09_TF_Developer_Certificate_Course.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "c72981f1-e1e7-4bd1-b793-9f1d948a97db",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "c72981f1-e1e7-4bd1-b793-9f1d948a97db",
    "outputId": "e685eac2-1885-4741-831a-206f3b9eed3e",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:09.631441858Z",
     "start_time": "2024-01-03T20:43:06.802232316Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] 'helper_functions.py' already exists, skipping download.\n",
      "2.13.0\n",
      "1 Physical GPUs, 1 Logical GPUs\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3'\n",
    "os.environ[\"KMP_DUPLICATE_LIB_OK\"]=\"TRUE\"\n",
    "import tensorflow as tf\n",
    "if not os.path.exists(\"helper_functions.py\"):\n",
    "  !wget https://raw.githubusercontent.com/mrdbourke/tensorflow-deep-learning/main/extras/helper_functions.py\n",
    "else:\n",
    "  print(\"[INFO] 'helper_functions.py' already exists, skipping download.\")\n",
    "from helper_functions import *\n",
    "# from tensorflow.keras import mixed_precision\n",
    "# mixed_precision.set_global_policy('mixed_float16')\n",
    "importTensorflow(memory=4090)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3117852a-2e2b-4fc5-8ae6-00444ca94e90",
   "metadata": {
    "id": "3117852a-2e2b-4fc5-8ae6-00444ca94e90"
   },
   "source": [
    "### Convert harder to read text into neatly classified texts(ex: research paper divided nicely into intro, backgtround, conclusion, etc)\n",
    "It is a many to one RNN architecture based problem"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "936eafeb-3d11-4dba-85e9-561307b79420",
   "metadata": {
    "id": "936eafeb-3d11-4dba-85e9-561307b79420"
   },
   "source": [
    "# SKIMLIT : Milestone Project"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7b0f4759-c644-4b86-8158-a49cce4fbb7c",
   "metadata": {
    "id": "7b0f4759-c644-4b86-8158-a49cce4fbb7c"
   },
   "source": [
    "#### Get the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "65fd93a3-69d9-4509-8beb-2a7a28af9a37",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "65fd93a3-69d9-4509-8beb-2a7a28af9a37",
    "outputId": "c6486d76-67a7-4d16-859e-f94312cb7079",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:09.772148188Z",
     "start_time": "2024-01-03T20:43:09.611044215Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PubMed_200k_RCT\r\n",
      "PubMed_200k_RCT_numbers_replaced_with_at_sign\r\n",
      "PubMed_20k_RCT\r\n",
      "PubMed_20k_RCT_numbers_replaced_with_at_sign\r\n",
      "README.md\r\n"
     ]
    }
   ],
   "source": [
    "if not os.path.exists(\"pubmed_rct\"):\n",
    "    !git clone https://github.com/Franck-Dernoncourt/pubmed-rct.git\n",
    "    !mv 'pubmed-rct' pubmed_rct\n",
    "!ls ./pubmed_rct"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "07e405cf-89e9-4a43-b50e-235a5b8587a9",
   "metadata": {
    "id": "07e405cf-89e9-4a43-b50e-235a5b8587a9"
   },
   "source": [
    "#### Work on pubmed-20k now and scale it later"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "611da2fb-5c86-4488-b029-2bef217e638c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "611da2fb-5c86-4488-b029-2bef217e638c",
    "outputId": "6451c2e2-9d4a-40e6-a784-d3301d452a7b",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:09.959565084Z",
     "start_time": "2024-01-03T20:43:09.773366761Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dev.txt  test.txt  train.txt\r\n"
     ]
    }
   ],
   "source": [
    "!ls pubmed_rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b1631215-e7a5-4b50-95d1-7d570e517b1a",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "b1631215-e7a5-4b50-95d1-7d570e517b1a",
    "outputId": "b990e9f5-f43a-47e0-8013-a866b2907a2e",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:10.009909458Z",
     "start_time": "2024-01-03T20:43:09.960891542Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "['./pubmed_rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/train.txt',\n './pubmed_rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/dev.txt',\n './pubmed_rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/test.txt']"
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_dir = './pubmed_rct/PubMed_20k_RCT_numbers_replaced_with_at_sign/'\n",
    "filenames = [data_dir+filename for filename in os.listdir(data_dir)]\n",
    "filenames"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c84f7c9e-cb36-4c8d-810b-76fb963edb06",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "c84f7c9e-cb36-4c8d-810b-76fb963edb06",
    "outputId": "1af68c4e-49a3-45f0-eaf1-baf75f66cb15",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:10.012889500Z",
     "start_time": "2024-01-03T20:43:10.002231399Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "['###24290286\\n',\n 'BACKGROUND\\tIgE sensitization to Aspergillus fumigatus and a positive sputum fungal culture result are common in patients with refractory asthma .\\n',\n 'BACKGROUND\\tIt is not clear whether these patients would benefit from antifungal treatment .\\n',\n 'OBJECTIVE\\tWe sought to determine whether a @-month course of voriconazole improved asthma-related outcomes in patients with asthma who are IgE sensitized to A fumigatus .\\n',\n 'METHODS\\tAsthmatic patients who were IgE sensitized to A fumigatus with a history of at least @ severe exacerbations in the previous @ months were treated for @ months with @ mg of voriconazole twice daily , followed by observation for @ months , in a double-blind , placebo-controlled , randomized design .\\n']"
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_lines(filepath):\n",
    "    with open(filepath, 'r') as f:\n",
    "        return f.readlines()\n",
    "get_lines(filenames[1])[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0833a4df-d145-4cd1-84f1-eef2017ee619",
   "metadata": {
    "id": "0833a4df-d145-4cd1-84f1-eef2017ee619",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:10.015193806Z",
     "start_time": "2024-01-03T20:43:10.002376280Z"
    }
   },
   "outputs": [],
   "source": [
    "def preprocess_text_with_line_numbers(filename):\n",
    "    input_lines = get_lines(filename)\n",
    "    abstract_lines = \"\"\n",
    "    abstract_samples = []\n",
    "    for line in input_lines:\n",
    "        if line.startswith(\"###\"):\n",
    "            abstract_id = line\n",
    "            abstract_lines=\"\"\n",
    "        elif line.isspace():\n",
    "            abstract_line_split = abstract_lines.splitlines()\n",
    "\n",
    "            for abstract_line_number, abstract_line in enumerate(abstract_line_split):\n",
    "                line_data = {}\n",
    "                target_text_split = abstract_line.split(\"\\t\")\n",
    "                line_data[\"target\"] = target_text_split[0]\n",
    "                line_data[\"text\"] = target_text_split[1].lower()\n",
    "                line_data[\"line_number\"] = abstract_line_number\n",
    "                line_data[\"total_lines\"] = len(abstract_line_split)-1\n",
    "                abstract_samples.append(line_data)\n",
    "        else:\n",
    "            abstract_lines+=line\n",
    "    return abstract_samples"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "447f70ef-26ac-4f56-b1af-50de5b96d606",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "447f70ef-26ac-4f56-b1af-50de5b96d606",
    "outputId": "2950b3fe-2f12-4f22-c6ae-0b2562bbdf20",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:10.316796583Z",
     "start_time": "2024-01-03T20:43:10.002429942Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "(180040, 30212, 30135)"
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_samples = preprocess_text_with_line_numbers(filenames[0])\n",
    "val_samples = preprocess_text_with_line_numbers(filenames[1])\n",
    "test_samples = preprocess_text_with_line_numbers(filenames[-1])\n",
    "len(train_samples), len(val_samples), len(test_samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c6047026-aa70-40e3-8a7a-bcc484620b49",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "c6047026-aa70-40e3-8a7a-bcc484620b49",
    "outputId": "0c02f8d8-27e2-4fc2-c4e4-d4f6d61a73ef",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:10.321864090Z",
     "start_time": "2024-01-03T20:43:10.316230964Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "[{'target': 'OBJECTIVE',\n  'text': 'to investigate the efficacy of @ weeks of daily low-dose oral prednisolone in improving pain , mobility , and systemic low-grade inflammation in the short term and whether the effect would be sustained at @ weeks in older adults with moderate to severe knee osteoarthritis ( oa ) .',\n  'line_number': 0,\n  'total_lines': 11},\n {'target': 'METHODS',\n  'text': 'a total of @ patients with primary knee oa were randomized @:@ ; @ received @ mg/day of prednisolone and @ received placebo for @ weeks .',\n  'line_number': 1,\n  'total_lines': 11},\n {'target': 'METHODS',\n  'text': 'outcome measures included pain reduction and improvement in function scores and systemic inflammation markers .',\n  'line_number': 2,\n  'total_lines': 11},\n {'target': 'METHODS',\n  'text': 'pain was assessed using the visual analog pain scale ( @-@ mm ) .',\n  'line_number': 3,\n  'total_lines': 11},\n {'target': 'METHODS',\n  'text': 'secondary outcome measures included the western ontario and mcmaster universities osteoarthritis index scores , patient global assessment ( pga ) of the severity of knee oa , and @-min walk distance ( @mwd ) .',\n  'line_number': 4,\n  'total_lines': 11}]"
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_samples[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "6fbc6d16-18db-4aad-9129-e69d386c1922",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 483
    },
    "id": "6fbc6d16-18db-4aad-9129-e69d386c1922",
    "outputId": "b8bc9c8e-9c14-4b47-90d4-11287f751a3d",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:10.685442159Z",
     "start_time": "2024-01-03T20:43:10.497021426Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "         target                                               text  \\\n0     OBJECTIVE  to investigate the efficacy of @ weeks of dail...   \n1       METHODS  a total of @ patients with primary knee oa wer...   \n2       METHODS  outcome measures included pain reduction and i...   \n3       METHODS  pain was assessed using the visual analog pain...   \n4       METHODS  secondary outcome measures included the wester...   \n5       METHODS  serum levels of interleukin @ ( il-@ ) , il-@ ...   \n6       RESULTS  there was a clinically relevant reduction in t...   \n7       RESULTS  the mean difference between treatment arms ( @...   \n8       RESULTS  further , there was a clinically relevant redu...   \n9       RESULTS  these differences remained significant at @ we...   \n10      RESULTS  the outcome measures in rheumatology clinical ...   \n11  CONCLUSIONS  low-dose oral prednisolone had both a short-te...   \n12   BACKGROUND  emotional eating is associated with overeating...   \n13   BACKGROUND  yet , empirical evidence for individual ( trai...   \n\n    line_number  total_lines  \n0             0           11  \n1             1           11  \n2             2           11  \n3             3           11  \n4             4           11  \n5             5           11  \n6             6           11  \n7             7           11  \n8             8           11  \n9             9           11  \n10           10           11  \n11           11           11  \n12            0           10  \n13            1           10  ",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>target</th>\n      <th>text</th>\n      <th>line_number</th>\n      <th>total_lines</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>OBJECTIVE</td>\n      <td>to investigate the efficacy of @ weeks of dail...</td>\n      <td>0</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>METHODS</td>\n      <td>a total of @ patients with primary knee oa wer...</td>\n      <td>1</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>METHODS</td>\n      <td>outcome measures included pain reduction and i...</td>\n      <td>2</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>METHODS</td>\n      <td>pain was assessed using the visual analog pain...</td>\n      <td>3</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>METHODS</td>\n      <td>secondary outcome measures included the wester...</td>\n      <td>4</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>METHODS</td>\n      <td>serum levels of interleukin @ ( il-@ ) , il-@ ...</td>\n      <td>5</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>RESULTS</td>\n      <td>there was a clinically relevant reduction in t...</td>\n      <td>6</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>RESULTS</td>\n      <td>the mean difference between treatment arms ( @...</td>\n      <td>7</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>RESULTS</td>\n      <td>further , there was a clinically relevant redu...</td>\n      <td>8</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>RESULTS</td>\n      <td>these differences remained significant at @ we...</td>\n      <td>9</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>10</th>\n      <td>RESULTS</td>\n      <td>the outcome measures in rheumatology clinical ...</td>\n      <td>10</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>11</th>\n      <td>CONCLUSIONS</td>\n      <td>low-dose oral prednisolone had both a short-te...</td>\n      <td>11</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>BACKGROUND</td>\n      <td>emotional eating is associated with overeating...</td>\n      <td>0</td>\n      <td>10</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>BACKGROUND</td>\n      <td>yet , empirical evidence for individual ( trai...</td>\n      <td>1</td>\n      <td>10</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "train_df = pd.DataFrame(train_samples)\n",
    "val_df = pd.DataFrame(val_samples)\n",
    "test_df = pd.DataFrame(test_samples)\n",
    "train_df.head(14)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "39c990fb-b652-4052-825c-849466c4a082",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "39c990fb-b652-4052-825c-849466c4a082",
    "outputId": "ce9d238f-5b62-4c4d-dd34-42192c2cfcc3",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:10.974085292Z",
     "start_time": "2024-01-03T20:43:10.970180020Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "target\nMETHODS        59353\nRESULTS        57953\nCONCLUSIONS    27168\nBACKGROUND     21727\nOBJECTIVE      13839\nName: count, dtype: int64"
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.target.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2c581d8b-4033-4ea3-8cde-ce4cb81008d2",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 448
    },
    "id": "2c581d8b-4033-4ea3-8cde-ce4cb81008d2",
    "outputId": "003f3d5b-be8a-41cc-86c2-3884f7e340aa",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:11.699409228Z",
     "start_time": "2024-01-03T20:43:11.472552486Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<Axes: ylabel='Frequency'>"
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_df.total_lines.plot.hist()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f3c09ac8-61c5-446f-abc6-d6a81e7bb35b",
   "metadata": {
    "id": "f3c09ac8-61c5-446f-abc6-d6a81e7bb35b"
   },
   "source": [
    "#### Get lists of sentences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e094ce7a-4321-457a-bbff-594d38216506",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "e094ce7a-4321-457a-bbff-594d38216506",
    "outputId": "8794ab47-023f-45d2-8609-06283700d4b8",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:12.355999632Z",
     "start_time": "2024-01-03T20:43:12.324656989Z"
    }
   },
   "outputs": [
    {
     "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 ) .']"
