Case_study_2.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zHepFGD_u3KO"
   },
   "source": [
    "# Chest X-Ray Medical Report Generation Using Deep Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1>1. Business Problem </h1>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> 1.1 Description </h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you go to the doctor with chest pain, a chest injury or shortness of breath, you will typically get a chest X-ray.The chest x-ray is the most frequently requested radiologic examination. The image helps your doctor determine whether you have heart problems, a collapsed lung, pneumonia, broken ribs, emphysema, cancer or any of several other conditions.\n",
    "  Chest x-ray is also helpful in diagnosing COVID-19 infection in lungs , because COVID-19 is primerily attacks our lungs.\n",
    "  Once you get your x-ray done your doctor will interpret it using his knowledge and instruct you further.\n",
    "  So our aim is to make these x-rays easy to read by generating reports depicting what a radiologist sees and observes in x-ray.\n",
    "   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<p style='font-size:18px'><b> Problem Statemtent </b></p>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "   Build a deep learning model to generate medical report from goven x-ray images. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<p style='font-size:18px'><b> Source:</b>  \n",
    "\n",
    "PNG Images : http://academictorrents.com/details/5a3a439df24931f410fac269b87b050203d9467d\n",
    "\n",
    "XML Reports : http://academictorrents.com/details/66450ba52ba3f83fbf82ef9c91f2bde0e845ab\n",
    "\n",
    "</p>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> 1.2 Real World / Business Objectives and Constraints </h2>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. The cost of a mis-classification can be very high.\n",
    "\n",
    "2. No strict latency concerns.\n",
    "\n",
    "3. Interpretability is partially important."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1>2. Machine Learning problem </h1>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> 2.1 Data </h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> 2.1.1 Data Overview </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "- Data will be in 2 TAR files:\n",
    "1. One folder will contain PNG images which will be x-rays.\n",
    "2. Ohter one has XML reports(More than one image could be associated with a report)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "CcHpN163u274"
   },
   "source": [
    "\n",
    "Size of Posts.xml - 14.6GB\n",
    "Number of files in XML report folder = 3955\n",
    "\n",
    "Number of files in XML report folder = 3955\n",
    "\n",
    "\n",
    "the dataframe that we are using has 5 columns:  'IMAGES','COMPARISON','INDICATION','FINDINGS','IMPRESSION'\n",
    "\n",
    "\n",
    "Data Field Explaination\n",
    "\n",
    "Dataset contains 3955 rows. The columns in the table are:\n",
    "IMAGES - contains paths of all the images related to a report(input variable)\n",
    "COMPARISON - contains comments about different images of a report \n",
    "INDICATION - contains discription of why the x-ray was suggested.\n",
    "FINDINGS -  initial findings from the x-ray images.\n",
    "IMPRESSION - detailed information about the x-ray(target vaiable)\n",
    "\n",
    "<b> Number of rows in dataframe = 3955  </b>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3>2.1.2 Example Data point </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "<?xml version=\"1.0\" encoding=\"utf-8\"?>\n",
    "<eCitation>\n",
    "   <meta type=\"rr\"/>\n",
    "   <uId id=\"CXR1772\"/>\n",
    "   <pmcId id=\"1772\"/>\n",
    "   <docSource>CXR</docSource>\n",
    "   <IUXRId id=\"1772\"/>\n",
    "   <licenseType>open-access</licenseType>\n",
    "   <licenseURL>http://creativecommons.org/licenses/by-nc-nd/4.0/</licenseURL>\n",
    "   <ccLicense>byncnd</ccLicense>\n",
    "   <articleURL/>\n",
    "   <articleDate>2013-08-01</articleDate>\n",
    "   <articleType>XR</articleType>\n",
    "   <publisher>Indiana University</publisher>\n",
    "   <title>Indiana University Chest X-ray Collection</title>\n",
    "   <note>The data are drawn from multiple hospital systems.</note>\n",
    "   <specialty>pulmonary diseases</specialty>\n",
    "   <subset>CXR</subset>\n",
    "   <MedlineCitation Owner=\"Indiana University\" Status=\"supplied by publisher\">\n",
    "   \n",
    "      <Article PubModel=\"Electronic\">\n",
    "      \n",
    "         <Journal>\n",
    "         \n",
    "            <JournalIssue>\n",
    "            \n",
    "               <PubDate>\n",
    "               \n",
    "                  <Year>2013</Year>\n",
    "               \n",
    "                  <Month>08</Month>\n",
    "               \n",
    "                  <Day>01</Day>\n",
    "            \n",
    "               </PubDate>\n",
    "         \n",
    "            </JournalIssue>\n",
    "      \n",
    "         </Journal>\n",
    "      \n",
    "         <ArticleTitle>Indiana University Chest X-ray Collection</ArticleTitle>\n",
    "      \n",
    "         <Abstract>\n",
    "         \n",
    "            <AbstractText Label=\"COMPARISON\">Radiograph Chest PA and Lateral XXXX, XXXX.</AbstractText>\n",
    "         \n",
    "            <AbstractText Label=\"INDICATION\">Chest pain and shortness of breath</AbstractText>\n",
    "         \n",
    "            <AbstractText Label=\"FINDINGS\">Mild cardiomegaly. The lungs are clear bilaterally. No large pleural effusion or pneumothorax. The XXXX are intact.</AbstractText>\n",
    "         \n",
    "            <AbstractText Label=\"IMPRESSION\">No acute cardiopulmonary abnormalities.</AbstractText>\n",
    "      \n",
    "         </Abstract>\n",
    "      \n",
    "         <Affiliation>Indiana University</Affiliation>\n",
    "      \n",
    "         <AuthorList CompleteYN=\"Y\">\n",
    "         \n",
    "            <Author ValidYN=\"Y\">\n",
    "            \n",
    "               <LastName>Kohli</LastName>\n",
    "            \n",
    "               <ForeName>Marc</ForeName>\n",
    "            \n",
    "               <Initials>MD</Initials>\n",
    "         \n",
