248  0.04209743 -0.01998476  0.00652485 -0.01513519 -0.04305659\n",
            "   0.05983948 -0.04112418 -0.03066362  0.04649432 -0.01333427 -0.01365321\n",
            "  -0.03455571 -0.04021921 -0.01873906 -0.04727636  0.01242243  0.00990467\n",
            "  -0.05759396  0.01265068 -0.027712   -0.08146599  0.04551631 -0.00995062\n",
            "  -0.05731947 -0.02886852 -0.00730847  0.06594404 -0.00079973  0.03874684\n",
            "  -0.02652788  0.02929918 -0.03214394  0.02181176 -0.02851197  0.08504303\n",
            "  -0.03626088 -0.01379551 -0.06899402 -0.04850094  0.03026449  0.00477673\n",
            "  -0.06269057  0.05251081 -0.0211375   0.00274357 -0.08921784 -0.0500516\n",
            "  -0.07235237  0.01250101 -0.02812791 -0.03961097  0.00633759  0.0688777\n",
            "   0.05815598 -0.07097053  0.03928106  0.01449976 -0.00145561  0.00042209\n",
            "  -0.04577707  0.03521949 -0.02219583 -0.04030516 -0.0335188   0.03252073\n",
            "  -0.06732298 -0.04809769  0.01639491 -0.05468457 -0.08400894  0.00120299\n",
            "  -0.05122733 -0.04521698  0.02773876 -0.05611512 -0.06682576 -0.04650849\n",
            "  -0.0068164   0.03803138  0.04722774 -0.01158178 -0.06651888  0.05278638\n",
            "   0.01567794 -0.04595082 -0.07006415  0.02691628  0.0331326  -0.00492572\n",
            "  -0.01633211  0.07506622 -0.06932224  0.00824634 -0.03480085 -0.0511357\n",
            "   0.01114459  0.0286223  -0.07397287 -0.02923896  0.03227289 -0.05118823\n",
            "   0.03456532 -0.0134306   0.0012642   0.02171657 -0.02362464  0.00059518\n",
            "   0.02619122  0.04229685  0.04066901  0.00894428 -0.05200322 -0.05956902\n",
            "   0.0278126  -0.01041684  0.02688926  0.08359204  0.07687952 -0.07772994\n",
            "  -0.03594743  0.03767167 -0.01335367 -0.02197474  0.00883473 -0.06036822\n",
            "   0.01641542 -0.03779118 -0.07662033 -0.05590115 -0.04340656 -0.0591604\n",
            "  -0.01256847 -0.07473843 -0.04395084 -0.02467102  0.01536686  0.01034586\n",
            "  -0.03637901 -0.00494601  0.05472495 -0.051578   -0.03652967  0.01020079\n",
            "  -0.03115343  0.00894155 -0.023258    0.00274938 -0.01517302  0.01282319\n",
            "   0.05521651 -0.00759074  0.09305223  0.00087549 -0.0233338  -0.02022741\n",
            "  -0.03935515 -0.03480283 -0.05660785  0.01597815  0.05053919 -0.02397225\n",
            "   0.00293668 -0.02199848 -0.05002479 -0.02796274  0.01527484 -0.01384418\n",
            "   0.03226756  0.0586704   0.05698703  0.0298252   0.05249213 -0.05072403\n",
            "   0.03141625  0.03773667  0.07527497  0.06593472 -0.00166525  0.0396852\n",
            "   0.03038916 -0.0139685  -0.00789248  0.02406214  0.00306804  0.01592704\n",
            "   0.04043958  0.09065128 -0.01152619 -0.04461971  0.03480529  0.02804287\n",
            "  -0.03865129 -0.03050387  0.03458373  0.01627247  0.07448224  0.048131\n",
            "  -0.02481474 -0.08248837  0.03879368  0.03330696 -0.00892428 -0.06620883\n",
            "  -0.05814323 -0.08289149 -0.01234821  0.02803981 -0.01181725  0.00387856\n",
            "  -0.04036615  0.01455573 -0.05774077 -0.01710967 -0.06415631 -0.05397844\n",
            "   0.00485027 -0.00548506 -0.04710032  0.00219872 -0.00858259  0.05618534\n",
            "   0.01607654  0.0196855   0.06117309  0.0315634  -0.00716114 -0.00247525\n",
            "  -0.06047356  0.07703559 -0.00438034  0.0534103  -0.09236425  0.03457633\n",
            "   0.03967819  0.00273603  0.08051584 -0.00034396  0.06496355  0.00156564\n",
            "  -0.07506599 -0.06926968  0.06287772  0.05396359 -0.08639121  0.02653282\n",
            "   0.0392168   0.00687397  0.00821634  0.0249399  -0.0018576  -0.07313524\n",
            "   0.00815776  0.02757236 -0.01001787  0.00057747 -0.07261707 -0.01671311\n",
            "   0.03577746  0.00908059 -0.03160043  0.06128781 -0.0374923  -0.05955745\n",
            "   0.06141972 -0.04895695  0.06842182  0.01852269 -0.07282653 -0.01216717\n",
            "  -0.01374816 -0.04788204  0.00572629  0.02013792 -0.01561133  0.05530243\n",
            "   0.06361717 -0.09424464  0.01902852  0.02865251 -0.06264448 -0.04417432\n",
            "  -0.02255045 -0.05799829]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Building model with NLP feature extraction model"
      ],
      "metadata": {
        "id": "KpkRBbxg4r-Z"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "inputs = layers.Input(shape = [], dtype = tf.string)\n",
        "\n",
        "pretrained_embedding = tf_hub_embedding_layer(inputs)\n",
        "x = layers.Dense(128, activation = \"relu\")(pretrained_embedding)\n",
        "outputs = layers.Dense(5, activation = \"softmax\")(x)\n",
        "\n",
        "model_2 = tf.keras.Model(inputs,outputs)\n",
        "\n",
        "model_2.compile(loss = \"categorical_crossentropy\",\n",
        "                metrics = ['accuracy'],\n",
        "                optimizer = tf.keras.optimizers.Adam())"
      ],
      "metadata": {
        "id": "nfuR2EKH49LL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lRWVCQlG5vGu",
        "outputId": "41c05861-d73c-4251-c388-43a40f9c404a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_2 (InputLayer)        [(None,)]                 0         \n",
            "                                                                 \n",
            " universal_sentence_encoder   (None, 512)              256797824 \n",
            " (KerasLayer)                                                    \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 128)               65664     \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 5)                 645       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 256,864,133\n",
            "Trainable params: 66,309\n",
            "Non-trainable params: 256,797,824\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2_history = model_2.fit(train_dataset,\n",
        "            steps_per_epoch=int(0.1 * len(train_dataset)),\n",
        "            epochs=10,\n",
        "            validation_data=valid_dataset,\n",
        "            validation_steps=int(0.1 * len(valid_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "26r3_lYq5yNA",
        "outputId": "a06d0ecc-ef0c-41b2-ce6d-841a9eeac77c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "562/562 [==============================] - 9s 12ms/step - loss: 0.9178 - accuracy: 0.6530 - val_loss: 0.7954 - val_accuracy: 0.6878\n",
            "Epoch 2/10\n",
            "562/562 [==============================] - 7s 12ms/step - loss: 0.7674 - accuracy: 0.7025 - val_loss: 0.7535 - val_accuracy: 0.7061\n",
            "Epoch 3/10\n",
            "562/562 [==============================] - 7s 12ms/step - loss: 0.7522 - accuracy: 0.7136 - val_loss: 0.7389 - val_accuracy: 0.7118\n",
            "Epoch 4/10\n",
            "562/562 [==============================] - 7s 12ms/step - loss: 0.7195 - accuracy: 0.7238 - val_loss: 0.7118 - val_accuracy: 0.7291\n",
            "Epoch 5/10\n",
            "562/562 [==============================] - 6s 11ms/step - loss: 0.7263 - accuracy: 0.7208 - val_loss: 0.6898 - val_accuracy: 0.7327\n",
            "Epoch 6/10\n",
            "562/562 [==============================] - 6s 12ms/step - loss: 0.7173 - accuracy: 0.7259 - val_loss: 0.6827 - val_accuracy: 0.7370\n",
            "Epoch 7/10\n",
            "562/562 [==============================] - 6s 11ms/step - loss: 0.6842 - accuracy: 0.7395 - val_loss: 0.6648 - val_accuracy: 0.7443\n",
            "Epoch 8/10\n",
            "562/562 [==============================] - 7s 12ms/step - loss: 0.6737 - accuracy: 0.7454 - val_loss: 0.6545 - val_accuracy: 0.7470\n",
            "Epoch 9/10\n",
            "562/562 [==============================] - 6s 11ms/step - loss: 0.6720 - accuracy: 0.7422 - val_loss: 0.6537 - val_accuracy: 0.7477\n",
            "Epoch 10/10\n",
            "562/562 [==============================] - 6s 11ms/step - loss: 0.6661 - accuracy: 0.7477 - val_loss: 0.6485 - val_accuracy: 0.7533\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7W5eKCMb6FAD",
        "outputId": "e0d462e5-54d2-4d20-a6cf-c6b65797f90a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_2 (InputLayer)        [(None,)]                 0         \n",
            "                                                                 \n",
            " universal_sentence_encoder   (None, 512)              256797824 \n",
            " (KerasLayer)                                                    \n",
            "                                                                 \n",
            " dense_1 (Dense)             (None, 128)               65664     \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 5)                 645       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 256,864,133\n",
            "Trainable params: 66,309\n",
            "Non-trainable params: 256,797,824\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.evaluate(valid_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ql5vTjA6AMiL",
        "outputId": "7e998ca9-c982-46a3-da17-20dfa15d6347"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 15s 16ms/step - loss: 0.6532 - accuracy: 0.7506\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.6531714797019958, 0.7506288886070251]"
            ]
          },
          "metadata": {},
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2_pred_probs = model_2.predict(val_sentences)\n",
        "model_2_pred_probs"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "U1yd-HiuAeIg",
        "outputId": "bc491368-f5e1-4d58-eb16-9866a38be443"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[3.4775996e-01, 5.1955056e-01, 5.6433043e-04, 1.2475245e-01,\n",
              "        7.3727127e-03],\n",
              "       [3.7425631e-01, 5.3511703e-01, 2.2494262e-03, 8.7468788e-02,\n",
              "        9.0837333e-04],\n",
              "       [4.9590975e-01, 4.0497880e-02, 2.5854072e-02, 3.9584160e-01,\n",
              "        4.1896634e-02],\n",
              "       ...,\n",