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_sentences = train_df.text.to_list()\n",
    "val_sentences = val_df.text.to_list()\n",
    "test_sentences = test_df.text.to_list()\n",
    "train_sentences[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "e5edabdf-eec4-4da1-80da-013cd6b2c0d4",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "e5edabdf-eec4-4da1-80da-013cd6b2c0d4",
    "outputId": "cb04415e-bf91-45bf-f72c-f87d2e193f82",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:15.710070984Z",
     "start_time": "2024-01-03T20:43:15.661779374Z"
    }
   },
   "outputs": [
    {
     "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.]])"
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# make numeric conversion of labels/targets - method 1\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "one_hot_encoder = OneHotEncoder(sparse=False)\n",
    "train_labels_one_hot = one_hot_encoder.fit_transform(train_df.target.to_numpy().reshape(-1,1))\n",
    "val_labels_one_hot = one_hot_encoder.fit_transform(val_df.target.to_numpy().reshape(-1,1))\n",
    "test_labels_one_hot = one_hot_encoder.fit_transform(test_df.target.to_numpy().reshape(-1,1))\n",
    "train_labels_one_hot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a4e6ddd5-e8c5-4223-bc0d-bf900bacccee",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "a4e6ddd5-e8c5-4223-bc0d-bf900bacccee",
    "outputId": "cf3995f0-5fb2-4a23-a56c-bff7d704c7de",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:16.628457117Z",
     "start_time": "2024-01-03T20:43:16.598503520Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "array([3, 2, 2, ..., 4, 1, 1])"
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# label encoding - method 2\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "label_encoder = LabelEncoder()\n",
    "train_labels_encoded = label_encoder.fit_transform(train_df.target.to_numpy())\n",
    "val_labels_encoded = label_encoder.fit_transform(val_df.target.to_numpy())\n",
    "test_labels_encoded = label_encoder.fit_transform(test_df.target.to_numpy())\n",
    "train_labels_encoded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "aaab5d0a-5484-4429-aa64-9cded8f44ec7",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "aaab5d0a-5484-4429-aa64-9cded8f44ec7",
    "outputId": "72fa7b11-f0af-4a34-a392-4aaa8e11daaf",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:17.404527695Z",
     "start_time": "2024-01-03T20:43:17.365660461Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "(5,\n array(['BACKGROUND', 'CONCLUSIONS', 'METHODS', 'OBJECTIVE', 'RESULTS'],\n       dtype=object))"
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get class names and number of classes back from LabelEncoder\n",
    "num_classes = len(label_encoder.classes_)\n",
    "class_names = label_encoder.classes_\n",
    "num_classes, class_names"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a282e5e7-65e5-443f-bc2f-100209532145",
   "metadata": {
    "id": "a282e5e7-65e5-443f-bc2f-100209532145"
   },
   "source": [
    "#### Model 0 - Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "efe31290-306b-4a05-a9d6-0e667392eb8b",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "efe31290-306b-4a05-a9d6-0e667392eb8b",
    "outputId": "c492c703-bca4-4499-98c3-f9844c08310c",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:21.331617328Z",
     "start_time": "2024-01-03T20:43:18.476114938Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "{'accuracy': 72.1832384482987,\n 'precision': 0.7186466952323352,\n 'recall': 0.7218323844829869,\n 'f1': 0.6989250353450294}"
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "# Create tokenization and modelling pipeline\n",
    "model_0 = Pipeline([\n",
    "                    (\"tfidf\", TfidfVectorizer()), # convert words to numbers using tfidf\n",
    "                    (\"clf\", MultinomialNB()) # model the text\n",
    "])\n",
    "model_0.fit(train_sentences, train_labels_encoded)\n",
    "model_0_results = calculate_results(val_labels_encoded, model_0.predict(val_sentences))\n",
    "model_0_results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4803ead6-054a-462b-af50-1e4fd8e78494",
   "metadata": {
    "id": "4803ead6-054a-462b-af50-1e4fd8e78494"
   },
   "source": [
    "#### Model 1 - Conv1D"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "01547241-b0fb-43d2-8682-75a4439c2c07",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "01547241-b0fb-43d2-8682-75a4439c2c07",
    "outputId": "71dc5c58-2b6e-40ad-be98-e25758b7c5e1",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:22.235506553Z",
     "start_time": "2024-01-03T20:43:21.940036619Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "26"
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# how long is each sentece of average\n",
    "sent_length = [len(sentence.split()) for sentence in train_sentences]\n",
    "tf.math.reduce_mean(sent_length).numpy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "eb43f603-2e11-49c1-822d-d4ae011167f8",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "eb43f603-2e11-49c1-822d-d4ae011167f8",
    "outputId": "579ea6ae-1a06-42cb-f7de-4c9a45518323",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:25.914808472Z",
     "start_time": "2024-01-03T20:43:25.877384781Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "55"
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# we want 95% of data to be covered, so whats the length of 95% of data\n",
    "import numpy as np\n",
    "int(np.percentile(sent_length, 95))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "367d97df-d552-41ea-b461-38d52bf7b5f2",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "367d97df-d552-41ea-b461-38d52bf7b5f2",
    "outputId": "dd0274b4-f0c4-4335-de89-ba2d6703c535",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:33.096086037Z",
     "start_time": "2024-01-03T20:43:27.090965729Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<tf.Tensor: shape=(1, 55), dtype=int64, numpy=\narray([[ 6193,    61,   898,    58, 10096,     0,     0,     0,     0,\n            0,     0,     0,     0,     0,     0,     0,     0,     0,\n            0,     0,     0,     0,     0,     0,     0,     0,     0,\n            0,     0,     0,     0,     0,     0,     0,     0,     0,\n            0,     0,     0,     0,     0,     0,     0,     0,     0,\n            0,     0,     0,     0,     0,     0,     0,     0,     0,\n            0]])>"
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create text vectorizer\n",
    "text_vectorizer = tf.keras.layers.TextVectorization(max_tokens = 68000,\n",
    "                                                output_sequence_length=55)\n",
    "text_vectorizer.adapt(train_sentences)\n",
    "sample_sentence = \"Hi there, how are you\"\n",
    "text_vectorizer([sample_sentence])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "7def48f0-0d00-42d8-864b-bcf98c918407",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "7def48f0-0d00-42d8-864b-bcf98c918407",
    "outputId": "20ee8886-b193-408d-e4f2-ab8de20a10fe",
    "ExecuteTime": {
     "end_time": "2024-01-03T20:43:33.101253413Z",
     "start_time": "2024-01-03T20:43:33.095291848Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "{'name': 'text_vectorization',\n 'trainable': True,\n 'dtype': 'string',\n 'batch_input_shape': (None,),\n 'max_tokens': 68000,\n 'standardize': 'lower_and_strip_punctuation',\n 'split': 'whitespace',\n 'ngrams': None,\n 'output_mode': 'int',\n 'output_sequence_length': 55,\n 'pad_to_max_tokens': False,\n 'sparse': False,\n 'ragged': False,\n 'vocabulary': None,\n 'idf_weights': None,\n 'encoding': 'utf-8',\n 'vocabulary_size': 64841}"
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "text_vectorizer.get_config()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "b3777b1d-9508-4a3e-a9fd-5f216fd3c5c4",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "b3777b1d-9508-4a3e-a9fd-5f216fd3c5c4",
    "outputId": "a6ff3aa5-7954-4eb4-93d7-5dcd4ecd5bf7",
    "ExecuteTime": {
     "end_time": "2024-01-03T17:29:25.065258571Z",
     "start_time": "2024-01-03T17:29:24.982891652Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<tf.Tensor: shape=(1, 55, 128), dtype=float32, numpy=\narray([[[ 0.00226015,  0.01180372, -0.03206461, ..., -0.03607845,\n          0.00944325, -0.0085087 ],\n        [-0.04890101,  0.03625152,  0.01611224, ..., -0.02481217,\n          0.04070003, -0.00928164],\n        [ 0.02036612,  0.0136939 , -0.01620729, ..., -0.02469981,\n          0.01816425,  0.02977945],\n        ...,\n        [ 0.04459692,  0.01344902, -0.02644463, ..., -0.04792669,\n         -0.00396391,  0.02958813],\n        [ 0.04459692,  0.01344902, -0.02644463, ..., -0.04792669,\n         -0.00396391,  0.02958813],\n        [ 0.04459692,  0.01344902, -0.02644463, ..., -0.04792669,\n         -0.00396391,  0.02958813]]], dtype=float32)>"
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create embedding layer\n",
    "embedding = tf.keras.layers.Embedding(input_dim=68000,\n",
    "                                     output_dim=128,\n",
    "                                      input_length=55,\n",
    "                                     mask_zero=True)\n",
    "embedding(text_vectorizer([sample_sentence]))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2986261-2805-4e70-b206-1934e93d0c3b",
   "metadata": {
    "id": "c2986261-2805-4e70-b206-1934e93d0c3b"
   },
   "source": [
    "#### Creating Fast loading dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "be8834d7-e652-44c3-84a7-becebf0f50cd",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "be8834d7-e652-44c3-84a7-becebf0f50cd",
    "outputId": "1b562269-b573-4411-e075-cb4b0d446ddc",
    "ExecuteTime": {
     "end_time": "2024-01-03T17:29:26.826890521Z",
     "start_time": "2024-01-03T17:29:26.336851029Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<_TensorSliceDataset element_spec=(TensorSpec(shape=(), dtype=tf.string, name=None), TensorSpec(shape=(5,), dtype=tf.float64, name=None))>"
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Turn data into tensorflow dataset\n",
    "train_dataset = tf.data.Dataset.from_tensor_slices((train_sentences, train_labels_one_hot))\n",
    "val_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"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "b309419a-f21a-4463-a1d8-024b59c26c64",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "b309419a-f21a-4463-a1d8-024b59c26c64",
    "outputId": "47c40c3c-98ec-440e-c50f-a183e7dff779",
    "ExecuteTime": {
     "end_time": "2024-01-03T17:29:27.465027717Z",
     "start_time": "2024-01-03T17:29:27.420088313Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<_PrefetchDataset element_spec=(TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>"
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# take tensor slice dadtaset and turn into prefetched datasets\n",
    "train_dataset = train_dataset.batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "val_dataset = val_dataset.batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "test_dataset = test_dataset.batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "\n",
    "train_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "a22c87d0-7177-4e27-9959-a696f1bd6126",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "a22c87d0-7177-4e27-9959-a696f1bd6126",
    "outputId": "f0ae6f44-6cbf-453a-f74d-1cdc2f9b9b86",
    "ExecuteTime": {
     "end_time": "2024-01-03T17:30:00.534580286Z",
     "start_time": "2024-01-03T17:29:27.969621790Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "281/281 [==============================] - 21s 66ms/step - loss: 1.0504 - accuracy: 0.5797 - val_loss: 0.7395 - val_accuracy: 0.7197\n",
      "Epoch 2/5\n",
      "281/281 [==============================] - 4s 15ms/step - loss: 0.6845 - accuracy: 0.7460 - val_loss: 0.6444 - val_accuracy: 0.7650\n",
      "Epoch 3/5\n",
      "281/281 [==============================] - 3s 11ms/step - loss: 0.6267 - accuracy: 0.7725 - val_loss: 0.6118 - val_accuracy: 0.7796\n",
      "Epoch 4/5\n",
      "281/281 [==============================] - 2s 7ms/step - loss: 0.5957 - accuracy: 0.7867 - val_loss: 0.5842 - val_accuracy: 0.7889\n",
      "Epoch 5/5\n",
      "281/281 [==============================] - 2s 8ms/step - loss: 0.5965 - accuracy: 0.7889 - val_loss: 0.5728 - val_accuracy: 0.7979\n"
     ]
    }
   ],
   "source": [
    "# create conv1d model\n",
    "inputs = tf.keras.layers.Input(shape=(1,), dtype=tf.string)\n",
    "x = text_vectorizer(inputs)\n",
    "x = embedding(x)\n",
    "x = tf.keras.layers.Conv1D(filters=64, kernel_size=5, strides=1, padding='same', activation='relu')(x)\n",
    "x = tf.keras.layers.GlobalAveragePooling1D()(x)\n",
    "outputs = tf.keras.layers.Dense(5, activation=tf.keras.activations.softmax)(x)\n",
    "\n",
    "model_1 = tf.keras.Model(inputs, outputs)\n",
    "model_1.compile(loss=tf.keras.losses.CategoricalCrossentropy(),\n",
    "               optimizer=tf.keras.optimizers.Adam(),\n",
    "               metrics=['accuracy'])\n",
    "history_1 = model_1.fit(train_dataset,\n",
    "                       epochs=5,\n",
    "                        steps_per_epoch=int(0.1*len(train_dataset)),\n",
    "                        validation_steps=int(0.1*len(val_dataset)),\n",
    "                       validation_data=val_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "87b29052-ce8d-43ef-a6f9-1c3544e43b71",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "87b29052-ce8d-43ef-a6f9-1c3544e43b71",
    "outputId": "a92c09b2-0c69-4b6a-a5a9-e7771950fdc6",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:24:53.767052834Z",
     "start_time": "2024-01-03T15:24:52.914455425Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 1s 2ms/step - loss: 0.5730 - accuracy: 0.7965\n"
     ]
    },
    {
     "data": {
      "text/plain": "[0.5730479955673218, 0.7965046763420105]"
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_1.evaluate(val_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "5e8aa6fd-7436-433b-b85f-ae86759bab31",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "5e8aa6fd-7436-433b-b85f-ae86759bab31",
    "outputId": "8b9a2fc3-61b1-4059-cd8d-b17a6ee4aea4",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:24:54.571475599Z",