    "            </Author>\n",
    "         \n",
    "            <Author ValidYN=\"Y\">\n",
    "            \n",
    "               <LastName>Rosenman</LastName>\n",
    "            \n",
    "               <ForeName>Marc</ForeName>\n",
    "            \n",
    "               <Initials>M</Initials>\n",
    "         \n",
    "            </Author>\n",
    "      \n",
    "         </AuthorList>\n",
    "      \n",
    "         <Language>eng</Language>\n",
    "      \n",
    "         <PublicationTypeList>\n",
    "         \n",
    "            <PublicationType>Radiology Report</PublicationType>\n",
    "      \n",
    "         </PublicationTypeList>\n",
    "      \n",
    "         <ArticleDate>\n",
    "         \n",
    "            <Year>2013</Year>\n",
    "         \n",
    "            <Month>08</Month>\n",
    "         \n",
    "            <Day>01</Day>\n",
    "      \n",
    "         </ArticleDate>\n",
    "   \n",
    "      </Article>\n",
    "   \n",
    "      <EssieArticleTitle>Indiana University Chest X-ray Collection</EssieArticleTitle>\n",
    "   \n",
    "      <IMedAuthor>Marc David Kohli MD</IMedAuthor>\n",
    "   \n",
    "      <IMedAuthor>Marc Rosenman M</IMedAuthor>\n",
    "\n",
    "   </MedlineCitation>\n",
    "   <MeSH>\n",
    "   \n",
    "      <major>Cardiomegaly/mild</major>\n",
    "\n",
    "   </MeSH>\n",
    "</eCitation>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>2.2 Mapping the real-world problem to a Machine Learning Problem </h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> 2.2.1 Type of Machine Learning Problem </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "1. This case study is basically based on generating a text given an inputimage. Specifically we want to generate a medical report given a chestx ray image.\n",
    "2. This falls under a Vision Language task , which is a mixture of computer vision domain and natural language processing domain.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3>2.2.2 Performance metric </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ujjpDMFRu26D"
   },
   "source": [
    "The metric that we will be using in this for evaluation will be BLEU score.\n",
    "\n",
    "Bleu is a most common metric used for machine translation. It uses\n",
    "something called modified precision to calculate the score, this score lies\n",
    "between 0-1, 1 meaning it's exactly the same or perfect and 0 means there is\n",
    "nothing in common.\n",
    "The BLEU score was proposed by Kishore Papineni, et al. in their 2002 paper “BLEU: a Method for Automatic Evaluation of Machine Translation“.\n",
    "\n",
    "link to paper: https://www.researchgate.net/publication/2588204_BLEU_a_Method_for_Automatic_Evaluation_of_Machine_Translation\n",
    "\n",
    "we will use nltk library to calculate BLEU score.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1> 3. Exploratory Data Analysis </h1>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> 3.1 Data Loading </h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "16fYbRvDaNzk"
   },
   "source": [
    "<h3> 3.1.1 downloading data and extracting the zip file </h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 109
    },
    "id": "uBUD1jgJw7fu",
    "outputId": "65446c6a-d557-4188-876b-c1f6f90e5c7d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading package lists... Done\n",
      "Building dependency tree       \n",
      "Reading state information... Done\n",
      "transmission-cli is already the newest version (2.92-3ubuntu2).\n",
      "0 upgraded, 0 newly installed, 0 to remove and 21 not upgraded.\n"
     ]
    }
   ],
   "source": [
    "#reference https://blog.gagahpangeran.com/torrent-to-google-drive-using-google-colab\n",
    "!apt-get install transmission-cli"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "P0wBlAU60lRt",
    "outputId": "2cfafdd1-2cf7-4b47-c93e-bad5fc9454a0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
     ]
    }
   ],
   "source": [
    "from google.colab import drive\n",
    "drive.mount('/content/drive')#mounting google drive "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "id": "5TZCumoc0oM5",
    "outputId": "3b71871c-8577-43a5-be48-71af4c3f9e5d"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "string"
      },
      "text/plain": [
       "\"''!transmission-cli -w '/content/drive/My Drive/case_study_2' https://academictorrents.com/download/66450ba52ba3f83fbf82ef9c91f2bde0e845aba9.torrent\""
      ]
     },
     "execution_count": 3,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''''!transmission-cli -w '/content/drive/My Drive/case_study_2' https://academictorrents.com/download/66450ba52ba3f83fbf82ef9c91f2bde0e845aba9.torrent'''''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 36
    },
    "id": "uD7hEVEX3S8H",
    "outputId": "9e1d5b93-4715-45d4-e837-3b43da67997a"
   },
   "outputs": [
    {
     "data": {
      "application/vnd.google.colaboratory.intrinsic+json": {
       "type": "string"
      },
      "text/plain": [
       "\"''!transmission-cli -w '/content/drive/My Drive/case_study_2' https://academictorrents.com/download/5a3a439df24931f410fac269b87b050203d9467d.torrent\""
      ]
     },
     "execution_count": 4,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "'''''!transmission-cli -w '/content/drive/My Drive/case_study_2' https://academictorrents.com/download/5a3a439df24931f410fac269b87b050203d9467d.torrent'''''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "vjiKx1Z98Kuy",
    "outputId": "506ac258-21d6-465c-adb6-b0cfa7c32887"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: patool in /usr/local/lib/python3.6/dist-packages (1.12)\n"
     ]
    }
   ],
   "source": [
    "#making directory where images need to be extracted\n",
    "!pip install patool # patool is a zip file extraction tool\n",
    "import patoolib\n",
    "import os \n",
    "path_of_images = \"images\"\n",
    "if path_of_images not in os.listdir():\n",
    "  os.mkdir(path_of_images)\n",
    "  patoolib.extract_archive(\"/content/drive/My Drive/case_study_2/NLMCXR_reports.tgz\")#extracting archive\n",
    "  patoolib.extract_archive('/content/drive/My Drive/case_study_2/NLMCXR_png.tgz',outdir=path_of_images)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> 3.1.2 parsing the xml file </h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "bKCXQVTOv_od"
