project_2_skimlit.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "mount_file_id": "1egxe8r_FL6cCNofoLJOMRp40yGiya4v1",
      "authorship_tag": "ABX9TyM4b8b1oTEg3u3eXzsg2u/z",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/RishitSingh10/SkimLit/blob/main/project_2_skimlit.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Project Description:\n",
        "\n",
        "The purpose of this project is to build an NLP model to make reading medical abstracts easier.\n",
        "\n",
        "The paper we're replicating (the source of the dataset that we'll be using) is available here: https://arxiv.org/abs/1710.06071\n",
        "\n",
        "And reading through the paper above, we see that the model architecture that they used to achieve the best results is available here: https://arxiv.org/abs/1612.05251"
      ],
      "metadata": {
        "id": "BOOVM8FsIsgj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import tensorflow as tf\n",
        "import numpy as np"
      ],
      "metadata": {
        "id": "xdmXdlNR8lhs"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Get data\n",
        "\n",
        "Since we'll be replicating the paper above (PubMed 200k RCT), let's download the dataset they used.\n",
        "\n",
        "We can do so from the authors GitHub: https://github.com/Franck-Dernoncourt/pubmed-rct"
      ],
      "metadata": {
        "id": "J3SdVuTgUp8E"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://github.com/Franck-Dernoncourt/pubmed-rct\n",
        "!ls pubmed-rct"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IlqzgWwDUzaQ",
        "outputId": "da423548-6867-4da4-c34b-34fca76e296b"
      },
      "execution_count": 2,
      "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",
            "Receiving objects: 100% (33/33), 177.08 MiB | 41.67 MiB/s, done.\n",
            "Resolving deltas: 100% (12/12), done.\n",
            "PubMed_200k_RCT\t\t\t\t       PubMed_20k_RCT_numbers_replaced_with_at_sign\n",
            "PubMed_200k_RCT_numbers_replaced_with_at_sign  README.md\n",
            "PubMed_20k_RCT\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Check what files are in the PubMed_20K dataset\n",
        "!ls pubmed-rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IoA3K0DRVpBb",
        "outputId": "c3f7d932-b66a-44bf-8d73-891012f50c22"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "dev.txt  test.txt  train.txt\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# dev.txt is same as validation set\n",
        "\n",
        "# Starting experiments using the 20k dataset with numbers replaced by \"@\" sign\n",
        "data_dir = \"/content/pubmed-rct/PubMed_20k_RCT_numbers_replaced_with_at_sign\""
      ],
      "metadata": {
        "id": "PClGGjKYWvuL"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Check all of the filenames in the target directory\n",
        "import os\n",
        "filenames = [data_dir + '/' + filename for filename in os.listdir(data_dir)]\n",
        "filenames"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8X3qmo3_Xv1G",
        "outputId": "77e4a27e-e60c-4a4c-cdfd-2dd7a3c66551"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['/content/pubmed-rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/dev.txt',\n",
              " '/content/pubmed-rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/test.txt',\n",
              " '/content/pubmed-rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/train.txt']"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Preprocess data\n",
        "\n",
        "Exploring the text data through visualisation."
      ],
      "metadata": {
        "id": "tkzPLoH_YB08"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# reading in all the lines of a target text file\n",
        "file = open(data_dir + '/train.txt' , 'r')\n",
        "train_text_list = file.readlines()"
      ],
      "metadata": {
        "id": "QVP39uY8aEj9"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(train_text_list[:10][1])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L3qq-1OObELQ",
        "outputId": "ecdcc369-9a3d-441e-fba6-f3386c6b9060"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "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"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def get_lines(filename):\n",
        "  \"\"\"\n",
        "  Reads filename (a text filename) and returns the lines of text as a list.\n",
        "\n",
        "  Args:\n",
        "    filename: a string containing the target filepath\n",
        "\n",
        "  Returns:\n",
        "    A list of strings with one string per line from the target from the target filename\n",
        "  \"\"\"\n",
        "  with open(filename, \"r\") as f:\n",
        "    return f.readlines()"
      ],
      "metadata": {
        "id": "BNd5m3G0bGRQ"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Let's read in the training lines\n",
        "train_lines = get_lines(data_dir + '/train.txt') # read the lines within the training file\n",
        "train_lines[:20]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VWiJqUaqdnlf",
        "outputId": "6aeb21e6-a6fa-404a-f055-58f475a97053"
      },
      "execution_count": 9,
      "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": [
        "len(train_lines)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bqQwwN9beFTT",
        "outputId": "75918feb-cbb1-4a3b-eded-b80baef6ffe6"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "210040"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_lines[1:20]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hk-0YDQ5fcSY",
        "outputId": "b6bf5e14-52d1-4299-c80d-cb4d39d2ff9e"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['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": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Our data would be best represented as:\n",
        "```\n",
        "[{'line_number' : 0,\n",
        "  'target' : 'BACKGROUND',\n",
        "  'text' : \"Emotional eating is associated with overeating and the development of obesity .\\n\"\n",
        "  'total_lines' : 11},\n",
        "  ...]\n",
        "```\n",
        "Writing a function to turn our data into the above format"
      ],
      "metadata": {
        "id": "NfubvKSbYGO_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def preprocess(filename):\n",
        "\n",
        "  lines = get_lines(filename)\n",
        "\n",
        "  abstract_lines = \"\"\n",
        "  abstract_samples = []\n",
        "\n",
        "  for line in lines:\n",
        "\n",
        "    if line.startswith(\"###\"):\n",
        "      abstract_ID = line\n",
        "      abstract_lines = \"\"\n",
        "\n",
        "    elif line.isspace(): # new abstract found\n",
        "\n",
        "      abstract_line_split = abstract_lines.splitlines()\n",
        "\n",
        "      for abstract_line_number, abstract_line in enumerate(abstract_line_split):\n",
        "\n",
        "        line_data = {}\n",
        "        target_line_split = abstract_line.split('\\t')\n",
        "\n",
        "        line_data['line_number'] = abstract_line_number\n",
        "        line_data['label'] = target_line_split[0]\n",
        "        line_data['text'] = target_line_split[1]\n",
        "        line_data['total_lines'] = len(abstract_line_split)-1\n",
        "        abstract_samples.append(line_data)\n",
        "\n",
        "    else:\n",
        "      abstract_lines+=line\n",
        "\n",
        "  return abstract_samples"
      ],
      "metadata": {
        "id": "oQpc36J6jy8k"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Get data from the file and preprocess it\n",
        "%%time\n",
        "train_samples = preprocess(data_dir + \"/train.txt\")\n",
        "val_samples = preprocess(data_dir + \"/dev.txt\") # dev is another name for valid\n",
        "test_samples = preprocess(data_dir + \"/test.txt\")\n",
        "print(len(train_samples), len(val_samples), len(test_samples))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hA5mSl1I8Eej",
        "outputId": "133e749b-aedc-4a9d-976b-6caea76f26b0"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "180040 30212 30135\n",
            "CPU times: user 553 ms, sys: 114 ms, total: 667 ms\n",
            "Wall time: 852 ms\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Check the first abstract of our training data\n",
        "train_samples[:10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "I6m7qXtg8HZN",
        "outputId": "2d819ca3-7067-46b6-8b1c-d3f3db7710d1"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[{'line_number': 0,\n",
              "  'label': '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",
              "  'label': '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",
              "  'label': '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",
              "  'label': 'METHODS',\n",
              "  'text': 'Pain was assessed using the visual analog pain scale ( @-@ mm ) .',\n",
              "  'total_lines': 11},\n",
              " {'line_number': 4,\n",
              "  'label': '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",
              "  'label': '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",
              "  'label': '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",
              "  'label': '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",
              "  'label': '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",
              "  'label': 'RESULTS',\n",
              "  'text': 'These differences remained significant at @ weeks .',\n",
              "  'total_lines': 11}]"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Turning the list of dictionaries into a DataFrame"
      ],
      "metadata": {
        "id": "t_6kHX-5AWMR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "\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(14)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 488
        },
        "id": "4DTNDvQIQGzg",
        "outputId": "f0801796-796a-4b02-8256-74e88c97da82"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    line_number        label  \\\n",
              "0             0    OBJECTIVE   \n",
              "1             1      METHODS   \n",
              "2             2      METHODS   \n",
              "3             3      METHODS   \n",
              "4             4      METHODS   \n",
              "5             5      METHODS   \n",
              "6             6      RESULTS   \n",
              "7             7      RESULTS   \n",
              "8             8      RESULTS   \n",
              "9             9      RESULTS   \n",
              "10           10      RESULTS   \n",
              "11           11  CONCLUSIONS   \n",
              "12            0   BACKGROUND   \n",
              "13            1   BACKGROUND   \n",
              "\n",
              "                                                 text  total_lines  \n",
              "0   To investigate the efficacy of @ weeks of dail...           11  \n",
              "1   A total of @ patients with primary knee OA wer...           11  \n",
              "2   Outcome measures included pain reduction and i...           11  \n",
              "3   Pain was assessed using the visual analog pain...           11  \n",
              "4   Secondary outcome measures included the Wester...           11  \n",
              "5   Serum levels of interleukin @ ( IL-@ ) , IL-@ ...           11  \n",
              "6   There was a clinically relevant reduction in t...           11  \n",
              "7   The mean difference between treatment arms ( @...           11  \n",
              "8   Further , there was a clinically relevant redu...           11  \n",
              "9   These differences remained significant at @ we...           11  \n",
              "10  The Outcome Measures in Rheumatology Clinical ...           11  \n",
              "11  Low-dose oral prednisolone had both a short-te...           11  \n",
              "12  Emotional eating is associated with overeating...           10  \n",
              "13  Yet , empirical evidence for individual ( trai...           10  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a8f69524-9d49-4723-80f4-563476603368\" 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>line_number</th>\n",
              "      <th>label</th>\n",
              "      <th>text</th>\n",
              "      <th>total_lines</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>OBJECTIVE</td>\n",
              "      <td>To investigate the efficacy of @ weeks of dail...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>A total of @ patients with primary knee OA wer...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>Outcome measures included pain reduction and i...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>Pain was assessed using the visual analog pain...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>Secondary outcome measures included the Wester...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>5</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>Serum levels of interleukin @ ( IL-@ ) , IL-@ ...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>6</td>\n",
              "      <td>RESULTS</td>\n",
              "      <td>There was a clinically relevant reduction in t...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>7</td>\n",
              "      <td>RESULTS</td>\n",
              "      <td>The mean difference between treatment arms ( @...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>8</td>\n",
              "      <td>RESULTS</td>\n",
              "      <td>Further , there was a clinically relevant redu...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>9</td>\n",
              "      <td>RESULTS</td>\n",
              "      <td>These differences remained significant at @ we...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>10</td>\n",
              "      <td>RESULTS</td>\n",
              "      <td>The Outcome Measures in Rheumatology Clinical ...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>11</td>\n",
              "      <td>CONCLUSIONS</td>\n",
              "      <td>Low-dose oral prednisolone had both a short-te...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>0</td>\n",
              "      <td>BACKGROUND</td>\n",
              "      <td>Emotional eating is associated with overeating...</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>1</td>\n",
              "      <td>BACKGROUND</td>\n",
              "      <td>Yet , empirical evidence for individual ( trai...</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a8f69524-9d49-4723-80f4-563476603368')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-a8f69524-9d49-4723-80f4-563476603368 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-a8f69524-9d49-4723-80f4-563476603368');\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",
              "\n",
              "\n",
              "<div id=\"df-a977f965-d83a-4a35-a922-ec2d17e2400f\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-a977f965-d83a-4a35-a922-ec2d17e2400f')\"\n",
              "            title=\"Suggest charts.\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-a977f965-d83a-4a35-a922-ec2d17e2400f button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Visualizing Data"
      ],
      "metadata": {
        "id": "N7DkEV1xdlqa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "E5SesihtdxDz"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Distribution of labels in training data\n",
        "df = pd.DataFrame(train_df['label'].value_counts())\n",
        "df.reset_index(inplace=True)\n",
        "# sns.barplot(x = df['Number of sentences'], y = df['index'])\n",
        "df.columns = ['label', 'Number of sentences']\n",
        "sns.barplot(x = df['Number of sentences'], y = df['label'])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 467
        },
        "id": "7WotlAIgQbuI",
        "outputId": "269d6ad4-8b1d-4237-b4ba-a9902d0857e1"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: xlabel='Number of sentences', ylabel='label'>"
            ]
          },
          "metadata": {},
          "execution_count": 17