     "start_time": "2024-01-03T15:24:53.766563742Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 1s 1ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "{'accuracy': 79.6504700119158,\n 'precision': 0.7951020285922893,\n 'recall': 0.796504700119158,\n 'f1': 0.7927107231102297}"
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_1_pred_probs = model_1.predict(val_dataset)\n",
    "model_1_preds = tf.argmax(model_1_pred_probs, axis=1).numpy()\n",
    "model_1_results = calculate_results(val_labels_encoded, model_1_preds)\n",
    "model_1_results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfbc07b2-e63f-441d-95da-f3a3024a285d",
   "metadata": {
    "id": "dfbc07b2-e63f-441d-95da-f3a3024a285d"
   },
   "source": [
    "#### Model 2 - USE - TF HUB Pretrained Feature extractor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "a092aa80-edcf-4ee8-be42-24906458f433",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "a092aa80-edcf-4ee8-be42-24906458f433",
    "outputId": "c615e343-639d-4eec-cb9b-d2c77e6a817e",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:24:58.692560176Z",
     "start_time": "2024-01-03T15:24:54.570261427Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "tf.Tensor(\n",
      "[-0.05182884 -0.04390026  0.05248979 -0.00642138  0.0454952  -0.02146592\n",
      "  0.03058733 -0.07073407  0.05152308  0.04701421  0.02491418  0.00674201\n",
      " -0.03172521  0.0080713   0.00499434 -0.09463525 -0.03576988  0.0037949\n",
      " -0.04158678 -0.07285322  0.0632658  -0.01339925  0.02972649  0.04720227\n",
      " -0.06259578 -0.0154889   0.039651    0.05203718 -0.00980406  0.05468165\n",
      "  0.0588766   0.0431632   0.05944403 -0.00140851 -0.01803795  0.00555375\n",
      "  0.01559336 -0.03206627  0.02260791  0.02096884  0.00670307  0.02270131\n",
      "  0.02106516  0.07490976  0.05131475  0.00114743 -0.09906604  0.0151604\n",
      "  0.01459556  0.02136121], shape=(50,), dtype=float32)\n"
     ]
    }
   ],
   "source": [
    "import tensorflow_hub as hub\n",
    "embed = hub.load(\"https://www.kaggle.com/models/google/universal-sentence-encoder/frameworks/TensorFlow2/variations/universal-sentence-encoder/versions/2\")\n",
    "embed_sample = embed([sample_sentence,\n",
    "                      \"When you can the unviersal sentence encider on a sentence, it turns into into numbers\"])\n",
    "print(embed_sample[0][:50])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "d0c6e4cb-16f9-471e-bfe2-e9d9de030ff1",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "d0c6e4cb-16f9-471e-bfe2-e9d9de030ff1",
    "outputId": "d255fdef-5c52-40ea-dd1c-826f5eaada55",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:13.623569572Z",
     "start_time": "2024-01-03T15:24:58.713213394Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "281/281 [==============================] - 5s 10ms/step - loss: 0.9761 - accuracy: 0.6309 - val_loss: 0.8065 - val_accuracy: 0.6845\n",
      "Epoch 2/5\n",
      "281/281 [==============================] - 2s 9ms/step - loss: 0.7840 - accuracy: 0.6933 - val_loss: 0.7674 - val_accuracy: 0.6988\n",
      "Epoch 3/5\n",
      "281/281 [==============================] - 2s 9ms/step - loss: 0.7701 - accuracy: 0.7045 - val_loss: 0.7573 - val_accuracy: 0.7045\n",
      "Epoch 4/5\n",
      "281/281 [==============================] - 3s 10ms/step - loss: 0.7393 - accuracy: 0.7157 - val_loss: 0.7344 - val_accuracy: 0.7134\n",
      "Epoch 5/5\n",
      "281/281 [==============================] - 2s 9ms/step - loss: 0.7509 - accuracy: 0.7124 - val_loss: 0.7158 - val_accuracy: 0.7204\n"
     ]
    }
   ],
   "source": [
    "# build USE model\n",
    "tf_hub_embedding_layer = hub.KerasLayer(embed,\n",
    "                                       trainable=False)\n",
    "\n",
    "inputs2 = tf.keras.layers.Input(shape=[], dtype=tf.string)\n",
    "x = tf_hub_embedding_layer(inputs2)\n",
    "x = tf.keras.layers.Dense(128, activation=tf.keras.activations.relu)(x)\n",
    "outputs2 = tf.keras.layers.Dense(5, activation=tf.keras.activations.softmax)(x)\n",
    "model_2 = tf.keras.Model(inputs2, outputs2)\n",
    "model_2.compile(loss=tf.keras.losses.CategoricalCrossentropy(),\n",
    "               optimizer=tf.keras.optimizers.Adam(),\n",
    "               metrics=['accuracy'])\n",
    "history_2 = model_2.fit(train_dataset,\n",
    "                       epochs=5,\n",
    "                        steps_per_epoch=int(0.1*len(train_dataset)),\n",
    "                        validation_steps=int(0.1*len(val_dataset)),\n",
    "                       validation_data=val_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "17ac87cc-beac-4709-b89e-718d4c7d539c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "17ac87cc-beac-4709-b89e-718d4c7d539c",
    "outputId": "23b2ca46-15a3-4448-b76f-4c61d5898eac",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:17.110422943Z",
     "start_time": "2024-01-03T15:25:13.623292445Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 3s 7ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "{'accuracy': 72.52085264133457,\n 'precision': 0.720505856191062,\n 'recall': 0.7252085264133457,\n 'f1': 0.719540778194056}"
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_2_preds = tf.argmax(model_2.predict(val_dataset),axis=1).numpy()\n",
    "model_2_results = calculate_results(val_labels_encoded, model_2_preds)\n",
    "model_2_results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c340e59d-d8a6-4ca2-945f-36322fdad444",
   "metadata": {
    "id": "c340e59d-d8a6-4ca2-945f-36322fdad444"
   },
   "source": [
    "#### Model3 - Conv1D with character embedding"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "fb869fcb-ff43-4fa8-83b4-8685932bab0d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 86
    },
    "id": "fb869fcb-ff43-4fa8-83b4-8685932bab0d",
    "outputId": "b7737226-95fe-46be-a1ee-23c3ad1d3384",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:17.112831038Z",
     "start_time": "2024-01-03T15:25:17.106355567Z"
    }
   },
   "outputs": [
    {
     "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   )   .'"
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# split into characters\n",
    "def split_chars(text):\n",
    "    return \" \".join(list(text))\n",
    "split_chars(train_sentences[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "95d4f98b-9bd7-45db-b589-1636048cdb0a",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 86
    },
    "id": "95d4f98b-9bd7-45db-b589-1636048cdb0a",
    "outputId": "b5646667-e1e3-4e61-99d9-f78a090f77c2",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:17.576652002Z",
     "start_time": "2024-01-03T15:25:17.106608945Z"
    }
   },
   "outputs": [
    {
     "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   )   .'"
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#do for whole dataset\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[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "5b91a299-65ab-473e-b297-3aeb27950e2e",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "5b91a299-65ab-473e-b297-3aeb27950e2e",
    "outputId": "3587bd62-c7aa-484e-c629-a1afc851361e",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:17.630081169Z",
     "start_time": "2024-01-03T15:25:17.570556916Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "149.3662574983337"
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# avg len of chars\n",
    "char_lens = [len(sentence) for sentence in train_sentences]\n",
    "avg_char_len = np.mean(char_lens)\n",
    "avg_char_len"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "2861474b-3240-4d6c-91c7-43a33b5efd8c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "2861474b-3240-4d6c-91c7-43a33b5efd8c",
    "outputId": "fdfc34d1-87ae-4e02-c621-0b0f1e55f06b",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:17.632873497Z",
     "start_time": "2024-01-03T15:25:17.610324259Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "290.0"
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.percentile(char_lens, 95)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "aefb4b30-059f-4020-8ddc-2bea9a330a96",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "aefb4b30-059f-4020-8ddc-2bea9a330a96",
    "outputId": "595f7bb8-7524-4473-84a6-629bc0332a49",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:17.734761074Z",
     "start_time": "2024-01-03T15:25:17.610697437Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "'abcdefghijklmnopqrstuvwxyz0123456789!\"#$%&\\'()*+,-./:;<=>?@[\\\\]^_`{|}~'"
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import string\n",
    "alphabet = string.ascii_lowercase+string.digits+string.punctuation\n",
    "alphabet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "448776f4-a8b1-48f9-9409-bb72911342f1",
   "metadata": {
    "id": "448776f4-a8b1-48f9-9409-bb72911342f1",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:17.737416149Z",
     "start_time": "2024-01-03T15:25:17.654435164Z"
    }
   },
   "outputs": [],
   "source": [
    "#Create char-level token vectorizer instance\n",
    "num_char_token = len(alphabet)+2\n",
    "char_vectorizer = tf.keras.layers.TextVectorization(max_tokens = num_char_token,\n",
    "                                                    standardize='lower_and_strip_punctuation',\n",
    "                                                output_sequence_length=290)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "15f9ef3a-e9e5-40bb-8f44-978bca77349b",
   "metadata": {
    "id": "15f9ef3a-e9e5-40bb-8f44-978bca77349b",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:27.033017333Z",
     "start_time": "2024-01-03T15:25:17.654848405Z"
    }
   },
   "outputs": [],
   "source": [
    "char_vectorizer.adapt(train_chars)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "98d09e3b-e472-4830-ad10-cd805ec1fc60",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "98d09e3b-e472-4830-ad10-cd805ec1fc60",
    "outputId": "35dd66fd-7887-408a-d99d-bfa52076a581",
    "ExecuteTime": {
     "start_time": "2024-01-03T15:25:27.047984176Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "['', '[UNK]', 'e', 't', 'i']"
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "char_vectorizer.get_vocabulary()[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "fLqeTlq9oQi9",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "fLqeTlq9oQi9",
    "outputId": "c4e270af-93d3-4831-e983-6eb72894b91e",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:27.147987958Z",
     "start_time": "2024-01-03T15:25:27.090464008Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<tf.Tensor: shape=(1, 290), dtype=int64, numpy=\narray([[17,  7,  8,  3, 19,  3, 20,  7, 14,  8,  5, 11,  3,  4, 11,  2,\n         9, 17,  8,  7, 15,  3, 13,  2, 14,  2, 10,  4,  5,  3,  8,  4,\n        11,  8,  2,  9,  2,  5,  8, 11, 13,  4,  6,  7, 17, 17,  4, 11,\n         2,  9,  2,  3,  3,  4,  6, 18,  9,  6,  2,  3, 20,  7,  8, 23,\n         7, 17,  3, 13,  2,  5, 15,  2,  8,  4, 11,  5,  6,  5, 11,  5,\n        10,  2, 15, 19,  7, 17, 14,  2, 10,  4,  5,  3,  8,  4, 11,  9,\n        20,  2,  8,  2,  8,  5,  6, 10,  7, 15, 12, 19,  5,  9,  9,  4,\n        18,  6,  2, 10,  3,  7,  7, 17, 18,  8,  7, 16, 14,  9,  0,  0,\n         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,\n         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,\n         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,\n         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,\n         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,\n         0,  0]])>"
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "char_vectorizer([train_chars[69]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "k5j1hAuSoiPD",
   "metadata": {
    "id": "k5j1hAuSoiPD",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:27.150527411Z",
     "start_time": "2024-01-03T15:25:27.134333201Z"
    }
   },
   "outputs": [],
   "source": [
    "# Create a character level embedding\n",
    "char_embed = tf.keras.layers.Embedding(input_dim=len(char_vectorizer.get_vocabulary()),\n",
    "                                       output_dim=25,\n",
    "                                       mask_zero=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "8OzYWQkNpiFO",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8OzYWQkNpiFO",
    "outputId": "0b3b483a-f8af-4440-ce5f-ef57e44896b4",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:27.196327743Z",
     "start_time": "2024-01-03T15:25:27.134944761Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<tf.Tensor: shape=(1, 290, 25), dtype=float32, numpy=\narray([[[-0.0326476 ,  0.0358884 , -0.04551125, ...,  0.02707442,\n         -0.02932605,  0.02666212],\n        [-0.00886161,  0.00471541,  0.01055075, ..., -0.00577623,\n         -0.02426082, -0.01620894],\n        [-0.04393995,  0.04564991,  0.00976259, ..., -0.0336604 ,\n          0.03632439, -0.04317206],\n        ...,\n        [ 0.02328774, -0.01656333,  0.01969042, ..., -0.03496037,\n          0.03137659, -0.03380005],\n        [ 0.02328774, -0.01656333,  0.01969042, ..., -0.03496037,\n          0.03137659, -0.03380005],\n        [ 0.02328774, -0.01656333,  0.01969042, ..., -0.03496037,\n          0.03137659, -0.03380005]]], dtype=float32)>"
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "char_embed(char_vectorizer([train_chars[69]]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "AwLYGon5rau-",
   "metadata": {
    "id": "AwLYGon5rau-",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:27.726749484Z",
     "start_time": "2024-01-03T15:25:27.192027934Z"
    }
   },
   "outputs": [],
   "source": [
    "train_char_dataset = tf.data.Dataset.from_tensor_slices((train_chars, train_labels_one_hot)).batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "val_char_dataset = tf.data.Dataset.from_tensor_slices((val_chars, val_labels_one_hot)).batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "test_char_dataset = tf.data.Dataset.from_tensor_slices((test_chars, test_labels_one_hot)).batch(64).prefetch(tf.data.AUTOTUNE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "ojuwJMWksKdY",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "ojuwJMWksKdY",