   },
   "outputs": [],
   "source": [
    "#https://stackoverflow.com/questions/1912434/how-do-i-parse-xml-in-python\n",
    "import xml.etree.ElementTree as ET\n",
    "root = ET.parse('/content/drive/My Drive/case_study_2/reports/ecgen-radiology/1.xml').getroot()#xml parsing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "FyHJqFsp6vLO"
   },
   "source": [
    "XML is an inherently hierarchical data format, and the most natural way to represent it is with a tree. ET has two classes for this purpose - ElementTree represents the whole XML document as a tree, and Element represents a single node in this tree. Interactions with the whole document (reading and writing to/from files) are usually done on the ElementTree level. Interactions with a single XML element and its sub-elements are done on the Element level."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> 3.1.3 extracting the intrested columns from xml file </h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "id": "I635Aik0v_o-",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import os\n",
    "ids = []\n",
    "COMPARISON = []\n",
    "INDICATION = []\n",
    "FINDINGS = []\n",
    "IMPRESSION = []\n",
    "images = []\n",
    "\n",
    "for filename in  sorted(os.listdir(os.path.join(os.getcwd(),'ecgen-radiology'))):#iterating over all the xml files in the directory\n",
    "    root = ET.parse(os.path.join(os.getcwd(),'ecgen-radiology',filename)).getroot()#creating parser\n",
    "    imageId = []\n",
    "\n",
    "    for child in root.iter():#iterating over root node\n",
    "\n",
    "\n",
    "        if child.tag == 'pmcId':#searching if tag has pmcId in it\n",
    "            ids.append(child.attrib['id'])# append that tag value in ids list\n",
    "        if child.tag == 'AbstractText':\n",
    "            if child.attrib['Label'] == 'COMPARISON':\n",
    "\n",
    "                COMPARISON.append(child.text)\n",
    "            if child.attrib['Label'] == 'INDICATION':\n",
    "\n",
    "                INDICATION.append(child.text)\n",
    "            if child.attrib['Label'] == 'FINDINGS':\n",
    "\n",
    "                FINDINGS.append(child.text)\n",
    "            if child.attrib['Label'] == 'IMPRESSION':\n",
    "\n",
    "                IMPRESSION.append(child.text)\n",
    "        if child.tag == 'parentImage':   \n",
    "            imageId.append(os.getcwd()+'/'+path_of_images+'/'+child.attrib['id']+'.png')#storing images in current working directory under path_of_images tag and filenames will be same as id atribute\n",
    "    images.append(imageId)# appending image_paths to images list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 109
    },
    "id": "Tni2Gw5wv_pU",
    "outputId": "0ea7a0d2-fd96-4c0a-96ee-62a255c86a8d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3955\n",
      "3955\n",
      "3955\n",
      "3955\n",
      "3955\n"
     ]
    }
   ],
   "source": [
    "print(len(images))\n",
    "print(len(COMPARISON))\n",
    "print(len(INDICATION))\n",
    "print(len(FINDINGS))\n",
    "print(len(IMPRESSION))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> 3.1.4 saving extracted columns into dataframe  </h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "NS2PUU0Mv_pf"
   },
   "outputs": [],
   "source": [
    "data = (zip(images,COMPARISON,INDICATION,FINDINGS,IMPRESSION))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "rxnYb-Y-v_pq"
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "df = pd.DataFrame(data=data,index = [int(i) for i in ids],columns = ['IMAGES','COMPARISON','INDICATION','FINDINGS','IMPRESSION'])#making dataframe out of data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 293
    },
    "id": "bV8tyc8Zv_px",
    "outputId": "e82e9b3d-cbd3-40dd-b87a-92d77ad945b7"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>IMAGES</th>\n",
       "      <th>COMPARISON</th>\n",
       "      <th>INDICATION</th>\n",
       "      <th>FINDINGS</th>\n",
       "      <th>IMPRESSION</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[/content/images/CXR1_1_IM-0001-3001.png, /con...</td>\n",
       "      <td>None.</td>\n",
       "      <td>Positive TB test</td>\n",
       "      <td>The cardiac silhouette and mediastinum size ar...</td>\n",
       "      <td>Normal chest x-XXXX.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>[/content/images/CXR10_IM-0002-1001.png, /cont...</td>\n",
       "      <td>Chest radiographs XXXX.</td>\n",
       "      <td>XXXX-year-old male, chest pain.</td>\n",
       "      <td>The cardiomediastinal silhouette is within nor...</td>\n",
       "      <td>No acute cardiopulmonary process.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>100</th>\n",
       "      <td>[/content/images/CXR100_IM-0002-1001.png, /con...</td>\n",
       "      <td>None.</td>\n",
       "      <td>None</td>\n",
       "      <td>Both lungs are clear and expanded. Heart and m...</td>\n",
       "      <td>No active disease.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1000</th>\n",
       "      <td>[/content/images/CXR1000_IM-0003-1001.png, /co...</td>\n",
       "      <td>XXXX PA and lateral chest radiographs</td>\n",
       "      <td>XXXX-year-old male, XXXX.</td>\n",
       "      <td>There is XXXX increased opacity within the rig...</td>\n",
       "      <td>1. Increased opacity in the right upper lobe w...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1001</th>\n",
       "      <td>[/content/images/CXR1001_IM-0004-1001.png, /co...</td>\n",
       "      <td>None</td>\n",
       "      <td>dyspnea, subjective fevers, arthritis, immigra...</td>\n",
       "      <td>Interstitial markings are diffusely prominent ...</td>\n",
       "      <td>Diffuse fibrosis. No visible focal acute disease.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 IMAGES  ...                                         IMPRESSION\n",
       "1     [/content/images/CXR1_1_IM-0001-3001.png, /con...  ...                               Normal chest x-XXXX.\n",
       "10    [/content/images/CXR10_IM-0002-1001.png, /cont...  ...                  No acute cardiopulmonary process.\n",
       "100   [/content/images/CXR100_IM-0002-1001.png, /con...  ...                                 No active disease.\n",