              "       [9.2517294e-04, 4.8851455e-04, 1.9870849e-02, 1.6115056e-04,\n",
              "        9.7855431e-01],\n",
              "       [4.3608118e-03, 6.4181827e-02, 2.4041805e-01, 1.6229691e-03,\n",
              "        6.8941635e-01],\n",
              "       [5.1823661e-02, 7.8632277e-01, 1.5275487e-01, 1.0317252e-03,\n",
              "        8.0670435e-03]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 64
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2_preds = tf.argmax(model_2_pred_probs, axis=1)\n",
        "model_2_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "diM0tZUQAuEo",
        "outputId": "18b8b098-15e4-4376-f793-85a4f75f57d0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(30212,), dtype=int64, numpy=array([1, 1, 0, ..., 4, 4, 1])>"
            ]
          },
          "metadata": {},
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2_results = calculate_results(y_true=val_labels_encoded,\n",
        "                                    y_pred=model_2_preds)\n",
        "model_2_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0FPPudGnAzrX",
        "outputId": "de3f88a0-f5e8-4fc6-cb50-378f53c874af"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 75.0628889183106,\n",
              " 'f1': 0.7458525193581951,\n",
              " 'precision': 0.7461287123421187,\n",
              " 'recall': 0.750628889183106}"
            ]
          },
          "metadata": {},
          "execution_count": 66
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Conv1D model with character embeddings\n",
        "\n",
        "![image.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "pRY9nVKVA45_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def split_chars(text):\n",
        "  return \" \".join(list(text))\n",
        "\n",
        "split_chars(random_training_sentence)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iohxFjPoCZhY",
        "outputId": "ab0942d4-7d6f-4ce5-f691-8431884f2728"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'m e t a t a r s u s   a d d u c t u s   i s   a   c o m m o n   p e d i a t r i c   f o o t   d e f o r m i t y   r e l a t e d   t o   i n t r a u t e r i n e   m o l d i n g   .'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_chars = [split_chars(sentence) for sentence in train_sentences]\n",
        "train_chars[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YkRAIzZxC8ap",
        "outputId": "e8e12ec3-8b58-4c28-ae2c-a2fb4bab552c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'t o   i n v e s t i g a t e   t h e   e f f i c a c y   o f   @   w e e k s   o f   d a i l y   l o w - d o s e   o r a l   p r e d n i s o l o n e   i n   i m p r o v i n g   p a i n   ,   m o b i l i t y   ,   a n d   s y s t e m i c   l o w - g r a d e   i n f l a m m a t i o n   i n   t h e   s h o r t   t e r m   a n d   w h e t h e r   t h e   e f f e c t   w o u l d   b e   s u s t a i n e d   a t   @   w e e k s   i n   o l d e r   a d u l t s   w i t h   m o d e r a t e   t o   s e v e r e   k n e e   o s t e o a r t h r i t i s   (   o a   )   .'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "val_chars = [split_chars(sentence) for sentence in val_sentences]\n",
        "test_chars = [split_chars(sentence) for sentence in test_sentences]"
      ],
      "metadata": {
        "id": "9FXnrynnDsdV"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "char_lens = [len(sentence) for sentence in train_sentences]\n",
        "mean_char_len = np.mean(char_lens)\n",
        "mean_char_len"
      ],
      "metadata": {
        "id": "T6-TUAKLDyIc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "290802ef-9950-4cfe-b985-5f9506af1774"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "149.3662574983337"
            ]
          },
          "metadata": {},
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "plt.hist(char_lens, bins=7)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LUdCTMMK0B41",
        "outputId": "8fb6c70d-51d5-4205-f247-2c59010a9c22"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([1.41175e+05, 3.71110e+04, 1.60000e+03, 1.27000e+02, 2.10000e+01,\n",
              "        5.00000e+00, 1.00000e+00]),\n",
              " array([1.00000000e+00, 1.98857143e+02, 3.96714286e+02, 5.94571429e+02,\n",
              "        7.92428571e+02, 9.90285714e+02, 1.18814286e+03, 1.38600000e+03]),\n",
              " <a list of 7 Patch objects>)"
            ]
          },
          "metadata": {},
          "execution_count": 71
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "output_seq_char_len = int(np.percentile(char_lens, 95))\n",
        "output_seq_char_len"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6WtCYZtW0VVx",
        "outputId": "d26a429d-d79a-4be0-ab74-e55df4576c5c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "290"
            ]
          },
          "metadata": {},
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import string\n",
        "alphabet = string.ascii_lowercase + string.digits + string.punctuation\n",
        "alphabet"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aPtu9hIH0elW",
        "outputId": "f4e33116-c35e-435a-be74-b51f59489de7"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'abcdefghijklmnopqrstuvwxyz0123456789!\"#$%&\\'()*+,-./:;<=>?@[\\\\]^_`{|}~'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 73
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "NUM_CHAR_TOKENS = len(alphabet) + 2\n",
        "char_vectorizer = TextVectorization(max_tokens = NUM_CHAR_TOKENS,\n",
        "                                    output_sequence_length = output_seq_char_len,\n",
        "                                    standardize = \"lower_and_strip_punctuation\",\n",
        "                                    name = 'char_vectorizer')\n",
        "char_vectorizer.adapt(train_chars)"
      ],
      "metadata": {
        "id": "YUdoGkVI03Yt"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "char_vocab = char_vectorizer.get_vocabulary()\n",
        "print(f\"Number of different characters in character vocab: {len(char_vocab)}\")\n",
        "print(f\"5 most common characters: {char_vocab[:5]}\")\n",
        "print(f\"5 least common characters: {char_vocab[-5:]}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4ybG3m6G4a6l",
        "outputId": "fe6e957d-24fa-4c7b-fd11-f98c7dfce086"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of different characters in character vocab: 28\n",
            "5 most common characters: ['', '[UNK]', 'e', 't', 'i']\n",
            "5 least common characters: ['k', 'x', 'z', 'q', 'j']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "random_train_chars = random.choice(train_chars)\n",
        "print(f\"Charified text:\\n{random_train_chars}\")\n",
        "print(f\"\\nLength of chars: {len(random_train_chars.split())}\")\n",
        "vectorized_chars = char_vectorizer([random_train_chars])\n",
        "print(f\"\\nVectorized chars:\\n{vectorized_chars}\")\n",
        "print(f\"\\nLength of vectorized chars: {len(vectorized_chars[0])}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kfSWHSBE5QTZ",
        "outputId": "86b6aedf-27cc-4903-d6d7-bd7298ac8c47"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Charified text:\n",
            "b a   c a n   a f f e c t   o u t c o m e   a f t e r   b i f u r c a t i o n   p c i   ;   @ d   a n g i o g r a p h i c   a n a l y s i s   p r o v i d e s   r e l i a b l e   b a   m e a s u r e m e n t s   .\n",
            "\n",
            "Length of chars: 91\n",
            "\n",
            "Vectorized chars:\n",
            "[[22  5 11  5  6  5 17 17  2 11  3  7 16  3 11  7 15  2  5 17  3  2  8 22\n",
            "   4 17 16  8 11  5  3  4  7  6 14 11  4 10  5  6 18  4  7 18  8  5 14 13\n",
            "   4 11  5  6  5 12 19  9  4  9 14  8  7 21  4 10  2  9  8  2 12  4  5 22\n",
            "  12  2 22  5 15  2  5  9 16  8  2 15  2  6  3  9  0  0  0  0  0  0  0  0\n",
            "   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "   0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
            "   0  0]]\n",
            "\n",
            "Length of vectorized chars: 290\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "char_embed = layers.Embedding(input_dim = NUM_CHAR_TOKENS,\n",
        "                              output_dim = 25,\n",
        "                              mask_zero = False,\n",
        "                              name = 'char_embed')\n",
        "print(f\"Charified text (before vectorization and embedding):\\n{random_train_chars}\\n\")\n",
        "char_embed_example = char_embed(char_vectorizer([random_train_chars]))\n",
        "print(f\"Embedded chars (after vectorization and embedding):\\n{char_embed_example}\\n\")\n",
        "print(f\"Character embedding shape: {char_embed_example.shape}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ex5S-58c5hKH",
        "outputId": "2b18c986-2452-45a0-e5c9-ef0e2dda8c5f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Charified text (before vectorization and embedding):\n",
            "b a   c a n   a f f e c t   o u t c o m e   a f t e r   b i f u r c a t i o n   p c i   ;   @ d   a n g i o g r a p h i c   a n a l y s i s   p r o v i d e s   r e l i a b l e   b a   m e a s u r e m e n t s   .\n",
            "\n",
            "Embedded chars (after vectorization and embedding):\n",
            "[[[ 0.01112474  0.03134895 -0.01081488 ... -0.03502323 -0.04227835\n",
            "    0.01871593]\n",
            "  [ 0.03765715 -0.04736936  0.03247071 ... -0.04509039 -0.02425269\n",
            "    0.03038658]\n",
            "  [ 0.00924825 -0.02854488 -0.04508573 ... -0.03506434 -0.02848865\n",
            "   -0.0113415 ]\n",
            "  ...\n",
            "  [ 0.01904103 -0.01485657 -0.04216234 ... -0.02119598 -0.02951763\n",
            "   -0.01226204]\n",
            "  [ 0.01904103 -0.01485657 -0.04216234 ... -0.02119598 -0.02951763\n",
            "   -0.01226204]\n",
            "  [ 0.01904103 -0.01485657 -0.04216234 ... -0.02119598 -0.02951763\n",
            "   -0.01226204]]]\n",
            "\n",