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAokAAAGwCAYAAADFSv/ZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA/x0lEQVR4nO3dfXzPdf////t7m52Zbc435yczJzMjipEcsoxEdUQOh8roTE0si0hHVAcbhyTlrGR8jzrQGSlZ5DRCGcM0Qs7KFjnZnDXM8/eH394f757DjFrsdr1cXpeL9+v5fD1fj9fzeOd9P15nHMYYIwAAAOAibkVdAAAAAP56CIkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFo+iLgA3jvPnz+vAgQMqVaqUHA5HUZcDAAAKwBij48ePq1KlSnJzK/j5QUIiCuzAgQOqWrVqUZcBAAAKYf/+/apSpUqB+xMSUWClSpWSdOFL5u/vX8TVAACAgsjOzlbVqlWdv+MFRUhEgeVdYvb39yckAgBwg7naW8V4cAUAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABg4elmXLU7Xpwldy+foi4DAICbRsp/HinqEiycSQQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBSbENiTEyMHA6H+vbta7XFxsbK4XAoJibGpe/vlw4dOmj58uX5tl28LF++XDNmzFBgYGC+tTgcDs2bN89l3eeff642bdqoVKlS8vX11a233qoZM2a49NmzZ4/LfkqVKqWwsDDFxsZqx44dLn1zc3OVmJioevXqycfHR2XKlFHz5s01bdq0wk4hAAC4iXkUdQFFqWrVqpo9e7Zef/11+fj4SJJ+++03/e9//1O1atVc+nbo0EFJSUku67y8vFSyZEllZGQ41w0YMEDZ2dkufcuUKaM9e/YUuK4333xTcXFxev755zV58mR5enrq008/Vd++fZWWlqaxY8e69P/qq68UFhamU6dOacuWLXrjjTcUERGhzz77TO3atZMkvfzyy5o6dareeustNWvWTNnZ2Vq/fr2OHj1a4LoAAEDxUaxD4i233KJdu3bpk08+Uc+ePSVJn3zyiapVq6aaNWu69PXy8lJQUFC+41y83sfHRzk5OZfseyX79+9XfHy84uLiNGrUKOf6+Ph4eXp6qn///urWrZuaN2/ubCtbtqxzf7Vq1VLnzp3Vrl07Pfroo9q1a5fc3d01f/58Pf300+rWrZtzu4iIiELVCAAAbn7F9nJznj59+ric9Zs+fbp69+5dZPV89NFHOnv2rJ577jmr7cknn5Sfn59mzZp12THc3Nw0YMAA7d27VykpKZIuBNmlS5fq0KFDBa4lJydH2dnZLgsAACgein1IfOihh7Rq1Srt3btXe/fu1erVq/XQQw9Z/T7//HP5+fm5LBef6SuIrKwsaww/Pz+XPj/88IMCAgIUHBxsbe/p6alatWrphx9+uOK+6tWrJ0nOy9zjxo3ToUOHFBQUpEaNGqlv375auHDhZcdISEhQQECAc6latWoBjxQAANzoivXlZkkqX768OnXqpBkzZsgYo06dOqlcuXJWv7Zt22ry5Mku68qUKXNV+ypVqpQ2bNhgra9Tp87VFV0AxhhJFx6KkaQGDRooLS1NKSkpWr16tVauXKnOnTsrJibmkg+vDB06VAMHDnR+zs7OJigCAFBMFPuQKF245NyvXz9J0sSJE/PtU7JkSYWEhFzTftzc3K44RmhoqLKysnTgwAFVqlTJpe3MmTPatWuX2rZte8V9paenS5LLvZVubm669dZbdeuttyouLk7vvfeeHn74YQ0bNsy6B1O6cB+ml5dXQQ4NAADcZIr95WbpwpPLZ86c0dmzZxUdHV2ktTzwwAMqUaKEXnvtNattypQpOnnypHr06HHZMc6fP68JEyaoZs2aatKkySX7NWjQQJJ08uTJaysaAADcdDiTKMnd3d155s3d3T3fPjk5OcrMzHRZ5+Hhke+l6WtRrVo1jRkzRvHx8fL29tbDDz+sEiVK6NNPP9ULL7yg+Ph4lyebJenw4cPKzMzUqVOnlJaWpvHjx+vbb7/VggULnMfTtWtXtWrVSi1btlRQUJB2796toUOHKjQ01Hn/IgAAQB5C4v/P39//su3JycnWwyR169bVtm3brnstcXFxqlWrlsaOHas33nhDubm5CgsL0+TJk/N98joqKkqS5Ovrq+rVq6tt27Z6++23XS5tR0dHa9asWUpISFBWVpaCgoJ05513asSIEfLw4GsAAABcOUzeEw7AFWRnZysgIEARz0yRu5dPUZcDAMBNI+U/j/xhY+f9fmdlZV3xpNjFuCcRAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABgISQCAADAQkgEAACAxaOoC8CNZ+W/e8jf37+oywAAAH8gziQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACweRV0Abjz7E1uolLd7UZcBAMBfWrWXthR1CdeEM4kAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIvIyYmRg6HQw6HQyVKlFDNmjU1ePBg/fbbb84+ee2/X2bPnu3s88477ygiIkJ+fn4KDAxUkyZNlJCQ4LKf++67z9r/8uXL5XA4dOzYMUnSjBkzFBgY6NKnRo0al6zB4XAoJiZGkrRixQrdeeedKlOmjHx9fVWnTh316tVLZ86cuW7zBQAAbh4eRV3AX12HDh2UlJSks2fPKiUlRb169ZLD4dDo0aOdfZKSktShQweX7fLC3PTp0xUXF6cJEyaoTZs2ysnJ0ebNm5WWlnZd6vvuu++Um5srSfrmm2/0wAMPaPv27fL395ck+fj46Pvvv1eHDh30zDPPaMKECfLx8dGOHTv08ccfO7cFAAC4GCHxCry8vBQUFCRJqlq1qqKiorR48WKXkBgYGOjs83vz58/Xgw8+qEcffdS5Liws7LrVV758eeefy5QpI0mqUKGCyxnHpKQkBQUFacyYMc51tWvXtoItAABAHi43X4W0tDR988038vT0LPA2QUFBWrt2rfbu3fsHVnblGjIyMrRy5cqr2i4nJ0fZ2dkuCwAAKB4IiVfw+eefy8/PT97e3goPD9fBgwc1aNAglz49evSQn5+fy7Jv3z5J0vDhwxUYGKgaNWqobt26iomJ0QcffKDz58//acfQrVs39ejRQ23atFFwcLDuv/9+vfXWW1cMfQkJCQoICHAuVatW/ZMqBgAARY2QeAVt27ZVamqq1q1bp169eql379564IEHXPq8/vrrSk1NdVkqVaokSQoODtaaNWu0ZcsWDRgwQOfOnVOvXr3UoUOHPy0ouru7KykpST/99JPGjBmjypUra9SoUQoLC1NGRsYltxs6dKiysrKcy/79+/+UegEAQNEjJF5ByZIlFRISooiICE2fPl3r1q3Tu+++69InKChIISEhLouHh+vtng0bNtTTTz+t9957T4sXL9bixYu1YsUKSZK/v7+ysrKsfR87dkzu7u4qWbLkdTmWypUr6+GHH9Zbb72lrVu36rffftOUKVMu2d/Ly0v+/v4uCwAAKB4IiVfBzc1NL7zwgl588UWdPn260OM0aNBAknTy5ElJUt26dbV161bl5OS49NuwYYNq1qypEiVKFL7oSyhdurSCg4OdNQAAAFyMkHiVunXrJnd3d02cONG57tixY8rMzHRZ8sLXU089pVdffVWrV6/W3r17tXbtWj3yyCMqX768IiMjJUk9e/aUw+HQI488opSUFO3cuVPTp0/X+PHjFR8f77L/3Nxc69J2enr6ZWueOnWqnnrqKS1atEi7du3S1q1b9fzzz2vr1q3q3LnzdZ4hAABwM+AVOFfJw8ND/fr105gxY/TUU09Jknr37m31S0hI0JAhQxQVFaXp06dr8uTJOnz4sMqVK6fIyEgtWbJEZcuWlXThFTpff/21hgwZoi5duigrK0shISEaN26cy6tzJOnEiRNq0qSJy7ratWtr586dl6z5tttu06pVq9S3b18dOHBAfn5+CgsL07x589SmTZtrnRIAAHATchhjTFEXgRtDdna2AgIClDa0vkp5uxd1OQAA/KVVe2lLUZcg6f9+v7Oysq7q+QIuNwMAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALB4FHUBuPFUHbJW/v7+RV0GAAD4A3EmEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABg8SjqAnDjuWvKXfLw4asDXG+rn1ld1CUAgBNnEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYPEoaMcJEyYUeND+/fsXqhgAAAD8NRQ4JL7++usF6udwOAiJAAAAN7gCh8Tdu3f/kXUAAADgL+Sa7kk8c+aMtm/frnPnzl2vegAAAPAXUKiQeOrUKT366KPy9fVVWFiY9u3bJ0l65plnlJiYeF0LBAAAwJ+vUCFx6NCh2rRpk5YvXy5vb2/n+qioKM2ZM+e6FQcAAICiUeB7Ei82b948zZkzRy1atJDD4XCuDwsL065du65bcQAAACgahTqTeOjQIVWoUMFaf/LkSZfQCAAAgBtToUJis2bNtGDBAufnvGA4bdo0RUZGXtVYmZmZeuaZZ1SrVi15eXmpatWq6ty5s5YsWeLs88033+juu+9W6dKl5e3trfDwcI0bN065ubkuYzkcDnl7e2vv3r0u6++77z7FxMRc9X5r1Kih8ePH51u3w+HQvHnzrPUxMTG67777nJ93796tf/7zn6pUqZK8vb1VpUoV3Xvvvdq2bdtlx/r888/Vpk0blSpVSr6+vrr11ls1Y8YMlz579uyRw+FQhQoVdPz4cZe2xo0ba8SIEVdVBwAAQJ5ChcRRo0bphRde0FNPPaVz587pjTfeUPv27ZWUlKSRI0cWeJw9e/aoadOmWrp0qf7zn/9oy5YtSk5OVtu2bRUbGytJmjt3rtq0aaMqVapo2bJl2rZtmwYMGKB///vf+sc//iFjjMuYDodDL7300jXv93o4e/as7rrrLmVlZemTTz7R9u3bNWfOHIWHh+vYsWOX3O7NN9/Uvffeq1atWmndunXavHmz/vGPf6hv37567rnnrP7Hjx/X2LFjr3sdAACg+CrUPYm33367UlNTlZiYqPDwcC1atEi33HKL1qxZo/Dw8AKP8/TTT8vhcOjbb79VyZIlnevDwsLUp08fnTx5Uo8//ri6dOmit99+29n+2GOPqWLFiurSpYs++OADde/e3dnWr18/jRs3ToMGDVLDhg0Ltd/rZevWrdq1a5eWLFmi6tWrS5KqV6+uVq1aXXKb/fv3Kz4+XnFxcRo1apRzfXx8vDw9PdW/f39169ZNzZs3d7Y988wzGjdunGJjY/O9DaAwdQAAgOKt0O9JrF27tt555x19++23+v777/Xee+9dVUA8cuSIkpOTFRsb6xLU8gQGBmrRokU6fPhwvmfPOnfurNDQUM2aNctlfatWrXTPPfdoyJAhhd7v9VK+fHm5ubnpo48+si6NX8pHH32ks2fP5nvMTz75pPz8/Kxj7tGjh0JCQvTKK69ctzokKScnR9nZ2S4LAAAoHgodEnNzc/XRRx/p1Vdf1auvvqqPP/74ql6qvXPnThljVK9evUv2+eGHHyRJ9evXz7e9Xr16zj4XS0hIUHJysr7++utC7fd6qVy5siZMmKCXXnpJpUuX1p133qlXX31VP/744yW3+eGHHxQQEKDg4GCrzdPTU7Vq1bKO2eFwKDExUW+//Xa+T5cXpg7pwjwGBAQ4l6pVqxbwyAEAwI2uUCFx69atCg0NVa9evTR37lzNnTtXvXr1Up06dZSWllagMX5/L+H16itJDRo00COPPJLv2cSrHetaxcbGKjMzU++//74iIyP14YcfKiwsTIsXL76u+4mOjtbtt9+uf/3rX9etjqFDhyorK8u57N+//7rWDAAA/roKFRIfe+wxhYWF6aefftKGDRu0YcMG7d+/X40aNdITTzxRoDHq1Kkjh8Nx2adrQ0NDJUnp6en5tqenpzv7/N7LL7+sDRs2WE8NF2S/BVGqVCllZWVZ648dO6aAgACrb+fOnTVy5Eht2rRJrVu31r///e98xw0NDVVWVpYOHDhgtZ05c0a7du265DEnJiZqzpw52rhx4yVrLmgdkuTl5SV/f3+XBQAAFA+FCompqalKSEhQ6dKlnetKly6tkSNHXjKg/F6ZMmUUHR2tiRMn6uTJk1b7sWPH1L59e5UpU0avvfaa1T5//nzt2LFDPXr0yHf8qlWrql+/fnrhhRdc7sMryH4Lom7dukpJSXFZl5ubq02bNl0yxEkXLg3Xq1cv331L0gMPPKASJUrke8xTpkzRyZMnL3nMt912m/7+979f8n7Mq6kDAAAUb4UKiaGhofrll1+s9QcPHlRISEiBx5k4caJyc3N122236eOPP9aOHTuUnp6uCRMmKDIyUiVLltTUqVP16aef6oknntDmzZu1Z88evfvuu4qJiVHXrl314IMPXnL8oUOH6sCBA/rqq6+uar8X+/nnn5WamuqyHD16VAMHDtS0adM0adIk7dixQ6mpqXriiSd09OhRPfbYY5IuhOl7771XH330kb7//nvt3LlT7777rqZPn657770335qrVaumMWPGaPz48Ro2bJi2bdumXbt2ady4cRo8eLDi4+Ndnmz+vZEjR2rp0qXavn27c11h6gAAAMVbgV+Bc/GTrQkJCerfv79GjBihFi1aSJLWrl2rV155RaNHjy7wzmvVqqUNGzZo5MiRio+PV0ZGhsqXL6+mTZtq8uTJkqSuXbtq2bJlGjlypFq3bq3ffvtNderU0bBhwxQXF3fZf+GlTJkyev755/XCCy9c9X7zjB071noH4X//+1899NBDMsZo3LhxGjJkiHx9fdW0aVOtXLlSFStWlCRVqVJFNWrU0Msvv+x88XXe52efffaSdcfFxalWrVoaO3as3njjDeXm5iosLEyTJ09W7969LzunoaGh6tOnj8srgwpbBwAAKL4cpoBPcri5ubkEsrzN8tZd/PlqXrOCG0d2drYCAgJ02+jb5OFTqFdsAriM1c+sLuoSANyE8n6/s7Kyrur5ggL/0i9btqxQhQEAAODGU+CQ2KZNmz+yDgAAAPyFXNM1w1OnTmnfvn06c+aMy/pGjRpdU1EAAAAoWoUKiYcOHVLv3r21cOHCfNu5JxEAAODGVqhX4MTFxenYsWNat26dfHx8lJycrJkzZ6pOnTqaP3/+9a4RAAAAf7JCnUlcunSpPv30UzVr1kxubm6qXr267rrrLvn7+yshIUGdOnW63nUCAADgT1SoM4knT55UhQoVJF34l1YOHTokSQoPD9eGDRuuX3UAAAAoEoUKiXXr1nX+ix4RERGaOnWqfv75Z02ZMkXBwcHXtUAAAAD8+Qp1uXnAgAHKyMiQJA0fPlwdOnTQe++9J09PT82cOfO6FggAAIA/X6FC4kMPPeT8c9OmTbV3715t27ZN1apVU7ly5a5bcQAAACgaBQ6JAwcOLPCg48aNK1QxAAAA+GsocEjcuHFjgfpd/O87AwAA4MbEv90MAAAAS6GebgYAAMDNjZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgKXA/+IKkGdx38Xy9/cv6jIAAMAfiDOJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWj6IuADeeVR06qqQHXx3cGNqsXFHUJQDADYkziQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAUaUiMiYmRw+FwLmXLllWHDh20efNmq++TTz4pd3d3ffjhh/mOtXPnTvXu3VtVqlSRl5eXatasqR49emj9+vXOPg6HQ/PmzXN+Pnv2rHr06KHKlSsrLS3NuX7ZsmW65557VL58eXl7e6t27drq3r27Vq5c6eyzfPlyl9rLly+vu+++W1u2bLFq279/v/r06aNKlSrJ09NT1atX14ABA3T48GGXfjVq1ND48eOt7UeMGKHGjRtb85aYmOjSb968eXI4HPnW6ObmpoCAADVp0kSDBw9WRkZGvvMIAAAg/QXOJHbo0EEZGRnKyMjQkiVL5OHhoXvuucelz6lTpzR79mwNHjxY06dPt8ZYv369mjZtqh9++EFTp07V999/r7lz56pevXqKj4/Pd7+nTp1Sly5d9N1332nVqlVq2LChJGnSpElq166dypYtqzlz5mj79u2aO3euWrZsqWeffdYaZ/v27crIyNCXX36pnJwcderUSWfOnHG2//jjj2rWrJl27NihWbNmaefOnZoyZYqWLFmiyMhIHTlypFDz5u3trdGjR+vo0aNX7Lt9+3YdOHBA3333nZ5//nl99dVXatiwYb6BFgAAQJI8iroALy8vBQUFSZKCgoI0ZMgQtW7dWocOHVL58uUlSR9++KEaNGigIUOGqFKlStq/f7+qVq0qSTLGKCYmRnXq1NHXX38tN7f/y72NGzfWgAEDrH0eO3ZMnTp10okTJ7Rq1Srn/vft26e4uDjFxcVp3LhxLts0atRI/fv3t8aqUKGCAgMDFRQUpLi4OHXp0kXbtm1To0aNJEmxsbHy9PTUokWL5OPjI0mqVq2amjRpotq1a2vYsGGaPHnyVc9bVFSUdu7cqYSEBI0ZM+ayfS+uMTQ0VPfee6+aNGmip556SqtWrbrqfQMAgJtfkZ9JvNiJEyf03nvvKSQkRGXLlnWuf/fdd/XQQw8pICBAHTt21IwZM5xtqamp2rp1q+Lj410CYp7AwECXz5mZmWrTpo0kacWKFc6AKEkff/yxzp49q8GDB+db38WXcn8vKytLs2fPliR5enpKko4cOaIvv/xSTz/9tDMg5gkKClLPnj01Z84cGWMuOe6luLu7a9SoUXrzzTf1008/XdW2Pj4+6tu3r1avXq2DBw9esl9OTo6ys7NdFgAAUDwUeUj8/PPP5efnJz8/P5UqVUrz58/XnDlznIFvx44dWrt2rbp37y5Jeuihh5SUlOQMVjt27JAk1atXr0D7GzBggM6cOaPFixdbAfKHH36Qv7+/FRzz6vPz87Mu0VapUkV+fn4KDAzU//73P3Xp0sVZy44dO2SMUf369fOtpX79+jp69KgOHTpUoNp/7/7771fjxo01fPjwq942r8Y9e/Zcsk9CQoICAgKcS97ZWwAAcPMr8pDYtm1bpaamKjU1Vd9++62io6PVsWNH7d27V5I0ffp0RUdHq1y5cpKku+++W1lZWVq6dKkkXfVZuHvuucd572J+fn+2MDo6WqmpqVqwYIFOnjyp3Nxcl/avv/5aKSkpmjFjhkJDQzVlyhRrzMKcKSyo0aNHa+bMmUpPT7+q7fJqutzZ0aFDhyorK8u57N+//5pqBQAAN44iD4klS5ZUSEiIQkJCdOutt2ratGk6efKk3nnnHeXm5mrmzJlasGCBPDw85OHhIV9fXx05csT5AEtoaKgkadu2bQXa38MPP6zp06frueees+47rFOnjrKyspSZmelc5+fnp5CQEFWvXj3f8WrWrKm6deuqV69eeuyxx5xnPCUpJCREDofjkgEuPT1dpUuXdt576e/vr6ysLKvfsWPHFBAQkO8Yd9xxh6KjozV06NDLH3g++5YuPFF9KV5eXvL393dZAABA8VDkIfH38l7Xcvr0aX3xxRc6fvy4Nm7c6DzbmJqaqlmzZumTTz7RsWPH1LhxYzVo0ECvvfaazp8/b4137Ngxa12vXr00Y8YMDR48WGPHjnWu79q1q0qUKKHRo0cXqvbY2FilpaVp7ty5kqSyZcvqrrvu0qRJk3T69GmXvpmZmXr//ffVvXt359m8unXrKiUlxRp3w4YNzjCcn8TERH322Wdas2ZNgeo8ffq03n77bd1xxx3OgAoAAHCxIn+6OScnx3nm7ujRo3rrrbd04sQJde7cWePHj1enTp0UERHhsk2DBg307LPP6v3331dsbKySkpIUFRWl1q1ba9iwYapXr55OnDihzz77TIsWLdKKFSus/T788MNyc3NTr169ZIzRoEGDVK1aNb322msaMGCAjhw5opiYGNWsWVNHjhzRe++9J+nCAyOX4uvrq8cff1zDhw/XfffdJ4fDobfeekstW7ZUdHS0/v3vf6tmzZraunWrBg0apMqVK2vkyJHO7Z999lm1bt1aI0eO1N///nfl5uZq1qxZWrNmjSZNmnTJ/YaHh6tnz56aMGFCvu0HDx7Ub7/9puPHjyslJUVjxozRr7/+qk8++eTS/8MAAIBircjPJCYnJys4OFjBwcFq3ry5vvvuO3344YeqX7++FixYoAceeMDaxs3NTffff7/effddSdJtt92m9evXKyQkRI8//rjq16+vLl26aOvWrfm+nDpPz5499d///ldDhw51nj185plntGjRIh06dEhdu3ZVnTp1dPfdd2v37t1KTk5WeHj4ZY+nX79+Sk9Pd770u06dOlq/fr1q1aqlBx98ULVr19YTTzyhtm3bas2aNSpTpoxz25YtW2rhwoVauHChWrVqpb/97W/65ptvtGTJEud7HC/llVdeyfdMqnThDGWlSpXUtGlTJSYmKioqSmlpaWrQoMFlxwQAAMWXw/yRT1XgppKdna2AgAAtiGypkh5FfhIaKJA2K+0rCQBQnOT9fmdlZV3V8wVFfiYRAAAAfz2ERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgMWjqAvAjef25IXy9/cv6jIAAMAfiDOJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWj6IuADeeqS8slI+Xb1GXgavQ77XORV0CAOAGw5lEAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAshEQAAABYCIkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWIpVSNy/f7/69OmjSpUqydPTU9WrV9eAAQN0+PBhZ5+//e1vcjgczqVixYrq1q2b9u7d6+yzZ88eORwOpaamunzOb1m7dq1zuzNnzmjMmDGKiIiQr6+vypUrp1atWikpKUlnz5695Bh5y4gRI1z2nZKSYu3jYu3atdPf//53SVJMTEy+Y3bo0OEPmGkAAHCj8yjqAv4sP/74oyIjIxUaGqpZs2apZs2a2rp1qwYNGqSFCxdq7dq1KlOmjCTp8ccf1yuvvCJjjPbu3au4uDg99NBD+vrrry+7j6+++kphYWEu68qWLSvpQkCMjo7Wpk2b9Oqrr6pVq1by9/fX2rVrNXbsWDVp0kQZGRnO7ebMmaOXXnpJ27dvd67z8/PTr7/+6vzctGlTRUREaPr06WrRooXLfvfs2aNly5bps88+c67r0KGDkpKSXPp5eXkVZPoAAEAxU2xCYmxsrDw9PbVo0SL5+PhIkqpVq6YmTZqodu3aGjZsmCZPnixJ8vX1VVBQkCQpODhY/fr105NPPnnFfZQtW9a53e+NHz9eK1eu1Pr169WkSRPn+lq1aqlbt246c+aMSpYs6VwfEBAgh8NhjXdxSJSkRx99VC+++KLGjx8vX19f5/oZM2YoODjY5Uyhl5fXJesDAAC4WLG43HzkyBF9+eWXevrpp50BMU9QUJB69uypOXPmyBiT77YffPCBmjdvfk01vP/++4qKinIJiHlKlCjhEhCvRs+ePZWTk6OPPvrIuc4Yo5kzZyomJkbu7u6FrjknJ0fZ2dkuCwAAKB6KRUjcsWOHjDGqX79+vu3169fX0aNHdejQIUnSpEmT5Ofnp5IlS6ps2bLavn27pk+ffsX9tGzZUn5+fi7LxTXUq1fv+hzQRcqUKaP777/fpb5ly5Zpz5496t27t0vfzz//3Kpv1KhRlxw7ISFBAQEBzqVq1arXvX4AAPDXVGwuN0vK90xhfnr27Klhw4ZJkn755ReNGjVK7du3V0pKikqVKnXJ7ebMmXPJIFrQfRdGnz59FB0drV27dql27dqaPn262rRpo5CQEJd+bdu2dV5Sz5N3H2Z+hg4dqoEDBzo/Z2dnExQBACgmikVIDAkJkcPhUHp6uu6//36rPT09XaVLl1b58uUlXbgfMC9ghYSE6N1331VwcLDmzJmjxx577JL7qVq1qhXM8oSGhmrbtm3X4Whs7dq1U7Vq1TRjxgwNGjRIn3zyiaZOnWr1K1my5CXry4+XlxcPtgAAUEwVi8vNZcuW1V133aVJkybp9OnTLm2ZmZl6//331b17dzkcjny3z7uv7/fbXo1//vOf+uqrr7Rx40ar7ezZszp58mShx3Zzc1Pv3r01c+ZM/e9//5Onp6e6du1a6PEAAACKRUiUpLfeeks5OTmKjo7WypUrtX//fiUnJ+uuu+5S5cqVNXLkSGffU6dOKTMzU5mZmdq0aZOeeuopeXt7q3379pfdx+HDh53b5S2//fabJCkuLk6tWrVSu3btNHHiRG3atEk//vijPvjgA7Vo0UI7duy4puPr3bu3fv75Z73wwgvq0aOH9YCOdOFBlN/X9/unpQEAAKRiFBLr1Kmj9evXq1atWnrwwQdVu3ZtPfHEE2rbtq3WrFnjcm/eO++8o+DgYAUHB6tt27b69ddf9cUXX6hu3bqSpPPnz0uSPDxcr9ZHRUU5t8tb5s2bJ+nCpdvFixdr8ODBmjp1qlq0aKFbb71VEyZMUP/+/dWwYcNrOr5q1aopKipKR48eVZ8+ffLtk5ycbNV3++23X9N+AQDAzclh/sgnKm5Sa9euVWRkpA4dOqRy5coVdTl/muzsbAUEBGhM7Gz5ePleeQP8ZfR7rXNRlwAAKCJ5v99ZWVny9/cv8HbF4sGV6+XcuXPas2eP/vOf/ygiIqJYBUQAAFC8FJvLzddDWlqaGjVqpIyMDP2///f/irocAACAPwxnEq9C48aNderUqaIuAwAA4A/HmUQAAABYCIkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAAAAsBASAQAAYCEkAgAAwEJIBAAAgIWQCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGDxKOoCcON5clRH+fv7F3UZAADgD8SZRAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALr8BBgRljJEnZ2dlFXAkAACiovN/tvN/xgiIkosAOHz4sSapatWoRVwIAAK7W8ePHFRAQUOD+hEQUWJkyZSRJ+/btu6ovGS78v7iqVatq//79vIi8EJi/a8P8FR5zd22Yv8K7nnNnjNHx48dVqVKlq9qOkIgCc3O7cAtrQEAA/7EXkr+/P3N3DZi/a8P8FR5zd22Yv8K7XnNXmJM7PLgCAAAACyERAAAAFkIiCszLy0vDhw+Xl5dXUZdyw2Hurg3zd22Yv8Jj7q4N81d4f4W5c5irfR4aAAAANz3OJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIKZOLEiapRo4a8vb3VvHlzffvtt0Vd0h9u5cqV6ty5sypVqiSHw6F58+a5tBtj9NJLLyk4OFg+Pj6KiorSjh07XPocOXJEPXv2lL+/vwIDA/Xoo4/qxIkTLn02b96s1q1by9vbW1WrVtWYMWOsWj788EPVq1dP3t7eCg8P1xdffHHdj/d6SkhI0K233qpSpUqpQoUKuu+++7R9+3aXPr/99ptiY2NVtmxZ+fn56YEHHtAvv/zi0mffvn3q1KmTfH19VaFCBQ0aNEjnzp1z6bN8+XLdcsst8vLyUkhIiGbMmGHVc6N9fydPnqxGjRo5X6IbGRmphQsXOtuZu4JLTEyUw+FQXFyccx3zd2kjRoyQw+FwWerVq+dsZ+4u7+eff9ZDDz2ksmXLysfHR+Hh4Vq/fr2z/Yb73TDAFcyePdt4enqa6dOnm61bt5rHH3/cBAYGml9++aWoS/tDffHFF2bYsGHmk08+MZLM3LlzXdoTExNNQECAmTdvntm0aZPp0qWLqVmzpjl9+rSzT4cOHUxERIRZu3at+frrr01ISIjp0aOHsz0rK8tUrFjR9OzZ06SlpZlZs2YZHx8fM3XqVGef1atXG3d3dzNmzBjz/fffmxdffNGUKFHCbNmy5Q+fg8KKjo42SUlJJi0tzaSmppq7777bVKtWzZw4ccLZp2/fvqZq1apmyZIlZv369aZFixamZcuWzvZz586Zhg0bmqioKLNx40bzxRdfmHLlypmhQ4c6+/z444/G19fXDBw40Hz//ffmzTffNO7u7iY5OdnZ50b8/s6fP98sWLDA/PDDD2b79u3mhRdeMCVKlDBpaWnGGOauoL799ltTo0YN06hRIzNgwADneubv0oYPH27CwsJMRkaGczl06JCznbm7tCNHjpjq1aubmJgYs27dOvPjjz+aL7/80uzcudPZ50b73SAk4opuu+02Exsb6/ycm5trKlWqZBISEoqwqj/X70Pi+fPnTVBQkPnPf/7jXHfs2DHj5eVlZs2aZYwx5vvvvzeSzHfffefss3DhQuNwOMzPP/9sjDFm0qRJpnTp0iYnJ8fZ5/nnnzd169Z1fn7wwQdNp06dXOpp3ry5efLJJ6/rMf6RDh48aCSZFStWGGMuzFWJEiXMhx9+6OyTnp5uJJk1a9YYYy6EdDc3N5OZmensM3nyZOPv7++cr8GDB5uwsDCXfXXv3t1ER0c7P98s39/SpUubadOmMXcFdPz4cVOnTh2zePFi06ZNG2dIZP4ub/jw4SYiIiLfNubu8p5//nlz++23X7L9Rvzd4HIzLuvMmTNKSUlRVFSUc52bm5uioqK0Zs2aIqysaO3evVuZmZku8xIQEKDmzZs752XNmjUKDAxUs2bNnH2ioqLk5uamdevWOfvccccd8vT0dPaJjo7W9u3bdfToUWefi/eT1+dGmv+srCxJUpkyZSRJKSkpOnv2rMtx1atXT9WqVXOZv/DwcFWsWNHZJzo6WtnZ2dq6dauzz+Xm5mb4/ubm5mr27Nk6efKkIiMjmbsCio2NVadOnaxjZP6ubMeOHapUqZJq1aqlnj17at++fZKYuyuZP3++mjVrpm7duqlChQpq0qSJ3nnnHWf7jfi7QUjEZf3666/Kzc11+Q9ekipWrKjMzMwiqqro5R375eYlMzNTFSpUcGn38PBQmTJlXPrkN8bF+7hUnxtl/s+fP6+4uDi1atVKDRs2lHThmDw9PRUYGOjS9/fzV9i5yc7O1unTp2/o7++WLVvk5+cnLy8v9e3bV3PnzlWDBg2YuwKYPXu2NmzYoISEBKuN+bu85s2ba8aMGUpOTtbkyZO1e/dutW7dWsePH2furuDHH3/U5MmTVadOHX355Zd66qmn1L9/f82cOVPSjfm74XFVvQHgKsXGxiotLU2rVq0q6lJuKHXr1lVqaqqysrL00UcfqVevXlqxYkVRl/WXt3//fg0YMECLFy+Wt7d3UZdzw+nYsaPzz40aNVLz5s1VvXp1ffDBB/Lx8SnCyv76zp8/r2bNmmnUqFGSpCZNmigtLU1TpkxRr169iri6wuFMIi6rXLlycnd3t55e++WXXxQUFFREVRW9vGO/3LwEBQXp4MGDLu3nzp3TkSNHXPrkN8bF+7hUnxth/vv166fPP/9cy5YtU5UqVZzrg4KCdObMGR07dsyl/+/nr7Bz4+/vLx8fnxv6++vp6amQkBA1bdpUCQkJioiI0BtvvMHcXUFKSooOHjyoW265RR4eHvLw8NCKFSs0YcIEeXh4qGLFiszfVQgMDFRoaKh27tzJd+8KgoOD1aBBA5d19evXd16uvxF/NwiJuCxPT081bdpUS5Ysca47f/68lixZosjIyCKsrGjVrFlTQUFBLvOSnZ2tdevWOeclMjJSx44dU0pKirPP0qVLdf78eTVv3tzZZ+XKlTp79qyzz+LFi1W3bl2VLl3a2efi/eT1+SvPvzFG/fr109y5c7V06VLVrFnTpb1p06YqUaKEy3Ft375d+/btc5m/LVu2uPyFuXjxYvn7+zv/Ir7S3NxM39/z588rJyeHubuCdu3aacuWLUpNTXUuzZo1U8+ePZ1/Zv4K7sSJE9q1a5eCg4P57l1Bq1atrFd9/fDDD6pevbqkG/R346oec0GxNHv2bOPl5WVmzJhhvv/+e/PEE0+YwMBAl6fXbkbHjx83GzduNBs3bjSSzLhx48zGjRvN3r17jTEXXmUQGBhoPv30U7N582Zz77335vsqgyZNmph169aZVatWmTp16ri8yuDYsWOmYsWK5uGHHzZpaWlm9uzZxtfX13qVgYeHhxk7dqxJT083w4cP/8u/Auepp54yAQEBZvny5S6v0jh16pSzT9++fU21atXM0qVLzfr1601kZKSJjIx0tue9SqN9+/YmNTXVJCcnm/Lly+f7Ko1BgwaZ9PR0M3HixHxfpXGjfX+HDBliVqxYYXbv3m02b95shgwZYhwOh1m0aJExhrm7Whc/3WwM83c58fHxZvny5Wb37t1m9erVJioqypQrV84cPHjQGMPcXc63335rPDw8zMiRI82OHTvM+++/b3x9fc17773n7HOj/W4QElEgb775pqlWrZrx9PQ0t912m1m7dm1Rl/SHW7ZsmZFkLb169TLGXHidwb/+9S9TsWJF4+XlZdq1a2e2b9/uMsbhw4dNjx49jJ+fn/H39ze9e/c2x48fd+mzadMmc/vttxsvLy9TuXJlk5iYaNXywQcfmNDQUOPp6WnCwsLMggUL/rDjvh7ymzdJJikpydnn9OnT5umnnzalS5c2vr6+5v777zcZGRku4+zZs8d07NjR+Pj4mHLlypn4+Hhz9uxZlz7Lli0zjRs3Np6enqZWrVou+8hzo31/+/TpY6pXr248PT1N+fLlTbt27ZwB0Rjm7mr9PiQyf5fWvXt3ExwcbDw9PU3lypVN9+7dXd7zx9xd3meffWYaNmxovLy8TL169czbb7/t0n6j/W44jDHm6s49AgAA4GbHPYkAAACwEBIBAABgISQCAADAQkgEAACAhZAIAAAACyERAAAAFkIiAAAALIREAAAAWAiJAPAn2LNnjxwOh1JTU4u6FKdt27apRYsW8vb2VuPGjYu6HAB/MYREAMVCTEyMHA6HEhMTXdbPmzdPDoejiKoqWsOHD1fJkiW1fft2LVmypEhqmDFjhgIDA4tk3wAuj5AIoNjw9vbW6NGjdfTo0aIu5bo5c+ZMobfdtWuXbr/9dlWvXl1ly5a9jlUBuBkQEgEUG1FRUQoKClJCQsIl+4wYMcK69Dp+/HjVqFHD+TkmJkb33XefRo0apYoVKyowMFCvvPKKzp07p0GDBqlMmTKqUqWKkpKSrPG3bdumli1bytvbWw0bNtSKFStc2tPS0tSxY0f5+fmpYsWKevjhh/Xrr7862//2t7+pX79+iouLU7ly5RQdHZ3vcZw/f16vvPKKqlSpIi8vLzVu3FjJycnOdofDoZSUFL3yyityOBwaMWJEvuN89NFHCg8Pl4+Pj8qWLauoqCidPHnS2T5t2jTVr19f3t7eqlevniZNmuRsy7vE/sknn6ht27by9fVVRESE1qxZI0lavny5evfuraysLDkcDpc6cnJy9Nxzz6ly5coqWbKkmjdvruXLlzvHzjsD+eWXX6p+/fry8/NThw4dlJGR4VL/9OnTFRYWJi8vLwUHB6tfv37OtmPHjumxxx5T+fLl5e/vrzvvvFObNm1ytm/atElt27ZVqVKl5O/vr6ZNm2r9+vX5zhNwMyIkAig23N3dNWrUKL355pv66aefrmmspUuX6sCBA1q5cqXGjRun4cOH65577lHp0qW1bt069e3bV08++aS1n0GDBik+Pl4bN25UZGSkOnfurMOHD0u6EFruvPNONWnSROvXr1dycrJ++eUXPfjggy5jzJw5U56enlq9erWmTJmSb31vvPGGXnvtNY0dO1abN29WdHS0unTpoh07dkiSMjIyFBYWpvj4eGVkZOi5556zxsjIyFCPHj3Up08fpaena/ny5fr73/8uY4wk6f3339dLL72kkSNHKj09XaNGjdK//vUvzZw502WcYcOG6bnnnlNqaqpCQ0PVo0cPnTt3Ti1bttT48ePl7++vjIwMlzr69eunNWvWaPbs2dq8ebO6deumDh06OOuXpFOnTmns2LH673//q5UrV2rfvn0uxzF58mTFxsbqiSee0JYtWzR//nyFhIQ427t166aDBw9q4cKFSklJ0S233KJ27drpyJEjkqSePXuqSpUq+u6775SSkqIhQ4aoRIkSl/lWADcZAwDFQK9evcy9995rjDGmRYsWpk+fPsYYY+bOnWsu/qtw+PDhJiIiwmXb119/3VSvXt1lrOrVq5vc3Fznurp165rWrVs7P587d86ULFnSzJo1yxhjzO7du40kk5iY6Oxz9uxZU6VKFTN69GhjjDGvvvqqad++vcu+9+/fbySZ7du3G2OMadOmjWnSpMkVj7dSpUpm5MiRLutuvfVW8/TTTzs/R0REmOHDh19yjJSUFCPJ7NmzJ9/22rVrm//9738u61599VUTGRlpjPm/Y542bZqzfevWrUaSSU9PN8YYk5SUZAICAlzG2Lt3r3F3dzc///yzy/p27dqZoUOHOreTZHbu3OlsnzhxoqlYsaLLHAwbNizf2r/++mvj7+9vfvvtN+uYpk6daowxplSpUmbGjBn5bg8UBx5FmE8BoEiMHj1ad955Z75nzwoqLCxMbm7/dzGmYsWKatiwofOzu7u7ypYtq4MHD7psFxkZ6fyzh4eHmjVrpvT0dEkXLm8uW7ZMfn5+1v527dql0NBQSVLTpk0vW1t2drYOHDigVq1auaxv1aqVy+XUK4mIiFC7du0UHh6u6OhotW/fXl27dlXp0qV18uRJ7dq1S48++qgef/xx5zbnzp1TQECAyziNGjVy/jk4OFiSdPDgQdWrVy/f/W7ZskW5ubnO482Tk5Pjcu+kr6+vateu7TJ23nwfPHhQBw4cULt27fLdx6ZNm3TixAnrXszTp09r165dkqSBAwfqscce03//+19FRUWpW7duLvsDbnaERADFzh133KHo6GgNHTpUMTExLm1ubm7Oy6l5zp49a43x+8uODocj33Xnz58vcF0nTpxQ586dNXr0aKstL1xJUsmSJQs85rVwd3fX4sWL9c0332jRokV68803NWzYMK1bt06+vr6SpHfeeUfNmze3trvYxfOS9yT55eblxIkTcnd3V0pKijXWxQE6v/nO+9/Ox8fnssd24sQJBQcHu9znmCfvaesRI0bon//8pxYsWKCFCxdq+PDhmj17tu6///7Ljg3cLAiJAIqlxMRENW7cWHXr1nVZX758eWVmZsoY4ww01/PdhmvXrtUdd9wh6cJZt5SUFOfDFLfccos+/vhj1ahRQx4ehf/r2d/fX5UqVdLq1avVpk0b5/rVq1frtttuu6qxHA6HWrVqpVatWumll15S9erVNXfuXA0cOFCVKlXSjz/+qJ49exa6Vk9PT+Xm5rqsa9KkiXJzc3Xw4EG1bt26UOOWKlVKNWrU0JIlS9S2bVur/ZZbblFmZqY8PDxcHkr6vdDQUIWGhurZZ59Vjx49lJSUREhEsUFIBFAshYeHq2fPnpowYYLL+r/97W86dOiQxowZo65duyo5OVkLFy6Uv7//ddnvxIkTVadOHdWvX1+vv/66jh49qj59+kiSYmNj9c4776hHjx4aPHiwypQpo507d2r27NmaNm2adVbtcgYNGqThw4erdu3aaty4sZKSkpSamqr333+/wGOsW7dOS5YsUfv27VWhQgWtW7dOhw4dUv369SVJL7/8svr376+AgAB16NBBOTk5Wr9+vY4ePaqBAwcWaB81atTQiRMntGTJEkVERMjX11ehoaHq2bOnHnnkEb322mtq0qSJDh06pCVLlqhRo0bq1KlTgcYeMWKE+vbtqwoVKqhjx446fvy4Vq9erWeeeUZRUVGKjIzUfffdpzFjxig0NFQHDhzQggULdP/99yssLEyDBg1S165dVbNmTf3000/67rvv9MADDxR4/oAbHU83Ayi2XnnlFeuyZ/369TVp0iRNnDhRERER+vbbb6/p3sXfS0xMVGJioiIiIrRq1SrNnz9f5cqVkyTn2b/c3Fy1b99e4eHhiouLU2BgoMv9jwXRv39/DRw4UPHx8QoPD1dycrLmz5+vOnXqFHgMf39/rVy5UnfffbdCQ0P14osv6rXXXlPHjh0lSY899pimTZumpKQkhYeHq02bNpoxY4Zq1qxZ4H20bNlSffv2Vffu3VW+fHmNGTNGkpSUlKRHHnlE8fHxqlu3ru677z599913qlatWoHH7tWrl8aPH69JkyYpLCxM99xzj/PpaIfDoS+++EJ33HGHevfurdDQUP3jH//Q3r17VbFiRbm7u+vw4cN65JFHFBoaqgcffFAdO3bUyy+/XOD9Azc6h/n9zTcAAAAo9jiTCAAAAAshEQAAABZCIgAAACyERAAAAFgIiQAAALAQEgEAAGAhJAIAAMBCSAQAAICFkAgAAAALIREAAAAWQiIAAAAs/x9hm2cJrPC2NwAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Distribution of number of sentences per abstract\n",