       "1000  [/content/images/CXR1000_IM-0003-1001.png, /co...  ...  1. Increased opacity in the right upper lobe w...\n",
       "1001  [/content/images/CXR1001_IM-0004-1001.png, /co...  ...  Diffuse fibrosis. No visible focal acute disease.\n",
       "\n",
       "[5 rows x 5 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "sSnqIQ21v_p4"
   },
   "outputs": [],
   "source": [
    "df.sort_index(inplace=True)#sorting index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 237
    },
    "id": "fbk86gbGv_qB",
    "outputId": "f9385ef0-55b2-4f57-ca4d-8af6d2e22c65"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 3955 entries, 1 to 3999\n",
      "Data columns (total 5 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   IMAGES      3955 non-null   object\n",
      " 1   COMPARISON  3333 non-null   object\n",
      " 2   INDICATION  3865 non-null   object\n",
      " 3   FINDINGS    3425 non-null   object\n",
      " 4   IMPRESSION  3921 non-null   object\n",
      "dtypes: object(5)\n",
      "memory usage: 185.4+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()#displaying info about the dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> 3.2 EDA on each columns </h2>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "vG4NU-Eyv_qJ"
   },
   "source": [
    "### 3.2.1 COLUMN IMAGES:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "9_7bX9_xv_qL"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "img_count = df.IMAGES.apply(lambda x:len(x))#counting number of images per datapoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 90
    },
    "id": "RgIRp6r1v_qS",
    "outputId": "f9f3f1db-b0e2-4068-8a42-6e0af254a03f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean of number  of images per report are : 1.8887484197218711\n",
      "median of number  of images per report are : 2.0\n",
      "maximum number  of images per report are : 5\n",
      "minimum number  of images per report are : 0\n"
     ]
    }
   ],
   "source": [
    "print('mean of number  of images per report are :',img_count.mean())\n",
    "print('median of number  of images per report are :',img_count.median())\n",
    "print('maximum number  of images per report are :',img_count.max())\n",
    "print('minimum number  of images per report are :',img_count.min())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 164
    },
    "id": "O9mICxErv_qd",
    "outputId": "cd9617b3-2717-4f58-e98b-4da2c5742312",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of images associated with each report\n",
      "2    3208\n",
      "1     446\n",
      "3     181\n",
      "0     104\n",
      "4      15\n",
      "5       1\n",
      "Name: IMAGES, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print('number of images associated with each report')\n",
    "print(img_count.value_counts())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 346
    },
    "id": "bXESHbNpv_ql",
    "outputId": "5a64cfcc-1455-4bc4-a7c8-2f2aacb2034f"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmoAAAFJCAYAAAArAgF6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de7xddX3n/9ebEEiACAJOigQIVbyA1CARdLRO0BERrVhLFUqRKIrTSr00tVJvMAq/H1rRqqNOQRCsNgFFhQIVI3oUVMpNxnCRmoEgQUDkmnCTwGf+2Cu4E06Sncvee52T1/Px2I/s/V1rfddnf88hebO+65KqQpIkSe2zybALkCRJ0ugMapIkSS1lUJMkSWopg5okSVJLGdQkSZJayqAmSZLUUgY1SeslyelJjh/SvpPky0nuSXLZKMsPS/LdYdQmSRuCQU0aZ5IsSvKbJFt2tb0tycgQy+qXlwKvBKZV1T4rL6yqr1XV/oMvS70YZsiXxgqDmjQ+TQDePewi1laSCWu5yS7Aoqp6oB/1jCdJNu1hnbUd/3U2yH1JY5lBTRqf/hH4uyTbrLwgyfQk1f0Pd5KRJG9r3s9O8uMkn05yb5Ibk/zXpv2W5mjdESt1u32S+UmWJPlhkl26+n5Os+zuJDckeWPXstOTfDHJBUkeAPYbpd6nJzm32X5hkrc37UcCXwJenGRpkv85yrazk1zS9bmS/HWSXza1fizJM5L8JMn9Sc5Kslmz7lOTnJfkzmZq9bwk07r62jXJj5p+vpfk80m+2rX8RU2/9yb5P0lmrVTXjc22NyU5bLQfYpLjknwjyZnNulclef5KY3N2U+NNSd41yrZfTXI/MHuU/p80/j32uap6ntv8Lt2b5Nokr1vNvo4EDgP+vvn5/dtoYyBt9KrKly9f4+gFLAL+O/BN4Pim7W3ASPN+OlDApl3bjABva97PBpYBb6FzZO544FfA54HNgf2BJcBWzfqnN59f1iz/DHBJs2xL4Jamr02BvYDfArt3bXsf8BI6/+M4aZTv8yPgC8AkYAZwJ/DyrlovWc1YrLC8+d7nAE8B9gAeAS4C/hDYGrgOOKJZdzvgz4AtgCnA14Fvd/X1U+CTwGZ0pmDvB77aLNsRuAs4sPler2w+P60Zk/uBZzfr7gDssYr6jwMeBQ4GJgJ/B9zUvN8EuBL4SFPDHwI3Aq9aadvXN+tOHqX/lcd/ix77HK2eicBC4APNti9vfi+evaqfddN2/LD/m/Hlq80vj6hJ49dHgL9J8rR12PamqvpyVT0GnAnsBHy0qh6pqu8CvwOe2bX++VX1o6p6BPggnaNcOwGvpTM1+eWqWlZVPwPOBv68a9tzqurHVfV4VT3cXUTTx0uA91fVw1V1NZ2jaG9eh++03Ceq6v6quha4BvhuVd1YVfcB/04nTFJVd1XV2VX1YFUtAU4A/ltT187AC4GPVNXvquoS4NyuffwlcEFVXdB8r/nAFXSCG8DjwPOSTK6q25paVuXKqvpGVT0KfIpOwHlRs/+nVdVHmxpuBE4BDuna9qdV9e2mhodW0f8T4w/s2UOfq6rnRcBWwInNtt8HzgMOHW1fK/+sJY1ujecsSBqbquqaJOcBxwDXr+Xmd3S9f6jpb+W2rbo+39K136VJ7gaeTuccsn2T3Nu17qbAv4y27SieDtzdBKXlbgZm9vIlVmHl77Hy5z8ASLIF8GngAOCpzfIpzblVy+t6sGvbW+gEWuh87z9P8iddyycCP6iqB5K8ic7RqFOT/BiYU1W/WEW93WP7eJLFzf4LePpKYzsBuHi0bVeje51d1qbPleoBuKUJfMvdTOfo4trUI6mLQU0a344FrgJO6mpbfuL9FnSm4KAJJ+theUAhyVbAtsCv6fzD/MOqeuVqtq3VLPs1sG2SKV1hbWfg1vWstxdzgGcD+1bV7UlmAD8DAtzW1LVFV1jbqWvbW4B/qaq3j9ZxVV0IXJhkMp2p5VOAP15FHd1juwkwjc64LKNz5HO31XyH1Y3taOvc0kOfq6oHYKckm3SFtZ2B/1xNPb3UJ23UnPqUxrGqWkhn6vJdXW130gk6f5lkQpK3As9Yz10dmOSlzYn4HwMurapb6Ex9PSvJ4UkmNq8XJnluj/XfAvwE+P+TTEryR3ROQv/q6rfcIKbQOcJ2b5Jt6YTe5XXdTGcq87gkmyV5MdB99OyrwJ8keVUzxpOSzEoyLcnUJAelc/uUR4CldKZCV2XvJG9I5+KP9zTbXApcBixJ8v4kk5v9PC/JC9fjO/fS56rq+Q/gQToXB0xsLp74E2DeavZ3B53z4CStgkFNGv8+SucE9m5vB95H5wT3PeiEofXxr3SCzN3A3nTO0aI5CrY/nXOcfg3cDnyczkUHvTqUzgUQvwa+BRxbVd9bz3p78U/AZDoXP1wKfGel5YcBL6YzhsfTCcSPwBMB8yA6J9bfSedI1fvo/J27CfC3dL7P3XTOe/ur1dRxDvAm4B7gcOANVfVoc/7ga+lcYHFTU+eX6FwUsU567HNV9fyOTjB7dbPdF4A3r2ZKF+BUYPfmKtFvr2vd0niWKo88S9L6SnIm8IuqOnaNK/fe53HAM6vqLzdUn+ujbfVIGwOPqEnSOmimcJ+RZJMkB9A5guZRIUkblBcTSNK6+QM696rbDlgM/FVz+xFJ2mCc+pQkSWoppz4lSZJayqAmSZLUUuPyHLXtt9++pk+fPuwy1skDDzzAlluufCcF9ZNjPniO+eA55oPnmA/eWB3zK6+88rdVNerj/sZlUJs+fTpXXHHFsMtYJyMjI8yaNWvYZWxUHPPBc8wHzzEfPMd88MbqmCe5eVXLnPqUJElqKYOaJElSSxnUJEmSWmpcnqMmSZIG69FHH2Xx4sU8/PDDQ6th66235vrrrx/a/tdk0qRJTJs2jYkTJ/a8jUFNkiStt8WLFzNlyhSmT59OkqHUsGTJEqZMmTKUfa9JVXHXXXexePFidt111563c+pTkiStt4cffpjttttuaCGt7ZKw3XbbrfURR4OaJEnaIAxpq7cu42NQkyRJ48IOO+wAwKJFi0jChz70oSeW/fa3v2XixIkcffTRK2wzY8YMDjnkkBXali1bxgc+8AF22203ZsyYwYwZMzjhhBOeWD5hwoQn2mfMmMGJJ54IwHnnncdee+3F85//fHbffXf++Z//eb2/k+eoSZKkDW76Medv0P4WnfiatVp/11135fzzz+f4448H4Otf/zp77LHHCutcf/31PPbYY1x88cUrPNXgQx/6ELfffjsLFixg0qRJLFmyhJNOOumJ7SZPnszVV1+9Ql+PPvooRx11FJdddhnTpk3jkUceYdGiRevwTVfkETVJkjTubLHFFjz3uc994klFZ555Jm984xtXWGfu3Lkcfvjh7L///pxzzjkAPPjgg5xyyil87nOfY9KkSQBMmTKF4447brX7W7JkCcuWLWO77bYDYPPNN+fZz372en8Pg5okSRqXDjnkEObNm8ctt9zChAkTePrTn77C8jPPPJNDDjmEQw89lLlz5wKwcOFCdt5559VePfrQQw+tMPV55plnsu222/K6172OXXbZhUMPPZSvfe1rPP744+v9HZz6XAcb+nButzl7LmN2n/pf28PGkiSNZQcccAAf/vCHmTp1Km9605tWWHbFFVew/fbbs/POO7Pjjjvy1re+lbvvvvtJfXz5y1/mM5/5DHfddRc/+clP2GmnnUad+gT40pe+xIIFC/je977HJz/5SebPn8/pp5++Xt/BI2qSJGlc2myzzdh777056aSTOPjgg1dYNnfuXH7xi18wffp0nvGMZ3D//fdz9tln88xnPpNf/epXLFmyBIC3vOUtXH311Wy99dY89thja9znnnvuyXvf+17mz5/P2Wefvd7fwaAmSZLGrTlz5vDxj3+cbbfd9om2xx9/nLPOOosFCxawaNEiFi1axDnnnMPcuXPZYostOPLIIzn66KOfuOfZY489xu9+97vV7mfp0qWMjIw88fnqq69ml112We/6nfqUJEnj1h577PGkqz0vvvhidtxxxxXOWXvZy17Gddddx2233cYJJ5zAhz/8YZ73vOcxZcoUJk+ezBFHHPHE+svPUVvugAMO4IMf/CCf+MQneMc73sHkyZPZcsst13vaEwxqkiSpD4ZxXvRtt90GwPTp07nmmmuetHz27NnMnj0bgEsvvXSFZRMmTOD2229/4vOJJ574xP3RVraqKdALLrhgXcpeLac+JUmSWsqgJkmS1FIGNUmSpJYyqEmSpA2iqoZdQquty/gY1CRJ0nqbNGkSd911l2FtFaqKu+6664nHUvXKqz4lSdJ6mzZtGosXL+bOO+8cWg0PP/zwWgehQZo0aRLTpk1bq20MapIkab1NnDiRXXfddag1jIyMsNdeew21hg3NqU9JkqSW6ltQSzIpyWVJ/k+Sa5P8z6Z91yT/kWRhkjOTbNa0b958Xtgsn97V1z807TckeVW/apYkSWqTfh5RewR4eVU9H5gBHJDkRcDHgU9X1TOBe4Ajm/WPBO5p2j/drEeS3YFDgD2AA4AvJJnQx7olSZJaoW9BrTqWNh8nNq8CXg58o2k/A3h98/6g5jPN8lckSdM+r6oeqaqbgIXAPv2qW5IkqS36ejFBc+TrSuCZwOeB/wvcW1XLmlUWAzs273cEbgGoqmVJ7gO2a9q7H8jVvU33vo4CjgKYOnXqCk+w39Dm7LlszSuto6mT+9d/P8dkLFu6dKljM2CO+eA55oPnmA/eeBzzvga1qnoMmJFkG+BbwHP6uK+TgZMBZs6cWbNmzerXrph9zPl963vOnss4aUF/fiyLDpvVl37HupGREfr5+6Inc8wHzzEfPMd88MbjmA/kqs+quhf4AfBiYJsky5PINODW5v2twE4AzfKtgbu620fZRpIkadzq51WfT2uOpJFkMvBK4Ho6ge3gZrUjgHOa9+c2n2mWf786tzc+FzikuSp0V2A34LJ+1S1JktQW/Zz63AE4ozlPbRPgrKo6L8l1wLwkxwM/A05t1j8V+JckC4G76VzpSVVdm+Qs4DpgGfDOZkpVkiRpXOtbUKuqnwNPuj1wVd3IKFdtVtXDwJ+voq8TgBM2dI2SJElt5pMJJEmSWsqgJkmS1FIGNUmSpJYyqEmSJLWUQU2SJKmlDGqSJEktZVCTJElqKYOaJElSSxnUJEmSWsqgJkmS1FIGNUmSpJYyqEmSJLWUQU2SJKmlDGqSJEktZVCTJElqKYOaJElSSxnUJEmSWsqgJkmS1FIGNUmSpJYyqEmSJLWUQU2SJKmlDGqSJEktZVCTJElqKYOaJElSSxnUJEmSWsqgJkmS1FIGNUmSpJYyqEmSJLWUQU2SJKmlDGqSJEktZVCTJElqqb4FtSQ7JflBkuuSXJvk3U37cUluTXJ18zqwa5t/SLIwyQ1JXtXVfkDTtjDJMf2qWZIkqU027WPfy4A5VXVVkinAlUnmN8s+XVWf7F45ye7AIcAewNOB7yV5VrP488ArgcXA5UnOrarr+li7JEnS0PUtqFXVbcBtzfslSa4HdlzNJgcB86rqEeCmJAuBfZplC6vqRoAk85p1DWqSJGlcG8g5akmmA3sB/9E0HZ3k50lOS/LUpm1H4JauzRY3batqlyRJGtdSVf3dQbIV8EPghKr6ZpKpwG+BAj4G7FBVb03yv4BLq+qrzXanAv/edHNAVb2taT8c2Leqjl5pP0cBRwFMnTp173nz5vXtOy249b6+9T11MtzxUH/63nPHrfvT8Ri3dOlSttpqq2GXsVFxzAfPMR88x3zwxuqY77fffldW1czRlvXzHDWSTATOBr5WVd8EqKo7upafApzXfLwV2Klr82lNG6tpf0JVnQycDDBz5syaNWvWhvkSo5h9zPl963vOnss4aUF/fiyLDpvVl37HupGREfr5+6Inc8wHzzEfPMd88MbjmPfzqs8ApwLXV9Wnutp36FrtT4FrmvfnAock2TzJrsBuwGXA5cBuSXZNshmdCw7O7VfdkiRJbdHPI2ovAQ4HFiS5umn7AHBokhl0pj4XAe8AqKprk5xF5yKBZcA7q+oxgCRHAxcCE4DTquraPtYtSZLUCv286vMSIKMsumA125wAnDBK+wWr206SJGk88skEkiRJLWVQkyRJaimDmiRJUksZ1CRJklrKoCZJktRSBjVJkqSWMqhJkiS1lEFNkiSppQxqkiRJLWVQkyRJaimDmiRJUksZ1CRJklrKoCZJktRSBjVJkqSWMqhJkiS1lEFNkiSppQxqkiRJLWVQkyRJaimDmiRJUksZ1CRJklrKoCZJktRSBjVJkqSWMqhJkiS1lEFNkiSppQxqkiRJLWVQkyRJaimDmiRJUksZ1CRJklrKoCZJktRSBjVJkqSWWmNQS7JrL22SJEnasHo5onb2KG3f2NCFSJIkaUWrDGpJnpPkz4Ctk7yh6zUbmLSmjpPslOQHSa5Lcm2Sdzft2yaZn+SXzZ9PbdqT5LNJFib5eZIXdPV1RLP+L5Mcsd7fWpIkaQzYdDXLng28FtgG+JOu9iXA23voexkwp6quSjIFuDLJfGA2cFFVnZjkGOAY4P3Aq4Hdmte+wBeBfZNsCxwLzASq6efcqrqn968pSZI09qwyqFXVOcA5SV5cVT9d246r6jbgtub9kiTXAzsCBwGzmtXOAEboBLWDgK9UVQGXJtkmyQ7NuvOr6m6AJuwdAMxd25okSZLGknRy0WpWSJ5G5wjadLqCXVW9teedJNOBHwHPA35VVds07QHuqaptkpwHnFhVlzTLLqIT4GYBk6rq+Kb9w8BDVfXJlfZxFHAUwNSpU/eeN29er+WttQW33te3vqdOhjse6k/fe+64dX86HuOWLl3KVlttNewyNiqO+eA55oPnmA/eWB3z/fbb78qqmjnastVNfS53DnAx8D3gsbXdeZKt6FyQ8J6qur+TzTqqqpKsPin2qKpOBk4GmDlzZs2aNWtDdDuq2cec37e+5+y5jJMW9PJjWXuLDpvVl37HupGREfr5+6Inc8wHzzEfPMd88MbjmPeSCLaoqvevS+dJJtIJaV+rqm82zXck2aGqbmumNn/TtN8K7NS1+bSm7VZ+P1W6vH1kXeqRJEkaS3q5Pcd5SQ5c246bac1Tgeur6lNdi84Fll+5eQSdI3bL29/cXP35IuC+5jy3C4H9kzy1uUJ0/6ZNkiRpXOvliNq7gQ8keQR4FAidWcunrGG7lwCHAwuSXN20fQA4ETgryZHAzcAbm2UXAAcCC4EHgbfQ2dHdST4GXN6s99HlFxZIkiSNZ2sMalU1ZV06bi4KyCoWv2KU9Qt45yr6Og04bV3qkCRJGqvWGNSSvGy09qr60YYvR5IkScv1MvX5vq73k4B9gCuBl/elIkmSJAG9TX12P5WAJDsB/9S3iiRJkgT0dtXnyhYDz93QhUiSJGlFvZyj9jk6z9iETrCbAVzVz6IkSZLU2zlqV3S9XwbMraof96keSZIkNXo5R+2MJJsBz2qabuhvSZIkSYLepj5nAWcAi+jcF22nJEd4ew5JkqT+6mXq8yRg/6q6ASDJs4C5wN79LEySJGlj18tVnxOXhzSAqvpPYGL/SpIkSRL0eDFBki8BX20+H8aKFxhIkiSpD3oJan9F5xmc72o+Xwx8oW8VSZIkCegtqG0KfKaqPgWQZAKweV+rkiRJUk/nqF0ETO76PBn4Xn/KkSRJ0nK9BLVJVbV0+Yfm/Rb9K0mSJEnQW1B7IMkLln9IsjfwUP9KkiRJEvR2jtp7gK8n+TWdG97+AfCmvlYlSZKknh4hdXmS5wDPbppuqKpH+1uWJEmSejmiRhPMrulzLZIkSerSyzlqkiRJGgKDmiRJUkutMail4y+TfKT5vHOSffpfmiRJ0satlyNqXwBeDBzafF4CfL5vFUmSJAno7WKCfavqBUl+BlBV9yTZrM91SZIkbfR6OaL2aPN8zwJI8jTg8b5WJUmSpJ6C2meBbwH/JckJwCXA/9fXqiRJktTTDW+/luRK4BV0nkzw+qq6vu+VSZIkbeTWGNSSbAv8Bpjb1TbRpxNIkiT1Vy9Tn1cBdwL/Cfyyeb8oyVXNA9olSZLUB70EtfnAgVW1fVVtB7waOA/4azq37pAkSVIf9BLUXlRVFy7/UFXfBV5cVZcCm/etMkmSpI1cL0HttiTvT7JL8/p74I7mlh2rvE1HktOS/CbJNV1txyW5NcnVzevArmX/kGRhkhuSvKqr/YCmbWGSY9bxe0qSJI05vQS1vwCmAd9uXjs3bROAN65mu9OBA0Zp/3RVzWheFwAk2R04BNij2eYLSSY0YfDzdKZbdwcObdaVJEka93q5Pcdvgb9ZxeKFq9nuR0mm91jHQcC8qnoEuCnJQmD580QXVtWNAEnmNete12O/kiRJY1Yvt+d4GvD3dI52TVreXlUvX8d9Hp3kzcAVwJyqugfYEbi0a53FTRvALSu177uKOo8CjgKYOnUqIyMj61jems3Zc1nf+p46uX/993NMxrKlS5c6NgPmmA+eYz54jvngjccx7+VZn18DzgReC/wP4Ag6t+hYF18EPkbncVQfA04C3rqOfa2gqk4GTgaYOXNmzZo1a0N0O6rZx5zft77n7LmMkxb08mNZe4sOm9WXfse6kZER+vn7oidzzAfPMR88x3zwxuOY93KO2nZVdSrwaFX9sKreCqzT0bSquqOqHquqx4FT+P305q3ATl2rTmvaVtUuSZI07vX0UPbmz9uSvCbJXsC267KzJDt0ffxTYPkVoecChyTZPMmuwG7AZcDlwG5Jdk2yGZ0LDs5dl31LkiSNNb3MsR2fZGtgDvA54CnAe9e0UZK5wCxg+ySLgWOBWUlm0Jn6XAS8A6Cqrk1yFp2LBJYB76yqx5p+jgYupHOV6WlVde3afEFJkqSxqperPs9r3t4H7Ndrx1V16CjNp65m/ROAE0ZpvwC4oNf9SpIkjRe9XPW5K53bc0zvXr+qXte/siRJktTL1Oe36RwJ+zdW8yQCSZIkbVi9BLWHq+qzfa9EkiRJK+glqH0mybHAd4FHljdW1VV9q0qSJEk9BbU9gcPp3Dtt+dRnsY73UpMkSVJveglqfw78YVX9rt/FSJIk6fd6ueHtNcA2/S5EkiRJK+rliNo2wC+SXM6K56h5ew5JkqQ+6iWoHdv3KiRJkvQkvTyZ4IeDKESSJEkrWmVQS3JJVb00yRI6V3k+sQioqnpK36uTJEnaiK0yqFXVS5s/pwyuHEmSJC3Xy1WfkiRJGgKDmiRJUksZ1CRJklrKoCZJktRSBjVJkqSWMqhJkiS1lEFNkiSppQxqkiRJLWVQkyRJaimDmiRJUksZ1CRJklrKoCZJktRSBjVJkqSWMqhJkiS1lEFNkiSppQxqkiRJLWVQkyRJaimDmiRJUksZ1CRJklqqb0EtyWlJfpPkmq62bZPMT/LL5s+nNu1J8tkkC5P8PMkLurY5oln/l0mO6Fe9kiRJbdPPI2qnAwes1HYMcFFV7QZc1HwGeDWwW/M6CvgidIIdcCywL7APcOzycCdJkjTe9S2oVdWPgLtXaj4IOKN5fwbw+q72r1THpcA2SXYAXgXMr6q7q+oeYD5PDn+SJEnj0qDPUZtaVbc1728HpjbvdwRu6VpvcdO2qnZJkqRxb9Nh7biqKkltqP6SHEVn2pSpU6cyMjKyobp+kjl7Lutb31Mn96//fo7JWLZ06VLHZsAc88FzzAfPMR+88Tjmgw5qdyTZoapua6Y2f9O03wrs1LXetKbtVmDWSu0jo3VcVScDJwPMnDmzZs2aNdpqG8TsY87vW99z9lzGSQv682NZdNisvvQ71o2MjNDP3xc9mWM+eI754Dnmgzcex3zQU5/nAsuv3DwCOKer/c3N1Z8vAu5rpkgvBPZP8tTmIoL9mzZJkqRxr29H1JLMpXM0bPski+lcvXkicFaSI4GbgTc2q18AHAgsBB4E3gJQVXcn+RhwebPeR6tq5QsUJEmSxqW+BbWqOnQVi14xyroFvHMV/ZwGnLYBS5MkSRoTfDKBJElSSxnUJEmSWsqgJkmS1FIGNUmSpJYyqEmSJLWUQU2SJKmlDGqSJEktZVCTJElqKYOaJElSSxnUJEmSWsqgJkmS1FIGNUmSpJYyqEmSJLWUQU2SJKmlDGqSJEktZVCTJElqKYOaJElSSxnUJEmSWsqgJkmS1FIGNUmSpJYyqEmSJLWUQU2SJKmlDGqSJEktZVCTJElqKYOaJElSSxnUJEmSWsqgJkmS1FIGNUmSpJYyqEmSJLWUQU2SJKmlDGqSJEktZVCTJElqqaEEtSSLkixIcnWSK5q2bZPMT/LL5s+nNu1J8tkkC5P8PMkLhlGzJEnSoA3ziNp+VTWjqmY2n48BLqqq3YCLms8ArwZ2a15HAV8ceKWSJElD0Kapz4OAM5r3ZwCv72r/SnVcCmyTZIdhFChJkjRIqarB7zS5CbgHKOCfq+rkJPdW1TbN8gD3VNU2Sc4DTqyqS5plFwHvr6orVurzKDpH3Jg6dere8+bN61v9C269r299T50MdzzUn7733HHr/nQ8xi1dupStttpq2GVsVBzzwXPMB88xH7yxOub77bfflV0zjCvYdNDFNF5aVbcm+S/A/CS/6F5YVZVkrRJkVZ0MnAwwc+bMmjVr1gYrdmWzjzm/b33P2XMZJy3oz49l0WGz+tLvWDcyMkI/f1/0ZI754Dnmg+eYD954HPOhTH1W1a3Nn78BvgXsA9yxfEqz+fM3zeq3Ajt1bT6taZMkSRrXBh7UkmyZZMry98D+wDXAucARzWpHAOc0788F3txc/fki4L6qum3AZUuSJA3cMKY+pwLf6pyGxqbAv1bVd5JcDpyV5EjgZuCNzfoXAAcCC4EHgbcMvmQN2/Q+Tzf3azp70Ymv6Uu/kqSNw8CDWlXdCDx/lPa7gFeM0l7AOwdQmiRJUqu06fYckiRJ6mJQkyRJaimDmiRJUksZ1CRJklrKoCZJktRSBjVJkqSWMqhJkiS1lEFNkiSppQxqkiRJLWVQkyRJaimDmiRJUksZ1CRJklrKoCZJktRSBjVJkqSWMqhJkiS1lEFNkiSppQxqkiRJLWVQkyRJaimDmiRJUp3GXRUAAAftSURBVEsZ1CRJklpq02EXIKmdph9zft/6nrPnMmb3qf9FJ76mL/1K0jB4RE2SJKmlDGqSJEktZVCTJElqKYOaJElSSxnUJEmSWsqrPiWpJbzSVtLKPKImSZLUUgY1SZKkljKoSZIktZRBTZIkqaXGTFBLckCSG5IsTHLMsOuRJEnqtzER1JJMAD4PvBrYHTg0ye7DrUqSJKm/xsrtOfYBFlbVjQBJ5gEHAdcNtSpJ0pjmLVHUdqmqYdewRkkOBg6oqrc1nw8H9q2qo7vWOQo4qvn4bOCGgRe6YWwP/HbYRWxkHPPBc8wHzzEfPMd88MbqmO9SVU8bbcFYOaK2RlV1MnDysOtYX0muqKqZw65jY+KYD55jPniO+eA55oM3Hsd8TJyjBtwK7NT1eVrTJkmSNG6NlaB2ObBbkl2TbAYcApw75JokSZL6akxMfVbVsiRHAxcCE4DTquraIZfVL2N++nYMcswHzzEfPMd88BzzwRt3Yz4mLiaQJEnaGI2VqU9JkqSNjkFNkiSppQxqkiRJLTUmLiYYz5I8B9gR+I+qWtrVfkBVfWd4lUkbTpJ9gKqqy5vHvx0A/KKqLhhyaeNW83fLQXT+foHOLY3Orarrh1fVxiPJV6rqzcOuY2OS5KV0nmR0TVV9d9j1bCheTDBESd4FvBO4HpgBvLuqzmmWXVVVLxhmfRubJG+pqi8Pu47xJsmxdJ7TuykwH9gX+AHwSuDCqjphiOWNS0neDxwKzAMWN83T6NzaaF5VnTis2sajJCvfLirAfsD3AarqdQMvaiOQ5LKq2qd5/3Y6/55+C9gf+Lfx8ntuUBuiJAuAF1fV0iTTgW8A/1JVn0nys6raa6gFbmSS/Kqqdh52HeNN83s+A9gcuB2YVlX3J5lM50jyHw21wHEoyX8Ce1TVoyu1bwZcW1W7Daey8SnJVXSePf0loOgEtbl0gjFV9cPhVTd+df87meRy4MCqujPJlsClVbXncCvcMJz6HK5Nlk93VtWiJLOAbyTZhc5/6NrAkvx8VYuAqYOsZSOyrKoeAx5M8n+r6n6AqnooyeNDrm28ehx4OnDzSu07NMu0Yc0E3g18EHhfVV2d5CEDWt9tkuSpdM63T1XdCVBVDyRZNtzSNhyD2nDdkWRGVV0N0BxZey1wGjAu/k+ghaYCrwLuWak9wE8GX85G4XdJtqiqB4G9lzcm2RpDQ7+8B7goyS+BW5q2nYFnAkcPrapxqqoeBz6d5OvNn3fgv6+DsDVwJZ2/vyvJDlV1W5KtGEcHO5z6HKIk0+gcbbh9lGUvqaofD6GscS3JqcCXq+qSUZb9a1X9xRDKGteSbF5Vj4zSvj2wQ1UtGEJZ416STeicWN19McHlzdFN9VGS1wAvqaoPDLuWjVGSLYCpVXXTsGvZEAxqkiRJLeV91CRJklrKoCZJktRSBjVJ406SkSQzB7CfdyW5PsnXVmqfmeSz/d6/pPHPq1IkqUuSTauq10v7/xr471W1uLuxqq4ArtjgxUna6HhETdJQJJneHI06Jcm1Sb7b3AR3hSNiSbZPsqh5PzvJt5PMT7IoydFJ/jbJz5JcmmTbrl0cnuTqJNc0j7AiyZZJTktyWbPNQV39npvk+8BFo9T6t00/1yR5T9P2v4E/BP49yXtXWn9WkvOa98clOSPJxUluTvKGJJ9IsiDJd5JMbNb7SJLLm32cnCRN+wuT/Lz5Lv+Y5JqmfULz+fJm+Tua9h2S/Kjru//xBvqRSRoCg5qkYdoN+HxV7QHcC/xZD9s8D3gD8ELgBODB5u7kPwW6n624RVXNoHPU67Sm7YPA95vHzuwH/GNzF3OAFwAHV9V/695Zkr2Bt9B59NWLgLcn2auq/gfwa2C/qvr0Gmp+BvBy4HXAV4EfNHdNfwh4TbPO/6qqF1bV84DJwGub9i8D72i+S/etNY4E7quqFzZj8fYkuwJ/QefRXDOA5wNXr6E2SS3m1KekYbpp+Q2f6dy4cnoP2/ygqpYAS5LcB/xb074A6H4c1VyAqvpRkqck2YbOMwBfl+TvmnUm0bkRLMD8qrp7lP29FPhWVT0AkOSbwB8DP+vlCzb+vaoebR6nNQH4TlfN05v3+yX5e2ALYFvg2iQXA1Oq6qfNOv/K7wPc/sAfJTm4+bw1neB7OXBac6Tu213jK2kMMqhJGqbuG+E+RudIEsAyfn/Ef9Jqtnm86/PjrPh32so3iVz+DMY/q6obuhck2Rd4YK0qXzuPQOcO9kkerd/fwPJxYNMkk4AvADOr6pYkx/Hk772yAH9TVRc+aUHyMjpH6k5P8qmq+sqG+iKSBsupT0lttIjfP27q4NWstzpvAkjyUjpThPcBFwJ/03X+11499HMx8PokWzTTpH/atG1Iy0PZb5vH3xwMUFX30jlyuG+z/JCubS4E/qrrHLdnNefg7QLcUVWn0HlI+As2cK2SBsgjapLa6JPAWUmOAs5fxz4eTvIzYCLw1qbtY8A/AT9vHrF0E7+fShxVVV2V5HTgsqbpS1W1NtOea1RV9yY5BbgGuJ3O9OVyRwKnpPMA+x8C9y2vg8606VVN8LwTeD0wC3hfkkeBpax43p6kMcZHSElSiyXZqqqWNu+PofN81HcPuSxJA+IRNUlqt9ck+Qc6f1/fDMwebjmSBskjapIkSS3lxQSSJEktZVCTJElqKYOaJElSSxnUJEmSWsqgJkmS1FIGNUmSpJb6f5Z/4ekK0wBjAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "df.IMAGES.apply(lambda x:len(x)).value_counts().plot(figsize = (10,5),kind = 'bar',legend =True,grid =True,title='Number of images per report')\n",
    "plt.xlabel('number of images')\n",
    "plt.ylabel('image count')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "id": "5d4Nxzxmv_qs"
   },
   "outputs": [],
   "source": [
    "img_count_0 = df.IMAGES.apply(lambda x:len(x)==0)#counting which datapoint does not have image attached to it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "1tG79AEsv_q8"
   },
   "outputs": [],
   "source": [
    "idx = []\n",
    "for i in img_count_0.index.values:\n",
    "    if img_count_0[i]==True:\n",
    "        idx.append(i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "id": "nGfKIqxfv_rC"
   },
   "outputs": [],
   "source": [
    "#dropping rows whocse which does not have images attached to it.\n",
    "df.drop(idx, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "xIiGWmtMv_rG",
    "outputId": "85ce2b80-57e0-46eb-a5ca-6e640d3523e5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3851\n"
     ]
    }
   ],
   "source": [
    "print(len(df))#length of dataframe after removing datapoints without images in it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 237
    },
    "id": "gF8Avtrwp9FB",
    "outputId": "655549ad-37db-45c6-dd25-822002aa020f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 3851 entries, 1 to 3999\n",
      "Data columns (total 5 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   IMAGES      3851 non-null   object\n",
      " 1   COMPARISON  3252 non-null   object\n",
      " 2   INDICATION  3765 non-null   object\n",
      " 3   FINDINGS    3337 non-null   object\n",
      " 4   IMPRESSION  3820 non-null   object\n",
      "dtypes: object(5)\n",
      "memory usage: 180.5+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 293
    },
    "id": "Sj0qkZfFv_rM",
    "outputId": "74f29cab-33a6-4843-b75f-d0c6080cda2a"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>IMAGES</th>\n",
       "      <th>COMPARISON</th>\n",
       "      <th>INDICATION</th>\n",