            "Character embedding shape: (1, 290, 25)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "inputs = layers.Input(shape=(1,), dtype=\"string\")\n",
        "char_vectors = char_vectorizer(inputs)\n",
        "char_embeddings = char_embed(char_vectors)\n",
        "x = layers.Conv1D(64, kernel_size=5, padding=\"same\", activation=\"relu\")(char_embeddings)\n",
        "x = layers.GlobalMaxPool1D()(x)\n",
        "outputs = layers.Dense(num_classes, activation=\"softmax\")(x)\n",
        "model_3 = tf.keras.Model(inputs=inputs,\n",
        "                         outputs=outputs,\n",
        "                         name=\"model_3_conv1D_char_embedding\")\n",
        "\n",
        "# Compile model\n",
        "model_3.compile(loss=\"categorical_crossentropy\",\n",
        "                optimizer=tf.keras.optimizers.Adam(),\n",
        "                metrics=[\"accuracy\"])"
      ],
      "metadata": {
        "id": "jDeeiAWL7aU9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_3.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "og5ik6e986jR",
        "outputId": "c35b5c95-ebef-4442-8071-f0c4a630ddd5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_3_conv1D_char_embedding\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " input_3 (InputLayer)        [(None, 1)]               0         \n",
            "                                                                 \n",
            " char_vectorizer (TextVector  (None, 290)              0         \n",
            " ization)                                                        \n",
            "                                                                 \n",
            " char_embed (Embedding)      (None, 290, 25)           1750      \n",
            "                                                                 \n",
            " conv1d_1 (Conv1D)           (None, 290, 64)           8064      \n",
            "                                                                 \n",
            " global_max_pooling1d (Globa  (None, 64)               0         \n",
            " lMaxPooling1D)                                                  \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 5)                 325       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 10,139\n",
            "Trainable params: 10,139\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_char_dataset = tf.data.Dataset.from_tensor_slices((train_chars, train_labels_one_hot)).batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "val_char_dataset = tf.data.Dataset.from_tensor_slices((val_chars, val_labels_one_hot)).batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "\n",
        "train_char_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ApHAiwrR9FHA",
        "outputId": "07748475-6ce5-4911-e5fd-b39d8a963191"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<PrefetchDataset element_spec=(TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>"
            ]
          },
          "metadata": {},
          "execution_count": 80
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3_history = model_3.fit(train_char_dataset,\n",
        "                              steps_per_epoch=int(0.1 * len(train_char_dataset)),\n",
        "                              epochs=10,\n",
        "                              validation_data=val_char_dataset,\n",
        "                              validation_steps=int(0.1 * len(val_char_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TDG_qaUx99JW",
        "outputId": "f12e02b0-937c-48af-e80e-518325e89b2b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "562/562 [==============================] - 4s 5ms/step - loss: 1.2506 - accuracy: 0.4917 - val_loss: 1.0543 - val_accuracy: 0.5791\n",
            "Epoch 2/10\n",
            "562/562 [==============================] - 3s 5ms/step - loss: 1.0228 - accuracy: 0.5895 - val_loss: 0.9661 - val_accuracy: 0.6213\n",
            "Epoch 3/10\n",
            "562/562 [==============================] - 3s 5ms/step - loss: 0.9457 - accuracy: 0.6269 - val_loss: 0.8919 - val_accuracy: 0.6569\n",
            "Epoch 4/10\n",
            "562/562 [==============================] - 3s 5ms/step - loss: 0.8861 - accuracy: 0.6580 - val_loss: 0.8542 - val_accuracy: 0.6782\n",
            "Epoch 5/10\n",
            "562/562 [==============================] - 3s 5ms/step - loss: 0.8642 - accuracy: 0.6642 - val_loss: 0.8263 - val_accuracy: 0.6875\n",
            "Epoch 6/10\n",
            "562/562 [==============================] - 3s 5ms/step - loss: 0.8404 - accuracy: 0.6797 - val_loss: 0.7921 - val_accuracy: 0.6945\n",
            "Epoch 7/10\n",
            "562/562 [==============================] - 4s 6ms/step - loss: 0.8262 - accuracy: 0.6822 - val_loss: 0.7798 - val_accuracy: 0.7011\n",
            "Epoch 8/10\n",
            "562/562 [==============================] - 5s 8ms/step - loss: 0.7933 - accuracy: 0.7011 - val_loss: 0.7726 - val_accuracy: 0.7061\n",
            "Epoch 9/10\n",
            "562/562 [==============================] - 5s 9ms/step - loss: 0.7783 - accuracy: 0.7060 - val_loss: 0.7743 - val_accuracy: 0.7061\n",
            "Epoch 10/10\n",
            "562/562 [==============================] - 5s 8ms/step - loss: 0.7839 - accuracy: 0.6994 - val_loss: 0.7382 - val_accuracy: 0.7241\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3.evaluate(val_char_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wPgCQd6H-L8S",
        "outputId": "abe1e588-e5c0-46ee-fa16-1061797d134f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 5s 5ms/step - loss: 0.7577 - accuracy: 0.7155\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.7576954960823059, 0.715510368347168]"
            ]
          },
          "metadata": {},
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3_pred_probs = model_3.predict(val_char_dataset)\n",
        "model_3_pred_probs"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hEEYa1Bm_V0P",
        "outputId": "e84aedd8-e7ee-48fe-bccd-f82a8df7f3df"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.2949905 , 0.4525005 , 0.01121626, 0.23792951, 0.00336324],\n",
              "       [0.12819879, 0.77970374, 0.00267248, 0.08471695, 0.00470801],\n",
              "       [0.21219814, 0.13365509, 0.05930118, 0.57733274, 0.01751287],\n",
              "       ...,\n",
              "       [0.00462163, 0.02148993, 0.03559697, 0.00306571, 0.9352257 ],\n",
              "       [0.02007607, 0.151781  , 0.05093847, 0.00785238, 0.7693521 ],\n",
              "       [0.11349985, 0.8371836 , 0.02861507, 0.00733084, 0.01337076]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 83
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3_preds = tf.argmax(model_3_pred_probs, axis=1)\n",
        "model_3_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Lx80y8IO_X_u",
        "outputId": "ff1790c5-fd34-46d0-bde7-c89e3997da92"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(30212,), dtype=int64, numpy=array([1, 1, 3, ..., 4, 4, 1])>"
            ]
          },
          "metadata": {},
          "execution_count": 84
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3_results = calculate_results(y_true=val_labels_encoded,\n",
        "                                        y_pred=model_3_preds)\n",
        "model_3_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FZFtCSyN_bFj",
        "outputId": "f6e3a3b4-7f91-49b1-c29f-a424adb1c6b2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 71.55103932212366,\n",
              " 'f1': 0.7087937584294924,\n",
              " 'precision': 0.7070884770687382,\n",
              " 'recall': 0.7155103932212366}"
            ]
          },
          "metadata": {},
          "execution_count": 85
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Pretrained char embeds plus token embeds"
      ],
      "metadata": {
        "id": "JS-yQIAh_c8T"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "token_inputs = layers.Input(shape = [], dtype = tf.string, name = 'token_input')\n",
        "token_embeddings = tf_hub_embedding_layer(token_inputs)\n",
        "token_output = layers.Dense(128, activation='relu')(token_embeddings)\n",
        "token_model = tf.keras.Model(inputs = token_inputs, outputs = token_output)\n",
        "\n",
        "char_inputs = layers.Input(shape=(1,), dtype = tf.string, name = 'char_input')\n",
        "char_vectors = char_vectorizer(char_inputs)\n",
        "char_embeddings = char_embed(char_vectors)\n",
        "char_bi_lstm = layers.Bidirectional(layers.LSTM(25))(char_embeddings)\n",
        "char_model = tf.keras.Model(inputs=char_inputs,\n",
        "                            outputs = char_bi_lstm)\n",
        "\n",
        "token_char_concat = layers.Concatenate(name = 'token_char_hybrid')([token_model.output, char_model.output])\n",
        "\n",
        "combined_dropout = layers.Dropout(0.2)(token_char_concat)\n",
        "combined_dense = layers.Dense(200, activation=\"relu\")(combined_dropout)\n",
        "final_dropout = layers.Dropout(0.2)(combined_dense)\n",
        "output_layer = layers.Dense(num_classes, activation = \"softmax\")(final_dropout)\n",
        "\n",
        "\n",
        "model_4 = tf.keras.Model(inputs = [token_model.input, char_model.input],\n",
        "                       outputs = output_layer,\n",
        "                       name = 'model_4_token_and_char_embeddings')\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "vZs-UFbhDZNi"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_4.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z608CajERf6q",
        "outputId": "cd7dc69c-de0f-4094-aefb-60ed4d835c09"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_4_token_and_char_embeddings\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " char_input (InputLayer)        [(None, 1)]          0           []                               \n",
            "                                                                                                  \n",
            " token_input (InputLayer)       [(None,)]            0           []                               \n",