        "sns.histplot(train_df['total_lines'], bins=20 )\n",
        "plt.xlabel('Number of sentences per abstract')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 469
        },
        "id": "3a393XbAQxhD",
        "outputId": "3fc51174-03b2-465d-8e1c-74198a9c9c54"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Number of sentences per abstract')"
            ]
          },
          "metadata": {},
          "execution_count": 18
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAGyCAYAAAD51vAJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA+20lEQVR4nO3dfVhUdf7/8Rcgd94MhCg3KaCV4L2JidR2Y5Jkblem9bW0Iu9aC92UzcpvJWbbWrZ5V5Tbjdp3N8tstS0tzTCtVTLFpbwBLMOwFBVLxhsEhc/vj36cdYL0aEMz6PNxXee6nPP5zPu8z+EoL2fOnPExxhgBAADglHw93QAAAEBDQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2NDI0w2cK6qrq7V79241a9ZMPj4+nm4HAADYYIzRoUOHFB0dLV/f07yWZDwoMzPTSHJZ4uPjrfHy8nJz3333mbCwMNOkSRMzcOBAU1JS4lLj22+/NTfccIMJDg42LVq0MA888IA5fvy4y5yPP/7YXHrppSYgIMBcdNFFZt68ebV6ef75501sbKwJDAw0PXv2NOvXrz+jfdm1a1etfWFhYWFhYWFpGMuuXbtO+7ve4680dezYUR999JH1uFGj/7Y0fvx4LVu2TIsWLVJISIjGjBmjgQMHau3atZKkqqoq9e/fX5GRkVq3bp327Nmju+66S/7+/vrLX/4iSSoqKlL//v01evRovf7668rOztbIkSMVFRWl1NRUSdLChQuVkZGhOXPmKCkpSTNnzlRqaqoKCwvVsmVLW/vRrFkzSdKuXbvkcDjccmwAAED9cjqdat26tfV7/JTO6OUUN8vMzDRdu3atc+zgwYPG39/fLFq0yFqXn59vJJmcnBxjjDHvv/++8fX1dXn16cUXXzQOh8NUVFQYY4x58MEHTceOHV1qDx482KSmplqPe/bsadLT063HVVVVJjo62kydOtX2vpSVlRlJpqyszPZzAACAZ53J72+PXwj+1VdfKTo6Wm3bttXQoUNVXFwsScrNzdXx48eVkpJizU1ISFBMTIxycnIkSTk5OercubMiIiKsOampqXI6ndq6das15+QaNXNqalRWVio3N9dljq+vr1JSUqw5damoqJDT6XRZAADAucujoSkpKUnz58/X8uXL9eKLL6qoqEhXXnmlDh06pJKSEgUEBCg0NNTlORERESopKZEklZSUuASmmvGasVPNcTqdKi8vV2lpqaqqquqcU1OjLlOnTlVISIi1tG7d+qyOAQAAaBg8ek1Tv379rD936dJFSUlJio2N1VtvvaXg4GAPdnZ6EydOVEZGhvW45j1RAABwbvL423MnCw0NVbt27fT1118rMjJSlZWVOnjwoMucvXv3KjIyUpIUGRmpvXv31hqvGTvVHIfDoeDgYIWHh8vPz6/OOTU16hIYGCiHw+GyAACAc5dXhabDhw9rx44dioqKUmJiovz9/ZWdnW2NFxYWqri4WMnJyZKk5ORkbd68Wfv27bPmrFy5Ug6HQx06dLDmnFyjZk5NjYCAACUmJrrMqa6uVnZ2tjUHAADAo5+e+9Of/mRWr15tioqKzNq1a01KSooJDw83+/btM8YYM3r0aBMTE2NWrVplNm7caJKTk01ycrL1/BMnTphOnTqZvn37mry8PLN8+XLTokULM3HiRGvON998Yxo3bmwmTJhg8vPzTVZWlvHz8zPLly+35rz55psmMDDQzJ8/32zbts3cc889JjQ0tNY9oU6FT88BANDwnMnvb49e0/Tdd9/p9ttv14EDB9SiRQv97ne/02effaYWLVpIkmbMmCFfX18NGjRIFRUVSk1N1QsvvGA938/PT0uXLtW9996r5ORkNWnSRGlpaZoyZYo1p02bNlq2bJnGjx+vWbNmqVWrVnrllVesezRJ0uDBg7V//35NmjRJJSUl6tatm5YvX17r4nAAAHD+8jHGGE83cS5wOp0KCQlRWVkZ1zcBANBAnMnvb6+6pgkAAMBbEZoAAABsIDQBAADYQGgCAACwgdAEAABgA6EJAADABo/epwk4XxQXF6u0tLReaoeHhysmJqZeagMA/ovQBNSz4uJiJSS0V3n50XqpHxzcWAUF+QQnAKhnhCagnpWWlqq8/KiShmfKERXn1trOPTu1fu7jKi0tJTQBQD0jNAG/EUdUnMJi4j3dBgDgLHEhOAAAgA2EJgAAABsITQAAADYQmgAAAGwgNAEAANhAaAIAALCB0AQAAGADoQkAAMAGQhMAAIANhCYAAAAbCE0AAAA2EJoAAABsIDQBAADYQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABsITQAAADYQmgAAAGwgNAEAANhAaAIAALCB0AQAAGADoQkAAMAGQhMAAIANhCYAAAAbCE0AAAA2EJoAAABsIDQBAADYQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABsITQAAADYQmgAAAGzwmtD01FNPycfHR+PGjbPWHTt2TOnp6WrevLmaNm2qQYMGae/evS7PKy4uVv/+/dW4cWO1bNlSEyZM0IkTJ1zmrF69Wt27d1dgYKAuvvhizZ8/v9b2s7KyFBcXp6CgICUlJenzzz+vj90EAAANlFeEpg0bNuhvf/ubunTp4rJ+/Pjxeu+997Ro0SKtWbNGu3fv1sCBA63xqqoq9e/fX5WVlVq3bp1ee+01zZ8/X5MmTbLmFBUVqX///urdu7fy8vI0btw4jRw5UitWrLDmLFy4UBkZGcrMzNSmTZvUtWtXpaamat++ffW/8wAAoEHweGg6fPiwhg4dqpdfflkXXHCBtb6srEyvvvqqpk+frmuvvVaJiYmaN2+e1q1bp88++0yS9OGHH2rbtm36xz/+oW7duqlfv3564oknlJWVpcrKSknSnDlz1KZNGz377LNq3769xowZo1tuuUUzZsywtjV9+nSNGjVKw4YNU4cOHTRnzhw1btxYc+fO/W0PBgAA8FoeD03p6enq37+/UlJSXNbn5ubq+PHjLusTEhIUExOjnJwcSVJOTo46d+6siIgIa05qaqqcTqe2bt1qzfl57dTUVKtGZWWlcnNzXeb4+voqJSXFmlOXiooKOZ1OlwUAAJy7Gnly42+++aY2bdqkDRs21BorKSlRQECAQkNDXdZHRESopKTEmnNyYKoZrxk71Ryn06ny8nL9+OOPqqqqqnNOQUHBL/Y+depUPf744/Z2FAAANHgee6Vp165duv/++/X6668rKCjIU22ctYkTJ6qsrMxadu3a5emWAABAPfJYaMrNzdW+ffvUvXt3NWrUSI0aNdKaNWs0e/ZsNWrUSBEREaqsrNTBgwddnrd3715FRkZKkiIjI2t9mq7m8enmOBwOBQcHKzw8XH5+fnXOqalRl8DAQDkcDpcFAACcuzwWmvr06aPNmzcrLy/PWnr06KGhQ4daf/b391d2drb1nMLCQhUXFys5OVmSlJycrM2bN7t8ym3lypVyOBzq0KGDNefkGjVzamoEBAQoMTHRZU51dbWys7OtOQAAAB67pqlZs2bq1KmTy7omTZqoefPm1voRI0YoIyNDYWFhcjgcGjt2rJKTk9WrVy9JUt++fdWhQwfdeeedmjZtmkpKSvToo48qPT1dgYGBkqTRo0fr+eef14MPPqjhw4dr1apVeuutt7Rs2TJruxkZGUpLS1OPHj3Us2dPzZw5U0eOHNGwYcN+o6MBAAC8nUcvBD+dGTNmyNfXV4MGDVJFRYVSU1P1wgsvWON+fn5aunSp7r33XiUnJ6tJkyZKS0vTlClTrDlt2rTRsmXLNH78eM2aNUutWrXSK6+8otTUVGvO4MGDtX//fk2aNEklJSXq1q2bli9fXuvicAAAcP7yMcYYTzdxLnA6nQoJCVFZWRnXN8HFpk2blJiYqOsemaewmHi31v6huFArnxym3Nxcde/e3a21AeB8cCa/vz1+nyYAAICGgNAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABsITQAAADYQmgAAAGwgNAEAANhAaAIAALCB0AQAAGADoQkAAMAGQhMAAIANhCYAAAAbCE0AAAA2EJoAAABsIDQBAADYQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABsITQAAADYQmgAAAGwgNAEAANhAaAIAALCB0AQAAGADoQkAAMAGQhMAAIANhCYAAAAbCE0AAAA2EJoAAABsIDQBAADYQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABsITQAAADZ4NDS9+OKL6tKlixwOhxwOh5KTk/XBBx9Y48eOHVN6erqaN2+upk2batCgQdq7d69LjeLiYvXv31+NGzdWy5YtNWHCBJ04ccJlzurVq9W9e3cFBgbq4osv1vz582v1kpWVpbi4OAUFBSkpKUmff/55vewzAABomDwamlq1aqWnnnpKubm52rhxo6699lrddNNN2rp1qyRp/Pjxeu+997Ro0SKtWbNGu3fv1sCBA63nV1VVqX///qqsrNS6dev02muvaf78+Zo0aZI1p6ioSP3791fv3r2Vl5encePGaeTIkVqxYoU1Z+HChcrIyFBmZqY2bdqkrl27KjU1Vfv27fvtDgYAAPBqPsYY4+kmThYWFqZnnnlGt9xyi1q0aKEFCxbolltukSQVFBSoffv2ysnJUa9evfTBBx/o97//vXbv3q2IiAhJ0pw5c/TQQw9p//79CggI0EMPPaRly5Zpy5Yt1jZuu+02HTx4UMuXL5ckJSUl6bLLLtPzzz8vSaqurlbr1q01duxYPfzww7b6djqdCgkJUVlZmRwOhzsPCRq4TZs2KTExUdc9Mk9hMfFurf1DcaFWPjlMubm56t69u1trA8D54Ex+f3vNNU1VVVV68803deTIESUnJys3N1fHjx9XSkqKNSchIUExMTHKycmRJOXk5Khz585WYJKk1NRUOZ1O69WqnJwclxo1c2pqVFZWKjc312WOr6+vUlJSrDl1qaiokNPpdFkAAMC5y+OhafPmzWratKkCAwM1evRoLVmyRB06dFBJSYkCAgIUGhrqMj8iIkIlJSWSpJKSEpfAVDNeM3aqOU6nU+Xl5SotLVVVVVWdc2pq1GXq1KkKCQmxltatW5/V/gMAgIbB46EpPj5eeXl5Wr9+ve69916lpaVp27Ztnm7rtCZOnKiysjJr2bVrl6dbAgAA9aiRpxsICAjQxRdfLElKTEzUhg0bNGvWLA0ePFiVlZU6ePCgy6tNe/fuVWRkpCQpMjKy1qfcaj5dd/Kcn3/ibu/evXI4HAoODpafn5/8/PzqnFNToy6BgYEKDAw8u50GAAANjsdfafq56upqVVRUKDExUf7+/srOzrbGCgsLVVxcrOTkZElScnKyNm/e7PIpt5UrV8rhcKhDhw7WnJNr1MypqREQEKDExESXOdXV1crOzrbmAAAAePSVpokTJ6pfv36KiYnRoUOHtGDBAq1evVorVqxQSEiIRowYoYyMDIWFhcnhcGjs2LFKTk5Wr169JEl9+/ZVhw4ddOedd2ratGkqKSnRo48+qvT0dOtVoNGjR+v555/Xgw8+qOHDh2vVqlV66623tGzZMquPjIwMpaWlqUePHurZs6dmzpypI0eOaNiwYR45LgAAwPt4NDTt27dPd911l/bs2aOQkBB16dJFK1as0HXXXSdJmjFjhnx9fTVo0CBVVFQoNTVVL7zwgvV8Pz8/LV26VPfee6+Sk5PVpEkTpaWlacqUKdacNm3aaNmyZRo/frxmzZqlVq1a6ZVXXlFqaqo1Z/Dgwdq/f78mTZqkkpISdevWTcuXL691cTgAADh/ed19mhoq7tOEX8J9mgDAezXI+zQBAAB4M0ITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA1nFZratm2rAwcO1Fp/8OBBtW3b9lc3BQAA4G3OKjTt3LlTVVVVtdZXVFTo+++//9VNAQAAeJtGZzL53Xfftf68YsUKhYSEWI+rqqqUnZ2tuLg4tzUHAADgLc4oNA0YMECS5OPjo7S0NJcxf39/xcXF6dlnn3VbcwAAAN7ijEJTdXW1JKlNmzbasGGDwsPD66UpAAAAb3NGoalGUVGRu/sAAADwamcVmiQpOztb2dnZ2rdvn/UKVI25c+f+6sYAAAC8yVmFpscff1xTpkxRjx49FBUVJR8fH3f3BQAA4FXOKjTNmTNH8+fP15133unufgAAALzSWd2nqbKyUpdffrm7ewEAAPBaZxWaRo4cqQULFri7FwAAAK91Vm/PHTt2TC+99JI++ugjdenSRf7+/i7j06dPd0tzAAAA3uKsQtOXX36pbt26SZK2bNniMsZF4QAA4Fx0VqHp448/dncfAAAAXu2srmkCAAA435zVK029e/c+5dtwq1atOuuGAAAAvNFZhaaa65lqHD9+XHl5edqyZUutL/IFgJ8rLi5WaWlpvdQODw9XTExMvdQGcH47q9A0Y8aMOtdPnjxZhw8f/lUNATi3FRcXKyGhvcrLj9ZL/eDgxiooyCc4AXC7s/7uubrccccd6tmzp/7617+6syyAc0hpaanKy48qaXimHFFxbq3t3LNT6+c+rtLSUkITALdza2jKyclRUFCQO0sCOEc5ouIUFhPv6TYAwLazCk0DBw50eWyM0Z49e7Rx40Y99thjbmkMAADAm5xVaAoJCXF57Ovrq/j4eE2ZMkV9+/Z1S2MAAADe5KxC07x589zdBwAAgFf7Vdc05ebmKj8/X5LUsWNHXXrppW5pCgAAwNucVWjat2+fbrvtNq1evVqhoaGSpIMHD6p3795688031aJFC3f2CAAA4HFn9TUqY8eO1aFDh7R161b98MMP+uGHH7RlyxY5nU798Y9/dHePAAAAHndWrzQtX75cH330kdq3b2+t69Chg7KysrgQHAAAnJPO6pWm6upq+fv711rv7++v6urqX90UAACAtzmr0HTttdfq/vvv1+7du61133//vcaPH68+ffq4rTkAAABvcVah6fnnn5fT6VRcXJwuuugiXXTRRWrTpo2cTqeee+45d/cIAADgcWd1TVPr1q21adMmffTRRyooKJAktW/fXikpKW5tDgAAwFuc0StNq1atUocOHeR0OuXj46PrrrtOY8eO1dixY3XZZZepY8eO+vTTT+urVwAAAI85o9A0c+ZMjRo1Sg6Ho9ZYSEiI/vCHP2j69Oluaw4AAMBbnFFo+uKLL3T99df/4njfvn2Vm5v7q5sCAADwNmcUmvbu3VvnrQZqNGrUSPv37//VTQEAAHibMwpNF154obZs2fKL419++aWioqJ+dVMAAADe5oxC0w033KDHHntMx44dqzVWXl6uzMxM/f73v3dbcwAAAN7ijG458Oijj2rx4sVq166dxowZo/j4eElSQUGBsrKyVFVVpUceeaReGgUAAPCkMwpNERERWrdune69915NnDhRxhhJko+Pj1JTU5WVlaWIiIh6aRQAAMCTzvjmlrGxsXr//ff1448/6uuvv5YxRpdccokuuOCC+ugPAADAK5zVHcEl6YILLtBll13mzl4AnKX8/Px6qRseHq6YmJh6qQ0ADc1ZhyYAnldedkCSj+644456qR8c3FgFBfkEJwAQoQlo0I4fPSTJqNuQh9SiTYJbazv37NT6uY+rtLSU0AQAIjQB54SmLWMUFhPv6TYA4Jx2RvdpAgAAOF8RmgAAAGwgNAEAANhAaAIAALCB0AQAAGADoQkAAMAGQhMAAIANhCYAAAAbCE0AAAA2eDQ0TZ06VZdddpmaNWumli1basCAASosLHSZc+zYMaWnp6t58+Zq2rSpBg0apL1797rMKS4uVv/+/dW4cWO1bNlSEyZM0IkTJ1zmrF69Wt27d1dgYKAuvvhizZ8/v1Y/WVlZiouLU1BQkJKSkvT555+7fZ8BAEDD5NHQtGbNGqWnp+uzzz7TypUrdfz4cfXt21dHjhyx5owfP17vvfeeFi1apDVr1mj37t0aOHCgNV5VVaX+/fursrJS69at02uvvab58+dr0qRJ1pyioiL1799fvXv3Vl5ensaNG6eRI0dqxYoV1pyFCxcqIyNDmZmZ2rRpk7p27arU1FTt27fvtzkYAADAq3n0u+eWL1/u8nj+/Plq2bKlcnNzddVVV6msrEyvvvqqFixYoGuvvVaSNG/ePLVv316fffaZevXqpQ8//FDbtm3TRx99pIiICHXr1k1PPPGEHnroIU2ePFkBAQGaM2eO2rRpo2effVaS1L59e/373//WjBkzlJqaKkmaPn26Ro0apWHDhkmS5syZo2XLlmnu3Ll6+OGHa/VeUVGhiooK67HT6ayXYwQAALyDV13TVFZWJkkKCwuTJOXm5ur48eNKSUmx5iQkJCgmJkY5OTmSpJycHHXu3FkRERHWnNTUVDmdTm3dutWac3KNmjk1NSorK5Wbm+syx9fXVykpKdacn5s6dapCQkKspXXr1r929wEAgBfzmtBUXV2tcePG6YorrlCnTp0kSSUlJQoICFBoaKjL3IiICJWUlFhzTg5MNeM1Y6ea43Q6VV5ertLSUlVVVdU5p6bGz02cOFFlZWXWsmvXrrPbcQAA0CB49O25k6Wnp2vLli3697//7elWbAkMDFRgYKCn2wAAAL8Rr3ilacyYMVq6dKk+/vhjtWrVylofGRmpyspKHTx40GX+3r17FRkZac35+afpah6fbo7D4VBwcLDCw8Pl5+dX55yaGgAA4Pzm0VeajDEaO3aslixZotWrV6tNmzYu44mJifL391d2drYGDRokSSosLFRxcbGSk5MlScnJyXryySe1b98+tWzZUpK0cuVKORwOdejQwZrz/vvvu9ReuXKlVSMgIECJiYnKzs7WgAEDJP30dmF2drbGjBlTb/sP71NcXKzS0lK31szPz3drPQCAZ3g0NKWnp2vBggX617/+pWbNmlnXD4WEhCg4OFghISEaMWKEMjIyFBYWJofDobFjxyo5OVm9evWSJPXt21cdOnTQnXfeqWnTpqmkpESPPvqo0tPTrbfPRo8ereeff14PPvighg8frlWrVumtt97SsmXLrF4yMjKUlpamHj16qGfPnpo5c6aOHDlifZoO577i4mIlJLRXefnReql/vKKyXuoCAH4bHg1NL774oiTpmmuucVk/b9483X333ZKkGTNmyNfXV4MGDVJFRYVSU1P1wgsvWHP9/Py0dOlS3XvvvUpOTlaTJk2UlpamKVOmWHPatGmjZcuWafz48Zo1a5ZatWqlV155xbrdgCQNHjxY+/fv16RJk1RSUqJu3bpp+fLltS4Ox7mrtLRU5eVHlTQ8U46oOLfV3bM5R1vefanWDVcBAA2Lx9+eO52goCBlZWUpKyvrF+fExsbWevvt56655hr95z//OeWcMWPG8HYc5IiKU1hMvNvqOffsdFstAIDneMWF4AAAAN6O0AQAAGADoQkAAMAGr7m5JQDv5O5bJnALBgANFaEJQJ3Kyw5I8tEdd9xRL/W5BQOAhobQBKBOx48ekmTUbchDatEmwW11uQUDgIaK0ATglJq2jOEWDAAgLgQHAACwhdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABsITQAAADYQmgAAAGwgNAEAANhAaAIAALCB0AQAAGADoQkAAMAGQhMAAIANhCYAAAAbCE0AAAA2EJoAAABsIDQBAADYQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABsITQAAADYQmgAAAGwgNAEAANhAaAIAALChkacbAM5UcXGxSktL3V43Pz/f7TUBAOcOQhMalOLiYiUktFd5+dF628bxisp6qw0AaLgITWhQSktLVV5+VEnDM+WIinNr7T2bc7Tl3Zd04sQJt9YFAJwbCE1okBxRcQqLiXdrTeeenW6tBwA4t3AhOAAAgA2EJgAAABsITQAAADYQmgAAAGwgNAEAANhAaAIAALCB0AQAAGAD92lCvamPrzvhq04AAJ5CaEK9qO+vO+GrTgAAvzVCE+pFfX3dCV91AgDwFI9e0/TJJ5/oxhtvVHR0tHx8fPTOO++4jBtjNGnSJEVFRSk4OFgpKSn66quvXOb88MMPGjp0qBwOh0JDQzVixAgdPnzYZc6XX36pK6+8UkFBQWrdurWmTZtWq5dFixYpISFBQUFB6ty5s95//3237+/5qObrTty1NAmP8vQuAQDOUx4NTUeOHFHXrl2VlZVV5/i0adM0e/ZszZkzR+vXr1eTJk2UmpqqY8eOWXOGDh2qrVu3auXKlVq6dKk++eQT3XPPPda40+lU3759FRsbq9zcXD3zzDOaPHmyXnrpJWvOunXrdPvtt2vEiBH6z3/+owEDBmjAgAHasmVL/e08AABoUDz69ly/fv3Ur1+/OseMMZo5c6YeffRR3XTTTZKk//u//1NERITeeecd3XbbbcrPz9fy5cu1YcMG9ejRQ5L03HPP6YYbbtBf//pXRUdH6/XXX1dlZaXmzp2rgIAAdezYUXl5eZo+fboVrmbNmqXrr79eEyZMkCQ98cQTWrlypZ5//nnNmTPnNzgSAADA23ntLQeKiopUUlKilJQUa11ISIiSkpKUk5MjScrJyVFoaKgVmCQpJSVFvr6+Wr9+vTXnqquuUkBAgDUnNTVVhYWF+vHHH605J2+nZk7NdupSUVEhp9PpsgAAgHOX14amkpISSVJERITL+oiICGuspKRELVu2dBlv1KiRwsLCXObUVePkbfzSnJrxukydOlUhISHW0rp16zPdRQAA0IB4bWjydhMnTlRZWZm17Nq1y9MtAQCAeuS1oSkyMlKStHfvXpf1e/futcYiIyO1b98+l/ETJ07ohx9+cJlTV42Tt/FLc2rG6xIYGCiHw+GyAACAc5fXhqY2bdooMjJS2dnZ1jqn06n169crOTlZkpScnKyDBw8qNzfXmrNq1SpVV1crKSnJmvPJJ5/o+PHj1pyVK1cqPj5eF1xwgTXn5O3UzKnZDgAAgEdD0+HDh5WXl6e8vDxJP138nZeXp+LiYvn4+GjcuHH685//rHfffVebN2/WXXfdpejoaA0YMECS1L59e11//fUaNWqUPv/8c61du1ZjxozRbbfdpujoaEnSkCFDFBAQoBEjRmjr1q1auHChZs2apYyMDKuP+++/X8uXL9ezzz6rgoICTZ48WRs3btSYMWN+60MCAAC8lEdvObBx40b17t3belwTZNLS0jR//nw9+OCDOnLkiO655x4dPHhQv/vd77R8+XIFBQVZz3n99dc1ZswY9enTR76+vho0aJBmz55tjYeEhOjDDz9Uenq6EhMTFR4erkmTJrncy+nyyy/XggUL9Oijj+p///d/dckll+idd95Rp06dfoOjAAAAGgKPhqZrrrlGxphfHPfx8dGUKVM0ZcqUX5wTFhamBQsWnHI7Xbp00aeffnrKObfeeqtuvfXWUzcMAADOW157TRMAAIA3ITQBAADYQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2ODRr1EBgPqQn5/v9prh4eGKiYlxe10ADQehCcA5o7zsgCQf3XHHHW6vHRzcWAUF+QQn4DxGaAJwzjh+9JAko25DHlKLNgluq+vcs1Pr5z6u0tJSQhNwHiM0ATjnNG0Zo7CYeE+3AeAcw4XgAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABu4uSUAeFhxcbFKS0vdXpfvywPci9AEAB5UXFyshIT2Ki8/6vbafF8e4F6EJgDwoNLSUpWXH1XS8Ew5ouLcVpfvywPcj9AEAF7AERXH9+UBXo4LwQEAAGzglSYAsCk/P79B1ARQPwhNAHAa5WUHJPnojjvuqLdtHK+orLfaANyD0AQAp3H86CFJRt2GPKQWbRLcWnvP5hxtefclnThxwq11AbgfoQkAbGraMsbtF2s79+x0az0A9YcLwQEAAGwgNAEAANhAaAIAALCB0AQAAGADoQkAAMAGQhMAAIANhCYAAAAbCE0AAAA2EJoAAABsIDQBAADYQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCEwAAgA2EJgAAABsITQAAADY08nQDAID6k5+fXy91w8PDFRMTUy+1AW9FaDrPFRcXq7S01O116+sfagD2lJcdkOSjO+64o17qBwc3VkFBPsEJ5xVC03msuLhYCQntVV5+tN62cbyist5qA/hlx48ekmTUbchDatEmwa21nXt2av3cx1VaWur20FRf/5GTeHUMvx6h6TxWWlqq8vKjShqeKUdUnFtr79mcoy3vvqQTJ064tS6AM9O0ZYzCYuI93YYt9f0fOV4dw69FaIIcUXFu/0fVuWenW+sB8D7ufhs+Pz+/3v4jV5+vjuH8QWgCAJyRer9eKiy6wbw6hvMLoQkAcEbq63op3taHtyM0/UxWVpaeeeYZlZSUqGvXrnruuefUs2dPT7cFAF7H3ddL8bY+vB2h6SQLFy5URkaG5syZo6SkJM2cOVOpqakqLCxUy5YtPdpbfXyihNsCADjf1Me/exUVFQoMDHR7XYlP/HkbQtNJpk+frlGjRmnYsGGSpDlz5mjZsmWaO3euHn74YY/1Vd+fKOG2AADOdfV6HZaPj2SM++tKCgwM0j//+baioqLcXru+wt65HCIJTf9fZWWlcnNzNXHiRGudr6+vUlJSlJOTU2t+RUWFKioqrMdlZWWSJKfT6fbedu7cqfLyo4q/bogah0W4re4PO/P17frlOvBtvnxU5ba6kuTc860kqez7r+TfyMfr69Zn7YbYc33WpuffpjY9uzqwY4sko7bX3KqQiFZuq1vz76i760pS2e5v9M2n/9Lvf/97t9ZtyIKCgrVx4wa1bt3abTVrfm8bO8HXwBhjzPfff28kmXXr1rmsnzBhgunZs2et+ZmZmUYSCwsLCwsLyzmw7Nq167RZgVeaztLEiROVkZFhPa6urtYPP/yg5s2by8fHvf9D8gZOp1OtW7fWrl275HA4PN1Og8fxdB+OpXtxPN2HY+le9XU8jTE6dOiQoqOjTzuX0PT/hYeHy8/PT3v37nVZv3fvXkVGRtaaHxgYWOs929DQ0Pps0Ss4HA7+8rsRx9N9OJbuxfF0H46le9XH8QwJCbE1z9etW23AAgIClJiYqOzsbGtddXW1srOzlZyc7MHOAACAN+CVppNkZGQoLS1NPXr0UM+ePTVz5kwdOXLE+jQdAAA4fxGaTjJ48GDt379fkyZNUklJibp166bly5crIsJ9n1hrqAIDA5WZmVlvHyM933A83Ydj6V4cT/fhWLqXNxxPH2Pq6eYSAAAA5xCuaQIAALCB0AQAAGADoQkAAMAGQhMAAIANhCac0uTJk+Xj4+OyJCQkeLqtBuGTTz7RjTfeqOjoaPn4+Oidd95xGTfGaNKkSYqKilJwcLBSUlL01VdfeabZBuB0x/Puu++uda5ef/31nmnWy02dOlWXXXaZmjVrppYtW2rAgAEqLCx0mXPs2DGlp6erefPmatq0qQYNGlTr5r+wdyyvueaaWufm6NGjPdSxd3vxxRfVpUsX6waWycnJ+uCDD6xxT5+XhCacVseOHbVnzx5r+fe//+3plhqEI0eOqGvXrsrKyqpzfNq0aZo9e7bmzJmj9evXq0mTJkpNTdWxY8d+404bhtMdT0m6/vrrXc7VN9544zfssOFYs2aN0tPT9dlnn2nlypU6fvy4+vbtqyNHjlhzxo8fr/fee0+LFi3SmjVrtHv3bg0cONCDXXsnO8dSkkaNGuVybk6bNs1DHXu3Vq1a6amnnlJubq42btyoa6+9VjfddJO2bt0qyQvOS7d82y3OWZmZmaZr166ebqPBk2SWLFliPa6urjaRkZHmmWeesdYdPHjQBAYGmjfeeMMDHTYsPz+exhiTlpZmbrrpJo/009Dt27fPSDJr1qwxxvx0Lvr7+5tFixZZc/Lz840kk5OT46k2G4SfH0tjjLn66qvN/fff77mmGrgLLrjAvPLKK15xXvJKE07rq6++UnR0tNq2bauhQ4equLjY0y01eEVFRSopKVFKSoq1LiQkRElJScrJyfFgZw3b6tWr1bJlS8XHx+vee+/VgQMHPN1Sg1BWViZJCgsLkyTl5ubq+PHjLudnQkKCYmJiOD9P4+fHssbrr7+u8PBwderUSRMnTtTRo0c90V6DUlVVpTfffFNHjhxRcnKyV5yX3BEcp5SUlKT58+crPj5ee/bs0eOPP64rr7xSW7ZsUbNmzTzdXoNVUlIiSbXuNh8REWGN4cxcf/31GjhwoNq0aaMdO3bof//3f9WvXz/l5OTIz8/P0+15rerqao0bN05XXHGFOnXqJOmn8zMgIKDWl5Bzfp5aXcdSkoYMGaLY2FhFR0fryy+/1EMPPaTCwkItXrzYg916r82bNys5OVnHjh1T06ZNtWTJEnXo0EF5eXkePy8JTTilfv36WX/u0qWLkpKSFBsbq7feeksjRozwYGeAq9tuu836c+fOndWlSxdddNFFWr16tfr06ePBzrxbenq6tmzZwrWKbvBLx/Kee+6x/ty5c2dFRUWpT58+2rFjhy666KLfuk2vFx8fr7y8PJWVlentt99WWlqa1qxZ4+m2JHEhOM5QaGio2rVrp6+//trTrTRokZGRklTrUx979+61xvDrtG3bVuHh4ZyrpzBmzBgtXbpUH3/8sVq1amWtj4yMVGVlpQ4ePOgyn/Pzl/3SsaxLUlKSJHFu/oKAgABdfPHFSkxM1NSpU9W1a1fNmjXLK85LQhPOyOHDh7Vjxw5FRUV5upUGrU2bNoqMjFR2dra1zul0av369UpOTvZgZ+eO7777TgcOHOBcrYMxRmPGjNGSJUu0atUqtWnTxmU8MTFR/v7+LudnYWGhiouLOT9/5nTHsi55eXmSxLlpU3V1tSoqKrzivOTtOZzSAw88oBtvvFGxsbHavXu3MjMz5efnp9tvv93TrXm9w4cPu/xPsqioSHl5eQoLC1NMTIzGjRunP//5z7rkkkvUpk0bPfbYY4qOjtaAAQM817QXO9XxDAsL0+OPP65BgwYpMjJSO3bs0IMPPqiLL75YqampHuzaO6Wnp2vBggX617/+pWbNmlnXg4SEhCg4OFghISEaMWKEMjIyFBYWJofDobFjxyo5OVm9evXycPfe5XTHcseOHVqwYIFuuOEGNW/eXF9++aXGjx+vq666Sl26dPFw995n4sSJ6tevn2JiYnTo0CEtWLBAq1ev1ooVK7zjvPxNPqOHBmvw4MEmKirKBAQEmAsvvNAMHjzYfP31155uq0H4+OOPjaRaS1pamjHmp9sOPPbYYyYiIsIEBgaaPn36mMLCQs827cVOdTyPHj1q+vbta1q0aGH8/f1NbGysGTVqlCkpKfF0216pruMoycybN8+aU15ebu677z5zwQUXmMaNG5ubb77Z7Nmzx3NNe6nTHcvi4mJz1VVXmbCwMBMYGGguvvhiM2HCBFNWVubZxr3U8OHDTWxsrAkICDAtWrQwffr0MR9++KE17unz0scYY36beAYAANBwcU0TAACADYQmAAAAGwhNAAAANhCaAAAAbCA0AQAA2EBoAgAAsIHQBAAAYAOhCQAAwAZCE3CO2blzp3x8fKzvt/IGBQUF6tWrl4KCgtStWzdPtwM3i4uL08yZMz3dBlDvCE2Am919993y8fHRU0895bL+nXfekY+Pj4e68qzMzEw1adJEhYWFLl+2+VuaP3++QkNDPbJtnNpvGbo4D/BrEJqAehAUFKSnn35aP/74o6dbcZvKysqzfu6OHTv0u9/9TrGxsWrevLkbu4K7/Jqf72+hqqpK1dXVnm4D5zlCE1APUlJSFBkZqalTp/7inMmTJ9d6q2rmzJmKi4uzHt99990aMGCA/vKXvygiIkKhoaGaMmWKTpw4oQkTJigsLEytWrXSvHnzatUvKCjQ5ZdfrqCgIHXq1Elr1qxxGd+yZYv69eunpk2bKiIiQnfeeadKS0ut8WuuuUZjxozRuHHjFB4ertTU1Dr3o7q6WlOmTFGrVq0UGBiobt26afny5da4j4+PcnNzNWXKFPn4+Gjy5Ml11nn77bfVuXNnBQcHq3nz5kpJSdGRI0es8VdeeUXt27dXUFCQEhIS9MILL1hjNW9JLl68WL1791bjxo3VtWtX5eTkSJJWr16tYcOGqaysTD4+Pi59VFRU6IEHHtCFF16oJk2aKCkpSatXr7Zq17wysWLFCrVv315NmzbV9ddfrz179rj0P3fuXHXs2FGBgYGKiorSmDFjrLGDBw9q5MiRatGihRwOh6699lp98cUX1vgXX3yh3r17q1mzZnI4HEpMTNTGjRvrPE41x/TFF19Uv379FBwcrLZt2+rtt992mbNr1y79z//8j0JDQxUWFqabbrpJO3futMZrzq0nn3xS0dHRio+Pr3NbO3bs0E033aSIiAg1bdpUl112mT766KNa8w4dOqTbb79dTZo00YUXXqisrCxrzBijyZMnKyYmRoGBgYqOjtYf//hHST+dZ99++63Gjx9v/WxOPu7vvvuuOnTooMDAQBUXF2vDhg267rrrFB4erpCQEF199dXatGmTSy8HDx7UH/7wB0VERFjn/