       "      <th>FINDINGS</th>\n",
       "      <th>IMPRESSION</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[/content/images/CXR1_1_IM-0001-3001.png, /con...</td>\n",
       "      <td>None.</td>\n",
       "      <td>Positive TB test</td>\n",
       "      <td>The cardiac silhouette and mediastinum size ar...</td>\n",
       "      <td>Normal chest x-XXXX.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[/content/images/CXR2_IM-0652-1001.png, /conte...</td>\n",
       "      <td>None.</td>\n",
       "      <td>Preop bariatric surgery.</td>\n",
       "      <td>Borderline cardiomegaly. Midline sternotomy XX...</td>\n",
       "      <td>No acute pulmonary findings.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[/content/images/CXR3_IM-1384-1001.png, /conte...</td>\n",
       "      <td>None</td>\n",
       "      <td>rib pain after a XXXX, XXXX XXXX steps this XX...</td>\n",
       "      <td>None</td>\n",
       "      <td>No displaced rib fractures, pneumothorax, or p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[/content/images/CXR4_IM-2050-1001.png, /conte...</td>\n",
       "      <td>None available</td>\n",
       "      <td>XXXX-year-old XXXX with XXXX.</td>\n",
       "      <td>There are diffuse bilateral interstitial and a...</td>\n",
       "      <td>1. Bullous emphysema and interstitial fibrosis...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[/content/images/CXR5_IM-2117-1003002.png, /co...</td>\n",
       "      <td>None</td>\n",
       "      <td>Chest and nasal congestion.</td>\n",
       "      <td>The cardiomediastinal silhouette and pulmonary...</td>\n",
       "      <td>No acute cardiopulmonary abnormality.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              IMAGES  ...                                         IMPRESSION\n",
       "1  [/content/images/CXR1_1_IM-0001-3001.png, /con...  ...                               Normal chest x-XXXX.\n",
       "2  [/content/images/CXR2_IM-0652-1001.png, /conte...  ...                       No acute pulmonary findings.\n",
       "3  [/content/images/CXR3_IM-1384-1001.png, /conte...  ...  No displaced rib fractures, pneumothorax, or p...\n",
       "4  [/content/images/CXR4_IM-2050-1001.png, /conte...  ...  1. Bullous emphysema and interstitial fibrosis...\n",
       "5  [/content/images/CXR5_IM-2117-1003002.png, /co...  ...              No acute cardiopulmonary abnormality.\n",
       "\n",
       "[5 rows x 5 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.reset_index()#resetting index\n",
    "df.sort_index(inplace=True)#sorting indeces\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 127
    },
    "id": "_BAMnPAov_rT",
    "outputId": "ab6a53a6-cf53-4100-f4e8-beca72088f2d"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "IMAGES          0\n",
       "COMPARISON    599\n",
       "INDICATION     86\n",
       "FINDINGS      514\n",
       "IMPRESSION     31\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isna().sum()#counting nan's in each column"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "wrFe3MQSv_re"
   },
   "source": [
    "### 3.2.2 COLUMN COMPARISON:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 219
    },
    "id": "8WMlX-kQv_rf",
    "outputId": "472fee24-77e0-411f-c4b5-fd54283015dd"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "None.                                 814\n",
       "NaN                                   599\n",
       "None                                  567\n",
       "XXXX                                  280\n",
       "XXXX, XXXX.                           220\n",
       "XXXX, XXXX                            159\n",
       "None available.                       126\n",
       "XXXX.                                 119\n",
       "None available                         81\n",
       "Two views of the chest dated XXXX.     23\n",
       "Name: COMPARISON, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#displaying top 10 datapoints with highest number of occurences\n",
    "df.COMPARISON.value_counts(dropna = False)[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "vYOcoFINv_rk",
    "outputId": "377d0931-4824-4c82-a108-75b31af99e6d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of None values in COMPARISION column: 2389\n"
     ]
    }
   ],
   "source": [
    "#summing up top 9 valuecounts to see how many are not useful\n",
    "print('number of None values in COMPARISION column:',df.COMPARISON.value_counts()[:9].sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YlZaIR_3v_rw"
   },
   "source": [
    "### 3.2.3 COLUMN INDICATION:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 311
    },
    "id": "f4cgVMoJv_rx",
    "outputId": "246bdb04-7157-4cce-d553-76810ee833cb"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Chest pain                               129\n",
       "XXXX                                     114\n",
       "Chest pain.                               88\n",
       "NaN                                       86\n",
       "chest pain                                62\n",
       "XXXX.                                     58\n",
       "XXXX-year-old female with chest pain.     39\n",
       "XXXX-year-old male with chest pain.       38\n",
       "Dyspnea                                   37\n",
       "Shortness of breath                       31\n",
       "dyspnea                                   29\n",
       "Shortness of breath.                      23\n",
       "XXXX-year-old female with chest pain      19\n",
       "XXXX-year-old woman with chest pain.      19\n",
       "XXXX-year-old female with XXXX.           19\n",
       "Name: INDICATION, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#displaying top 15 datapoints with highest number of occurences\n",
    "df.INDICATION.value_counts(dropna = False)[:15]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bL9t77Avv_r2"
   },
   "source": [
    "### 3.2.4 COLUMN FINDINGS:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 239
    },
    "id": "J2djAxJYv_r6",
    "outputId": "3f595080-6ade-4682-c066-2f847e3f9f1b",
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NaN                                                                                                                                                                                                                                                 514\n",