            "                                                                                                  \n",
            " char_vectorizer (TextVectoriza  (None, 290)         0           ['char_input[0][0]']             \n",
            " tion)                                                                                            \n",
            "                                                                                                  \n",
            " universal_sentence_encoder (Ke  (None, 512)         256797824   ['token_input[0][0]']            \n",
            " rasLayer)                                                                                        \n",
            "                                                                                                  \n",
            " char_embed (Embedding)         (None, 290, 25)      1750        ['char_vectorizer[1][0]']        \n",
            "                                                                                                  \n",
            " dense_4 (Dense)                (None, 128)          65664       ['universal_sentence_encoder[1][0\n",
            "                                                                 ]']                              \n",
            "                                                                                                  \n",
            " bidirectional (Bidirectional)  (None, 50)           10200       ['char_embed[1][0]']             \n",
            "                                                                                                  \n",
            " token_char_hybrid (Concatenate  (None, 178)         0           ['dense_4[0][0]',                \n",
            " )                                                                'bidirectional[0][0]']          \n",
            "                                                                                                  \n",
            " dropout (Dropout)              (None, 178)          0           ['token_char_hybrid[0][0]']      \n",
            "                                                                                                  \n",
            " dense_5 (Dense)                (None, 200)          35800       ['dropout[0][0]']                \n",
            "                                                                                                  \n",
            " dropout_1 (Dropout)            (None, 200)          0           ['dense_5[0][0]']                \n",
            "                                                                                                  \n",
            " dense_6 (Dense)                (None, 5)            1005        ['dropout_1[0][0]']              \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 256,912,243\n",
            "Trainable params: 114,419\n",
            "Non-trainable params: 256,797,824\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.utils import plot_model\n",
        "plot_model(model_4)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 833
        },
        "id": "VyQo8oFRRybE",
        "outputId": "c6b2329b-6093-48d7-9922-6b2dc06810ea"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<IPython.core.display.Image object>"
            ],
            "image/png": "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\n"
          },
          "metadata": {},
          "execution_count": 88
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_model(model_1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "ljph-mPgR9Is",
        "outputId": "1a751e03-bed3-4634-e18f-705de4519159"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<IPython.core.display.Image object>"
            ],
            "image/png": "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\n"
          },
          "metadata": {},
          "execution_count": 89
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_model(model_2)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 369
        },
        "id": "PaL1Xq8wSMeE",
        "outputId": "da39969e-54e5-41ae-fe4b-f7efd6d84515"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<IPython.core.display.Image object>"
            ],
            "image/png": "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\n"
          },
          "metadata": {},
          "execution_count": 90
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_model(model_3)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 564
        },
        "id": "PFMq3ku1STnL",
        "outputId": "7cc9ee0f-8500-4c19-f1a3-a348f465ab76"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<IPython.core.display.Image object>"
            ],
            "image/png": "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\n"
          },
          "metadata": {},
          "execution_count": 91
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_4.compile(loss = 'categorical_crossentropy',\n",
        "                optimizer = tf.keras.optimizers.Adam(),\n",
        "                metrics = ['accuracy'])"
      ],
      "metadata": {
        "id": "G0alLkraSXK8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_char_token_data = tf.data.Dataset.from_tensor_slices((train_sentences, train_chars))\n",
        "train_char_token_labels = tf.data.Dataset.from_tensor_slices(train_labels_one_hot)\n",
        "train_char_token_dataset = tf.data.Dataset.zip((train_char_token_data, train_char_token_labels))\n",
        "\n",
        "train_char_token_dataset = train_char_token_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "\n"
      ],
      "metadata": {
        "id": "EfCkjPjNSihD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "val_char_token_data = tf.data.Dataset.from_tensor_slices((val_sentences, val_chars))\n",
        "val_char_token_labels = tf.data.Dataset.from_tensor_slices(val_labels_one_hot)\n",
        "val_char_token_dataset = tf.data.Dataset.zip((val_char_token_data, val_char_token_labels))\n",
        "val_char_token_dataset = val_char_token_dataset.batch(32).prefetch(tf.data.AUTOTUNE)"
      ],
      "metadata": {
        "id": "9LlbUNy4v5_l"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_char_token_dataset, val_char_token_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m7nIT4t_wV6J",
        "outputId": "ffbb9ba4-ca7e-4ef6-d2a8-b9e91268d2ea"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<PrefetchDataset element_spec=((TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None,), dtype=tf.string, name=None)), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>,\n",
              " <PrefetchDataset element_spec=((TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None,), dtype=tf.string, name=None)), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>)"
            ]
          },
          "metadata": {},
          "execution_count": 96
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_4_history = model_4.fit(train_char_token_dataset,\n",
        "                              steps_per_epoch=int(0.1 * len(train_char_token_dataset)),\n",
        "                              epochs=10,\n",
        "                              validation_data=val_char_token_dataset,\n",
        "                              validation_steps=int(0.1 * len(val_char_token_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "te3TuQ4QwasL",
        "outputId": "298f9e30-517b-4b9c-eee2-78ce01c17796"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "562/562 [==============================] - 23s 40ms/step - loss: 0.6787 - accuracy: 0.7430 - val_loss: 0.6442 - val_accuracy: 0.7580\n",
            "Epoch 2/10\n",
            "562/562 [==============================] - 19s 33ms/step - loss: 0.6475 - accuracy: 0.7504 - val_loss: 0.6223 - val_accuracy: 0.7673\n",
            "Epoch 3/10\n",
            "562/562 [==============================] - 19s 34ms/step - loss: 0.6449 - accuracy: 0.7501 - val_loss: 0.6125 - val_accuracy: 0.7696\n",
            "Epoch 4/10\n",
            "562/562 [==============================] - 19s 33ms/step - loss: 0.6464 - accuracy: 0.7557 - val_loss: 0.6072 - val_accuracy: 0.7706\n",
            "Epoch 5/10\n",
            "562/562 [==============================] - 19s 35ms/step - loss: 0.6585 - accuracy: 0.7507 - val_loss: 0.6054 - val_accuracy: 0.7713\n",
            "Epoch 6/10\n",
            "562/562 [==============================] - 19s 34ms/step - loss: 0.6582 - accuracy: 0.7474 - val_loss: 0.5956 - val_accuracy: 0.7826\n",
            "Epoch 7/10\n",
            "562/562 [==============================] - 19s 35ms/step - loss: 0.6319 - accuracy: 0.7592 - val_loss: 0.5985 - val_accuracy: 0.7779\n",
            "Epoch 8/10\n",
            "562/562 [==============================] - 19s 34ms/step - loss: 0.6263 - accuracy: 0.7631 - val_loss: 0.5873 - val_accuracy: 0.7783\n",
            "Epoch 9/10\n",
            "562/562 [==============================] - 19s 34ms/step - loss: 0.6253 - accuracy: 0.7574 - val_loss: 0.5892 - val_accuracy: 0.7783\n",
            "Epoch 10/10\n",
            "562/562 [==============================] - 19s 34ms/step - loss: 0.6220 - accuracy: 0.7642 - val_loss: 0.5881 - val_accuracy: 0.7799\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_4.evaluate(val_char_token_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ugzBLPQEwxlD",
        "outputId": "e5971401-ff73-472a-8418-bff2e8654bd2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 20s 22ms/step - loss: 0.5855 - accuracy: 0.7771\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.5855470895767212, 0.7770753502845764]"
            ]
          },
          "metadata": {},
          "execution_count": 99
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_4_pred_probs = model_4.predict(val_char_token_dataset)\n",
        "model_4_pred_probs"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e09fCx7gx4FX",
        "outputId": "de1833a8-0391-4f41-d93a-c7a02db46d89"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[5.12424350e-01, 2.95821905e-01, 1.40130299e-03, 1.86791122e-01,\n",
              "        3.56134214e-03],\n",