    "outputId": "2cd4239e-e956-48f9-b2c5-c45399fa6efc",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:27.732751971Z",
     "start_time": "2024-01-03T15:25:27.727944511Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<_PrefetchDataset element_spec=(TensorSpec(shape=(None,), dtype=tf.string, name=None), TensorSpec(shape=(None, 5), dtype=tf.float64, name=None))>"
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_char_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "Mfe2n50OpxQb",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Mfe2n50OpxQb",
    "outputId": "7ff3ff23-92f5-4b0c-a921-6173ff02df75",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:36.178554995Z",
     "start_time": "2024-01-03T15:25:27.737354276Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "281/281 [==============================] - 3s 7ms/step - loss: 1.2878 - accuracy: 0.4866 - val_loss: 1.0355 - val_accuracy: 0.5864\n",
      "Epoch 2/5\n",
      "281/281 [==============================] - 1s 4ms/step - loss: 0.9831 - accuracy: 0.6085 - val_loss: 0.9097 - val_accuracy: 0.6473\n",
      "Epoch 3/5\n",
      "281/281 [==============================] - 1s 5ms/step - loss: 0.8905 - accuracy: 0.6535 - val_loss: 0.8282 - val_accuracy: 0.6855\n",
      "Epoch 4/5\n",
      "281/281 [==============================] - 1s 4ms/step - loss: 0.8330 - accuracy: 0.6808 - val_loss: 0.7984 - val_accuracy: 0.6898\n",
      "Epoch 5/5\n",
      "281/281 [==============================] - 1s 4ms/step - loss: 0.8126 - accuracy: 0.6853 - val_loss: 0.7634 - val_accuracy: 0.7108\n"
     ]
    }
   ],
   "source": [
    "# Model 3 - Conv1D with character embedding\n",
    "inputs3 = tf.keras.Input(shape=(1,), dtype=tf.string)\n",
    "x = char_vectorizer(inputs3)\n",
    "x = char_embed(x)\n",
    "x = tf.keras.layers.Conv1D(filters=128, kernel_size=5, strides=1)(x)\n",
    "x = tf.keras.layers.GlobalMaxPool1D()(x)\n",
    "outputs3 = tf.keras.layers.Dense(5, activation=tf.keras.activations.softmax)(x)\n",
    "\n",
    "model_3 =tf.keras.Model(inputs3, outputs3)\n",
    "model_3.compile(loss=tf.keras.losses.CategoricalCrossentropy(),\n",
    "               optimizer=tf.keras.optimizers.Adam(),\n",
    "               metrics=['accuracy'])\n",
    "history_3 = model_3.fit(train_char_dataset,\n",
    "                        epochs=5,\n",
    "                        steps_per_epoch=int(0.1*len(train_char_dataset)),\n",
    "                        validation_data=val_char_dataset,\n",
    "                        validation_steps=int(0.1*len(val_char_dataset)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "mqL2MiTJq6me",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "mqL2MiTJq6me",
    "outputId": "b2923297-34f6-4192-c477-bfba72b29a8b",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:37.181645977Z",
     "start_time": "2024-01-03T15:25:36.178476534Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 1s 2ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "{'accuracy': 70.38593936184297,\n 'precision': 0.693403307249158,\n 'recall': 0.7038593936184298,\n 'f1': 0.6947280052718605}"
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_3_preds = tf.argmax(model_3.predict(val_char_dataset), axis=1).numpy()\n",
    "model_3_results = calculate_results(val_labels_encoded, model_3_preds)\n",
    "model_3_results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "AiffRpH5Qlrn",
   "metadata": {
    "id": "AiffRpH5Qlrn"
   },
   "source": [
    "#### Model 4 - Combine token embedding and character embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "mGJdTdxUWeLN",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "mGJdTdxUWeLN",
    "outputId": "94ab80ff-1661-4816-d867-f5a0238dc3d1",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:38.202377101Z",
     "start_time": "2024-01-03T15:25:37.174457131Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"token_and_char_embedding\"\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",
      " text_vectorization_1 (Text  (None, 290)                  0         ['char_input[0][0]']          \n",
      " Vectorization)                                                                                   \n",
      "                                                                                                  \n",
      " keras_layer (KerasLayer)    (None, 512)                  2567978   ['token_input[0][0]']         \n",
      "                                                          24                                      \n",
      "                                                                                                  \n",
      " embedding_1 (Embedding)     (None, 290, 25)              700       ['text_vectorization_1[1][0]']\n",
      "                                                                                                  \n",
      " token_outputs (Dense)       (None, 128)                  65664     ['keras_layer[1][0]']         \n",
      "                                                                                                  \n",
      " char_bidirectional (Bidire  (None, 48)                   9600      ['embedding_1[1][0]']         \n",
      " ctional)                                                                                         \n",
      "                                                                                                  \n",
      " token_char_hybrid (Concate  (None, 176)                  0         ['token_outputs[0][0]',       \n",
      " nate)                                                               'char_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"
     ]
    }
   ],
   "source": [
    "# setup token inputs\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,\n",
    "                                     activation=tf.keras.activations.relu,\n",
    "                                     name='token_outputs')(token_embeddings)\n",
    "token_model = tf.keras.Model(token_inputs, token_outputs)\n",
    "\n",
    "# setup char inputs\n",
    "char_inputs = tf.keras.layers.Input(shape=(1,), dtype=tf.string, name='char_input')\n",
    "char_vector = char_vectorizer(char_inputs)\n",
    "char_embeddings = char_embed(char_vector)\n",
    "char_bi_lstm = tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(24), name='char_bidirectional')(char_embeddings)\n",
    "char_model = tf.keras.Model(char_inputs, char_bi_lstm)\n",
    "\n",
    "#concatenate the inputs\n",
    "token_char_concat = tf.keras.layers.Concatenate(name='token_char_hybrid')([token_model.output,\n",
    "                                                                           char_model.output])\n",
    "# create output layers - adding dropouts\n",
    "combined_dropout = tf.keras.layers.Dropout(0.5)(token_char_concat)\n",
    "combined_dense = tf.keras.layers.Dense(128, activation = tf.keras.activations.relu)(combined_dropout)\n",
    "final_dropout = tf.keras.layers.Dropout(0.5)(combined_dense)\n",
    "output_layer = tf.keras.layers.Dense(5, activation=tf.keras.activations.softmax)(final_dropout)\n",
    "\n",
    "model_4 = tf.keras.Model(inputs=[token_model.input, char_model.input],\n",
    "                         outputs=output_layer,\n",
    "                         name='token_and_char_embedding')\n",
    "\n",
    "model_4.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "12rA6tMjZVaj",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 976
    },
    "id": "12rA6tMjZVaj",
    "outputId": "7be7aa8e-c24d-4e9d-e132-a9b5ce20438b",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:38.470167895Z",
     "start_time": "2024-01-03T15:25:38.199036159Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": "<IPython.core.display.Image object>"
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# plot the model summary\n",
    "from keras.utils import plot_model\n",
    "plot_model(model_4, show_shapes=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "4aoyOjRgaXMZ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "4aoyOjRgaXMZ",
    "outputId": "262da2e5-914d-4f48-e28d-34d9442d7ce8",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:25:39.820449180Z",
     "start_time": "2024-01-03T15:25:38.474600294Z"
    }
   },
   "outputs": [
    {
     "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))>)"
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# compile the model\n",
    "model_4.compile(loss=tf.keras.losses.CategoricalCrossentropy(),\n",
    "                optimizer=tf.keras.optimizers.Adam(),\n",
    "                metrics=['accuracy'])\n",
    "# create ttensorflwo dataset for the combined input\n",
    "train_char_token_data = tf.data.Dataset.from_tensor_slices((train_sentences,train_chars))\n",
    "train_char_token_labels = tf.data.Dataset.from_tensor_slices(train_labels_one_hot)\n",
    "train_char_token_dataset = tf.data.Dataset.zip((train_char_token_data, train_char_token_labels))\n",
    "train_char_token_dataset = train_char_token_dataset.batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "\n",
    "val_char_token_data = tf.data.Dataset.from_tensor_slices((val_sentences,val_chars))\n",
    "val_char_token_labels = tf.data.Dataset.from_tensor_slices(val_labels_one_hot)\n",
    "val_char_token_dataset = tf.data.Dataset.zip((val_char_token_data, val_char_token_labels))\n",
    "val_char_token_dataset = val_char_token_dataset.batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "\n",
    "test_char_token_data = tf.data.Dataset.from_tensor_slices((test_sentences,test_chars))\n",
    "test_char_token_labels = tf.data.Dataset.from_tensor_slices(test_labels_one_hot)\n",
    "test_char_token_dataset = tf.data.Dataset.zip((test_char_token_data, test_char_token_labels))\n",
    "test_char_token_dataset = test_char_token_dataset.batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "\n",
    "train_char_token_dataset, val_char_token_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "2HsZgYOCdbId",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "2HsZgYOCdbId",
    "outputId": "30be1bbb-9748-40c9-fbeb-ef54f708c6cc",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:17.910120840Z",
     "start_time": "2024-01-03T15:25:39.820045382Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "281/281 [==============================] - 14s 27ms/step - loss: 1.0407 - accuracy: 0.5819 - val_loss: 0.7940 - val_accuracy: 0.6935\n",
      "Epoch 2/5\n",
      "281/281 [==============================] - 6s 22ms/step - loss: 0.8191 - accuracy: 0.6813 - val_loss: 0.7364 - val_accuracy: 0.7214\n",
      "Epoch 3/5\n",
      "281/281 [==============================] - 5s 19ms/step - loss: 0.7873 - accuracy: 0.7013 - val_loss: 0.7126 - val_accuracy: 0.7281\n",
      "Epoch 4/5\n",
      "281/281 [==============================] - 6s 23ms/step - loss: 0.7509 - accuracy: 0.7162 - val_loss: 0.6830 - val_accuracy: 0.7443\n",
      "Epoch 5/5\n",
      "281/281 [==============================] - 6s 20ms/step - loss: 0.7572 - accuracy: 0.7124 - val_loss: 0.6624 - val_accuracy: 0.7507\n"
     ]
    }
   ],
   "source": [
    "# fit model 4\n",
    "history_4 = model_4.fit(train_char_token_dataset,\n",
    "                        epochs=5,\n",
    "                        validation_data=val_char_token_dataset,\n",
    "                        steps_per_epoch=int(0.1*len(train_char_token_dataset)),\n",
    "                        validation_steps=int(0.1*len(val_char_token_dataset)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "LD2urEC4eEJv",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "LD2urEC4eEJv",
    "outputId": "ad6830df-7a29-4069-f102-a5596f2bbe7e",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:22.018732527Z",
     "start_time": "2024-01-03T15:26:17.910056327Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 4s 9ms/step - loss: 0.6715 - accuracy: 0.7441\n"
     ]
    },
    {
     "data": {
      "text/plain": "[0.6715497970581055, 0.7441413998603821]"
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#evaluate\n",
    "model_4.evaluate(val_char_token_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "sAmSk5E8f9s0",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "sAmSk5E8f9s0",
    "outputId": "155d7f7f-5de7-4b6a-d0da-26fe90d4611d",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:28.224789527Z",
     "start_time": "2024-01-03T15:26:22.019007601Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 6s 9ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "{'accuracy': 74.41414007679069,\n 'precision': 0.7407433815757226,\n 'recall': 0.7441414007679068,\n 'f1': 0.7388932019267636}"
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_4_preds = tf.argmax(model_4.predict(val_char_token_dataset), axis=1).numpy()\n",
    "model_4_results = calculate_results(val_labels_encoded, model_4_preds)\n",
    "model_4_results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "Eu3hIpk1Qd5U",
   "metadata": {
    "id": "Eu3hIpk1Qd5U"
   },
   "source": [
    "#### Model 5 - Token embeddings + character embeddings + positional embeddings\n",
    "\n",
    "maybe use line numbers for embeddings for better training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "UZh7wP2ygQgr",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 483
    },
    "id": "UZh7wP2ygQgr",
    "outputId": "2bdf69f0-77c2-4533-8764-b3d8d9342f6e",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:28.247721287Z",