9KlS095HgC2/GZfDQycJ9LS0sxNN91kFi9ebIKCgsyuXbuMMcYsWbLEnPxXLjMz03Tt2tXluTNmzDCxsbEutZo1a2bS09NNQUGBefXVV40kk5qaap588kmzfft288QTTxh/f39rO0VFRUaSadWqlXn77bfNtm3bzMiRI02zZs1MaWmpMcaYH3/80bRo0cJMnDjR5Ofnm02bNpnrrrvO9O7d29r21VdfbZo2bWomTJhgCgoKTEFBQZ37O336dONwOMwbb7xhCgoKzIMPPmj8/f3N9u3bjTHG7Nmzx3Ts2NH86U9/Mnv27DGHDh2qVWP37t2mUaNGZvr06aaoqMh8+eWXJisry5r7j3/8w0RFRZl//vOf5ptvvjH//Oc/TVhYmJk/f77LPickJJilS5eawsJCc8stt5jY2Fhz/PhxU1FRYWbOnGkcDofZs2ePSx8jR440l19+ufnkk0/M119/bZ555hkTGBho9T9v3jzj7+9vUlJSzIYNG0xubq5p3769GTJkiNX/Cy+8YIKCgszMmTNNYWGh+fzzz82MGTOs8ZSUFHPjjTeaDRs2mO3bt5s//elPpnnz5ubAgQPGGGM6duxo7rjjDpOfn2+2b99u3nrrLZOXl1fn8TbGGEmmefPm5uWXXzaFhYXm0UcfNX5+fmbbtm3GGGMqKytN+/btzfDhw82XX35ptm3bZoYMGWLi4+NNRUWFdW41bdrU3HnnnWbLli1my5YtdW4rLy/PzJkzx2zevNls377dPProoyYoKMh8++231pzY2FjTrFkzM3XqVFNYWGhmz55t/Pz8rG+nX7RokXE4HOb999833377rVm/fr156aWXjDHGHDhwwLRq1cpMmTLF+tmcfNwvv/xys3btWlNQUGCOHDlisrOzzd///neTn59vtm3bZkaMGGEiIiKM0+k0xhhTVVVlevXqZTp27Gg+/PBDs2PHDvPee++Z999//5TnAWAHoQlws5rQZIwxvXr1MsOHDzfGnH1oio2NNVVVVda6+Ph4c+WVV1qPT5w4YZo0aWLeeOMNY8x/A8RTTz1lzTl+/Lhp1aqVefrpp40xxjzxxBOmb9++LtvetWuXkWQKCwuNMT+FpksvvfS0+xsdHW2efPJJl3WXXXaZue+++6zHXbt2NZmZmb9YIzc310gyO3furHP8oosuMgsWLHBZ98QTT5jk5GRjzH/3+ZVXXrHGt27daiSZ/Px8Y8xPv4RDQkJcanz77bfGz8/PfP/99y7r+/TpYyZOnGg9T5L5+uuvrfGsrCwTERHhcgweeeSROnv/9NNPjcPhMMeOHau1T3/729+MMcY0a9bMCoB2SDKjR492WZeUlGTuvfdeY4wxf//73018fLyprq62xisqKkxwcLBZsWKFMeancysiIsIKUWeiY8eO5rnnnrMex8bGmuuvv95lzuDBg02/fv2MMcY8++yzpl27dqaysrLOerGxsS4h05j/HvdThUdjfgpJzZo1M++9954xxpgVK1YYX19f6zz+ubrOA8Au3p4D6tHTTz+t1157Tfn5+Wddo2PHjvL1/e9f1YiICHXu3Nl67Ofnp+bNm2vfvn0uz0tOTrb+3KhRI/Xo0cPq44svvtDHH3+spk2bWktCQoKkn96OqZGYmHjK3pxOp3bv3q0rrrjCZf0VV1xxRvvctWtX9enTR507d9att96ql19+2boe7MiRI9qxY4dGjBjh0u+f//xnl14lqUuXLtafo6KiJKnWcTnZ5s2bVVVVpXbt2rnUXrNmjUvtxo0b66KLLnKpXVN337592r17t/r06VPnNr744gsdPnxYzZs3d9lGUVGRtY2MjAyNHDlSKSkpeuqpp2rtV11O/vnWPD755/v111+rWbNm1vbCwsJ07Ngxl9qdO3dWQEDAKbdz+PBhPfDAA2rfvr1CQ0PVtGlT5efnq7i42HY/t956q8rLy9W2bVuNGjVKS5Ys0YkTJ067jwEBAS4/U0nau3evRo0apUsuuUQhISFyOBw6fPiw1U9eXp5atWqldu3anbY+cKYaeboB4Fx21VVXKTU1VRMnTtTdd9/tMubr6ytjjMu648eP16rh7+/v8tjHx6fOdWdykezhw4d144036umnn641VhM2JKlJkya2a/4afn5+WrlypdatW6cPP/xQzz33nB555BGtX79ejRs3liS9/PLLSkpKqvW8k518XGqujTnVcTl8+LD8/PyUm5tbq1bTpk3rrFtTu+ZnFxwcfMp9O3z4sKKiolyuk6pR8ymuyZMna8iQIVq2bJk++OADZWZm6s0339TNN998ytqn2mZiYqJef/31WmMtWrSw/mzn5/vAAw9o5cqV+utf/6qLL75YwcHBuuWWW87owvHWrVursLBQH330kVauXKn77rtPzzzzjNasWVPr2J4sODi41idO09LSdODAAc2aNUuxsbEKDAxUcnKy1c/pfh7Ar0FoAurZU089pW7dutW60LZFixYqKSmRMcb6xeDOeyt99tlnuuqqqyRJJ06cUG5urnVxcvfu3fXPf/5TcXFxatTo7P8ZcDgcio6O1tq1a3X11Vdb69euXauePXueUS0fHx9dccUVuuKKKzRp0iTFxsZqyZIlysjIUHR0tL755hsNHTr0rHsNCAhQVVWVy7pLL71UVVVV2rdvn6688sqzqtusWTPFxcUpOztbvXv3rjXevXt3lZSUqFGjRi4X+f9cu3bt1K5dO40fP16333675s2bd8rQ9Nlnn+muu+5yeXzppZda21y4cKFatmwph8NxVvtVY+3atbr77rutXg4fPuxyQfnJ2//54/bt21uPg4ODdeONN+rGG29Uenq6EhIStHnzZnXv3r3On82p+nnhhRd0ww03SPrpgveTP8DQpUsXfffdd9q+fXudrzadybaAn+PtOaCede7cWUOHDtXs2bNd1l9zzTXav3+/pk2bph07digrK0sffPCB27ablZWlJUuWqKCgQOnp6frxxx81fPhwSVJ6erp++OEH3X777dqwYYN27NihFStWaNiwYWf8C2XChAl6+umntXDhQhUWFurhhx9WXl6e7r//fts11q9fr7/85S/auHGjiouLtXjxYu3fv9/6pfv4449r6tSpmj17trZv367Nmzdr3rx5mj59uu1txMXF6fDhw8rOzlZpaamOHj2qdu3aaejQobrrrru0ePFiFRUV6fPPP9fUqVO1bNky27UnT56sZ599VrNnz9ZXX32lTZs26bnnnpP00ycpk5OTNWDAAH344YfauXOn1q1bp0ceeUQbN25UeXm5xowZo9WrV+vbb7/V2rVrtWHDBpfAUZdFixZp7ty52r59uzIzM/X5559boXjo0KEKDw/XTTfdpE8//VRFRUVavXq1/vjHP+q7776zvV+SdMkll2jx4sXKy8vTF198oSFDhtT56t3atWs1bdo0bd++XVlZWVq0aJF1DsyfP1+vvvqqtmzZom+++Ub/+Mc/FBwcrNjYWEk//Ww++eQTff/99y4B6Jf6+fvf/678/HytX79eQ4cOdXl16eqrr9ZVV12lQYMGaeXKlSoqKtIHH3xgfaKzrvMAsM3D11QB55yTLwSvUVRUZAICAszP/8q9+OKLpnXr1qZJkybmrrvuMk8++WStC8F/Xuvqq682999/v8u6ky+krbkoesGCBaZnz54mICDAdOjQwaxatcrlOdu3bzc333yzCQ0NNcHBwSYhIcGMGzfOuni4ru3UpaqqykyePNlceOGFxt/f33Tt2tV88MEHLnNOdyH4tm3bTGpqqmnRooUJDAw07dq1c7nQ2BhjXn/9ddOtWzcTEBBgLrjgAnPVVVeZxYsXu+zzf/7zH2v+jz/+aCSZjz/+2Fo3evRo07x5cyPJ6qeystJMmjTJxMXFGX9/fxMVFWVuvvlm8+WXXxpj6r5w+OcX9RtjzJw5c0x8fLxVY+zYsdaY0+k0Y8eONdHR0cbf39+0bt3aDB061BQXF5uKigpz2223mdatW5uAgAATHR1txowZY8rLy3/xeEkyWVlZ5rrrrjOBgYEmLi7OLFy40GXOnj17zF133WXCw8NNYGCgadu2rRk1apQpKyszxtR9btWlqKjI9O7d2wQHB5vWrVub559/vta5ERsbax5//HFz6623msaNG5vIyEgza9Ysl+OVlJRkHA6HadKkienVq5f56KOPrPGcnBzTpUsXExgYaB3XX7pge9OmTaZHjx4mKCjIXHLJJWbRokW1LiQ/cOCAGTZsmGnevLkJCgoynTp1MkuXLrXG6zoPADt8jPnZRRUAAK/m4+OjJUuWaMCAAZ5uBTiv8PYcAACADYQmAAAAG/j0HAA0MFxVAXgGrzQBAADYQGgCAACwgdAEAABgA6EJAADABkITAACADYQmAAAAGwhNAAAANhCaAAAAbPh//sT6Y6AKgiAAAAAASUVORK5CYII=\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Get lists of sentences"
      ],
      "metadata": {
        "id": "mp0xekKjRKHb"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Convert abstract text lines into lists\n",
        "train_sentences = train_df['text'].tolist()\n",
        "val_sentences = val_df['text'].tolist()\n",
        "test_sentences = test_df['text'].tolist()\n",
        "\n",
        "len(train_sentences), len(val_sentences), len(test_sentences)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "k9T_QGyYRSn7",
        "outputId": "76e532da-5c00-4e43-b84a-149286614111"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(180040, 30212, 30135)"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_sentences[:10]"
      ],
      "metadata": {
        "id": "hd6jc2HpR99d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "537d28b3-b5a9-4d64-8390-0b6f9ffaa4bf"
      },
      "execution_count": 20,
      "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": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Distribution of number of tokens per sentence\n",
        "num_tokens = []\n",
        "for sentence in train_sentences:\n",
        "  tokens = sentence.split()\n",
        "  num_tokens.append(len(tokens))\n",
        "\n",
        "sns.histplot(num_tokens, bins=70)\n",
        "plt.xlabel('Number of tokens per sentence')"
      ],
      "metadata": {
        "id": "JrrrJ55TZRnU",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 467
        },
        "outputId": "a2a2cf3e-c26a-4dfa-82dc-182887caaaf6"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Number of tokens per sentence')"
            ]
          },
          "metadata": {},
          "execution_count": 21
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Make numeric labels"
      ],
      "metadata": {
        "id": "AOcaxvbHbi7x"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# One hot encode labels\n",
        "from sklearn.preprocessing import OneHotEncoder\n",
        "one_hot_encoder = OneHotEncoder(sparse_output = False)\n",
        "train_labels_one_hot = one_hot_encoder.fit_transform(train_df['label'].to_numpy().reshape(-1,1))\n",
        "val_labels_one_hot = one_hot_encoder.transform(val_df['label'].to_numpy().reshape(-1,1))\n",
        "test_labels_one_hot = one_hot_encoder.transform(test_df['label'].to_numpy().reshape(-1,1))\n",
        "\n",
        "train_labels_one_hot"
      ],
      "metadata": {
        "id": "4BO10fdVcIYt",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3c4bd5d3-e331-4b5a-fa34-cb50a06be9b9"
      },
      "execution_count": 22,
      "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": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Label encode labels : encoding labels as integers\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "label_encoder = LabelEncoder()\n",
        "train_labels_encoded = label_encoder.fit_transform(train_df['label'].to_numpy())\n",
        "val_labels_encoded = label_encoder.transform(val_df['label'].to_numpy())\n",
        "test_labels_encoded = label_encoder.transform(test_df['label'].to_numpy())\n",
        "\n",
        "train_labels_encoded"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mawbyf-0k5QN",
        "outputId": "153e7fee-1c76-4c56-bbc8-b4ab438241a7"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([3, 2, 2, ..., 4, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Get class names and number of classes from LabelEncoder instance\n",
        "num_classes = len(label_encoder.classes_)\n",
        "class_names = label_encoder.classes_\n",
        "\n",
        "num_classes, class_names"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2ZcQBWL_l9h6",
        "outputId": "95fb635c-43e2-4590-e136-1a49a5439093"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(5,\n",
              " array(['BACKGROUND', 'CONCLUSIONS', 'METHODS', 'OBJECTIVE', 'RESULTS'],\n",
              "       dtype=object))"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model 0: Naive Bayes"
      ],
      "metadata": {
        "id": "j7kWCo3zmV1D"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.naive_bayes import MultinomialNB\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.metrics import confusion_matrix\n",
        "from sklearn.pipeline import Pipeline\n",
        "\n",
        "tfidf = TfidfVectorizer()\n",
        "\n",
        "X_train = tfidf.fit_transform(train_sentences)\n",
        "\n",
        "model_0 = MultinomialNB()\n",
        "\n",
        "model_0.fit(X_train, train_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 74
        },
        "id": "Z5Lvwxf5o5D6",
        "outputId": "4a59adea-9ad4-49a4-dd12-6707297932d9"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "MultinomialNB()"
            ],
            "text/html": [
              "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>MultinomialNB()</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">MultinomialNB</label><div class=\"sk-toggleable__content\"><pre>MultinomialNB()</pre></div></div></div></div></div>"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X_val = tfidf.transform(val_sentences)\n",
        "y_pred = model_0.predict(X_val)\n",
        "\n",
        "confusion_matrix(val_labels_encoded, y_pred)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EDzzNwc0o_hu",
        "outputId": "b6dea020-740e-4975-ee82-52cb715eee88"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1683,  794,  694,   97,  181],\n",
              "       [ 244, 2687,  432,    5, 1214],\n",
              "       [ 116,   75, 8670,    7, 1096],\n",
              "       [ 505,  422,  978,  325,  146],\n",
              "       [  20,  181, 1196,    1, 8443]])"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_0.score(X_val, val_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gyL_jOiOrfUv",
        "outputId": "23b97997-247c-484a-8ec9-3ec5ad3babda"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.7218323844829869"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://raw.githubusercontent.com/RishitSingh10/Multiclass-Image-Classification/main/helper_functions.py"
      ],
      "metadata": {
        "id": "UZzP5tns2ypP",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "57416ca0-42ac-4990-f319-8d1b78ef2d6f"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2023-10-25 05:16:30--  https://raw.githubusercontent.com/RishitSingh10/Multiclass-Image-Classification/main/helper_functions.py\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.109.133, 185.199.111.133, 185.199.108.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.109.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 11043 (11K) [text/plain]\n",
            "Saving to: ‘helper_functions.py’\n",
            "\n",
            "helper_functions.py 100%[===================>]  10.78K  --.-KB/s    in 0.001s  \n",
            "\n",
            "2023-10-25 05:16:31 (17.8 MB/s) - ‘helper_functions.py’ saved [11043/11043]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculate baseline results\n",
        "from helper_functions import print_evaluation_metrics\n",
        "baseline_results = print_evaluation_metrics(y_pred,val_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m8TyzlJHCy65",
        "outputId": "ba716da6-73ef-420a-d6ba-5a18618f25b1"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Confusion matrix:  [[1683  794  694   97  181]\n",
            " [ 244 2687  432    5 1214]\n",
            " [ 116   75 8670    7 1096]\n",
            " [ 505  422  978  325  146]\n",
            " [  20  181 1196    1 8443]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "baseline_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XWymGolYF5BA",
        "outputId": "ca84f863-ddd2-48bb-8170-06100f151211"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.7218323844829869,\n",
              " 'precision': 0.646068766530416,\n",
              " 'recall': 0.58642514185945,\n",
              " 'f1-score': 0.6148037981924266}"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Tokenization"
      ],
      "metadata": {
        "id": "hGlP2a-lF7nv"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Average number of tokens in each training sentence\n",
        "import tensorflow as tf\n",
        "tf.round(sum(num_tokens)/len(train_sentences))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kFadZMMVIof8",
        "outputId": "171ffa90-d56c-4cc3-f8c8-67877a148980"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(), dtype=float32, numpy=26.0>"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# How long of sentence length covers 95% of examples\n",
        "np.percentile(num_tokens, 95)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Px02J6ZNU5IX",
        "outputId": "fc329c24-8f92-4db6-a16b-f21a0eebac4e"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "55.0"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Maximum sequence length in the training set\n",
        "max(num_tokens)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cgZbgWj3VYz4",
        "outputId": "f3489644-6cf1-41c8-ea2e-0f43b2fb6fe2"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "296"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.layers.experimental.preprocessing import TextVectorization\n",
        "\n",
        "max_vocab_length = 68000 # maximum number of words to have in our vocabulary (taken from table 2 of the paper)\n",
        "max_length = 55 # max length of the sequence (how many words our model would see at time from a sequence)\n",
        "\n",
        "text_vectorizer = TextVectorization(max_tokens = max_vocab_length,\n",
        "                                    output_sequence_length = max_length,\n",
        "                                    output_mode = 'int')"
      ],
      "metadata": {
        "id": "AaVfPTy5G-bM"
      },
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Adapt text vectorizer on random sentences\n",
        "text_vectorizer.adapt(train_sentences)"
      ],
      "metadata": {
        "id": "9MfVzN7oHXwD"
      },
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sample_sentence = \"Research paper implementation using deep learning\"\n",
        "text_vectorizer([sample_sentence])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IC61uf3_IGSS",
        "outputId": "26a4c06d-329e-4db1-fea0-ccff51720094"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(1, 55), dtype=int64, numpy=\n",
              "array([[ 285, 1716,  754,   59, 1676,  953,    0,    0,    0,    0,    0,\n",
              "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
              "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
              "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0,\n",
              "           0,    0,    0,    0,    0,    0,    0,    0,    0,    0,    0]])>"
            ]
          },
          "metadata": {},
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Testing out text vectorizer on random sentences\n",
        "import random\n",
        "\n",
        "random_sentence = random.choice(train_sentences)\n",
        "print(\"Text:\\n\", random_sentence)\n",
        "print(\"\\nLength of text:\\n\", len(random_sentence.split()))\n",
        "print(\"\\nTokens:\\n\", text_vectorizer([random_sentence]))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qy2mHkl_WKoW",
        "outputId": "2effa199-5c2e-485f-c874-506d633f4375"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Text:\n",
            " Workplace sitting time ( minutes/@-hour workday ) assessed objectively via activPAL@ devices worn for @ days at baseline and @ months ( end-of-intervention ) .\n",
            "\n",
            "Length of text:\n",
            " 25\n",
            "\n",
            "Tokens:\n",
            " tf.Tensor(\n",
            "[[ 3737  2875    63 25728 13564   113  3503   598 23973  1342  7598    11\n",
            "     84    15    49     3    41 31479     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     0     0     0     0     0\n",
            "      0     0     0     0     0     0     0]], shape=(1, 55), dtype=int64)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# How many words in our training vocabulary\n",
        "rct_20k_text_vocab = text_vectorizer.get_vocabulary()\n",
        "print(\"Number of words in vocab: \", len(text_vectorizer.get_vocabulary()))\n",
        "print(\"\\nMost common words: \", rct_20k_text_vocab[:5])\n",
        "print(\"\\nLeast common words: \", rct_20k_text_vocab[-5:])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nYGUZseBXCiX",
        "outputId": "426de67e-0dab-4098-c7d2-40b32adf800d"
      },
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of words in vocab:  64841\n",
            "\n",
            "Most common words:  ['', '[UNK]', 'the', 'and', 'of']\n",
            "\n",
            "Least common words:  ['aainduced', 'aaigroup', 'aachener', 'aachen', 'aaacp']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Creating an embedding using Embedding layer"
      ],
      "metadata": {
        "id": "bTxPyiaXLLGB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "embedding = tf.keras.layers.Embedding(input_dim = len(rct_20k_text_vocab),\n",
        "                                      output_dim = 128,\n",
        "                                      mask_zero = True, # save space\n",
        "                                      input_length = max_length,\n",
        "                                      name=\"token_embedding\")\n",
        "\n",
        "embedding(text_vectorizer([sample_sentence]))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aj5Cry50IS2d",