              "       [3.95313352e-01, 5.12508512e-01, 2.80806585e-03, 8.82644877e-02,\n",
              "        1.10550236e-03],\n",
              "       [5.73641419e-01, 2.77354615e-03, 1.21124042e-02, 4.10227001e-01,\n",
              "        1.24561659e-03],\n",
              "       ...,\n",
              "       [6.08190167e-05, 1.03454855e-04, 2.03501992e-03, 1.43581965e-05,\n",
              "        9.97786283e-01],\n",
              "       [1.75771397e-02, 1.19096681e-01, 5.30819334e-02, 4.58072731e-03,\n",
              "        8.05663526e-01],\n",
              "       [9.05806758e-03, 9.84305978e-01, 4.69435425e-03, 1.72536864e-04,\n",
              "        1.76910730e-03]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 100
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_4_preds = tf.argmax(model_4_pred_probs, axis=1)\n",
        "model_4_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9yXO5jE_yEYA",
        "outputId": "907ac1ca-6e6c-464c-e177-189b17d35f69"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(30212,), dtype=int64, numpy=array([0, 1, 0, ..., 4, 4, 1])>"
            ]
          },
          "metadata": {},
          "execution_count": 101
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_4_results = calculate_results(y_true=val_labels_encoded,\n",
        "                                    y_pred=model_4_preds)\n",
        "model_4_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "g73AyL17yLmq",
        "outputId": "d0ee7a4b-7f80-4683-cae2-6878c6b82684"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 77.707533430425,\n",
              " 'f1': 0.7735720227344046,\n",
              " 'precision': 0.7745035948165857,\n",
              " 'recall': 0.77707533430425}"
            ]
          },
          "metadata": {},
          "execution_count": 106
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "val_labels_encoded"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AWq1Fls7y7uH",
        "outputId": "915d77db-495a-470b-f74c-2a74a0da9743"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 0, 3, ..., 4, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Trasfer learning with pretrained token embeddings + character embeddings + positional embeddings"
      ],
      "metadata": {
        "id": "BiV4wfxYzCHm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 302
        },
        "id": "IMpytqpkAP6C",
        "outputId": "845e097c-56a9-4003-cfd5-28cf7ef0c202"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "      target                                               text  line_number  \\\n",
              "0  OBJECTIVE  to investigate the efficacy of @ weeks of dail...            0   \n",
              "1    METHODS  a total of @ patients with primary knee oa wer...            1   \n",
              "2    METHODS  outcome measures included pain reduction and i...            2   \n",
              "3    METHODS  pain was assessed using the visual analog pain...            3   \n",
              "4    METHODS  secondary outcome measures included the wester...            4   \n",
              "\n",
              "   total_lines  \n",
              "0           11  \n",
              "1           11  \n",
              "2           11  \n",
              "3           11  \n",
              "4           11  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-6a740d01-7c63-4b85-9e01-7680e6e3e451\">\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>target</th>\n",
              "      <th>text</th>\n",
              "      <th>line_number</th>\n",
              "      <th>total_lines</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>OBJECTIVE</td>\n",
              "      <td>to investigate the efficacy of @ weeks of dail...</td>\n",
              "      <td>0</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>a total of @ patients with primary knee oa wer...</td>\n",
              "      <td>1</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>outcome measures included pain reduction and i...</td>\n",
              "      <td>2</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>pain was assessed using the visual analog pain...</td>\n",
              "      <td>3</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>METHODS</td>\n",
              "      <td>secondary outcome measures included the wester...</td>\n",
              "      <td>4</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-6a740d01-7c63-4b85-9e01-7680e6e3e451')\"\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-6a740d01-7c63-4b85-9e01-7680e6e3e451 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-6a740d01-7c63-4b85-9e01-7680e6e3e451');\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",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 107
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df[\"line_number\"].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cEwLG5cTAz99",
        "outputId": "03b0516a-59c0-43c9-86dc-522e44592433"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0     15000\n",
              "1     15000\n",
              "2     15000\n",
              "3     15000\n",
              "4     14992\n",
              "5     14949\n",
              "6     14758\n",
              "7     14279\n",
              "8     13346\n",
              "9     11981\n",
              "10    10041\n",
              "11     7892\n",
              "12     5853\n",
              "13     4152\n",
              "14     2835\n",
              "15     1861\n",
              "16     1188\n",
              "17      751\n",
              "18      462\n",
              "19      286\n",
              "20      162\n",
              "21      101\n",
              "22       66\n",
              "23       33\n",
              "24       22\n",
              "25       14\n",
              "26        7\n",
              "27        4\n",
              "28        3\n",
              "29        1\n",
              "30        1\n",
              "Name: line_number, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 108
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_line_numbers_one_hot = tf.one_hot(train_df['line_number'].to_numpy(), depth=15)\n",
        "val_line_numbers_one_hot = tf.one_hot(val_df[\"line_number\"].to_numpy(), depth=15)\n",
        "test_line_numbers_one_hot = tf.one_hot(test_df[\"line_number\"].to_numpy(), depth=15)\n"
      ],
      "metadata": {
        "id": "4__gpw4DBBFv"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_line_numbers_one_hot.shape, train_line_numbers_one_hot[:20]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Cb0IG8SCB1nA",
        "outputId": "1f117424-0779-4dd6-9271-147effca8ff5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(TensorShape([180040, 15]), <tf.Tensor: shape=(20, 15), dtype=float32, numpy=\n",
              " array([[1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0.],\n",
              "        [1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0., 0., 0., 0.]],\n",
              "       dtype=float32)>)"
            ]
          },
          "metadata": {},
          "execution_count": 110
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df[\"total_lines\"].value_counts()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "exI2RC7nCiM1",
        "outputId": "d2699fca-e20b-4a15-d1c6-7d322a03ce1c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "11    24468\n",
              "10    23639\n",
              "12    22113\n",
              "9     19400\n",
              "13    18438\n",
              "14    14610\n",
              "8     12285\n",
              "15    10768\n",
              "7      7464\n",
              "16     7429\n",
              "17     5202\n",
              "6      3353\n",
              "18     3344\n",
              "19     2480\n",
              "20     1281\n",
              "5      1146\n",
              "21      770\n",
              "22      759\n",
              "23      264\n",
              "4       215\n",
              "24      200\n",
              "25      182\n",
              "26       81\n",
              "28       58\n",
              "3        32\n",
              "30       31\n",
              "27       28\n",
              "Name: total_lines, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 114
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.total_lines.plot.hist();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 267
        },
        "id": "5vUga-LPDBbv",
        "outputId": "ff0ea1a9-869d-471a-a9c0-c0a3ed428d92"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "np.percentile(train_df.total_lines, 98)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UlE00eS9DNov",
        "outputId": "bd42f84d-746f-42c0-e7d1-9e3505be4054"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "20.0"
            ]
          },
          "metadata": {},
          "execution_count": 116
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_total_lines_one_hot = tf.one_hot(train_df[\"total_lines\"].to_numpy(), depth=20)\n",
        "val_total_lines_one_hot = tf.one_hot(val_df[\"total_lines\"].to_numpy(), depth=20)\n",
        "test_total_lines_one_hot = tf.one_hot(test_df[\"total_lines\"].to_numpy(), depth=20)"
      ],
      "metadata": {
        "id": "FihDv6P7DQIR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_total_lines_one_hot.shape, train_total_lines_one_hot[:10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IS6leOquDS5U",
        "outputId": "a98941ca-ee38-491d-cb33-ba72eca8c1c9"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(TensorShape([180040, 20]), <tf.Tensor: shape=(10, 20), dtype=float32, numpy=\n",
              " array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.],\n",
              "        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 1., 0., 0., 0., 0.,\n",
              "         0., 0., 0., 0.]], dtype=float32)>)"
            ]
          },
          "metadata": {},
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "token_inputs = layers.Input(shape = [], dtype = \"string\", name = \"token_inputs\")\n",