     "start_time": "2024-01-03T15:26:28.222346399Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "         target                                               text  \\\n0     OBJECTIVE  to investigate the efficacy of @ weeks of dail...   \n1       METHODS  a total of @ patients with primary knee oa wer...   \n2       METHODS  outcome measures included pain reduction and i...   \n3       METHODS  pain was assessed using the visual analog pain...   \n4       METHODS  secondary outcome measures included the wester...   \n5       METHODS  serum levels of interleukin @ ( il-@ ) , il-@ ...   \n6       RESULTS  there was a clinically relevant reduction in t...   \n7       RESULTS  the mean difference between treatment arms ( @...   \n8       RESULTS  further , there was a clinically relevant redu...   \n9       RESULTS  these differences remained significant at @ we...   \n10      RESULTS  the outcome measures in rheumatology clinical ...   \n11  CONCLUSIONS  low-dose oral prednisolone had both a short-te...   \n12   BACKGROUND  emotional eating is associated with overeating...   \n13   BACKGROUND  yet , empirical evidence for individual ( trai...   \n\n    line_number  total_lines  \n0             0           11  \n1             1           11  \n2             2           11  \n3             3           11  \n4             4           11  \n5             5           11  \n6             6           11  \n7             7           11  \n8             8           11  \n9             9           11  \n10           10           11  \n11           11           11  \n12            0           10  \n13            1           10  ",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>target</th>\n      <th>text</th>\n      <th>line_number</th>\n      <th>total_lines</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>OBJECTIVE</td>\n      <td>to investigate the efficacy of @ weeks of dail...</td>\n      <td>0</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>METHODS</td>\n      <td>a total of @ patients with primary knee oa wer...</td>\n      <td>1</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>METHODS</td>\n      <td>outcome measures included pain reduction and i...</td>\n      <td>2</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>METHODS</td>\n      <td>pain was assessed using the visual analog pain...</td>\n      <td>3</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>METHODS</td>\n      <td>secondary outcome measures included the wester...</td>\n      <td>4</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>METHODS</td>\n      <td>serum levels of interleukin @ ( il-@ ) , il-@ ...</td>\n      <td>5</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>RESULTS</td>\n      <td>there was a clinically relevant reduction in t...</td>\n      <td>6</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>RESULTS</td>\n      <td>the mean difference between treatment arms ( @...</td>\n      <td>7</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>RESULTS</td>\n      <td>further , there was a clinically relevant redu...</td>\n      <td>8</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>RESULTS</td>\n      <td>these differences remained significant at @ we...</td>\n      <td>9</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>10</th>\n      <td>RESULTS</td>\n      <td>the outcome measures in rheumatology clinical ...</td>\n      <td>10</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>11</th>\n      <td>CONCLUSIONS</td>\n      <td>low-dose oral prednisolone had both a short-te...</td>\n      <td>11</td>\n      <td>11</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>BACKGROUND</td>\n      <td>emotional eating is associated with overeating...</td>\n      <td>0</td>\n      <td>10</td>\n    </tr>\n    <tr>\n      <th>13</th>\n      <td>BACKGROUND</td>\n      <td>yet , empirical evidence for individual ( trai...</td>\n      <td>1</td>\n      <td>10</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.head(14)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9Mxaw_mhVdUL",
   "metadata": {
    "id": "9Mxaw_mhVdUL"
   },
   "source": [
    "Any engineered features usd to train the model need to be available at test time, in our case, line number and total number of lines needs to be available"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "0BTG33kzRl1G",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "0BTG33kzRl1G",
    "outputId": "8a1db4c3-90bb-488e-e43a-80a08558d8f2",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:28.250760551Z",
     "start_time": "2024-01-03T15:26:28.222813812Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "line_number\n0     15000\n1     15000\n2     15000\n3     15000\n4     14992\n5     14949\n6     14758\n7     14279\n8     13346\n9     11981\n10    10041\n11     7892\n12     5853\n13     4152\n14     2835\n15     1861\n16     1188\n17      751\n18      462\n19      286\n20      162\n21      101\n22       66\n23       33\n24       22\n25       14\n26        7\n27        4\n28        3\n29        1\n30        1\nName: count, dtype: int64"
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# How many different line numbers are there\n",
    "train_df.line_number.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "_0LiguLoXlXH",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 447
    },
    "id": "_0LiguLoXlXH",
    "outputId": "95d682da-b894-4558-a1fe-b7fda1cbc809",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:28.591645101Z",
     "start_time": "2024-01-03T15:26:28.223036084Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<Axes: ylabel='Frequency'>"
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# check distribution of line_numbers column\n",
    "train_df.line_number.plot.hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "aYHowe43kLMz",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "aYHowe43kLMz",
    "outputId": "30355de6-62f1-4df5-8b97-e371fe209c25",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:28.598863336Z",
     "start_time": "2024-01-03T15:26:28.434460736Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<tf.Tensor: shape=(10, 15), dtype=float32, numpy=\narray([[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)>"
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use tensorflow to create one hot encoded tensors of line_numbers\n",
    "train_line_numbers_one_hot = tf.one_hot(train_df.line_number.to_numpy(), depth=15)\n",
    "val_line_numbers_one_hot = tf.one_hot(val_df.line_number.to_numpy(), depth=15)\n",
    "test_line_numbers_one_hot = tf.one_hot(test_df.line_number.to_numpy(), depth=15)\n",
    "train_line_numbers_one_hot[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "wxvkX5tmkgza",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 924
    },
    "id": "wxvkX5tmkgza",
    "outputId": "932f46ca-b499-4ae8-da04-53badea01039",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:28.719790258Z",
     "start_time": "2024-01-03T15:26:28.434938953Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "(total_lines\n 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: count, dtype: int64,\n 18.0)"
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": "<Figure size 640x480 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_df.total_lines.plot.hist()\n",
    "train_df.total_lines.value_counts(), np.percentile(train_df.total_lines, 95)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "_baWhp5ilCxl",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "_baWhp5ilCxl",
    "outputId": "092d898a-9e1f-41e4-d318-f4264deff55f",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:28.722876247Z",
     "start_time": "2024-01-03T15:26:28.614493115Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<tf.Tensor: shape=(10, 20), dtype=float32, numpy=\narray([[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)>"
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use tensorflow to one hot encode tensors\n",
    "train_total_line_one_hot = tf.one_hot(train_df.total_lines.to_numpy(), depth=20)\n",
    "val_total_line_one_hot = tf.one_hot(val_df.total_lines.to_numpy(), depth=20)\n",
    "test_total_line_one_hot = tf.one_hot(test_df.total_lines.to_numpy(), depth=20)\n",
    "train_total_line_one_hot[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bK_PMIZOmn3X",
   "metadata": {
    "id": "bK_PMIZOmn3X"
   },
   "source": [
    "#### build model 5 - tribrid embedding model\n",
    "1. Build a token level model\n",
    "2. Build a character level model\n",
    "3. Build a model for \"line_numbers\" feature\n",
    "4. Build a model for \"total_lines\" feature\n",
    "5. Concatenate mdoel 1 & 2\n",
    "6. Concatenate model 3, 4 & 5\n",
    "7. Create output layer that accepts tribed input and gives output label probabilities\n",
    "8. Combine the inputs of 1, 2, 3 & 4 and outputs of 7 into a model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "58TlK9_ml-bR",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "58TlK9_ml-bR",
    "outputId": "c403530e-2615-484c-c9f9-f07e4847c30a",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:30.166829089Z",
     "start_time": "2024-01-03T15:26:28.614955648Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"tribid_model\"\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",
      " text_vectorization_1 (Text  (None, 290)                  0         ['char_input[0][0]']          \n",
      " Vectorization)                                                                                   \n",
      "                                                                                                  \n",
      " keras_layer (KerasLayer)    (None, 512)                  2567978   ['token_input[0][0]']         \n",
      "                                                          24                                      \n",
      "                                                                                                  \n",
      " embedding_1 (Embedding)     (None, 290, 25)              700       ['text_vectorization_1[2][0]']\n",
      "                                                                                                  \n",
      " token_output (Dense)        (None, 128)                  65664     ['keras_layer[2][0]']         \n",
      "                                                                                                  \n",
      " char_bi_lstm (Bidirectiona  (None, 48)                   9600      ['embedding_1[2][0]']         \n",
      " l)                                                                                               \n",
      "                                                                                                  \n",
      " token_char_concat (Concate  (None, 176)                  0         ['token_output[0][0]',        \n",
      " nate)                                                               'char_bi_lstm[0][0]']        \n",
      "                                                                                                  \n",
      " line_number_inputs (InputL  [(None, 15)]                 0         []                            \n",
      " ayer)                                                                                            \n",
      "                                                                                                  \n",
      " total_lines_input (InputLa  [(None, 20)]                 0         []                            \n",
      " yer)                                                                                             \n",
      "                                                                                                  \n",
      " dense_6 (Dense)             (None, 256)                  45312     ['token_char_concat[0][0]']   \n",
      "                                                                                                  \n",
      " line_number_dense (Dense)   (None, 32)                   512       ['line_number_inputs[0][0]']  \n",
      "                                                                                                  \n",
      " total_lines_dense (Dense)   (None, 32)                   672       ['total_lines_input[0][0]']   \n",
      "                                                                                                  \n",
      " dropout_2 (Dropout)         (None, 256)                  0         ['dense_6[0][0]']             \n",
      "                                                                                                  \n",
      " all_inputs_concatenate (Co  (None, 320)                  0         ['line_number_dense[0][0]',   \n",
      " ncatenate)                                                          'total_lines_dense[0][0]',   \n",
      "                                                                     'dropout_2[0][0]']           \n",
      "                                                                                                  \n",
      " dense_7 (Dense)             (None, 5)                    1605      ['all_inputs_concatenate[0][0]\n",
      "                                                                    ']                            \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 256921889 (980.08 MB)\n",
      "Trainable params: 124065 (484.63 KB)\n",
      "Non-trainable params: 256797824 (979.61 MB)\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# setup token level 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=tf.keras.activations.relu, name='token_output')(token_embeddings)\n",
    "token_model = tf.keras.Model(inputs = token_inputs,\n",
    "                             outputs = token_outputs)\n",
    "\n",
    "# setup character level model\n",
    "char_inputs = tf.keras.layers.Input(shape=(1,), dtype=tf.string, name='char_input')\n",
    "char_vector = char_vectorizer(char_inputs)\n",
    "char_embeddings = char_embed(char_vector)\n",
    "char_bi_lstm = tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(24), name='char_bi_lstm')(char_embeddings)\n",
    "char_model = tf.keras.Model(inputs = char_inputs,\n",
    "                            outputs = char_bi_lstm)\n",
    "\n",
    "#setup model for line_number feature\n",
    "line_number_input = tf.keras.layers.Input(shape=(15, ), dtype=tf.float32, name='line_number_inputs')\n",
    "line_number_dense = tf.keras.layers.Dense(32, activation=tf.keras.activations.relu, name='line_number_dense')(line_number_input)\n",
    "line_number_model = tf.keras.Model(inputs = line_number_input,\n",
    "                                   outputs = line_number_dense)\n",
    "\n",
    "#setup model for total_lines feature\n",
    "total_lines_input = tf.keras.layers.Input(shape=(20, ), dtype=tf.float32, name='total_lines_input')\n",
    "total_lines_dense = tf.keras.layers.Dense(32, activation=tf.keras.activations.relu, name='total_lines_dense')(total_lines_input)\n",
    "total_lines_model = tf.keras.Model(inputs=total_lines_input,\n",
    "                                  outputs = total_lines_dense)\n",
    "\n",
    "#concatenate token and char embeddings\n",