        "outputId": "4f3b1f06-268a-4352-9d7c-0b03d020c17a"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(1, 55, 128), dtype=float32, numpy=\n",
              "array([[[ 0.00672934,  0.02650677, -0.02087051, ...,  0.02165282,\n",
              "          0.02051536, -0.04587524],\n",
              "        [ 0.04917165, -0.03862307,  0.03479278, ...,  0.01722375,\n",
              "         -0.04982775,  0.03939562],\n",
              "        [-0.02381491,  0.04400786,  0.00745978, ...,  0.01406615,\n",
              "          0.03512119, -0.02777462],\n",
              "        ...,\n",
              "        [ 0.01683818, -0.00310458, -0.03984574, ..., -0.03851879,\n",
              "          0.02161192,  0.03595467],\n",
              "        [ 0.01683818, -0.00310458, -0.03984574, ..., -0.03851879,\n",
              "          0.02161192,  0.03595467],\n",
              "        [ 0.01683818, -0.00310458, -0.03984574, ..., -0.03851879,\n",
              "          0.02161192,  0.03595467]]], dtype=float32)>"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Creating datasets (making sure our data loads as fast as possible)\n",
        "Setting up our data to run as fast as possible with the TensorFlow tf.data API."
      ],
      "metadata": {
        "id": "E1EiPkHG73BU"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Turn our data into TensorFLow Datasets\n",
        "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": "pVgGYSlT8Qme",
        "outputId": "15c8deef-5c49-41da-c0bf-89f2551dae11"
      },
      "execution_count": 40,
      "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": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Take the TensorSliceDataset's and turn them into prefetched datasets\n",
        "train_dataset = train_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "valid_dataset = valid_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "test_dataset = test_dataset.batch(32).prefetch(tf.data.AUTOTUNE)"
      ],
      "metadata": {
        "id": "dcLnMPcd-Ial"
      },
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model_1: Conv1D"
      ],
      "metadata": {
        "id": "6wG2OWQxLZbh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating a Sequential model\n",
        "model_1 = tf.keras.Sequential([\n",
        "    tf.keras.layers.Input(shape = (1,), dtype = tf.string),\n",
        "    text_vectorizer,\n",
        "    embedding,\n",
        "    tf.keras.layers.Conv1D(32, 5, activation='relu'),\n",
        "    tf.keras.layers.GlobalAveragePooling1D(),\n",
        "    tf.keras.layers.Dense(5, activation = 'softmax')\n",
        "])\n",
        "\n",
        "# Compiling the model\n",
        "model_1.compile(loss = 'categorical_crossentropy',\n",
        "                optimizer = tf.keras.optimizers.Adam(),\n",
        "                metrics = ['accuracy'])\n",
        "\n",
        "# fit the model\n",
        "history_1 = model_1.fit(train_dataset,\n",
        "                        epochs = 3,\n",
        "                        validation_data = valid_dataset,\n",
        "                        validation_steps = int(0.15*(len(val_sentences))))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xvnWT0qsJfTh",
        "outputId": "11d92b6f-3e2f-4239-fec5-47bf0ada3ca6"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/3\n",
            "5627/5627 [==============================] - ETA: 0s - loss: 0.6388 - accuracy: 0.7670"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:tensorflow:Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 4531 batches). You may need to use the repeat() function when building your dataset.\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\b\r5627/5627 [==============================] - 69s 11ms/step - loss: 0.6388 - accuracy: 0.7670 - val_loss: 0.5436 - val_accuracy: 0.8059\n",
            "Epoch 2/3\n",
            "5627/5627 [==============================] - 36s 6ms/step - loss: 0.4649 - accuracy: 0.8366\n",
            "Epoch 3/3\n",
            "5627/5627 [==============================] - 36s 6ms/step - loss: 0.3863 - accuracy: 0.8660\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-i4Y8K-cN36N",
        "outputId": "066f5ee1-a1bf-47d3-a2ec-c083bef72488"
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " text_vectorization (TextVe  (None, 55)                0         \n",
            " ctorization)                                                    \n",
            "                                                                 \n",
            " token_embedding (Embedding  (None, 55, 128)           8299648   \n",
            " )                                                               \n",
            "                                                                 \n",
            " conv1d (Conv1D)             (None, 51, 32)            20512     \n",
            "                                                                 \n",
            " global_average_pooling1d (  (None, 32)                0         \n",
            " GlobalAveragePooling1D)                                         \n",
            "                                                                 \n",
            " dense (Dense)               (None, 5)                 165       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 8320325 (31.74 MB)\n",
            "Trainable params: 8320325 (31.74 MB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluate on whole validation dataset\n",
        "model_1.evaluate(valid_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fqQpBon8PPTy",
        "outputId": "86382e3a-53ff-40bf-fca7-6c875d09fc52"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 3s 3ms/step - loss: 0.5580 - accuracy: 0.8073\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.5579712390899658, 0.8072620034217834]"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Make predictions\n",
        "model_1_pred_probs = model_1.predict(valid_dataset)\n",
        "model_1_pred_probs, model_1_pred_probs.shape"
      ],
      "metadata": {
        "id": "0i8EFMegZjcb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2da8349c-e07a-46b7-95da-ae64ce7f3cfe"
      },
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 3s 3ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[6.71610653e-01, 2.85100434e-02, 1.18667297e-01, 1.01118095e-01,\n",
              "         8.00938532e-02],\n",
              "        [5.87751150e-01, 1.69166774e-01, 7.20219081e-03, 2.02442870e-01,\n",
              "         3.34370360e-02],\n",
              "        [1.15211353e-01, 3.71399405e-03, 6.03280962e-04, 8.80461156e-01,\n",
              "         1.01502992e-05],\n",
              "        ...,\n",
              "        [2.62011440e-07, 4.09022368e-06, 1.45982963e-03, 4.84112775e-07,\n",
              "         9.98535275e-01],\n",
              "        [5.05341440e-02, 6.55810177e-01, 8.91496763e-02, 7.83277117e-03,\n",
              "         1.96673200e-01],\n",
              "        [7.34659145e-03, 9.90247130e-01, 2.12443550e-03, 4.55240661e-05,\n",
              "         2.36386768e-04]], dtype=float32),\n",
              " (30212, 5))"
            ]
          },
          "metadata": {},
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Convert pred probs to classes\n",
        "model_1_preds = model_1_pred_probs.argmax(axis=1)\n",
        "model_1_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8QHYHjS4BtDO",
        "outputId": "3f338eaf-e183-45a4-85e4-e045f6c03418"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 0, 3, ..., 4, 1, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Calculate model_1 results\n",
        "model_1_results = print_evaluation_metrics(val_labels_encoded,\n",
        "                                           model_1_preds)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wCxpVly6CIXd",
        "outputId": "65d5e4b0-5f66-4e19-f9bb-319b114ba9f2"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Confusion matrix:  [[2223  377  125  713   45]\n",
            " [ 499 3319  102  199  363]\n",
            " [ 256  150 8845  178  664]\n",
            " [ 399   56   63 1247   14]\n",
            " [  72  680  829   39 8755]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_1_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vDMB__ovCQPw",
        "outputId": "cfcd226d-991f-4f0f-be29-b70018c52f35"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.8072620150933404,\n",
              " 'precision': 0.7243561763422086,\n",
              " 'recall': 0.7405176260597948,\n",
              " 'f1-score': 0.7323477493380407}"
            ]
          },
          "metadata": {},
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "baseline_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VVmn3B5GD0u6",
        "outputId": "6051a33f-aa1c-4e76-ab60-8ef999e0086f"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.7218323844829869,\n",
              " 'precision': 0.646068766530416,\n",
              " 'recall': 0.58642514185945,\n",
              " 'f1-score': 0.6148037981924266}"
            ]
          },
          "metadata": {},
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model 2: Feature extraction with pretrained token embeddings\n",
        "Using pretrained word embeddings from TensorFlow Hub, Universal Sentence Encoder (USE)"
      ],
      "metadata": {
        "id": "iHKv0TTJD2sM"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Download pretrained TnesorFlow Hub USE\n",
        "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\")"
      ],
      "metadata": {
        "id": "nMOn_rV0O6Tj"
      },
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Testing out the pretrained embedding on a random sentence\n",
        "random_train_sentence = random.choice(train_sentences)\n",
        "print(f\"Random sentence:\\n {random_train_sentence}\")\n",
        "use_embedded_sentence = tf_hub_embedding_layer([random_train_sentence])\n",
        "print(f\"Sentence after embedding:\\n{use_embedded_sentence[0][:30]}\\n\")\n",
        "print(f\"Length of sentence embedding: {len(use_embedded_sentence[0])}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3TkGjXCOQSsN",
        "outputId": "8089ee02-cf3b-4b29-f2c2-68510631ba80"
      },
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Random sentence:\n",
            " Group @ was instructed to self-weigh at the same time once per day and group @ was instructed to self-weigh immediately after waking up in the morning and immediately before going to bed every day for twelve weeks .\n",
            "Sentence after embedding:\n",
            "[ 0.02976039 -0.06562553 -0.02510425  0.07249914 -0.04580202 -0.00053815\n",
            " -0.0338756  -0.03716788 -0.03966338  0.05528587  0.00341866  0.05516094\n",
            "  0.00951776 -0.05806515  0.0044857   0.04410116 -0.00364564  0.00965701\n",
            "  0.02384832  0.02972046 -0.04654155  0.07259403 -0.04849371 -0.07879612\n",
            " -0.02733787 -0.01986016 -0.04512879 -0.02203325 -0.08342368 -0.01413936]\n",
            "\n",
            "Length of sentence embedding: 512\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from keras.src.layers import GlobalAveragePooling1D\n",
        "# Creating model 2\n",
        "model_2 = tf.keras.Sequential([\n",
        "    tf_hub_embedding_layer,\n",
        "    tf.keras.layers.Dense(5, activation = 'softmax')\n",
        "])\n",
        "\n",
        "# Compiling the model\n",
        "model_2.compile(loss = 'categorical_crossentropy',\n",
        "                optimizer = tf.keras.optimizers.Adam(),\n",
        "                metrics = ['accuracy'])"
      ],
      "metadata": {
        "id": "NcAtANYZW4QJ"
      },
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Fitting the model\n",
        "hsitory_2 = model_2.fit(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": "pMG4nhL4akoH",
        "outputId": "0bbcb3ae-dfda-4a1b-be84-4078acbfe34e"
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/3\n",
            "5627/5627 [==============================] - 73s 12ms/step - loss: 0.9026 - accuracy: 0.6644 - val_loss: 0.7982 - val_accuracy: 0.6981\n",
            "Epoch 2/3\n",
            "5627/5627 [==============================] - 71s 13ms/step - loss: 0.7867 - accuracy: 0.7006 - val_loss: 0.7744 - val_accuracy: 0.7045\n",
            "Epoch 3/3\n",
            "5627/5627 [==============================] - 71s 13ms/step - loss: 0.7721 - accuracy: 0.7054 - val_loss: 0.7651 - val_accuracy: 0.7101\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2.evaluate(valid_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v4iFwJJbb7AC",
        "outputId": "f2e55785-34ae-41b6-9ac6-ee2583fe6276"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 12s 12ms/step - loss: 0.7623 - accuracy: 0.7094\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.7623432874679565, 0.7093870043754578]"
            ]
          },
          "metadata": {},
          "execution_count": 54
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluating the model\n",
        "model_2_pred_probs = model_2.predict(valid_dataset)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cpbOWY1vefOR",
        "outputId": "fbdb4784-13f3-4eb0-e993-74608867a22d"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 10s 10ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2_preds = model_2_pred_probs.argmax(axis=1)\n",
        "model_2_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3w0tRse6fOeI",
        "outputId": "8c8f1ca0-4db9-4480-ef6f-4c0e39da4060"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 1, 3, ..., 4, 4, 2])"
            ]
          },
          "metadata": {},
          "execution_count": 56
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_2_results = print_evaluation_metrics(model_2_preds, val_labels_encoded)\n",
        "model_2_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JVD7_nDrOI0Z",
        "outputId": "7ea3e6a8-6257-4ead-8221-f358af4df9ea"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Confusion matrix:  [[1845  702  397  361  144]\n",
            " [ 663 2568  423  140  788]\n",
            " [ 282  197 8101  108 1276]\n",
            " [ 520  326  355 1073  102]\n",
            " [  92  394 1488   22 7845]]\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.7093869985436251,\n",
              " 'precision': 0.6133269644136613,\n",
              " 'recall': 0.5604539502400698,\n",
              " 'f1-score': 0.5856996236743072}"
            ]
          },
          "metadata": {},
          "execution_count": 57
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "baseline_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZSiOvzslOktO",
        "outputId": "8c0d3566-3e64-4e9e-dfe1-b9b1f4b0b7db"
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.7218323844829869,\n",
              " 'precision': 0.646068766530416,\n",
              " 'recall': 0.58642514185945,\n",
              " 'f1-score': 0.6148037981924266}"
            ]
          },
          "metadata": {},
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model 3: Conv1D with character embeddings\n",
        "\n",
        "The paper states they used a combination of token and character-level embedding"
      ],
      "metadata": {
        "id": "YRfKEL2zfZB1"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Creating a character-level tokenizer"
      ],
      "metadata": {
        "id": "JFKLF0CBzh9M"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Function to split sentences into characters\n",
        "def split_chars(text):\n",
        "  return \" \".join(list(text))\n",
        "\n",
        "split_chars(random_train_sentence)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        },
        "id": "66omdxGw0AAz",
        "outputId": "cf36224f-9132-40ba-a32a-f71efb58f1da"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'G r o u p   @   w a s   i n s t r u c t e d   t o   s e l f - w e i g h   a t   t h e   s a m e   t i m e   o n c e   p e r   d a y   a n d   g r o u p   @   w a s   i n s t r u c t e d   t o   s e l f - w e i g h   i m m e d i a t e l y   a f t e r   w a k i n g   u p   i n   t h e   m o r n i n g   a n d   i m m e d i a t e l y   b e f o r e   g o i n g   t o   b e d   e v e r y   d a y   f o r   t w e l v e   w e e k s   .'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Split sequence-level data splits into character-level data splits\n",
        "train_chars = [split_chars(sentence) for sentence in train_sentences]\n",
        "val_chars = [split_chars(sentence) for sentence in val_sentences]\n",
        "test_chars = [split_chars(sentence) for sentence in test_sentences]\n",
        "train_chars[:5]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HcwI5SVp1BlC",
        "outputId": "5cb8f324-6224-46d1-d9a2-0605e2248916"
      },
      "execution_count": 60,
      "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   )   .',\n",
              " 'A   t o t a l   o f   @   p a t i e n t s   w i t h   p r i m a r y   k n e e   O A   w e r e   r a n d o m i z e d   @ : @   ;   @   r e c e i v e d   @   m g / d a y   o f   p r e d n i s o l o n e   a n d   @   r e c e i v e d   p l a c e b o   f o r   @   w e e k s   .',\n",
              " 'O u t c o m e   m e a s u r e s   i n c l u d e d   p a i n   r e d u c t i o n   a n d   i m p r o v e m e n t   i n   f u n c t i o n   s c o r e s   a n d   s y s t e m i c   i n f l a m m a t i o n   m a r k e r s   .',\n",
              " 'P a i n   w a s   a s s e s s e d   u s i n g   t h e   v i s u a l   a n a l o g   p a i n   s c a l e   (   @ - @   m m   )   .',\n",
              " 'S e c o n d a r y   o u t c o m e   m e a s u r e s   i n c l u d e d   t h e   W e s t e r n   O n t a r i o   a n d   M c M a s t e r   U n i v e r s i t i e s   O s t e o a r t h r i t i s   I n d e x   s c o r e s   ,   p a t i e n t   g l o b a l   a s s e s s m e n t   (   P G A   )   o f   t h e   s e v e r i t y   o f   k n e e   O A   ,   a n d   @ - m i n   w a l k   d i s t a n c e   (   @ M W D   )   .']"