        "token_embeddings = tf_hub_embedding_layer(token_inputs)\n",
        "token_outputs = layers.Dense(128, activation = \"relu\")(token_embeddings)\n",
        "token_model = tf.keras.Model(inputs = token_inputs, outputs = token_outputs)\n",
        "\n",
        "char_inputs = layers.Input(shape=(1,), dtype=\"string\", name=\"char_inputs\")\n",
        "char_vectors = char_vectorizer(char_inputs)\n",
        "char_embeddings = char_embed(char_vectors)\n",
        "char_bi_lstm = layers.Bidirectional(layers.LSTM(32))(char_embeddings)\n",
        "char_model = tf.keras.Model(char_inputs, char_bi_lstm)\n",
        "\n",
        "line_number_inputs = layers.Input(shape = (15,), dtype = tf.int32, name=\"line_number_input\")\n",
        "x = layers.Dense(32, activation=\"relu\")(line_number_inputs)\n",
        "line_number_model = tf.keras.Model(line_number_inputs, x)\n",
        "\n",
        "total_lines_inputs = layers.Input(shape=(20,), dtype = tf.int32, name = \"total_lines_input\")\n",
        "y = layers.Dense(32, activation=\"relu\")(total_lines_inputs)\n",
        "total_line_model = tf.keras.Model(total_lines_inputs, y)\n",
        "\n",
        "combined_embeddings = layers.Concatenate(name = \"token_char_hybrid_embedding\")([token_model.output, char_model.output])\n",
        "\n",
        "z = layers.Dense(256, activation=\"relu\")(combined_embeddings)\n",
        "z = layers.Dropout(0.2)(z)\n",
        "z = layers.Concatenate(name = \"token_char_positional_embedding\")([line_number_model.output, total_line_model.output,z])\n",
        "\n",
        "output_layer = layers.Dense(5, activation=\"softmax\", name=\"output_layer\")(z)\n",
        "\n",
        "model_5 = tf.keras.Model(inputs = [line_number_model.input, total_line_model.input, token_model.input,\n",
        "                                   char_model.input], outputs = output_layer)\n",
        "\n"
      ],
      "metadata": {
        "id": "ngOplVshDU59"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_5.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CjALRMMeM9Mv",
        "outputId": "12c8ac06-8467-418f-ac47-e6f8fa2a3043"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_8\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " char_inputs (InputLayer)       [(None, 1)]          0           []                               \n",
            "                                                                                                  \n",
            " token_inputs (InputLayer)      [(None,)]            0           []                               \n",
            "                                                                                                  \n",
            " char_vectorizer (TextVectoriza  (None, 290)         0           ['char_inputs[0][0]']            \n",
            " tion)                                                                                            \n",
            "                                                                                                  \n",
            " universal_sentence_encoder (Ke  (None, 512)         256797824   ['token_inputs[0][0]']           \n",
            " rasLayer)                                                                                        \n",
            "                                                                                                  \n",
            " char_embed (Embedding)         (None, 290, 25)      1750        ['char_vectorizer[2][0]']        \n",
            "                                                                                                  \n",
            " dense_7 (Dense)                (None, 128)          65664       ['universal_sentence_encoder[2][0\n",
            "                                                                 ]']                              \n",
            "                                                                                                  \n",
            " bidirectional_1 (Bidirectional  (None, 64)          14848       ['char_embed[2][0]']             \n",
            " )                                                                                                \n",
            "                                                                                                  \n",
            " token_char_hybrid_embedding (C  (None, 192)         0           ['dense_7[0][0]',                \n",
            " oncatenate)                                                      'bidirectional_1[0][0]']        \n",
            "                                                                                                  \n",
            " line_number_input (InputLayer)  [(None, 15)]        0           []                               \n",
            "                                                                                                  \n",
            " total_lines_input (InputLayer)  [(None, 20)]        0           []                               \n",
            "                                                                                                  \n",
            " dense_10 (Dense)               (None, 256)          49408       ['token_char_hybrid_embedding[0][\n",
            "                                                                 0]']                             \n",
            "                                                                                                  \n",
            " dense_8 (Dense)                (None, 32)           512         ['line_number_input[0][0]']      \n",
            "                                                                                                  \n",
            " dense_9 (Dense)                (None, 32)           672         ['total_lines_input[0][0]']      \n",
            "                                                                                                  \n",
            " dropout_2 (Dropout)            (None, 256)          0           ['dense_10[0][0]']               \n",
            "                                                                                                  \n",
            " token_char_positional_embeddin  (None, 320)         0           ['dense_8[0][0]',                \n",
            " g (Concatenate)                                                  'dense_9[0][0]',                \n",
            "                                                                  'dropout_2[0][0]']              \n",
            "                                                                                                  \n",
            " output_layer (Dense)           (None, 5)            1605        ['token_char_positional_embedding\n",
            "                                                                 [0][0]']                         \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 256,932,283\n",
            "Trainable params: 134,459\n",
            "Non-trainable params: 256,797,824\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_model(model_5)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 538
        },
        "id": "FekvHzDnM__5",
        "outputId": "dd11bf98-1c94-4327-b4ba-3963d63be00f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<IPython.core.display.Image object>"
            ],
            "image/png": "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\n"
          },
          "metadata": {},
          "execution_count": 123
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "for layer in model_5.layers:\n",
        "  print(layer, layer.trainable)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3HwTsWRnNCpN",
        "outputId": "8d477c27-32da-40f7-fea4-9bc5ec66ee7e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<keras.engine.input_layer.InputLayer object at 0x7f2debb625d0> True\n",
            "<keras.engine.input_layer.InputLayer object at 0x7f2deb9e4390> True\n",
            "<keras.layers.preprocessing.text_vectorization.TextVectorization object at 0x7f2e0cfb2450> True\n",
            "<tensorflow_hub.keras_layer.KerasLayer object at 0x7f2edaa4c350> False\n",
            "<keras.layers.embeddings.Embedding object at 0x7f2e0cf73850> True\n",
            "<keras.layers.core.dense.Dense object at 0x7f2debbce450> True\n",
            "<keras.layers.wrappers.Bidirectional object at 0x7f2debcfc210> True\n",
            "<keras.layers.merge.Concatenate object at 0x7f2e0ac69e90> True\n",
            "<keras.engine.input_layer.InputLayer object at 0x7f2deb8be7d0> True\n",
            "<keras.engine.input_layer.InputLayer object at 0x7f2debc08150> True\n",
            "<keras.layers.core.dense.Dense object at 0x7f2deba466d0> True\n",
            "<keras.layers.core.dense.Dense object at 0x7f2debb07910> True\n",
            "<keras.layers.core.dense.Dense object at 0x7f2e0aa1f490> True\n",
            "<keras.layers.core.dropout.Dropout object at 0x7f2deba46550> True\n",
            "<keras.layers.merge.Concatenate object at 0x7f2deb9af890> True\n",
            "<keras.layers.core.dense.Dense object at 0x7f2e0ac697d0> True\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_5.compile(loss=tf.keras.losses.CategoricalCrossentropy(label_smoothing=0.2), \n",
        "                optimizer=tf.keras.optimizers.Adam(),\n",
        "                metrics=[\"accuracy\"])"
      ],
      "metadata": {
        "id": "sCMcWQwTNZib"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_pos_char_token_data = tf.data.Dataset.from_tensor_slices((train_line_numbers_one_hot, \n",
        "                                                                train_total_lines_one_hot, \n",
        "                                                                train_sentences, \n",
        "                                                                train_chars))\n",
        "train_pos_char_token_labels = tf.data.Dataset.from_tensor_slices(train_labels_one_hot)\n",
        "train_pos_char_token_dataset = tf.data.Dataset.zip((train_pos_char_token_data, train_pos_char_token_labels))\n",
        "train_pos_char_token_dataset = train_pos_char_token_dataset.batch(32).prefetch(tf.data.AUTOTUNE)"