    "token_char_concat = tf.keras.layers.Concatenate(name='token_char_concat')([token_model.output,\n",
    "                                                                           char_model.output])\n",
    "token_char_dense = tf.keras.layers.Dense(256, activation=tf.keras.activations.relu)(token_char_concat)\n",
    "token_char_dropout = tf.keras.layers.Dropout(0.5)(token_char_dense)\n",
    "\n",
    "#concatenate previous concatenated layer with features\n",
    "all_concat = tf.keras.layers.Concatenate(name='all_inputs_concatenate')([line_number_model.output, total_lines_model.output, token_char_dropout])\n",
    "\n",
    "# create output layer that accepts tribed input\n",
    "output_layer = tf.keras.layers.Dense(5, activation=tf.keras.activations.softmax)(all_concat)\n",
    "\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,\n",
    "                             name='tribid_model')\n",
    "model_5.compile(loss=tf.keras.losses.CategoricalCrossentropy(label_smoothing=0.2), # reduces overfitting in label probability predictions\n",
    "                optimizer=tf.keras.optimizers.Adam(),\n",
    "                metrics=['accuracy'])\n",
    "\n",
    "model_5.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "7H7m1A0JrZKl",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 982
    },
    "id": "7H7m1A0JrZKl",
    "outputId": "40e184ec-de84-410d-e295-27fd2cfdf970",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:30.396623893Z",
     "start_time": "2024-01-03T15:26:30.102732915Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": "<IPython.core.display.Image object>"
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.utils import plot_model\n",
    "plot_model(model_5, show_shapes=True, show_dtype=True, expand_nested=True, show_layer_activations=True, dpi=75)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "wTKUqq5QtDOT",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "wTKUqq5QtDOT",
    "outputId": "51aa8039-09d4-4361-fa5d-5a541f420c70",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:26:32.164575793Z",
     "start_time": "2024-01-03T15:26:30.426531667Z"
    }
   },
   "outputs": [
    {
     "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))>"
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create data API for multi inputs\n",
    "train_all_dataset_x = tf.data.Dataset.from_tensor_slices((train_line_numbers_one_hot, train_total_line_one_hot, train_sentences, train_chars))\n",
    "train_all_dataset_y = tf.data.Dataset.from_tensor_slices(train_labels_one_hot)\n",
    "train_all_dataset = tf.data.Dataset.zip((train_all_dataset_x, train_all_dataset_y)).batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "\n",
    "val_all_dataset_x = tf.data.Dataset.from_tensor_slices((val_line_numbers_one_hot, val_total_line_one_hot, val_sentences, val_chars))\n",
    "val_all_dataset_y = tf.data.Dataset.from_tensor_slices(val_labels_one_hot)\n",
    "val_all_dataset = tf.data.Dataset.zip((val_all_dataset_x, val_all_dataset_y)).batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "\n",
    "test_all_dataset_x = tf.data.Dataset.from_tensor_slices((test_line_numbers_one_hot, test_total_line_one_hot, test_sentences, test_chars))\n",
    "test_all_dataset_y = tf.data.Dataset.from_tensor_slices(test_labels_one_hot)\n",
    "test_all_dataset = tf.data.Dataset.zip((test_all_dataset_x, test_all_dataset_y)).batch(64).prefetch(tf.data.AUTOTUNE)\n",
    "\n",
    "train_all_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "J_CC5jGIxSGp",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "J_CC5jGIxSGp",
    "outputId": "25c24cf6-beff-4945-e184-31d62bd3477f",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:14.149234566Z",
     "start_time": "2024-01-03T15:26:32.165311863Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "281/281 [==============================] - 15s 29ms/step - loss: 1.1451 - accuracy: 0.6903 - val_loss: 1.0103 - val_accuracy: 0.7882\n",
      "Epoch 2/5\n",
      "281/281 [==============================] - 8s 27ms/step - loss: 0.9818 - accuracy: 0.8055 - val_loss: 0.9625 - val_accuracy: 0.8178\n",
      "Epoch 3/5\n",
      "281/281 [==============================] - 6s 22ms/step - loss: 0.9591 - accuracy: 0.8181 - val_loss: 0.9469 - val_accuracy: 0.8278\n",
      "Epoch 4/5\n",
      "281/281 [==============================] - 7s 26ms/step - loss: 0.9461 - accuracy: 0.8306 - val_loss: 0.9374 - val_accuracy: 0.8311\n",
      "Epoch 5/5\n",
      "281/281 [==============================] - 6s 22ms/step - loss: 0.9420 - accuracy: 0.8320 - val_loss: 0.9296 - val_accuracy: 0.8414\n"
     ]
    }
   ],
   "source": [
    "# Fit the model\n",
    "history_5 = model_5.fit(train_all_dataset,\n",
    "                        epochs=5,\n",
    "                        validation_data=val_all_dataset,\n",
    "                        steps_per_epoch=int(0.1*len(train_all_dataset)),\n",
    "                        validation_steps=int(0.1*len(val_all_dataset)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "E5qda-9vzh-Z",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "E5qda-9vzh-Z",
    "outputId": "81276e41-db13-458d-e1e1-f8b2fcd33f61",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:19.597691197Z",
     "start_time": "2024-01-03T15:27:14.157891858Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 5s 8ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "{'accuracy': 83.8706474248643,\n 'precision': 0.8431532947654939,\n 'recall': 0.838706474248643,\n 'f1': 0.8343531343947536}"
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# make predictions\n",
    "model_5_preds = tf.argmax(model_5.predict(val_all_dataset), axis=1).numpy()\n",
    "model_5_results = calculate_results(val_labels_encoded, model_5_preds)\n",
    "model_5_results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "FShqw_mX41kU",
   "metadata": {
    "id": "FShqw_mX41kU"
   },
   "source": [
    "#### Compare model results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "VV90f4Ql0PRT",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 235
    },
    "id": "VV90f4Ql0PRT",
    "outputId": "91f108ab-9cdc-4b87-f62a-1c849a7060d0",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:19.601245847Z",
     "start_time": "2024-01-03T15:27:19.594382833Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "                                                accuracy  precision    recall  \\\nmodel_0_baseline                                0.721832   0.718647  0.721832   \nmodel_1_custom_token_embedding                  0.796505   0.795102  0.796505   \nmodel_2_pretrained_token_embedding              0.725209   0.720506  0.725209   \nmodel_3_char_embedding                          0.703859   0.693403  0.703859   \nmodel_4_hybrid_token_char_embedding             0.744141   0.740743  0.744141   \nmodel_5_hybrid_token_char_positional_embedding  0.838706   0.843153  0.838706   \n\n                                                      f1  \nmodel_0_baseline                                0.698925  \nmodel_1_custom_token_embedding                  0.792711  \nmodel_2_pretrained_token_embedding              0.719541  \nmodel_3_char_embedding                          0.694728  \nmodel_4_hybrid_token_char_embedding             0.738893  \nmodel_5_hybrid_token_char_positional_embedding  0.834353  ",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>accuracy</th>\n      <th>precision</th>\n      <th>recall</th>\n      <th>f1</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>model_0_baseline</th>\n      <td>0.721832</td>\n      <td>0.718647</td>\n      <td>0.721832</td>\n      <td>0.698925</td>\n    </tr>\n    <tr>\n      <th>model_1_custom_token_embedding</th>\n      <td>0.796505</td>\n      <td>0.795102</td>\n      <td>0.796505</td>\n      <td>0.792711</td>\n    </tr>\n    <tr>\n      <th>model_2_pretrained_token_embedding</th>\n      <td>0.725209</td>\n      <td>0.720506</td>\n      <td>0.725209</td>\n      <td>0.719541</td>\n    </tr>\n    <tr>\n      <th>model_3_char_embedding</th>\n      <td>0.703859</td>\n      <td>0.693403</td>\n      <td>0.703859</td>\n      <td>0.694728</td>\n    </tr>\n    <tr>\n      <th>model_4_hybrid_token_char_embedding</th>\n      <td>0.744141</td>\n      <td>0.740743</td>\n      <td>0.744141</td>\n      <td>0.738893</td>\n    </tr>\n    <tr>\n      <th>model_5_hybrid_token_char_positional_embedding</th>\n      <td>0.838706</td>\n      <td>0.843153</td>\n      <td>0.838706</td>\n      <td>0.834353</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Combine model results into data frame\n",
    "all_model_results = pd.DataFrame({\"model_0_baseline\":model_0_results,\n",
    "                                  \"model_1_custom_token_embedding\":model_1_results,\n",
    "                                  \"model_2_pretrained_token_embedding\":model_2_results,\n",
    "                                  \"model_3_char_embedding\":model_3_results,\n",
    "                                  \"model_4_hybrid_token_char_embedding\":model_4_results,\n",
    "                                  \"model_5_hybrid_token_char_positional_embedding\":model_5_results})\n",
    "all_model_results = all_model_results.transpose()\n",
    "all_model_results.accuracy = all_model_results.accuracy/100\n",
    "all_model_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "Vx4IbKl45myZ",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 956
    },
    "id": "Vx4IbKl45myZ",
    "outputId": "8feab1b1-2b79-4db5-df9e-c148d81f49ac",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:19.958410386Z",
     "start_time": "2024-01-03T15:27:19.594784549Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<matplotlib.legend.Legend at 0x7fad1369c5e0>"
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": "<Figure size 1000x700 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "all_model_results.plot(kind='bar', figsize=(10,7)).legend(bbox_to_anchor=(1.0, 1.0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "1e0B9ag76SPG",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 956
    },
    "id": "1e0B9ag76SPG",
    "outputId": "56b0ae4a-dcfb-4a89-f6ad-9e3e5adabb2f",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:20.016783047Z",
     "start_time": "2024-01-03T15:27:19.874656111Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<matplotlib.legend.Legend at 0x7fad133aa310>"
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": "<Figure size 1000x700 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# get f1 score\n",
    "all_model_results.sort_values(\"f1\", ascending=True).f1.plot(kind='bar', figsize=(10,7)).legend(bbox_to_anchor=(1.0, 1.0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "AJc_gQYS6k43",
   "metadata": {
    "id": "AJc_gQYS6k43",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:35.325584788Z",
     "start_time": "2024-01-03T15:27:19.989404585Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: skimlit_tribrid_model_10percent/assets\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:tensorflow:Assets written to: skimlit_tribrid_model_10percent/assets\n"
     ]
    }
   ],
   "source": [
    "# save the model\n",
    "model_5.save(\"skimlit_tribrid_model_10percent\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "F2PEG8Ok8YLs",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "F2PEG8Ok8YLs",
    "outputId": "bcd5346c-c9ac-4884-8f80-f225c01d24da",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:43.628952178Z",
     "start_time": "2024-01-03T15:27:35.329379833Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"tribid_model\"\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",
      " text_vectorization_1 (Text  (None, 290)                  0         ['char_input[0][0]']          \n",
      " Vectorization)                                                                                   \n",
      "                                                                                                  \n",
      " keras_layer (KerasLayer)    (None, 512)                  2567978   ['token_input[0][0]']         \n",
      "                                                          24                                      \n",
      "                                                                                                  \n",
      " embedding_1 (Embedding)     (None, 290, 25)              700       ['text_vectorization_1[0][0]']\n",
      "                                                                                                  \n",
      " token_output (Dense)        (None, 128)                  65664     ['keras_layer[0][0]']         \n",
      "                                                                                                  \n",
      " char_bi_lstm (Bidirectiona  (None, 48)                   9600      ['embedding_1[0][0]']         \n",
      " l)                                                                                               \n",
      "                                                                                                  \n",
      " token_char_concat (Concate  (None, 176)                  0         ['token_output[0][0]',        \n",
      " nate)                                                               'char_bi_lstm[0][0]']        \n",
      "                                                                                                  \n",
      " line_number_inputs (InputL  [(None, 15)]                 0         []                            \n",
      " ayer)                                                                                            \n",
      "                                                                                                  \n",
      " total_lines_input (InputLa  [(None, 20)]                 0         []                            \n",
      " yer)                                                                                             \n",
      "                                                                                                  \n",
      " dense_6 (Dense)             (None, 256)                  45312     ['token_char_concat[0][0]']   \n",
      "                                                                                                  \n",