            ]
          },
          "metadata": {},
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# What's the average character length\n",
        "num_chars = [len(sentence) for sentence in train_sentences]\n",
        "mean_char_len = np.mean(num_chars)\n",
        "mean_char_len"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XJ4RAFNX1qzM",
        "outputId": "f84348c1-a730-49bd-fc16-810e520bd664"
      },
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "149.3662574983337"
            ]
          },
          "metadata": {},
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sns.histplot(num_chars, bins = 70)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 448
        },
        "id": "5ujGCBnG32Lb",
        "outputId": "cf27641c-fdee-4c19-c56c-e5fbbe638fab"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: ylabel='Count'>"
            ]
          },
          "metadata": {},
          "execution_count": 62
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Find what character length covers 95% of sequences\n",
        "output_seq_char_len = int(np.percentile(num_chars, 95))\n",
        "output_seq_char_len"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oA2RSpVA4H3Q",
        "outputId": "6009da27-e793-4c2f-cc0a-d007a7808cc5"
      },
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "290"
            ]
          },
          "metadata": {},
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Get all keyboard characters\n",
        "import string\n",
        "alphabet = string.ascii_lowercase + string.digits + string.punctuation\n",
        "alphabet"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "nuB1i3pl5Epe",
        "outputId": "ce0c8288-70a6-48ce-bf5f-f102bd6209b3"
      },
      "execution_count": 64,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'abcdefghijklmnopqrstuvwxyz0123456789!\"#$%&\\'()*+,-./:;<=>?@[\\\\]^_`{|}~'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 64
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Create char-level token vectorizer instance\n",
        "NUM_CHAR_TOKENS = len(alphabet) + 2 # add 2 for space and OOV token (OOV = out of vocab, '{UNK}')\n",
        "char_vectorizer = TextVectorization(max_tokens = NUM_CHAR_TOKENS,\n",
        "                                      output_sequence_length = output_seq_char_len,\n",
        "                                      name = \"char_vectorizer\")"
      ],
      "metadata": {
        "id": "uRYKjP7O5pce"
      },
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Adapt character vectorizer to training character\n",
        "char_vectorizer.adapt(train_chars)"
      ],
      "metadata": {
        "id": "rBz5w16b5tjn"
      },
      "execution_count": 66,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Check character vocab stats\n",
        "char_vocab = char_vectorizer.get_vocabulary()\n",
        "print(f\"Number of different charcaters 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": "dEljv-Sk7r-L",
        "outputId": "eabae336-8d4a-43e0-e103-7b0a8b5b6c35"
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of different charcaters 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": [
        "# Test out character vectorizer\n",
        "random_train_chars = random.choice(train_chars)\n",
        "print(f\"Charified text:\\n {random_train_chars}\")\n",
        "print(f\"Length of random_train_chars: {len(random_train_chars.split())}\")\n",
        "vectorized_chars = char_vectorizer([random_train_chars])\n",
        "print(f\"Vectorized chars:\\n {vectorized_chars}\")\n",
        "print(f\"\\nLength of vectorized chars: {len(vectorized_chars[0])}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "onaiDk9t74PK",
        "outputId": "2426bc14-fc50-4ade-ece9-28949046239f"
      },
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Charified text:\n",
            " T o t a l   ,   p e l v i c   ,   a n d   a r m - l e g   B M C   d e c r e a s e d   d u r i n g   w e i g h t   m a i n t e n a n c e   i n   t h e   c o n t r o l   g r o u p   (   P   <   @   )   ,   b u t   n o t   s i g n i f i c a n t l y   i n   t h e   l i r a g l u t i d e   g r o u p   .\n",
            "Length of random_train_chars: 122\n",
            "Vectorized chars:\n",
            " [[ 3  7  3  5 12 14  2 12 21  4 11  5  6 10  5  8 15 12  2 18 22 15 11 10\n",
            "   2 11  8  2  5  9  2 10 10 16  8  4  6 18 20  2  4 18 13  3 15  5  4  6\n",
            "   3  2  6  5  6 11  2  4  6  3 13  2 11  7  6  3  8  7 12 18  8  7 16 14\n",
            "  14 22 16  3  6  7  3  9  4 18  6  4 17  4 11  5  6  3 12 19  4  6  3 13\n",
            "   2 12  4  8  5 18 12 16  3  4 10  2 18  8  7 16 14  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": "markdown",
      "source": [
        "### Creating a character-level embedding"
      ],
      "metadata": {
        "id": "NnT1XQ4e-NpV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating char embedding layer\n",
        "char_embed = tf.keras.layers.Embedding(input_dim = len(char_vocab), # number of different characters\n",
        "                                       output_dim = 25, # this is the size of the char embedding in the paper: https://arxiv.org/pdf/1612.05251.pdf (figure 1)\n",
        "                                       mask_zero = True,\n",
        "                                       name = 'char_embed')"
      ],
      "metadata": {
        "id": "uhlIdWDFyqsA"
      },
      "execution_count": 69,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Test our character embedding layer\n",
        "print(f\"Charified text:\\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",
        "print(f\"Character embedding shape: {char_embed_example.shape}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AAcXdNAk0C3T",
        "outputId": "48321548-c238-4fea-c3d6-c9e00ad01372"
      },
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Charified text:\n",
            "T o t a l   ,   p e l v i c   ,   a n d   a r m - l e g   B M C   d e c r e a s e d   d u r i n g   w e i g h t   m a i n t e n a n c e   i n   t h e   c o n t r o l   g r o u p   (   P   <   @   )   ,   b u t   n o t   s i g n i f i c a n t l y   i n   t h e   l i r a g l u t i d e   g r o u p   .\n",
            "\n",
            "Embedded chars. (after vectorization and embedding):\n",
            " [[[-0.01565876  0.01432519  0.03274189 ...  0.03292527  0.04242125\n",
            "    0.04404254]\n",
            "  [ 0.04897071 -0.04538006 -0.03327845 ... -0.03483269  0.04951486\n",
            "   -0.00011095]\n",
            "  [-0.01565876  0.01432519  0.03274189 ...  0.03292527  0.04242125\n",
            "    0.04404254]\n",
            "  ...\n",
            "  [ 0.0397389  -0.02455981 -0.03255949 ... -0.01956769 -0.02067075\n",
            "   -0.00363652]\n",
            "  [ 0.0397389  -0.02455981 -0.03255949 ... -0.01956769 -0.02067075\n",
            "   -0.00363652]\n",
            "  [ 0.0397389  -0.02455981 -0.03255949 ... -0.01956769 -0.02067075\n",
            "   -0.00363652]]]\n",
            "Character embedding shape: (1, 290, 25)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Building a Conv1D model to fit on character embeddings\n"
      ],
      "metadata": {
        "id": "qqUY5HfI4e1k"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_3 = tf.keras.Sequential([\n",
        "    tf.keras.layers.Input(shape = (1,), dtype = tf.string ),\n",
        "    char_vectorizer,\n",
        "    char_embed,\n",
        "    tf.keras.layers.Conv1D(64, 5,padding = 'same', activation = 'relu'),\n",
        "    tf.keras.layers.GlobalMaxPooling1D(),\n",
        "    tf.keras.layers.Dense(5, activation = 'softmax')\n",
        "])\n",
        "\n",
        "model_3.compile(loss = 'categorical_crossentropy',\n",
        "                optimizer = 'adam',\n",
        "                metrics = ['accuracy'])"
      ],
      "metadata": {
        "id": "idT40IzQ0zSj"
      },
      "execution_count": 71,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_3.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bXGXf4vJ408K",
        "outputId": "86ecf636-4f4c-4f28-9c8f-0bbe1d1d59d3"
      },
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_2\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " char_vectorizer (TextVecto  (None, 290)               0         \n",
            " rization)                                                       \n",
            "                                                                 \n",
            " char_embed (Embedding)      (None, 290, 25)           700       \n",
            "                                                                 \n",
            " conv1d_1 (Conv1D)           (None, 290, 64)           8064      \n",
            "                                                                 \n",
            " global_max_pooling1d (Glob  (None, 64)                0         \n",
            " alMaxPooling1D)                                                 \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 5)                 325       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 9089 (35.50 KB)\n",
            "Trainable params: 9089 (35.50 KB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Create char level datasets\n",
        "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",
        "test_char_dataset = tf.data.Dataset.from_tensor_slices((test_chars, test_labels_one_hot)).batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "\n",
        "train_char_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "b0uEWApu5U6V",
        "outputId": "cd4e217e-fdf9-4548-f559-94afc51377fe"
      },
      "execution_count": 73,
      "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": 73
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# fit the model\n",
        "history_3 = model_3.fit(train_char_dataset,\n",
        "                        epochs = 3,\n",
        "                        validation_data = val_char_dataset,\n",
        "                        validation_steps = int(.1*len(valid_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C5NGehjP1CtY",
        "outputId": "8f6a6aad-47bb-4fb9-ed79-bb5482e68941"
      },
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/3\n",
            "5627/5627 [==============================] - 34s 6ms/step - loss: 0.8950 - accuracy: 0.6478 - val_loss: 0.7656 - val_accuracy: 0.7055\n",
            "Epoch 2/3\n",
            "5627/5627 [==============================] - 31s 6ms/step - loss: 0.7483 - accuracy: 0.7147 - val_loss: 0.7253 - val_accuracy: 0.7188\n",
            "Epoch 3/3\n",
            "5627/5627 [==============================] - 32s 6ms/step - loss: 0.7144 - accuracy: 0.7290 - val_loss: 0.7107 - val_accuracy: 0.7291\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3.evaluate(val_char_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pCUvoLK57Vxk",
        "outputId": "9db3661a-ce7a-4319-8c5e-17fcc7a2691a"
      },
      "execution_count": 75,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 5s 5ms/step - loss: 0.7158 - accuracy: 0.7289\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.7158170342445374, 0.7288825511932373]"
            ]
          },
          "metadata": {},
          "execution_count": 75
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Make predictions with character model only\n",
        "model_3_pred_probs = model_3.predict(val_char_dataset)\n",
        "model_3_pred_probs"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0h2KZodi4AmG",
        "outputId": "a8861f7f-ee20-47cf-b1da-39aa55966a0f"
      },
      "execution_count": 76,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 3s 3ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.33066198, 0.12632932, 0.06196222, 0.46654278, 0.01450371],\n",
              "       [0.37370494, 0.5595301 , 0.00131499, 0.06190281, 0.00354715],\n",
              "       [0.24820685, 0.02495987, 0.04088579, 0.6832481 , 0.00269943],\n",
              "       ...,\n",
              "       [0.00554553, 0.014908  , 0.01419033, 0.00305665, 0.96229947],\n",
              "       [0.0364136 , 0.01680026, 0.45804924, 0.01587128, 0.47286564],\n",
              "       [0.10020638, 0.87967336, 0.01502796, 0.00353095, 0.00156137]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 76
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3_preds = model_3_pred_probs.argmax(axis=1)\n",
        "model_3_preds"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E0Y6ouT07VNb",
        "outputId": "484735c8-3433-4a62-f36f-676189f5e207"
      },
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([3, 1, 3, ..., 4, 4, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 77
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_3_results = print_evaluation_metrics(model_3_preds,\n",
        "                                           val_labels_encoded)\n",
        "model_3_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xE3qJ1Sj734V",
        "outputId": "b2d57912-0633-4994-db2f-41c53da87f36"
      },
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Confusion matrix:  [[1841  730  328  320  230]\n",
            " [ 556 2601  260   80 1085]\n",
            " [ 162  155 7895   74 1678]\n",
            " [ 653  338  248 1002  135]\n",
            " [  72  317  765    5 8682]]\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.7288825632199126,\n",
              " 'precision': 0.6281091523786525,\n",
              " 'recall': 0.567656045395024,\n",
              " 'f1-score': 0.5963544652069243}"
            ]
          },
          "metadata": {},
          "execution_count": 78
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "baseline_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ETE018G68NpT",
        "outputId": "1b859180-49a0-4ad7-96d4-fb0fa586f34f"
      },
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.7218323844829869,\n",
              " 'precision': 0.646068766530416,\n",
              " 'recall': 0.58642514185945,\n",
              " 'f1-score': 0.6148037981924266}"
            ]
          },
          "metadata": {},
          "execution_count": 79
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model 4: Combining pretrained token embeddings +  characters embeddings (hybrid embedding layer)\n",
        "\n",
        "1. Create a token level embedding model (similar `model_1`)\n",
        "2. Create a character level embedding model (similar to `model_3` with a slight modification)\n",
        "3. Combine 1 & 2 with a concatenate (`layers.concatenate`)\n",
        "4. Build a series of output layers on top of 3 similar to Figure 1 and section 4.2 of https://arxiv.org/pdf/1612.05251.pdf\n",
        "5. Construct a model which takes token and character-level sequences as input and produces sequence label probabilities as output"
      ],
      "metadata": {
        "id": "3WZRmJ3f8QRz"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# 1. Setup token inputs/model\n",
        "token_inputs = tf.keras.layers.Input(shape = [], dtype = tf.string, name = 'token_input')\n",
        "token_embeddings = tf_hub_embedding_layer(token_inputs)\n",
        "token_outputs = tf.keras.layers.Dense(128, activation = 'relu')(token_embeddings)\n",
        "token_model = tf.keras.Model(inputs = token_inputs,\n",
        "                             outputs = token_outputs)\n",
        "\n",
        "# 2. Setup char inputs/model\n",
        "char_inputs = tf.keras.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 = tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(24))(char_embeddings) # bi-LSTM shown in Figure 1 of https://arxiv.org/pdf/1612.05251.pdf\n",
        "char_model = tf.keras.Model(inputs = char_inputs,\n",
        "                            outputs = char_bi_lstm)\n",
        "\n",
        "# 3. Concatenate token and char inputs (create a hybrid token embedding)\n",
        "token_char_concat = tf.keras.layers.Concatenate(name = 'token_char_hybrid')([token_model.output,\n",
        "                                                                             char_model.output])\n",
        "\n",
        "\n",
        "# 4. Create output layers - adding in Dropout, discussed in section 4.2 of https://arxiv.org/pdf/1612.05251.pdf\n",
        "combined_dropout = tf.keras.layers.Dropout(0.5)(token_char_concat)\n",
        "combined_dense = tf.keras.layers.Dense(128, activation = 'relu')(combined_dropout)\n",
        "final_dropout = tf.keras.layers.Dropout(0.5)(combined_dense)\n",
        "output_layer = tf.keras.layers.Dense(num_classes, activation = 'softmax')(final_dropout)\n",
        "\n",
        "# 5. Construct model with char and token inputs\n",
        "model_4 = tf.keras.Model(inputs = [token_model.input, char_model.input],\n",
        "                         outputs = output_layer,\n",
        "                         name = 'model_4_token_char_embeddings')"
      ],
      "metadata": {
        "id": "PfCJngy7MNBo"
      },
      "execution_count": 80,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_4.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Tgs1WTOkR6wD",
        "outputId": "f0cc7ec0-a105-430f-b40f-2b63fb1346a7"
      },
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_4_token_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 (TextVecto  (None, 290)                  0         ['char_input[0][0]']          \n",
            " rization)                                                                                        \n",
            "                                                                                                  \n",
            " universal_sentence_encoder  (None, 512)                  2567978   ['token_input[0][0]']         \n",
            "  (KerasLayer)                                            24                                      \n",
            "                                                                                                  \n",
            " char_embed (Embedding)      (None, 290, 25)              700       ['char_vectorizer[1][0]']     \n",
            "                                                                                                  \n",
            " dense_3 (Dense)             (None, 128)                  65664     ['universal_sentence_encoder[1\n",
            "                                                                    ][0]']                        \n",
            "                                                                                                  \n",
            " bidirectional (Bidirection  (None, 48)                   9600      ['char_embed[1][0]']          \n",
            " al)                                                                                              \n",
            "                                                                                                  \n",
            " token_char_hybrid (Concate  (None, 176)                  0         ['dense_3[0][0]',             \n",
            " nate)                                                               'bidirectional[0][0]']       \n",
            "                                                                                                  \n",
            " dropout (Dropout)           (None, 176)                  0         ['token_char_hybrid[0][0]']   \n",
            "                                                                                                  \n",
            " dense_4 (Dense)             (None, 128)                  22656     ['dropout[0][0]']             \n",
            "                                                                                                  \n",
            " dropout_1 (Dropout)         (None, 128)                  0         ['dense_4[0][0]']             \n",
            "                                                                                                  \n",
            " dense_5 (Dense)             (None, 5)                    645       ['dropout_1[0][0]']           \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 256897089 (979.98 MB)\n",
            "Trainable params: 99265 (387.75 KB)\n",
            "Non-trainable params: 256797824 (979.61 MB)\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot hybrid token and character model\n",
        "from keras.utils import plot_model\n",
        "plot_model(model_4, show_shapes = True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 976
        },
        "id": "X2561P2MSEO9",
        "outputId": "1169c22a-a3ad-40ad-e845-e98de122ad35"
      },
      "execution_count": 82,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile token char model\n",
        "model_4.compile(loss = 'categorical_crossentropy',\n",
        "                optimizer = tf.keras.optimizers.Adam(),\n",
        "                metrics = ['accuracy'])"
      ],
      "metadata": {
        "id": "GokDgAGJVXc_"
      },
      "execution_count": 83,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Combining token and character data into a tf.data Dataset"
      ],
      "metadata": {
        "id": "-QPivjlhWn9q"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Combine chars and tokens into dataset\n",
        "train_char_token_data = tf.data.Dataset.from_tensor_slices((train_sentences, train_chars)) # make data\n",
        "train_char_token_labels = tf.data.Dataset.from_tensor_slices(train_labels_one_hot) # make labels\n",
        "train_char_token_dataset = tf.data.Dataset.zip((train_char_token_data, train_char_token_labels))\n",
        "\n",
        "# Prefetch and batch train data\n",
        "train_char_token_dataset = train_char_token_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "\n",
        "# Repeating the above steps for validation data\n",
        "\n",
        "# Combine chars and tokens into dataset\n",
        "val_char_token_data = tf.data.Dataset.from_tensor_slices((val_sentences, val_chars)) # make data\n",
        "val_char_token_labels = tf.data.Dataset.from_tensor_slices(val_labels_one_hot) # make labels\n",
        "val_char_token_dataset = tf.data.Dataset.zip((val_char_token_data, val_char_token_labels))\n",
        "\n",
        "# Prefetch and batch train data\n",
        "val_char_token_dataset = val_char_token_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n"
      ],
      "metadata": {
        "id": "KGce0ugYXH7G"
      },
      "execution_count": 84,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Check our training dataset\n",
        "train_char_token_dataset, val_char_token_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WWbjCzHFY3w8",
        "outputId": "9f4310f5-90d7-44d1-9535-921bd5f21b3c"
      },
      "execution_count": 85,
      "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": 85
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Fitting a model on token and character-level sequences"
      ],
      "metadata": {
        "id": "n6JAgwa6aZq5"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Fitting model_4 (hybrid model)\n",
        "history_4 = model_4.fit(train_char_token_dataset,\n",
        "                        epochs = 3,\n",
        "                        validation_data = val_char_token_dataset,\n",
        "                        validation_steps = int(.1*len(val_char_token_dataset)))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FkCqVXsUZQMm",
        "outputId": "d0c235b0-a071-49c5-82f1-094902084e35"
      },
      "execution_count": 86,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/3\n",
            "5627/5627 [==============================] - 274s 46ms/step - loss: 0.7609 - accuracy: 0.7093 - val_loss: 0.6181 - val_accuracy: 0.7670\n",
            "Epoch 2/3\n",
            "5627/5627 [==============================] - 249s 44ms/step - loss: 0.6735 - accuracy: 0.7464 - val_loss: 0.5882 - val_accuracy: 0.7856\n",
            "Epoch 3/3\n",
            "5627/5627 [==============================] - 250s 44ms/step - loss: 0.6505 - accuracy: 0.7558 - val_loss: 0.5782 - val_accuracy: 0.7886\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluating the model on whole validation dataset\n",
        "model_4.evaluate(val_char_token_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "myjMPCSpZ7wh",
        "outputId": "419fa562-01bf-465e-ebb4-f7fe566cee50"
      },
      "execution_count": 87,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 19s 20ms/step - loss: 0.5785 - accuracy: 0.7816\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.578478991985321, 0.7816430330276489]"
            ]
          },
          "metadata": {},
          "execution_count": 87
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Making predictions\n",
        "model_4_pred_probs = model_4.predict(val_char_token_dataset)\n",
        "model_4_preds = model_4_pred_probs.argmax(axis = 1)\n",
        "\n",
        "model_4_results = print_evaluation_metrics(model_4_preds, val_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Nt_leuLybWaR",
        "outputId": "438ffc47-8870-4f8f-e278-cb8b9a45f81c"
      },
      "execution_count": 88,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 21s 20ms/step\n",
            "Confusion matrix:  [[2135  679  245  282  108]\n",
            " [ 447 3093  212   19  811]\n",
            " [ 176  146 8578   57 1007]\n",
            " [ 732  242  163 1194   45]\n",
            " [  52  322  849    3 8615]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_4_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HuKf-r1zbydJ",
        "outputId": "6f68bf9b-def1-4d87-bb7d-8c1c38dc6eea"
      },
      "execution_count": 89,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.7816430557394413,\n",
              " 'precision': 0.6900937081659974,\n",
              " 'recall': 0.6750327367961588,\n",
              " 'f1-score': 0.6824801412180054}"
            ]
          },
          "metadata": {},
          "execution_count": 89
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Model 5: Transfer learning with pretrained token embeddings + character embeddings + positional embeddings\n"
      ],
      "metadata": {
        "id": "_4CNg3N-dpHb"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "9sVoGyQghjR2",
        "outputId": "63e0ca5e-df81-4dcc-c267-3c2443618dc9"
      },
      "execution_count": 90,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   line_number      label                                               text  \\\n",
              "0            0  OBJECTIVE  To investigate the efficacy of @ weeks of dail...   \n",
              "1            1    METHODS  A total of @ patients with primary knee OA wer...   \n",
              "2            2    METHODS  Outcome measures included pain reduction and i...   \n",
              "3            3    METHODS  Pain was assessed using the visual analog pain...   \n",
              "4            4    METHODS  Secondary outcome measures included the Wester...   \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-14e53c0c-ae6e-471d-ac77-4908d9920d83\" 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>line_number</th>\n",
              "      <th>label</th>\n",
              "      <th>text</th>\n",
              "      <th>total_lines</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>OBJECTIVE</td>\n",
              "      <td>To investigate the efficacy of @ weeks of dail...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>A total of @ patients with primary knee OA wer...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>Outcome measures included pain reduction and i...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>Pain was assessed using the visual analog pain...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>METHODS</td>\n",
              "      <td>Secondary outcome measures included the Wester...</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-14e53c0c-ae6e-471d-ac77-4908d9920d83')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-14e53c0c-ae6e-471d-ac77-4908d9920d83 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-14e53c0c-ae6e-471d-ac77-4908d9920d83');\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",
              "\n",
              "\n",
              "<div id=\"df-2e46eb38-8805-4a1c-b9c6-53974c2b8c35\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-2e46eb38-8805-4a1c-b9c6-53974c2b8c35')\"\n",
              "            title=\"Suggest charts.\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-2e46eb38-8805-4a1c-b9c6-53974c2b8c35 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 90
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "> **Note:** Any engineered features used to train a model needs to be available at test time. In our case, line numbers and total lines are available"
      ],
      "metadata": {
        "id": "9J0K8OBghm9s"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Create positional embeddings"
      ],
      "metadata": {
        "id": "Jrr7UQJgjTE_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# How many different line numbers are there\n",
        "train_df['line_number'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "I3PJBaC0jjcB",
        "outputId": "8dbcf449-ddad-49fe-8b81-5b601475bf5f"
      },
      "execution_count": 91,
      "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": 91
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Check distribution of line number df\n",
        "train_df.line_number.plot.hist()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 448
        },
        "id": "fpAu8u8rjxmN",