      ],
      "metadata": {
        "id": "8Bhpv3w3Nyxh"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "val_pos_char_token_data = tf.data.Dataset.from_tensor_slices((val_line_numbers_one_hot,\n",
        "                                                              val_total_lines_one_hot,\n",
        "                                                              val_sentences,\n",
        "                                                              val_chars))\n",
        "val_pos_char_token_labels = tf.data.Dataset.from_tensor_slices(val_labels_one_hot)\n",
        "val_pos_char_token_dataset = tf.data.Dataset.zip((val_pos_char_token_data, val_pos_char_token_labels))\n",
        "val_pos_char_token_dataset = val_pos_char_token_dataset.batch(32).prefetch(tf.data.AUTOTUNE)"
      ],
      "metadata": {
        "id": "ZYAEvtTYOwpF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_pos_char_token_dataset, val_pos_char_token_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S8pF1L-iPKuS",
        "outputId": "53d5af14-2240-4626-a006-0f1908232649"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<PrefetchDataset element_spec=((TensorSpec(shape=(None, 15), dtype=tf.float32, name=None), TensorSpec(shape=(None, 20), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None,), dtype=tf.string, name=None)), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>,\n",
              " <PrefetchDataset element_spec=((TensorSpec(shape=(None, 15), dtype=tf.float32, name=None), TensorSpec(shape=(None, 20), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None,), dtype=tf.string, name=None)), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>)"
            ]
          },
          "metadata": {},
          "execution_count": 129
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history_model_5 = model_5.fit(train_pos_char_token_dataset,\n",
        "                              steps_per_epoch=int(0.1 * len(train_pos_char_token_dataset)),\n",
        "                              epochs=10,\n",
        "                              validation_data=val_pos_char_token_dataset,\n",
        "                              validation_steps=int(0.1 * len(val_pos_char_token_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rSBhAsEkPO0c",
        "outputId": "29d34365-aa1b-4168-b08c-d98e4baa9972"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "562/562 [==============================] - 23s 35ms/step - loss: 1.0061 - accuracy: 0.7826 - val_loss: 0.9691 - val_accuracy: 0.8142\n",
            "Epoch 2/10\n",
            "562/562 [==============================] - 18s 33ms/step - loss: 0.9550 - accuracy: 0.8226 - val_loss: 0.9450 - val_accuracy: 0.8308\n",
            "Epoch 3/10\n",
            "562/562 [==============================] - 18s 33ms/step - loss: 0.9404 - accuracy: 0.8276 - val_loss: 0.9333 - val_accuracy: 0.8321\n",
            "Epoch 4/10\n",
            "562/562 [==============================] - 18s 33ms/step - loss: 0.9310 - accuracy: 0.8385 - val_loss: 0.9248 - val_accuracy: 0.8391\n",
            "Epoch 5/10\n",
            "562/562 [==============================] - 18s 33ms/step - loss: 0.9294 - accuracy: 0.8372 - val_loss: 0.9202 - val_accuracy: 0.8414\n",
            "Epoch 6/10\n",
            "562/562 [==============================] - 19s 33ms/step - loss: 0.9320 - accuracy: 0.8366 - val_loss: 0.9162 - val_accuracy: 0.8457\n",
            "Epoch 7/10\n",
            "562/562 [==============================] - 19s 33ms/step - loss: 0.9204 - accuracy: 0.8419 - val_loss: 0.9208 - val_accuracy: 0.8381\n",
            "Epoch 8/10\n",
            "562/562 [==============================] - 18s 33ms/step - loss: 0.9156 - accuracy: 0.8460 - val_loss: 0.9045 - val_accuracy: 0.8507\n",
            "Epoch 9/10\n",
            "562/562 [==============================] - 18s 33ms/step - loss: 0.9166 - accuracy: 0.8469 - val_loss: 0.9171 - val_accuracy: 0.8434\n",
            "Epoch 10/10\n",
            "562/562 [==============================] - 18s 33ms/step - loss: 0.9136 - accuracy: 0.8485 - val_loss: 0.9069 - val_accuracy: 0.8461\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "history_full_model_5 = model_5.fit(train_pos_char_token_dataset,\n",
        "                              steps_per_epoch=int( len(train_pos_char_token_dataset)),\n",
        "                              epochs=10,\n",
        "                              validation_data=val_pos_char_token_dataset,\n",
        "                              validation_steps=int( len(val_pos_char_token_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vGdHvZSNQorF",
        "outputId": "6bf112bc-2b03-4a2c-b929-7d5653c4ad4a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "5627/5627 [==============================] - 186s 33ms/step - loss: 0.9010 - accuracy: 0.8571 - val_loss: 0.8970 - val_accuracy: 0.8547\n",
            "Epoch 2/10\n",
            "5627/5627 [==============================] - 187s 33ms/step - loss: 0.8879 - accuracy: 0.8664 - val_loss: 0.8941 - val_accuracy: 0.8572\n",
            "Epoch 3/10\n",
            "5627/5627 [==============================] - 186s 33ms/step - loss: 0.8782 - accuracy: 0.8738 - val_loss: 0.8956 - val_accuracy: 0.8557\n",
            "Epoch 4/10\n",
            "5627/5627 [==============================] - 187s 33ms/step - loss: 0.8706 - accuracy: 0.8793 - val_loss: 0.8952 - val_accuracy: 0.8565\n",
            "Epoch 5/10\n",
            "5627/5627 [==============================] - 188s 33ms/step - loss: 0.8640 - accuracy: 0.8848 - val_loss: 0.8964 - val_accuracy: 0.8554\n",
            "Epoch 6/10\n",
            "5627/5627 [==============================] - 189s 34ms/step - loss: 0.8581 - accuracy: 0.8885 - val_loss: 0.8973 - val_accuracy: 0.8543\n",
            "Epoch 7/10\n",
            "5627/5627 [==============================] - 190s 34ms/step - loss: 0.8531 - accuracy: 0.8929 - val_loss: 0.8978 - val_accuracy: 0.8557\n",
            "Epoch 8/10\n",
            "5627/5627 [==============================] - 191s 34ms/step - loss: 0.8487 - accuracy: 0.8965 - val_loss: 0.8961 - val_accuracy: 0.8565\n",
            "Epoch 9/10\n",
            "5627/5627 [==============================] - 191s 34ms/step - loss: 0.8444 - accuracy: 0.9000 - val_loss: 0.9028 - val_accuracy: 0.8528\n",
            "Epoch 10/10\n",
            "5627/5627 [==============================] - 191s 34ms/step - loss: 0.8402 - accuracy: 0.9037 - val_loss: 0.9029 - val_accuracy: 0.8538\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_5.save(\"pos_char_token_model\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2w1k3orMTdWs",
        "outputId": "e2bd419f-722c-451d-8f3f-cc213c6dd6cd"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:Found untraced functions such as lstm_cell_4_layer_call_fn, lstm_cell_4_layer_call_and_return_conditional_losses, lstm_cell_5_layer_call_fn, lstm_cell_5_layer_call_and_return_conditional_losses while saving (showing 4 of 4). These functions will not be directly callable after loading.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "INFO:tensorflow:Assets written to: pos_char_token_model/assets\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:tensorflow:Assets written to: pos_char_token_model/assets\n",
            "WARNING:absl:<keras.layers.recurrent.LSTMCell object at 0x7f2deba08f90> has the same name 'LSTMCell' as a built-in Keras object. Consider renaming <class 'keras.layers.recurrent.LSTMCell'> to avoid naming conflicts when loading with `tf.keras.models.load_model`. If renaming is not possible, pass the object in the `custom_objects` parameter of the load function.\n",
            "WARNING:absl:<keras.layers.recurrent.LSTMCell object at 0x7f2deb8c5710> has the same name 'LSTMCell' as a built-in Keras object. Consider renaming <class 'keras.layers.recurrent.LSTMCell'> to avoid naming conflicts when loading with `tf.keras.models.load_model`. If renaming is not possible, pass the object in the `custom_objects` parameter of the load function.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_5.save(\"pos_char_token_model.pkl\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "K0_ZYsGbfCjt",
        "outputId": "d1ca92ae-4403-4499-f108-f262b497f069"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:Found untraced functions such as lstm_cell_4_layer_call_fn, lstm_cell_4_layer_call_and_return_conditional_losses, lstm_cell_5_layer_call_fn, lstm_cell_5_layer_call_and_return_conditional_losses while saving (showing 4 of 4). These functions will not be directly callable after loading.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "INFO:tensorflow:Assets written to: pos_char_token_model.pkl/assets\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "INFO:tensorflow:Assets written to: pos_char_token_model.pkl/assets\n",
            "WARNING:absl:<keras.layers.recurrent.LSTMCell object at 0x7f2deba08f90> has the same name 'LSTMCell' as a built-in Keras object. Consider renaming <class 'keras.layers.recurrent.LSTMCell'> to avoid naming conflicts when loading with `tf.keras.models.load_model`. If renaming is not possible, pass the object in the `custom_objects` parameter of the load function.\n",
            "WARNING:absl:<keras.layers.recurrent.LSTMCell object at 0x7f2deb8c5710> has the same name 'LSTMCell' as a built-in Keras object. Consider renaming <class 'keras.layers.recurrent.LSTMCell'> to avoid naming conflicts when loading with `tf.keras.models.load_model`. If renaming is not possible, pass the object in the `custom_objects` parameter of the load function.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!cp pos_char_token_model -r /content/drive/MyDrive/Sravanth/"
      ],
      "metadata": {
        "id": "J_dWQUxefXor"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sEwmmpAEtB1y",
        "outputId": "bac250d2-f0ce-41c1-bb54-a3f3bec20833"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow_hub as hub\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.layers import TextVectorization\n",
        "\n",
        "model_path = \"pos_char_token_model\"\n",
        "\n",