      " line_number_dense (Dense)   (None, 32)                   512       ['line_number_inputs[0][0]']  \n",
      "                                                                                                  \n",
      " total_lines_dense (Dense)   (None, 32)                   672       ['total_lines_input[0][0]']   \n",
      "                                                                                                  \n",
      " dropout_2 (Dropout)         (None, 256)                  0         ['dense_6[0][0]']             \n",
      "                                                                                                  \n",
      " all_inputs_concatenate (Co  (None, 320)                  0         ['line_number_dense[0][0]',   \n",
      " ncatenate)                                                          'total_lines_dense[0][0]',   \n",
      "                                                                     'dropout_2[0][0]']           \n",
      "                                                                                                  \n",
      " dense_7 (Dense)             (None, 5)                    1605      ['all_inputs_concatenate[0][0]\n",
      "                                                                    ']                            \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 256921889 (980.08 MB)\n",
      "Trainable params: 124065 (484.63 KB)\n",
      "Non-trainable params: 256797824 (979.61 MB)\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "loaded_model = tf.keras.models.load_model(\"skimlit_tribrid_model_10percent\")\n",
    "loaded_model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "WNjxQksX8e2a",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "WNjxQksX8e2a",
    "outputId": "22c47b1d-f842-4bd9-a06b-61c4c3d10cea",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:49.220984953Z",
     "start_time": "2024-01-03T15:27:43.590856990Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 6s 8ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "{'accuracy': 83.8706474248643,\n 'precision': 0.8431532947654939,\n 'recall': 0.838706474248643,\n 'f1': 0.8343531343947536}"
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "loaded_model_preds = tf.argmax(loaded_model.predict(val_all_dataset), axis=1).numpy()\n",
    "loaded_model_result = calculate_results(val_labels_encoded, loaded_model_preds)\n",
    "loaded_model_result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "6em0MZrK80ad",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "6em0MZrK80ad",
    "outputId": "8b2bb5a6-fd24-454c-cdc5-1e581eb958ba",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:27:49.223434850Z",
     "start_time": "2024-01-03T15:27:49.218519623Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "{'accuracy': 83.8706474248643,\n 'precision': 0.8431532947654939,\n 'recall': 0.838706474248643,\n 'f1': 0.8343531343947536}"
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_5_results"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "u5w41hTq_yh0",
   "metadata": {
    "id": "u5w41hTq_yh0"
   },
   "source": [
    "## Exercise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "GBOsmy7f9-o_",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "GBOsmy7f9-o_",
    "outputId": "0da46338-525a-4501-83db-6abf78f2861f",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:36:48.288831416Z",
     "start_time": "2024-01-03T15:27:49.218998157Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n",
      "2814/2814 [==============================] - ETA: 0s - loss: 0.9630 - accuracy: 0.8168\n",
      "Epoch 1: val_loss improved from inf to 0.91511, saving model to model_5_entire/\n",
      "2814/2814 [==============================] - 71s 23ms/step - loss: 0.9630 - accuracy: 0.8168 - val_loss: 0.9151 - val_accuracy: 0.8434\n",
      "Epoch 2/100\n",
      "2812/2814 [============================>.] - ETA: 0s - loss: 0.9101 - accuracy: 0.8524\n",
      "Epoch 2: val_loss improved from 0.91511 to 0.90311, saving model to model_5_entire/\n",
      "2814/2814 [==============================] - 66s 23ms/step - loss: 0.9101 - accuracy: 0.8524 - val_loss: 0.9031 - val_accuracy: 0.8491\n",
      "Epoch 3/100\n",
      "2812/2814 [============================>.] - ETA: 0s - loss: 0.8965 - accuracy: 0.8624\n",
      "Epoch 3: val_loss improved from 0.90311 to 0.89678, saving model to model_5_entire/\n",
      "2814/2814 [==============================] - 64s 23ms/step - loss: 0.8966 - accuracy: 0.8623 - val_loss: 0.8968 - val_accuracy: 0.8547\n",
      "Epoch 4/100\n",
      "2812/2814 [============================>.] - ETA: 0s - loss: 0.8871 - accuracy: 0.8693\n",
      "Epoch 4: val_loss improved from 0.89678 to 0.89419, saving model to model_5_entire/\n",
      "2814/2814 [==============================] - 67s 24ms/step - loss: 0.8871 - accuracy: 0.8693 - val_loss: 0.8942 - val_accuracy: 0.8587\n",
      "Epoch 5/100\n",
      "2812/2814 [============================>.] - ETA: 0s - loss: 0.8804 - accuracy: 0.8743\n",
      "Epoch 5: val_loss improved from 0.89419 to 0.89227, saving model to model_5_entire/\n",
      "2814/2814 [==============================] - 68s 24ms/step - loss: 0.8804 - accuracy: 0.8742 - val_loss: 0.8923 - val_accuracy: 0.8594\n",
      "Epoch 6/100\n",
      "2812/2814 [============================>.] - ETA: 0s - loss: 0.8746 - accuracy: 0.8786\n",
      "Epoch 6: val_loss improved from 0.89227 to 0.89226, saving model to model_5_entire/\n",
      "2814/2814 [==============================] - 67s 24ms/step - loss: 0.8747 - accuracy: 0.8785 - val_loss: 0.8923 - val_accuracy: 0.8600\n",
      "Epoch 7/100\n",
      "2812/2814 [============================>.] - ETA: 0s - loss: 0.8697 - accuracy: 0.8823\n",
      "Epoch 7: val_loss improved from 0.89226 to 0.89222, saving model to model_5_entire/\n",
      "2814/2814 [==============================] - 67s 24ms/step - loss: 0.8697 - accuracy: 0.8823 - val_loss: 0.8922 - val_accuracy: 0.8617\n",
      "Epoch 8/100\n",
      "2812/2814 [============================>.] - ETA: 0s - loss: 0.8653 - accuracy: 0.8861\n",
      "Epoch 8: val_loss improved from 0.89222 to 0.89150, saving model to model_5_entire/\n",
      "Restoring model weights from the end of the best epoch: 5.\n",
      "2814/2814 [==============================] - 68s 24ms/step - loss: 0.8653 - accuracy: 0.8860 - val_loss: 0.8915 - val_accuracy: 0.8600\n",
      "Epoch 8: early stopping\n"
     ]
    }
   ],
   "source": [
    "# 1. Train model_5 on all of the data in the training dataset for as many epochs until it stops improving.\n",
    "#lets compile the model again(this time with some callbacks)\n",
    "\n",
    "#callbacks\n",
    "model_checkpoint = tf.keras.callbacks.ModelCheckpoint(filepath='model_5_entire/',\n",
    "                                                      verbose=1,\n",
    "                                                      save_best_only=True,\n",
    "                                                      save_weights_only=True)\n",
    "early_stopping = tf.keras.callbacks.EarlyStopping(patience=3,\n",
    "                                                  verbose=1,\n",
    "                                                  restore_best_weights=True,\n",
    "                                                  min_delta=0.001)\n",
    "\n",
    "# setup token level 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=tf.keras.activations.relu, name='token_output')(token_embeddings)\n",
    "token_model = tf.keras.Model(inputs = token_inputs,\n",
    "                             outputs = token_outputs)\n",
    "\n",
    "# setup character level model\n",
    "char_inputs = tf.keras.layers.Input(shape=(1,), dtype=tf.string, name='char_input')\n",
    "char_vector = char_vectorizer(char_inputs)\n",
    "char_embeddings = char_embed(char_vector)\n",
    "char_bi_lstm = tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(24), name='char_bi_lstm')(char_embeddings)\n",
    "char_model = tf.keras.Model(inputs = char_inputs,\n",
    "                            outputs = char_bi_lstm)\n",
    "\n",
    "#setup model for line_number feature\n",
    "line_number_input = tf.keras.layers.Input(shape=(15, ), dtype=tf.float32, name='line_number_inputs')\n",
    "line_number_dense = tf.keras.layers.Dense(32, activation=tf.keras.activations.relu, name='line_number_dense')(line_number_input)\n",
    "line_number_model = tf.keras.Model(inputs = line_number_input,\n",
    "                                   outputs = line_number_dense)\n",
    "\n",
    "#setup model for total_lines feature\n",
    "total_lines_input = tf.keras.layers.Input(shape=(20, ), dtype=tf.float32, name='total_lines_input')\n",
    "total_lines_dense = tf.keras.layers.Dense(32, activation=tf.keras.activations.relu, name='total_lines_dense')(total_lines_input)\n",
    "total_lines_model = tf.keras.Model(inputs=total_lines_input,\n",
    "                                  outputs = total_lines_dense)\n",
    "\n",
    "#concatenate token and char embeddings\n",
    "token_char_concat = tf.keras.layers.Concatenate(name='token_char_concat')([token_model.output,\n",
    "                                                                           char_model.output])\n",
    "token_char_dense = tf.keras.layers.Dense(256, activation=tf.keras.activations.relu)(token_char_concat)\n",
    "token_char_dropout = tf.keras.layers.Dropout(0.5)(token_char_dense)\n",
    "\n",
    "#concatenate previous concatenated layer with features\n",
    "all_concat = tf.keras.layers.Concatenate(name='all_inputs_concatenate')([line_number_model.output, total_lines_model.output, token_char_dropout])\n",
    "\n",
    "# create output layer that accepts triple input\n",
    "output_layer = tf.keras.layers.Dense(5, activation=tf.keras.activations.softmax)(all_concat)\n",
    "\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,\n",
    "                             name='tribid_model')\n",
    "model_5.compile(loss=tf.keras.losses.CategoricalCrossentropy(label_smoothing=0.2), # reduces overfitting in label probability predictions\n",
    "                optimizer=tf.keras.optimizers.Adam(),\n",
    "                metrics=['accuracy'])\n",
    "\n",
    "history_5_exercise = model_5.fit(train_all_dataset,\n",
    "                                 epochs=100,\n",
    "                                 validation_data=val_all_dataset,\n",
    "                                 validation_steps=int(0.1*len(val_all_dataset)),\n",
    "                                 callbacks=[model_checkpoint, early_stopping])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "3tUCckQeVHwf",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "3tUCckQeVHwf",
    "outputId": "cc8a3c5f-317c-4a42-f1b7-9f86e113c175",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:36:54.736986470Z",
     "start_time": "2024-01-03T15:36:48.291572530Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 6s 9ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "{'accuracy': 85.86985303852774,\n 'precision': 0.8602748646717919,\n 'recall': 0.8586985303852773,\n 'f1': 0.8560176717765344}"
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#make predictions and compare\n",
    "model_5_exercise_preds = tf.argmax(model_5.predict(val_all_dataset), axis=1).numpy()\n",
    "model_5_exercise_results = calculate_results(val_labels_encoded, model_5_exercise_preds)\n",
    "model_5_exercise_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "Snk9l1jPgOCw",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Snk9l1jPgOCw",
    "outputId": "942b12ab-ea17-4e08-c4fc-94aa4a2a7f96",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:36:58.377099741Z",
     "start_time": "2024-01-03T15:36:54.734566938Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 400000 word vectors.\n"
     ]
    }
   ],
   "source": [
    "#2. Checkout the Keras guide on using pretrained GloVe embeddings. Can you get this working with one of our models?\n",
    "if not os.path.exists(\"glove.6B.zip\"):\n",
    "    !wget https://downloads.cs.stanford.edu/nlp/data/glove.6B.zip\n",
    "    !unzip -q glove.6B.zip\n",
    "\n",
    "path_to_glove_file = \"glove.6B.100d.txt\"\n",
    "\n",
    "embeddings_index = {}\n",
    "with open(path_to_glove_file) as f:\n",
    "    for line in f:\n",
    "        word, coefs = line.split(maxsplit=1)\n",
    "        coefs = np.fromstring(coefs, \"f\", sep=\" \")\n",
    "        embeddings_index[word] = coefs\n",
    "\n",
    "print(\"Found %s word vectors.\" % len(embeddings_index))\n",
    "voc = char_vectorizer.get_vocabulary()\n",
    "word_index = dict(zip(voc, range(len(voc))))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "fYu85DOPmiWl",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "fYu85DOPmiWl",
    "outputId": "f4bf79fe-5760-4613-f4e2-b5508346ac0c",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:36:58.391150995Z",
     "start_time": "2024-01-03T15:36:58.382607420Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Converted 26 words (2 misses)\n"
     ]
    }
   ],
   "source": [
    "num_tokens = len(voc) + 2\n",
    "embedding_dim = 100\n",
    "hits = 0\n",
    "misses = 0\n",
    "\n",
    "# Prepare embedding matrix\n",
    "embedding_matrix = np.zeros((num_tokens, embedding_dim))\n",
    "for word, i in word_index.items():\n",
    "    embedding_vector = embeddings_index.get(word)\n",
    "    if embedding_vector is not None:\n",
    "        # Words not found in embedding index will be all-zeros.\n",
    "        # This includes the representation for \"padding\" and \"OOV\"\n",
    "        embedding_matrix[i] = embedding_vector\n",
    "        hits += 1\n",
    "    else:\n",
    "        misses += 1\n",
    "print(\"Converted %d words (%d misses)\" % (hits, misses))\n",
    "\n",
    "embedding_layer = tf.keras.layers.Embedding(\n",
    "    num_tokens,\n",
    "    embedding_dim,\n",
    "    trainable=False,\n",
    ")\n",
    "embedding_layer.build((1,))\n",
    "embedding_layer.set_weights([embedding_matrix])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "DbeQ7W7NnZIw",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "DbeQ7W7NnZIw",
    "outputId": "638dc5e1-a636-43e7-f0e8-3480dceaabeb",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:06.510330908Z",
     "start_time": "2024-01-03T15:36:58.399635555Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/5\n",
      "281/281 [==============================] - 3s 6ms/step - loss: 1.1557 - accuracy: 0.5522 - val_loss: 0.9646 - val_accuracy: 0.6210\n",
      "Epoch 2/5\n",
      "281/281 [==============================] - 1s 5ms/step - loss: 0.9314 - accuracy: 0.6356 - val_loss: 0.8441 - val_accuracy: 0.6769\n",
      "Epoch 3/5\n",
      "281/281 [==============================] - 1s 5ms/step - loss: 0.8802 - accuracy: 0.6573 - val_loss: 0.8282 - val_accuracy: 0.6872\n",
      "Epoch 4/5\n",
      "281/281 [==============================] - 1s 5ms/step - loss: 0.8396 - accuracy: 0.6763 - val_loss: 0.7797 - val_accuracy: 0.7074\n",
      "Epoch 5/5\n",
      "281/281 [==============================] - 1s 5ms/step - loss: 0.8125 - accuracy: 0.6872 - val_loss: 0.7796 - val_accuracy: 0.7084\n"
     ]
    }
   ],
   "source": [
    "inputs_exercise = tf.keras.Input(shape=(1,), dtype=tf.string)\n",
    "x = char_vectorizer(inputs_exercise)\n",
    "x = embedding_layer(x)\n",
    "x = tf.keras.layers.Conv1D(filters=128, kernel_size=5)(x)\n",
    "x = tf.keras.layers.GlobalMaxPool1D()(x)\n",
    "outputs_exercise = tf.keras.layers.Dense(5, activation=tf.keras.activations.softmax)(x)\n",
    "\n",
    "model_exercise_2 =tf.keras.Model(inputs_exercise, outputs_exercise)\n",
    "model_exercise_2.compile(loss=tf.keras.losses.CategoricalCrossentropy(),\n",
    "               optimizer=tf.keras.optimizers.Adam(),\n",
    "               metrics=['accuracy'])\n",
    "history_exercise_2 = model_exercise_2.fit(train_char_dataset,\n",
    "                        epochs=5,\n",
    "                        steps_per_epoch=int(0.1*len(train_char_dataset)),\n",
    "                        validation_data=val_char_dataset,\n",
    "                        validation_steps=int(0.1*len(val_char_dataset)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "kh7g0Y-1n1lw",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "kh7g0Y-1n1lw",
    "outputId": "3d5f5e63-4634-4284-a07c-54dabd0697b0",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:07.735372495Z",
     "start_time": "2024-01-03T15:37:06.510263188Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "473/473 [==============================] - 1s 2ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "{'accuracy': 69.89937773070302,\n 'precision': 0.7154760731216794,\n 'recall': 0.6989937773070303,\n 'f1': 0.7024797137785422}"
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_exercise_2_preds = tf.argmax(model_exercise_2.predict(val_char_dataset), axis=1).numpy()\n",
    "model_exercise_2_results = calculate_results(val_labels_encoded, model_exercise_2_preds)\n",
    "model_exercise_2_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "q0sOf7A1ok7b",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "q0sOf7A1ok7b",
    "outputId": "babe227b-4b93-47f9-d27b-a46a257ffbb7",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:07.737809464Z",
     "start_time": "2024-01-03T15:37:07.734281181Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "{'accuracy': 70.38593936184297,\n 'precision': 0.693403307249158,\n 'recall': 0.7038593936184298,\n 'f1': 0.6947280052718605}"
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_3_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "bcGh6QXqondr",
   "metadata": {
    "id": "bcGh6QXqondr",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:07.746483404Z",
     "start_time": "2024-01-03T15:37:07.734486360Z"
    }
   },
   "outputs": [],
   "source": [
    "#3. Try replacing the TensorFlow Hub Universal Sentence Encoder pretrained embedding for the TensorFlow Hub\n",
    "# BERT PubMed expert (a language model pretrained on PubMed texts) pretrained embedding. Does this effect results?\n",
    "# !pip install --quiet tensorflow-text\n",
    "# tf.compat.v1.register_op('CaseFoldUTF8')\n",
    "# preprocess = hub.load('https://kaggle.com/models/tensorflow/bert/frameworks/TensorFlow2/variations/en-uncased-preprocess/versions/3')\n",
    "# bert = hub.load('https://www.kaggle.com/models/google/experts-bert/frameworks/TensorFlow2/variations/pubmed/versions/2')\n",
    "# bert_inputs = preprocess(train_sentences)\n",
    "# bert_outputs = bert(bert_inputs, training=False)\n",
    "# pooled_output = bert_outputs['pooled_output']\n",
    "# sequence_output = bert_outputs['sequence_output']\n",
    "# sequence_outputs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "rCQ3tVx8pwSh",
   "metadata": {
    "id": "rCQ3tVx8pwSh",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:07.748916647Z",
     "start_time": "2024-01-03T15:37:07.734554589Z"
    }
   },
   "outputs": [],
   "source": [
    "#4. What happens if you were to merge our line_number and total_lines features for each sequence?\n",
    "# For example, created a X_of_Y feature instead? Does this effect model performance?\n",
    "# train_df_exercise = train_df\n",
    "# train_df_exercise['lines'] = train_df_exercise.line_number.astype(str)+\"_of_\"+train_df_exercise.total_lines.astype(str)\n",
    "\n",
    "# val_df_exercise = val_df\n",
    "# val_df_exercise['lines'] = val_df_exercise.line_number.astype(str)+\"_of_\"+val_df_exercise.total_lines.astype(str)\n",
    "\n",
    "# test_df_exercise = test_df\n",
    "# test_df_exercise['lines'] = test_df_exercise.line_number.astype(str)+\"_of_\"+test_df_exercise.total_lines.astype(str)\n",
    "\n",
    "# train_df_exercise.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "MBbpKAajwT80",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 452
    },
    "id": "MBbpKAajwT80",
    "outputId": "bf738a97-c03c-494f-e247-9a81e95fe146",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:07.794138645Z",
     "start_time": "2024-01-03T15:37:07.734611203Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "                                                 text  line_number  \\\n0   Hepatitis C virus (HCV) and alcoholic liver di...            0   \n1   There is no safe level of drinking in HCV-infe...            1   \n2   Baclofen, a GABA(B) receptor agonist, represen...            2   \n3   Previously, we performed a randomized clinical...            3   \n4   The goal of this post-hoc analysis was to expl...            4   \n5   Any patient with HCV infection was selected fo...            5   \n6   Among the 84 subjects randomized in the main t...            6   \n7                and 12 received placebo for 12-weeks            7   \n8   With respect to the placebo group (3/12, 25.0%...            8   \n9   Furthermore, in the baclofen group, compared t...            9   \n10  In conclusion, baclofen was safe and significa...           10   \n11  albumin, INR) in alcohol-dependent HCV-infecte...           11   \n12  Baclofen may represent a clinically relevant a...           12   \n\n    total_lines  \n0            13  \n1            13  \n2            13  \n3            13  \n4            13  \n5            13  \n6            13  \n7            13  \n8            13  \n9            13  \n10           13  \n11           13  \n12           13  ",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>text</th>\n      <th>line_number</th>\n      <th>total_lines</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Hepatitis C virus (HCV) and alcoholic liver di...</td>\n      <td>0</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>There is no safe level of drinking in HCV-infe...</td>\n      <td>1</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Baclofen, a GABA(B) receptor agonist, represen...</td>\n      <td>2</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Previously, we performed a randomized clinical...</td>\n      <td>3</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>The goal of this post-hoc analysis was to expl...</td>\n      <td>4</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>5</th>\n      <td>Any patient with HCV infection was selected fo...</td>\n      <td>5</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>6</th>\n      <td>Among the 84 subjects randomized in the main t...</td>\n      <td>6</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>7</th>\n      <td>and 12 received placebo for 12-weeks</td>\n      <td>7</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>8</th>\n      <td>With respect to the placebo group (3/12, 25.0%...</td>\n      <td>8</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>9</th>\n      <td>Furthermore, in the baclofen group, compared t...</td>\n      <td>9</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>10</th>\n      <td>In conclusion, baclofen was safe and significa...</td>\n      <td>10</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>11</th>\n      <td>albumin, INR) in alcohol-dependent HCV-infecte...</td>\n      <td>11</td>\n      <td>13</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td>Baclofen may represent a clinically relevant a...</td>\n      <td>12</td>\n      <td>13</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#5. Write a function (or series of functions) to take a sample abstract string, preprocess it (in the same way our model has been trained),\n",
    "# make a prediction on each sequence in the abstract and return the abstract\n",
    "sentence=\"Hepatitis C virus (HCV) and alcoholic liver disease (ALD), either alone or in combination, count for more than two thirds of all liver diseases in the Western world. There is no safe level of drinking in HCV-infected patients and the most effective goal for these patients is total abstinence. Baclofen, a GABA(B) receptor agonist, represents a promising pharmacotherapy for alcohol dependence (AD). Previously, we performed a randomized clinical trial (RCT), which demonstrated the safety and efficacy of baclofen in patients affected by AD and cirrhosis. The goal of this post-hoc analysis was to explore baclofen's effect in a subgroup of alcohol-dependent HCV-infected cirrhotic patients. Any patient with HCV infection was selected for this analysis. Among the 84 subjects randomized in the main trial, 24 alcohol-dependent cirrhotic patients had a HCV infection; 12 received baclofen 10mg t.i.d. and 12 received placebo for 12-weeks. With respect to the placebo group (3/12, 25.0%), a significantly higher number of patients who achieved and maintained total alcohol abstinence was found in the baclofen group (10/12, 83.3%; p=0.0123). Furthermore, in the baclofen group, compared to placebo, there was a significantly higher increase in albumin values from baseline (p=0.0132) and a trend toward a significant reduction in INR levels from baseline (p=0.0716). In conclusion, baclofen was safe and significantly more effective than placebo in promoting alcohol abstinence, and improving some Liver Function Tests (LFTs) (i.e. albumin, INR) in alcohol-dependent HCV-infected cirrhotic patients. Baclofen may represent a clinically relevant alcohol pharmacotherapy for these patients.\"\n",
    "sentence = sentence.strip().split(\". \")\n",
    "char_test = [split_chars(line) for line in sentence]\n",
    "line_number_test = [i for i in range(len(sentence))]\n",
    "total_lines_test = len(sentence)\n",
    "test_df_exercise = pd.DataFrame({\"text\":sentence,\n",
    "                                 \"line_number\":line_number_test,\n",
    "                                 \"total_lines\":total_lines_test})\n",
    "test_df_exercise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "Xx0Qd8d8zAGO",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "Xx0Qd8d8zAGO",
    "outputId": "820733ae-733e-41af-9bfe-34ff4720b16c",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:07.800502617Z",
     "start_time": "2024-01-03T15:37:07.778330477Z"
    }
   },
   "outputs": [
    {
     "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))>"
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "val_all_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "kAtnPwxtwmM3",
   "metadata": {
    "id": "kAtnPwxtwmM3",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:07.882108745Z",
     "start_time": "2024-01-03T15:37:07.778565787Z"
    }
   },
   "outputs": [],
   "source": [
    "test_sentences=test_df_exercise.text\n",
    "test_line_number=test_df_exercise.line_number\n",
    "test_total_lines=test_df_exercise.total_lines\n",
    "\n",
    "test_chars = [split_chars(sentence) for sentence in test_sentences]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "aTG5C6OXzplq",
   "metadata": {
    "id": "aTG5C6OXzplq",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:08.016214775Z",
     "start_time": "2024-01-03T15:37:07.849569307Z"
    }
   },
   "outputs": [],
   "source": [
    "test_total_lines_one_hot = tf.one_hot(test_df_exercise.total_lines.to_numpy(), depth=14)\n",
    "test_line_numbers_one_hot = tf.one_hot(test_df_exercise.line_number.to_numpy(), depth=13)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "KcdWj7NL0R1H",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "KcdWj7NL0R1H",
    "outputId": "c2e32adf-eee1-4933-9805-1cd36381405c",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:08.019822713Z",
     "start_time": "2024-01-03T15:37:07.890223691Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": "<_TensorSliceDataset element_spec=(TensorSpec(shape=(13,), dtype=tf.float32, name=None), TensorSpec(shape=(14,), dtype=tf.float32, name=None), TensorSpec(shape=(), dtype=tf.string, name=None), TensorSpec(shape=(), dtype=tf.string, name=None))>"
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_all_dataset_exercise = tf.data.Dataset.from_tensor_slices((test_line_numbers_one_hot, test_total_lines_one_hot, test_sentences, test_chars))\n",
    "test_all_dataset_exercise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "zKyQG7J313QY",
   "metadata": {
    "id": "zKyQG7J313QY",
    "ExecuteTime": {
     "end_time": "2024-01-03T15:37:08.021821349Z",
     "start_time": "2024-01-03T15:37:07.934295416Z"
    }
   },
   "outputs": [],
   "source": [
    "# model_5.predict(test_all_dataset_exercise)"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "gpuType": "T4",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}