        "outputId": "14099dbe-e579-4991-8a00-fc56872605d0"
      },
      "execution_count": 92,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: ylabel='Frequency'>"
            ]
          },
          "metadata": {},
          "execution_count": 92
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Use TensorFlow to create one-hot-encoded tensors of out \"line_number\" column\n",
        "train_line_number_one_hot = tf.one_hot(train_df['line_number'].to_numpy(), depth = 15)\n",
        "val_line_number_one_hot = tf.one_hot(val_df['line_number'].to_numpy(), depth = 15)\n",
        "test_line_number_one_hot = tf.one_hot(test_df['line_number'].to_numpy(), depth = 15)\n",
        "train_line_number_one_hot[:10], train_line_number_one_hot.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sURleA2zj_1-",
        "outputId": "7b636c1b-722c-4052-ceb6-4ee9b43ca27e"
      },
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(<tf.Tensor: shape=(10, 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",
              "       dtype=float32)>,\n",
              " TensorShape([180040, 15]))"
            ]
          },
          "metadata": {},
          "execution_count": 93
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# How many different total lines are there\n",
        "train_df['total_lines'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OGjOwyZhkjAA",
        "outputId": "5fd8e904-7968-471a-8465-6d8dd9ab5728"
      },
      "execution_count": 94,
      "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": 94
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Check distribution of total lines df\n",
        "train_df['total_lines'].plot.hist()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "id": "PYiHZzP5lfeb",
        "outputId": "79b1b1b9-cbf1-4e02-c135-ce6b8abac1c8"
      },
      "execution_count": 95,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: ylabel='Frequency'>"
            ]
          },
          "metadata": {},
          "execution_count": 95
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Check the coverage of a 'total_lines' value of 20\n",
        "np.percentile(train_df.total_lines, 98)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ofz0HFXnnE3w",
        "outputId": "10b9b474-bc4e-4650-ab69-88a5c10a4abe"
      },
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "20.0"
            ]
          },
          "metadata": {},
          "execution_count": 96
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating one-hot encoded tensors of our 'total_lines' feature\n",
        "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)\n",
        "train_total_lines_one_hot.shape, train_total_lines_one_hot[:10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RZk80b6gl2Ny",
        "outputId": "6443c822-2feb-4324-dc59-143a5d16906b"
      },
      "execution_count": 97,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(TensorShape([180040, 20]),\n",
              " <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": 97
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Building a tribrid embedding model\n",
        "\n",
        "1. Create a token-level model\n",
        "2. Create a character-level model\n",
        "3. Create a model for the \"line_number\" feature\n",
        "4. Create a model for the \"total_lines\" feature\n",
        "5. Combine the outputs of 1 & 2 using tf.keras.layers.Concatenate\n",
        "6. Combine the outputs of 3, 4 & 5 using tf.keras.layers.Concatenate\n",
        "7. Create an output layer to accept the tribried embedding and output label probabilities\n",
        "8. Combine the inputs of 1,2,3,4 and outputs of 7 into tf.keras.Model"
      ],
      "metadata": {
        "id": "9ohRVF20npyt"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# 1. Token inputs\n",
        "token_inputs = tf.keras.layers.Input(shape=[], dtype=tf.string, name='token_inputs')\n",
        "token_embeddings = tf_hub_embedding_layer(token_inputs)\n",
        "token_outputs = tf.keras.layers.Dense(128, activation = 'relu')(token_embeddings)\n",
        "token_model = tf.keras.Model(inputs = token_inputs,\n",
        "                             outputs = token_outputs)\n",
        "\n",
        "# 2. Character inputs\n",
        "char_inputs = tf.keras.layers.Input(shape=(1,), dtype = tf.string, name = 'char_inputs')\n",
        "char_vectors = char_vectorizer(char_inputs)\n",
        "char_embeddings = char_embed(char_vectors)\n",
        "char_bi_lstm = tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(24))(char_embeddings)\n",
        "char_model = tf.keras.Model(inputs = char_inputs,\n",
        "                            outputs = char_bi_lstm)\n",
        "\n",
        "# 3. Line Number inputs\n",
        "line_number_inputs = tf.keras.layers.Input(shape = (15,), dtype = tf.float32, name = 'line_number_inputs')\n",
        "line_number_outputs = tf.keras.layers.Dense(32, activation = 'relu')(line_number_inputs)\n",
        "line_number_model = tf.keras.Model(inputs = line_number_inputs,\n",
        "                                   outputs = line_number_outputs)\n",
        "\n",
        "# 4. Total lines inputs\n",
        "total_lines_inputs = tf.keras.layers.Input(shape = (20,), dtype = tf.float32, name = 'total_lines_inputs')\n",
        "total_lines_outputs = tf.keras.layers.Dense(32, activation = 'relu')(total_lines_inputs)\n",
        "total_lines_model = tf.keras.Model(inputs = total_lines_inputs,\n",
        "                                   outputs = total_lines_outputs)\n",
        "\n",
        "# 5. Combine token and char embeddings into a hybrid embeddings\n",
        "combined_embeddings = tf.keras.layers.Concatenate(name = 'char_token_hybrid_embedding')([token_model.output,\n",
        "                                                                                         char_model.output])\n",
        "# 6. Adding a dropout layer as mentioned in the paper\n",
        "combined_dropout = tf.keras.layers.Dropout(0.5)(combined_embeddings)\n",
        "combined_dense = tf.keras.layers.Dense(128, activation = 'relu')(combined_dropout)\n",
        "final_dropout = tf.keras.layers.Dropout(0.5)(combined_dense)\n",
        "\n",
        "# 7. Combine positional embeddings with combined token and char embeddings\n",
        "tribrid_embeddings = tf.keras.layers.Concatenate(name = 'char_token_positional_embedding')([line_number_model.output,\n",
        "                                                                                            total_lines_model.output,\n",
        "                                                                                            final_dropout])\n",
        "\n",
        "# 8. Create output layer\n",
        "output_layer = tf.keras.layers.Dense(5, activation = 'softmax', name = 'output_layer')(tribrid_embeddings)\n",
        "\n",
        "# 9. Put together model with all kind of inputs\n",
        "model_5 = tf.keras.Model(inputs = [line_number_model.input,\n",
        "                                   total_lines_model.input,\n",
        "                                   token_model.input,\n",
        "                                   char_model.input],\n",
        "                         outputs = output_layer)"
      ],
      "metadata": {
        "id": "wnPqKRlfpXwB"
      },
      "execution_count": 98,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_5.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BLUjtdHltBZB",
        "outputId": "899be26d-dd74-4b5a-d9d4-8eb1d9e5b3b1"
      },
      "execution_count": 99,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_6\"\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 (TextVecto  (None, 290)                  0         ['char_inputs[0][0]']         \n",
            " rization)                                                                                        \n",
            "                                                                                                  \n",
            " universal_sentence_encoder  (None, 512)                  2567978   ['token_inputs[0][0]']        \n",
            "  (KerasLayer)                                            24                                      \n",
            "                                                                                                  \n",
            " char_embed (Embedding)      (None, 290, 25)              700       ['char_vectorizer[2][0]']     \n",
            "                                                                                                  \n",
            " dense_6 (Dense)             (None, 128)                  65664     ['universal_sentence_encoder[2\n",
            "                                                                    ][0]']                        \n",
            "                                                                                                  \n",
            " bidirectional_1 (Bidirecti  (None, 48)                   9600      ['char_embed[2][0]']          \n",
            " onal)                                                                                            \n",
            "                                                                                                  \n",
            " char_token_hybrid_embeddin  (None, 176)                  0         ['dense_6[0][0]',             \n",
            " g (Concatenate)                                                     'bidirectional_1[0][0]']     \n",
            "                                                                                                  \n",
            " dropout_2 (Dropout)         (None, 176)                  0         ['char_token_hybrid_embedding[\n",
            "                                                                    0][0]']                       \n",
            "                                                                                                  \n",
            " line_number_inputs (InputL  [(None, 15)]                 0         []                            \n",
            " ayer)                                                                                            \n",
            "                                                                                                  \n",
            " total_lines_inputs (InputL  [(None, 20)]                 0         []                            \n",
            " ayer)                                                                                            \n",
            "                                                                                                  \n",
            " dense_9 (Dense)             (None, 128)                  22656     ['dropout_2[0][0]']           \n",
            "                                                                                                  \n",
            " dense_7 (Dense)             (None, 32)                   512       ['line_number_inputs[0][0]']  \n",
            "                                                                                                  \n",
            " dense_8 (Dense)             (None, 32)                   672       ['total_lines_inputs[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_3 (Dropout)         (None, 128)                  0         ['dense_9[0][0]']             \n",
            "                                                                                                  \n",
            " char_token_positional_embe  (None, 192)                  0         ['dense_7[0][0]',             \n",
            " dding (Concatenate)                                                 'dense_8[0][0]',             \n",
            "                                                                     'dropout_3[0][0]']           \n",
            "                                                                                                  \n",
            " output_layer (Dense)        (None, 5)                    965       ['char_token_positional_embedd\n",
            "                                                                    ing[0][0]']                   \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 256898593 (979.99 MB)\n",
            "Trainable params: 100769 (393.63 KB)\n",
            "Non-trainable params: 256797824 (979.61 MB)\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Exploring model_5 visually\n",
        "from tensorflow.keras.utils import plot_model\n",
        "\n",
        "plot_model(model_5, show_shapes = True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "pxmtkUTdwuiM",
        "outputId": "f04435fa-dbd5-4eb3-aa94-103f4348e564"
      },
      "execution_count": 100,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 100
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "What is label smoothing?\n",
        "\n",
        "For example, if our model gets too confident on single class (e.g. its prediction probability is really high), it may get stuck on that class and not consider other classes...\n",
        "\n",
        "Really confident: `[0.0, 0.0, 1.0, 0.0]`\n",
        "\n",
        "What label smoothing does is it assigns some of the value from the highest pred prob to other classes, in turn, hopefully improving generalisation: `[0.01, 0.01, .96, 0.01]`"
      ],
      "metadata": {
        "id": "n4hCVVaOyNub"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Compile model_5\n",
        "model_5.compile(loss = tf.keras.losses.CategoricalCrossentropy(label_smoothing = .2),\n",
        "                optimizer = 'adam',\n",
        "                metrics = ['accuracy'])"
      ],
      "metadata": {
        "id": "p2YXxxI0w3vE"
      },
      "execution_count": 101,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Create tribrid embedding datasets using tf.data\n",
        "\n"
      ],
      "metadata": {
        "id": "vnG9lSa_zrpB"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating training and validation dataset (with all four kinds of data)\n",
        "train_char_token_pos_data = tf.data.Dataset.from_tensor_slices((train_line_number_one_hot,\n",
        "                                                               train_total_lines_one_hot,\n",
        "                                                                train_sentences,\n",
        "                                                                train_chars))\n",
        "train_char_token_pos_labels = tf.data.Dataset.from_tensor_slices(train_labels_one_hot)\n",
        "train_char_token_pos_dataset = tf.data.Dataset.zip(train_char_token_pos_data, train_char_token_pos_labels)\n",
        "train_char_token_pos_dataset = train_char_token_pos_dataset.batch(32).prefetch(tf.data.AUTOTUNE)\n",
        "\n",
        "# Do the same as above but for validation dataset\n",
        "val_char_token_pos_data = tf.data.Dataset.from_tensor_slices((val_line_number_one_hot,\n",
        "                                                               val_total_lines_one_hot,\n",
        "                                                                val_sentences,\n",
        "                                                                val_chars))\n",
        "val_char_token_pos_labels = tf.data.Dataset.from_tensor_slices(val_labels_one_hot)\n",
        "val_char_token_pos_dataset = tf.data.Dataset.zip(val_char_token_pos_data, val_char_token_pos_labels)\n",
        "val_char_token_pos_dataset = val_char_token_pos_dataset.batch(32).prefetch(tf.data.AUTOTUNE)"
      ],
      "metadata": {
        "id": "ObrRu7UA064W"
      },
      "execution_count": 102,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_char_token_pos_dataset, val_char_token_pos_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rbmHDFDi3PZy",
        "outputId": "92a2174b-0467-4ef5-c1d9-175756b4877b"
      },
      "execution_count": 103,
      "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": 103
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Fitting, evaluating and making predictions with model_5"
      ],
      "metadata": {
        "id": "bmb6CZLZ3iZP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Fitting the model\n",
        "history_5 = model_5.fit(train_char_token_pos_dataset,\n",
        "                        epochs = 3,\n",
        "                        validation_data = val_char_token_pos_dataset,\n",
        "                        validation_steps = int(.1*(len(val_char_token_pos_dataset))))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6UhWHXmh4DiU",
        "outputId": "2e0d2678-09a7-4f68-af20-33ac5ce3a1d8"
      },
      "execution_count": 104,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/3\n",
            "5627/5627 [==============================] - 264s 45ms/step - loss: 0.9800 - accuracy: 0.8068 - val_loss: 0.9182 - val_accuracy: 0.8408\n",
            "Epoch 2/3\n",
            "5627/5627 [==============================] - 250s 44ms/step - loss: 0.9362 - accuracy: 0.8377 - val_loss: 0.9110 - val_accuracy: 0.8454\n",
            "Epoch 3/3\n",
            "5627/5627 [==============================] - 256s 45ms/step - loss: 0.9279 - accuracy: 0.8436 - val_loss: 0.9064 - val_accuracy: 0.8484\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_5.evaluate(val_char_token_pos_dataset)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nI-evxf-46Yr",
        "outputId": "bd306822-bb3a-4a72-faf4-4cac0c02adc7"
      },
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 20s 22ms/step - loss: 0.9010 - accuracy: 0.8532\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.9010348320007324, 0.8532371520996094]"
            ]
          },
          "metadata": {},
          "execution_count": 105
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Making predictions\n",
        "model_5_pred_probs = model_5.predict(val_char_token_pos_dataset)\n",
        "model_5_preds = model_5_pred_probs.argmax(axis = 1)\n",
        "model_5_results = print_evaluation_metrics(model_5_preds,\n",
        "                                           val_labels_encoded)\n",
        "model_5_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PHlmEf-J5CkF",
        "outputId": "5e8c739c-b1ca-4d89-9b75-1c89d6b32209"
      },
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 23s 21ms/step\n",
            "Confusion matrix:  [[2854   80  265  228   22]\n",
            " [  39 3875   43    1  624]\n",
            " [  87  116 9075   12  674]\n",
            " [1055    1  173 1144    3]\n",
            " [  11  358  642    0 8830]]\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.8532371243214617,\n",
              " 'precision': 0.8747178329571106,\n",
              " 'recall': 0.8457005674378001,\n",
              " 'f1-score': 0.8599644917887262}"
            ]
          },
          "metadata": {},
          "execution_count": 106
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Comparing Model Results"
      ],
      "metadata": {
        "id": "YZ0jSOkG6YV8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Combine model results into a dataframe\n",
        "all_model_results = pd.DataFrame({\"model_0_baseline\": baseline_results,\n",
        "                                  \"model_1_custom_token_embedding\": model_1_results,\n",
        "                                  \"model_2_pretrained_token_embedding\": model_2_results,\n",
        "                                  \"model_3_custom_char_embedding\": model_3_results,\n",
        "                                  \"model_4_hybrid_char_token_embedding\": model_4_results,\n",
        "                                  \"model_5_pos_char_token_embedding\": model_5_results})\n",
        "all_model_results = all_model_results.transpose()\n",
        "all_model_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "7D1nR5zhbQxP",
        "outputId": "1ef30bbe-38f4-433a-da2b-e360d263798d"
      },
      "execution_count": 110,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                     accuracy  precision    recall  f1-score\n",
              "model_0_baseline                     0.721832   0.646069  0.586425  0.614804\n",
              "model_1_custom_token_embedding       0.807262   0.724356  0.740518  0.732348\n",
              "model_2_pretrained_token_embedding   0.709387   0.613327  0.560454  0.585700\n",
              "model_3_custom_char_embedding        0.728883   0.628109  0.567656  0.596354\n",
              "model_4_hybrid_char_token_embedding  0.781643   0.690094  0.675033  0.682480\n",
              "model_5_pos_char_token_embedding     0.853237   0.874718  0.845701  0.859964"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-fa3e5da9-5371-4d09-be6b-e64282fcf982\" 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>accuracy</th>\n",
              "      <th>precision</th>\n",
              "      <th>recall</th>\n",
              "      <th>f1-score</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>model_0_baseline</th>\n",
              "      <td>0.721832</td>\n",
              "      <td>0.646069</td>\n",
              "      <td>0.586425</td>\n",
              "      <td>0.614804</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>model_1_custom_token_embedding</th>\n",
              "      <td>0.807262</td>\n",
              "      <td>0.724356</td>\n",
              "      <td>0.740518</td>\n",
              "      <td>0.732348</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>model_2_pretrained_token_embedding</th>\n",
              "      <td>0.709387</td>\n",
              "      <td>0.613327</td>\n",
              "      <td>0.560454</td>\n",
              "      <td>0.585700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>model_3_custom_char_embedding</th>\n",
              "      <td>0.728883</td>\n",
              "      <td>0.628109</td>\n",
              "      <td>0.567656</td>\n",
              "      <td>0.596354</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>model_4_hybrid_char_token_embedding</th>\n",
              "      <td>0.781643</td>\n",
              "      <td>0.690094</td>\n",
              "      <td>0.675033</td>\n",
              "      <td>0.682480</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>model_5_pos_char_token_embedding</th>\n",
              "      <td>0.853237</td>\n",
              "      <td>0.874718</td>\n",
              "      <td>0.845701</td>\n",
              "      <td>0.859964</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fa3e5da9-5371-4d09-be6b-e64282fcf982')\"\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 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\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",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\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-fa3e5da9-5371-4d09-be6b-e64282fcf982 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-fa3e5da9-5371-4d09-be6b-e64282fcf982');\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",
              "\n",
              "\n",
              "<div id=\"df-94ab27c4-d302-464d-b5d0-697fd86f447c\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-94ab27c4-d302-464d-b5d0-697fd86f447c')\"\n",
              "            title=\"Suggest charts.\"\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",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-94ab27c4-d302-464d-b5d0-697fd86f447c button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 110
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Plot and compare all model results\n",
        "all_model_results.plot(kind = 'bar', figsize = (10,7)).legend(bbox_to_anchor = (1.0, 1.0))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 883
        },
        "id": "xEHqt9JWco9z",
        "outputId": "79955341-9bdf-4526-fe3b-18dc1dd995bf"
      },
      "execution_count": 111,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.legend.Legend at 0x78dfc0e1ba00>"
            ]
          },
          "metadata": {},
          "execution_count": 111
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Sort models results by f1-score\n",
        "all_model_results.sort_values(by = 'f1-score', ascending = True)['f1-score'].plot(kind = 'bar', figsize = (10,7))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 883
        },
        "id": "hZvnDw_HdFcR",
        "outputId": "0294fe09-87a2-4d3c-9fef-19e0df112df1"
      },
      "execution_count": 112,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "metadata": {},
          "execution_count": 112
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1000x700 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Save and load model"
      ],
      "metadata": {
        "id": "X8qYgztcdsCa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Save the best performing model to SavedModel format (default)\n",
        "model_5.save(\"/content/drive/MyDrive/skimlit_NLP_project/skimlit_tribrid_model\")"
      ],
      "metadata": {
        "id": "Fs9lCdXSd_mU"
      },
      "execution_count": 121,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Load in best performing model\n",
        "loaded_model = tf.keras.models.load_model(\"/content/drive/MyDrive/skimlit_NLP_project/skimlit_tribrid_model\")"
      ],
      "metadata": {
        "id": "rsHsRs_EeT-D"
      },
      "execution_count": 122,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Making predictions with the loaded model on the validation set\n",
        "loaded_pred_probs = loaded_model.predict(val_char_token_pos_dataset)\n",
        "loaded_preds = loaded_pred_probs.argmax(axis=1)\n",
        "loaded_model_results = print_evaluation_metrics(loaded_preds,\n",
        "                                                val_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lL5L2a5MeyJL",
        "outputId": "913c7779-9b49-422f-97e9-4da144bc9fe9"
      },
      "execution_count": 123,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "945/945 [==============================] - 24s 21ms/step\n",
            "Confusion matrix:  [[2854   80  265  228   22]\n",
            " [  39 3875   43    1  624]\n",
            " [  87  116 9075   12  674]\n",
            " [1055    1  173 1144    3]\n",
            " [  11  358  642    0 8830]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "loaded_model_results == model_5_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O4BcSjgWfav9",
        "outputId": "c414761f-970a-4841-f852-f5c5e1e17a8d"
      },
      "execution_count": 124,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 124
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Evaluating model on test dataset"
      ],
      "metadata": {
        "id": "UDnSA9-3fgVr"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating a test dataset batch and prefetched\n",
        "test_pos_char_token_data = tf.data.Dataset.from_tensor_slices((test_line_number_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",
        "test_pos_char_token_dataset"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dfR1u1GLkbrh",
        "outputId": "38ed74be-674c-457e-f0c4-5a697a93f6f0"
      },
      "execution_count": 125,
      "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": 125
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Making predictions\n",
        "test_pred_probs = loaded_model.predict(test_pos_char_token_dataset)\n",
        "test_preds = test_pred_probs.argmax(axis=1)\n",
        "test_results = print_evaluation_metrics(test_preds,\n",
        "                                        test_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2hzGhKu6l3eI",
        "outputId": "50f68d3e-915a-4ea5-d641-fc5f66a1fd3d"
      },
      "execution_count": 126,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "942/942 [==============================] - 24s 26ms/step\n",
            "Confusion matrix:  [[2967   88  299  247   20]\n",
            " [  29 3868   29    0  645]\n",
            " [  88  104 9014   16  675]\n",
            " [ 997    4  182 1140   10]\n",
            " [  11  348  729    2 8623]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gN5kZs0kmWrc",
        "outputId": "68a9577c-67c5-4b35-cef0-ae8a4a0291d0"
      },
      "execution_count": 127,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.8499087439853991,\n",
              " 'precision': 0.8766999093381687,\n",
              " 'recall': 0.8462043316560928,\n",
              " 'f1-score': 0.8611822331069798}"
            ]
          },
          "metadata": {},
          "execution_count": 127
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "It seems our best model so far still doesn't match the performance of the results in the paper (their model gets 90.0 F1-score on the test dataset, where as ours get ~86.1 F1-score)"
      ],
      "metadata": {
        "id": "3bQIjvhwmXlh"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Trying to improve model_5 performance through:\n",
        "\n",
        "1. Training `model_5` for as many epochs until it stops improving. Using:\n",
        "* `ModelCheckpoint`\n",
        "* `Earlystopping`\n",
        "\n",
        "2. Fine tuning our pretrained embeddings (the paper fine-tunes GloVe embeddings).\n",
        "\n",
        "3. Using pretrained GloVe embeddings. If need be fine-tuning them.\n",
        "\n",
        "4. Replacing the TensorFlow Hub Unsiversal Sentence Encoder pretrained embedding for the TensorFlow Hub BERT PubMed Expert (a language model pretrained on PubMed texts) pretrained embedding.\n",
        "\n",
        "5. Merging our `line_number` and `total_lines` features for each sequence."