        "# Load downloaded model from Google Storage\n",
        "loaded_model = tf.keras.models.load_model(model_path)#"
      ],
      "metadata": {
        "id": "LmIF0Jxkts6a"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "loaded_pred_probs = loaded_model.predict(val_pos_char_token_dataset, verbose=1)\n",
        "loaded_preds = tf.argmax(loaded_pred_probs, axis=1)\n",
        "loaded_preds[:10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PAyDuFATuRTL",
        "outputId": "ff540322-607a-4c1c-8289-511e7ca09fe4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 35s 35ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(10,), dtype=int64, numpy=array([0, 0, 3, 2, 2, 4, 4, 4, 4, 1])>"
            ]
          },
          "metadata": {},
          "execution_count": 144
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loaded_model_results = calculate_results(val_labels_encoded,\n",
        "                                         loaded_preds)\n",
        "loaded_model_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NOMOjL2euXmO",
        "outputId": "60e80ab6-9a27-4c53-9f11-625e284d46f4"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 85.37667152124983,\n",
              " 'f1': 0.8512136321062118,\n",
              " 'precision': 0.8546920735990701,\n",
              " 'recall': 0.8537667152124984}"
            ]
          },
          "metadata": {},
          "execution_count": 145
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loaded_model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "41LL2VQWulXd",
        "outputId": "f85e5cd2-d878-4fbf-d586-f0926245d132"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_8\"\n",
            "__________________________________________________________________________________________________\n",
            " Layer (type)                   Output Shape         Param #     Connected to                     \n",
            "==================================================================================================\n",
            " char_inputs (InputLayer)       [(None, 1)]          0           []                               \n",
            "                                                                                                  \n",
            " token_inputs (InputLayer)      [(None,)]            0           []                               \n",
            "                                                                                                  \n",
            " char_vectorizer (TextVectoriza  (None, 290)         0           ['char_inputs[0][0]']            \n",
            " tion)                                                                                            \n",
            "                                                                                                  \n",
            " universal_sentence_encoder (Ke  (None, 512)         256797824   ['token_inputs[0][0]']           \n",
            " rasLayer)                                                                                        \n",
            "                                                                                                  \n",
            " char_embed (Embedding)         (None, 290, 25)      1750        ['char_vectorizer[0][0]']        \n",
            "                                                                                                  \n",
            " dense_7 (Dense)                (None, 128)          65664       ['universal_sentence_encoder[0][0\n",
            "                                                                 ]']                              \n",
            "                                                                                                  \n",
            " bidirectional_1 (Bidirectional  (None, 64)          14848       ['char_embed[0][0]']             \n",
            " )                                                                                                \n",
            "                                                                                                  \n",
            " token_char_hybrid_embedding (C  (None, 192)         0           ['dense_7[0][0]',                \n",
            " oncatenate)                                                      'bidirectional_1[0][0]']        \n",
            "                                                                                                  \n",
            " line_number_input (InputLayer)  [(None, 15)]        0           []                               \n",
            "                                                                                                  \n",
            " total_lines_input (InputLayer)  [(None, 20)]        0           []                               \n",
            "                                                                                                  \n",
            " dense_10 (Dense)               (None, 256)          49408       ['token_char_hybrid_embedding[0][\n",
            "                                                                 0]']                             \n",
            "                                                                                                  \n",
            " dense_8 (Dense)                (None, 32)           512         ['line_number_input[0][0]']      \n",
            "                                                                                                  \n",
            " dense_9 (Dense)                (None, 32)           672         ['total_lines_input[0][0]']      \n",
            "                                                                                                  \n",
            " dropout_2 (Dropout)            (None, 256)          0           ['dense_10[0][0]']               \n",
            "                                                                                                  \n",
            " token_char_positional_embeddin  (None, 320)         0           ['dense_8[0][0]',                \n",
            " g (Concatenate)                                                  'dense_9[0][0]',                \n",
            "                                                                  'dropout_2[0][0]']              \n",
            "                                                                                                  \n",
            " output_layer (Dense)           (None, 5)            1605        ['token_char_positional_embedding\n",
            "                                                                 [0][0]']                         \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 256,932,283\n",
            "Trainable params: 134,459\n",
            "Non-trainable params: 256,797,824\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_pos_char_token_data = tf.data.Dataset.from_tensor_slices((test_line_numbers_one_hot,\n",
        "                                                               test_total_lines_one_hot,\n",
        "                                                               test_sentences,\n",
        "                                                               test_chars))\n",
        "test_pos_char_token_labels = tf.data.Dataset.from_tensor_slices(test_labels_one_hot)\n",
        "test_pos_char_token_dataset = tf.data.Dataset.zip((test_pos_char_token_data, test_pos_char_token_labels))\n",
        "test_pos_char_token_dataset = test_pos_char_token_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "\n",
        "# Check shapes\n",
        "test_pos_char_token_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "067Cs8mJun2v",
        "outputId": "6f492c22-a54e-4db2-ac0a-dd1b74c32c75"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<PrefetchDataset element_spec=((TensorSpec(shape=(None, 15), dtype=tf.float32, name=None), TensorSpec(shape=(None, 20), dtype=tf.float32, name=None), TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None,), dtype=tf.string, name=None)), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>"
            ]
          },
          "metadata": {},
          "execution_count": 147
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_pred_probs = loaded_model.predict(test_pos_char_token_dataset,\n",
        "                                       verbose=1)\n",
        "test_preds = tf.argmax(test_pred_probs, axis=1)\n",
        "test_preds[:10]"
      ],
      "metadata": {
        "id": "NTdV0yOtuv49",
        "outputId": "1da138ba-2648-4bf1-c1d3-c8a8df03b728",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "942/942 [==============================] - 21s 22ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(10,), dtype=int64, numpy=array([3, 2, 2, 2, 4, 4, 4, 1, 1, 0])>"
            ]
          },
          "metadata": {},
          "execution_count": 148
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loaded_model_test_results = calculate_results(y_true=test_labels_encoded,\n",
        "                                              y_pred=test_preds)\n",
        "loaded_model_test_results"
      ],
      "metadata": {
        "id": "YQAZqNJuuxYz",
        "outputId": "47c09547-0017-4390-8980-d3e6a9d84837",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 84.9145511863282,\n",
              " 'f1': 0.8465995590060523,\n",
              " 'precision': 0.8492134832630014,\n",
              " 'recall': 0.8491455118632819}"
            ]
          },
          "metadata": {},
          "execution_count": 149
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Transfer Learning with bert will be on new notebook.\n"
      ],
      "metadata": {
        "id": "y_Q-8pq1EE2v"
      }
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "KbKnoyoyuzxR"
      },
      "execution_count": 160,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "tz2TQaE9BmKt"
      },
      "execution_count": 160,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "nroVwPawBuMo"
      },
      "execution_count": 160,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "cg9_Bnf0EMM0"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}