      ],
      "metadata": {
        "id": "j6bzlcQbooN5"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 1. Training for more epochs"
      ],
      "metadata": {
        "id": "ZXt1yKq0opd_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Re-compiling model_5\n",
        "model_5.compile(loss = tf.keras.losses.CategoricalCrossentropy(label_smoothing = .2),\n",
        "                optimizer = 'adam',\n",
        "                metrics = ['accuracy'])\n",
        "\n",
        "# Preparing callbacks\n",
        "\n",
        "# ModelCheckpoint callback\n",
        "import os\n",
        "checkpoint_path = '/content/drive/MyDrive/skimlit_NLP_project/model_checkpoints/'\n",
        "# os.mkdir(checkpoint_path)\n",
        "\n",
        "model_checkpoint = tf.keras.callbacks.ModelCheckpoint(checkpoint_path,\n",
        "                                                      monitor = 'val_accuracy',\n",
        "                                                      save_best_only = True,\n",
        "                                                      save_weights_only = True)\n",
        "\n",
        "# EarlyStopping Callback\n",
        "early_stopping = tf.keras.callbacks.EarlyStopping(monitor = 'val_accuracy',\n",
        "                                                 mode = 'max',\n",
        "                                                 patience = 3)"
      ],
      "metadata": {
        "id": "_pqTVcv4sCnU"
      },
      "execution_count": 133,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Fitting the model\n",
        "history_6 = model_5.fit(train_char_token_pos_dataset,\n",
        "                        epochs = 100,\n",
        "                        validation_data = val_char_token_pos_dataset,\n",
        "                        validation_steps = int(.1*len(val_char_token_pos_dataset)),\n",
        "                        callbacks = [model_checkpoint,\n",
        "                                     early_stopping])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "625pCh65uK-I",
        "outputId": "a9f367e4-3c22-43d8-bc5b-e67b85def98e"
      },
      "execution_count": 134,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "5627/5627 [==============================] - 328s 56ms/step - loss: 0.9227 - accuracy: 0.8472 - val_loss: 0.9078 - val_accuracy: 0.8477\n",
            "Epoch 2/100\n",
            "5627/5627 [==============================] - 319s 57ms/step - loss: 0.9188 - accuracy: 0.8505 - val_loss: 0.8986 - val_accuracy: 0.8567\n",
            "Epoch 3/100\n",
            "5627/5627 [==============================] - 362s 64ms/step - loss: 0.9155 - accuracy: 0.8525 - val_loss: 0.9004 - val_accuracy: 0.8570\n",
            "Epoch 4/100\n",
            "5627/5627 [==============================] - 287s 51ms/step - loss: 0.9143 - accuracy: 0.8532 - val_loss: 0.8977 - val_accuracy: 0.8541\n",
            "Epoch 5/100\n",
            "5627/5627 [==============================] - 347s 62ms/step - loss: 0.9114 - accuracy: 0.8548 - val_loss: 0.8938 - val_accuracy: 0.8580\n",
            "Epoch 6/100\n",
            "5627/5627 [==============================] - 290s 51ms/step - loss: 0.9090 - accuracy: 0.8565 - val_loss: 0.8957 - val_accuracy: 0.8561\n",
            "Epoch 7/100\n",
            "5627/5627 [==============================] - 350s 62ms/step - loss: 0.9082 - accuracy: 0.8576 - val_loss: 0.8914 - val_accuracy: 0.8590\n",
            "Epoch 8/100\n",
            "5627/5627 [==============================] - 368s 65ms/step - loss: 0.9068 - accuracy: 0.8579 - val_loss: 0.8882 - val_accuracy: 0.8607\n",
            "Epoch 9/100\n",
            "5627/5627 [==============================] - 286s 51ms/step - loss: 0.9041 - accuracy: 0.8601 - val_loss: 0.8906 - val_accuracy: 0.8604\n",
            "Epoch 10/100\n",
            "5627/5627 [==============================] - 352s 62ms/step - loss: 0.9034 - accuracy: 0.8610 - val_loss: 0.8866 - val_accuracy: 0.8650\n",
            "Epoch 11/100\n",
            "5627/5627 [==============================] - 287s 51ms/step - loss: 0.9008 - accuracy: 0.8622 - val_loss: 0.8853 - val_accuracy: 0.8624\n",
            "Epoch 12/100\n",
            "5627/5627 [==============================] - 271s 48ms/step - loss: 0.9001 - accuracy: 0.8627 - val_loss: 0.8865 - val_accuracy: 0.8650\n",
            "Epoch 13/100\n",
            "5627/5627 [==============================] - 271s 48ms/step - loss: 0.8986 - accuracy: 0.8641 - val_loss: 0.8865 - val_accuracy: 0.8634\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Evaluating model_5 on test dataset\n",
        "test_pred_probs = model_5.predict(test_pos_char_token_dataset)\n",
        "test_preds = test_pred_probs.argmax(axis=1)\n",
        "test_results = print_evaluation_metrics(test_preds,\n",
        "                                        test_labels_encoded)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Q9dcc7I_u01S",
        "outputId": "e6f86e90-42f6-4ab6-e537-277995871d52"
      },
      "execution_count": 165,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "942/942 [==============================] - 21s 23ms/step\n",
            "Confusion matrix:  [[3023   71  262  234   31]\n",
            " [  23 3863   35    0  650]\n",
            " [  97   87 9159   16  538]\n",
            " [ 999    0  176 1148   10]\n",
            " [  13  326  706    1 8667]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_results"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-W3CnzIH_rSn",
        "outputId": "f5073392-da84-4b43-f9a3-a63443ffba03"
      },
      "execution_count": 166,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'accuracy': 0.8581383773021404,\n",
              " 'precision': 0.8886588451805844,\n",
              " 'recall': 0.8451104791074163,\n",
              " 'f1-score': 0.8663377438887643}"
            ]
          },
          "metadata": {},
          "execution_count": 166
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### 2. Fine tuning pre-trained embedding (Model_6: Fine Tuned Model)"
      ],
      "metadata": {
        "id": "DVULOUcU_tIa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# 1. Token inputs\n",
        "token_inputs = tf.keras.layers.Input(shape=[], dtype=tf.string, name='token_inputs')\n",
        "token_embeddings = tf_hub_embedding_layer(token_inputs)\n",
        "token_outputs = tf.keras.layers.Dense(128, activation = 'relu')(token_embeddings)\n",
        "token_model = tf.keras.Model(inputs = token_inputs,\n",
        "                             outputs = token_outputs)\n",
        "\n",
        "# 2. Character inputs\n",
        "char_inputs = tf.keras.layers.Input(shape=(1,), dtype = tf.string, name = 'char_inputs')\n",
        "char_vectors = char_vectorizer(char_inputs)\n",
        "char_embeddings = char_embed(char_vectors)\n",
        "char_bi_lstm = tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(24))(char_embeddings)\n",
        "char_model = tf.keras.Model(inputs = char_inputs,\n",
        "                            outputs = char_bi_lstm)\n",
        "\n",
        "# 3. Line Number inputs\n",
        "line_number_inputs = tf.keras.layers.Input(shape = (15,), dtype = tf.float32, name = 'line_number_inputs')\n",
        "line_number_outputs = tf.keras.layers.Dense(32, activation = 'relu')(line_number_inputs)\n",
        "line_number_model = tf.keras.Model(inputs = line_number_inputs,\n",
        "                                   outputs = line_number_outputs)\n",
        "\n",
        "# 4. Total lines inputs\n",
        "total_lines_inputs = tf.keras.layers.Input(shape = (20,), dtype = tf.float32, name = 'total_lines_inputs')\n",
        "total_lines_outputs = tf.keras.layers.Dense(32, activation = 'relu')(total_lines_inputs)\n",
        "total_lines_model = tf.keras.Model(inputs = total_lines_inputs,\n",
        "                                   outputs = total_lines_outputs)\n",
        "\n",
        "# 5. Combine token and char embeddings into a hybrid embeddings\n",
        "combined_embeddings = tf.keras.layers.Concatenate(name = 'char_token_hybrid_embedding')([token_model.output,\n",
        "                                                                                         char_model.output])\n",
        "# 6. Adding a dropout layer as mentioned in the paper\n",
        "combined_dropout = tf.keras.layers.Dropout(0.5)(combined_embeddings)\n",
        "combined_dense = tf.keras.layers.Dense(128, activation = 'relu')(combined_dropout)\n",
        "final_dropout = tf.keras.layers.Dropout(0.5)(combined_dense)\n",
        "\n",
        "# 7. Combine positional embeddings with combined token and char embeddings\n",
        "tribrid_embeddings = tf.keras.layers.Concatenate(name = 'char_token_positional_embedding')([line_number_model.output,\n",
        "                                                                                            total_lines_model.output,\n",
        "                                                                                            final_dropout])\n",
        "\n",
        "# 8. Create output layer\n",
        "output_layer = tf.keras.layers.Dense(5, activation = 'softmax', name = 'output_layer')(tribrid_embeddings)\n",
        "\n",
        "# 9. Put together model with all kind of inputs\n",
        "model_6 = tf.keras.Model(inputs = [line_number_model.input,\n",
        "                                   total_lines_model.input,\n",
        "                                   token_model.input,\n",
        "                                   char_model.input],\n",
        "                         outputs = output_layer)"
      ],
      "metadata": {
        "id": "MltE4aAvEbkJ"
      },
      "execution_count": 182,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model_6.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pc3taD8dEx3z",
        "outputId": "acee2157-f012-4df9-952a-a9ab9ee5a9fa"
      },
      "execution_count": 183,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"model_11\"\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 (TextVecto  (None, 290)                  0         ['char_inputs[0][0]']         \n",
            " rization)                                                                                        \n",
            "                                                                                                  \n",
            " universal_sentence_encoder  (None, 512)                  2567978   ['token_inputs[0][0]']        \n",
            "  (KerasLayer)                                            24                                      \n",
            "                                                                                                  \n",
            " char_embed (Embedding)      (None, 290, 25)              700       ['char_vectorizer[3][0]']     \n",
            "                                                                                                  \n",
            " dense_10 (Dense)            (None, 128)                  65664     ['universal_sentence_encoder[3\n",
            "                                                                    ][0]']                        \n",
            "                                                                                                  \n",
            " bidirectional_2 (Bidirecti  (None, 48)                   9600      ['char_embed[3][0]']          \n",
            " onal)                                                                                            \n",
            "                                                                                                  \n",
            " char_token_hybrid_embeddin  (None, 176)                  0         ['dense_10[0][0]',            \n",
            " g (Concatenate)                                                     'bidirectional_2[0][0]']     \n",
            "                                                                                                  \n",
            " dropout_4 (Dropout)         (None, 176)                  0         ['char_token_hybrid_embedding[\n",
            "                                                                    0][0]']                       \n",
            "                                                                                                  \n",
            " line_number_inputs (InputL  [(None, 15)]                 0         []                            \n",
            " ayer)                                                                                            \n",
            "                                                                                                  \n",
            " total_lines_inputs (InputL  [(None, 20)]                 0         []                            \n",
            " ayer)                                                                                            \n",
            "                                                                                                  \n",
            " dense_13 (Dense)            (None, 128)                  22656     ['dropout_4[0][0]']           \n",
            "                                                                                                  \n",
            " dense_11 (Dense)            (None, 32)                   512       ['line_number_inputs[0][0]']  \n",
            "                                                                                                  \n",
            " dense_12 (Dense)            (None, 32)                   672       ['total_lines_inputs[0][0]']  \n",
            "                                                                                                  \n",
            " dropout_5 (Dropout)         (None, 128)                  0         ['dense_13[0][0]']            \n",
            "                                                                                                  \n",
            " char_token_positional_embe  (None, 192)                  0         ['dense_11[0][0]',            \n",
            " dding (Concatenate)                                                 'dense_12[0][0]',            \n",
            "                                                                     'dropout_5[0][0]']           \n",
            "                                                                                                  \n",
            " output_layer (Dense)        (None, 5)                    965       ['char_token_positional_embedd\n",
            "                                                                    ing[0][0]']                   \n",
            "                                                                                                  \n",
            "==================================================================================================\n",
            "Total params: 256898593 (979.99 MB)\n",
            "Trainable params: 100769 (393.63 KB)\n",
            "Non-trainable params: 256797824 (979.61 MB)\n",
            "__________________________________________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Checking the layers in our model\n",
        "for i, layer in enumerate(model_6.layers):\n",
        "  print(layer.name, layer.trainable)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PgaWw2t5AFUf",
        "outputId": "968170fb-ead3-4788-d4d4-f598f80fe865"
      },
      "execution_count": 184,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "char_inputs True\n",
            "token_inputs True\n",
            "char_vectorizer True\n",
            "universal_sentence_encoder False\n",
            "char_embed True\n",
            "dense_10 True\n",
            "bidirectional_2 True\n",
            "char_token_hybrid_embedding True\n",
            "dropout_4 True\n",
            "line_number_inputs True\n",
            "total_lines_inputs True\n",
            "dense_13 True\n",
            "dense_11 True\n",
            "dense_12 True\n",
            "dropout_5 True\n",
            "char_token_positional_embedding True\n",
            "output_layer True\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Making the universal sentence encoder embedding layer trainable for fine tuning\n",
        "model_6.layers[3].trainable = True\n",
        "\n",
        "for i, layer in enumerate(model_6.layers):\n",
        "  print(layer.name, layer.trainable)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0Wis0LzaAWSy",
        "outputId": "7434efa9-041b-4648-a523-8b2e12fc3f28"
      },
      "execution_count": 185,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "char_inputs True\n",
            "token_inputs True\n",
            "char_vectorizer True\n",
            "universal_sentence_encoder True\n",
            "char_embed True\n",
            "dense_10 True\n",
            "bidirectional_2 True\n",
            "char_token_hybrid_embedding True\n",
            "dropout_4 True\n",
            "line_number_inputs True\n",
            "total_lines_inputs True\n",
            "dense_13 True\n",
            "dense_11 True\n",
            "dense_12 True\n",
            "dropout_5 True\n",
            "char_token_positional_embedding True\n",
            "output_layer True\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Compiling the model\n",
        "model_6.compile(loss = tf.keras.losses.CategoricalCrossentropy(label_smoothing = .2),\n",
        "                optimizer = tf.keras.optimizers.Adam(learning_rate = .0001), # lower the learining rate by 10x to avoid overfitting\n",
        "                metrics = ['accuracy'])"
      ],
      "metadata": {
        "id": "N3LsCpWoD78m"
      },
      "execution_count": 189,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Updating the callbacks\n",
        "\n",
        "checkpoint_path = '/content/drive/MyDrive/skimlit_NLP_project/model_checkpoints/fine_tune_checkpoint'\n",
        "\n",
        "# os.mkdir(checkpoint_path)\n",
        "\n",
        "model_checkpoint = tf.keras.callbacks.ModelCheckpoint(checkpoint_path,\n",
        "                                                      monitor = 'val_loss',\n",
        "                                                      save_best_only = True,\n",
        "                                                      save_weights_only = True)\n",
        "\n",
        "early_stopping = tf.keras.callbacks.EarlyStopping(monitor = 'val_loss',\n",
        "                                                  mode = 'min',\n",
        "                                                  patience = 3)\n",
        "\n",
        "# Creating learning rate reduction callback\n",
        "reduce_lr = tf.keras.callbacks.ReduceLROnPlateau(monitor = \"val_loss\",\n",
        "                                                 factor = 0.2,\n",
        "                                                 patience = 2,\n",
        "                                                 verbose = 1,\n",
        "                                                 min_lr = 1e-7)"
      ],
      "metadata": {
        "id": "2dYQuaJxF4ZD"
      },
      "execution_count": 190,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Fitting the model\n",
        "history_7 = model_6.fit(train_char_token_pos_dataset,\n",
        "                        epochs = 100,\n",
        "                        validation_data = val_char_token_pos_dataset,\n",
        "                        validation_steps = int(.1*len(val_char_token_pos_dataset)),\n",
        "                        callbacks = [model_checkpoint,\n",
        "                                     early_stopping,\n",
        "                                     reduce_lr])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 894
        },
        "id": "e6Y87t5OHOnE",
        "outputId": "0fdb18a5-274d-45a4-c67d-62242b335d6c"
      },
      "execution_count": 191,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n"
          ]
        },
        {
          "output_type": "error",
          "ename": "ResourceExhaustedError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mResourceExhaustedError\u001b[0m                    Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-191-5e5bb986aef4>\u001b[0m in \u001b[0;36m<cell line: 2>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;31m# Fitting the model\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m history_7 = model_6.fit(train_char_token_pos_dataset,\n\u001b[0m\u001b[1;32m      3\u001b[0m                         \u001b[0mepochs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m100\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m                         \u001b[0mvalidation_data\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mval_char_token_pos_dataset\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m                         \u001b[0mvalidation_steps\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m.1\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mval_char_token_pos_dataset\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/keras/src/utils/traceback_utils.py\u001b[0m in \u001b[0;36merror_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     68\u001b[0m             \u001b[0;31m# To get the full stack trace, call:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     69\u001b[0m             \u001b[0;31m# `tf.debugging.disable_traceback_filtering()`\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 70\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwith_traceback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfiltered_tb\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     71\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     72\u001b[0m             \u001b[0;32mdel\u001b[0m \u001b[0mfiltered_tb\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/polymorphic_function/autograph_util.py\u001b[0m in \u001b[0;36mautograph_handler\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m     50\u001b[0m     \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m  \u001b[0;31m# pylint:disable=broad-except\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     51\u001b[0m       \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"ag_error_metadata\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 52\u001b[0;31m         \u001b[0;32mraise\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mag_error_metadata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_exception\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0me\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     53\u001b[0m       \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     54\u001b[0m         \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mResourceExhaustedError\u001b[0m: in user code:\n\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py\", line 1377, in train_function  *\n        return step_function(self, iterator)\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py\", line 1360, in step_function  **\n        outputs = model.distribute_strategy.run(run_step, args=(data,))\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py\", line 1349, in run_step  **\n        outputs = model.train_step(data)\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/engine/training.py\", line 1130, in train_step\n        self.optimizer.minimize(loss, self.trainable_variables, tape=tape)\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/optimizers/optimizer.py\", line 544, in minimize\n        self.apply_gradients(grads_and_vars)\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/optimizers/optimizer.py\", line 1223, in apply_gradients\n        return super().apply_gradients(grads_and_vars, name=name)\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/optimizers/optimizer.py\", line 638, in apply_gradients\n        self.build(trainable_variables)\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/optimizers/adam.py\", line 145, in build\n        self.add_variable_from_reference(\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/optimizers/optimizer.py\", line 1125, in add_variable_from_reference\n        return super().add_variable_from_reference(\n    File \"/usr/local/lib/python3.10/dist-packages/keras/src/optimizers/optimizer.py\", line 508, in add_variable_from_reference\n        initial_value = tf.zeros(\n    File \"/usr/local/lib/python3.10/dist-packages/tensorflow/dtensor/python/api.py\", line 64, in call_with_layout\n        return fn(*args, **kwargs)\n\n    ResourceExhaustedError: {{function_node __wrapped__Fill_device_/job:localhost/replica:0/task:0/device:GPU:0}} OOM when allocating tensor with shape[26667,320] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc [Op:Fill] name: \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "gmcpngcCHeXy"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}