       "The heart and lungs have XXXX XXXX in the interval. Both lungs are clear and expanded. Heart and mediastinum normal.                                                                                                                                 51\n",
       "The heart is normal in size. The mediastinum is unremarkable. The lungs are clear.                                                                                                                                                                   51\n",
       "Heart size normal. Lungs are clear. XXXX are normal. No pneumonia, effusions, edema, pneumothorax, adenopathy, nodules or masses.                                                                                                                    46\n",
       "The lungs are clear bilaterally. Specifically, no evidence of focal consolidation, pneumothorax, or pleural effusion.. Cardio mediastinal silhouette is unremarkable. Visualized osseous structures of the thorax are without acute abnormality.     45\n",
       "Cardiac and mediastinal contours are within normal limits. The lungs are clear. Bony structures are intact.                                                                                                                                          35\n",
       "Both lungs are clear and expanded. Heart and mediastinum normal.                                                                                                                                                                                     33\n",
       "The cardiomediastinal silhouette and pulmonary vasculature are within normal limits in size. The lungs are clear of focal airspace disease, pneumothorax, or pleural effusion. There are no acute bony findings.                                     26\n",
       "Lungs are clear bilaterally. Cardiac and mediastinal silhouettes are normal. Pulmonary vasculature is normal. No pneumothorax or pleural effusion. No acute bony abnormality.                                                                        21\n",
       "The heart size and pulmonary vascularity appear within normal limits. The lungs are free of focal airspace disease. No pleural effusion or pneumothorax is seen.                                                                                     20\n",
       "Name: FINDINGS, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#displaying 10 datapoints with highest number of occurences\n",
    "df.FINDINGS.value_counts(dropna = False)[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "oBIwCsH8v_sE"
   },
   "source": [
    "### 3.2.5 COLUMN IMPRESSION(TARGET):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 403
    },
    "id": "CYucM678v_sE",
    "outputId": "43ec00e7-bad0-44f2-bb57-c25eb58159f0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "No acute cardiopulmonary abnormality.                301\n",
       "No active disease.                                   127\n",
       "No acute cardiopulmonary abnormalities.              112\n",
       "No acute cardiopulmonary findings.                   111\n",
       "No acute disease.                                    109\n",
       "No acute cardiopulmonary disease.                     99\n",
       "No acute cardiopulmonary process.                     66\n",
       "1. No acute radiographic cardiopulmonary process.     66\n",
       "No evidence of active disease.                        47\n",
       "No acute cardiopulmonary abnormality. .               47\n",
       "1. No evidence of active disease.                     44\n",
       "No acute pulmonary disease.                           44\n",
       "No acute cardiopulmonary abnormality..                43\n",
       "Normal chest                                          36\n",
       "No acute cardiopulmonary findings. .                  35\n",
       "NaN                                                   31\n",
       "No acute findings                                     30\n",
       "Negative for acute abnormality.                       30\n",
       "No acute cardiopulmonary finding.                     29\n",
       "No acute findings.                                    29\n",
       "Name: IMPRESSION, dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#displaying 10 datapoints with highest number of occurences\n",
    "df.IMPRESSION.value_counts(dropna = False)[:20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 239
    },
    "id": "EmSmj7wkxKIf",
    "outputId": "94dde82a-6604-44dd-c2bc-32d4b50cb0ec"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "No focal lung infiltrates.                                                                                                                                                                                                                        1\n",
       "Stable left lower lung XXXX calcified granuloma. Otherwise lungs are clear. Left-sided perihilar calcified lymph node, unchanged. No XXXX of the pleural effusions. No XXXX of active cardiopulmonary disease. No XXXX of active tuberculosis.    1\n",
       "1. Mildly elevated right hemidiaphragm. Otherwise no acute cardiopulmonary abnormality seen.                                                                                                                                                      1\n",
       "Negative preoperative chest x-XXXX.                                                                                                                                                                                                               1\n",
       "1. Vascular congestion and diffuse interstitial edema 2. Stable cardiomegaly 3. Low lung volumes                                                                                                                                                  1\n",
       "1. Blunting of the right costophrenic XXXX. This may represent pleural effusion or pleural reaction. 2. Clearing of left base airspace disease. 3. Outward XXXX of PICC line. Tip now projects over upper SVC.                                    1\n",
       "Slight cardiomegaly. Lungs are clear. No rib abnormalities are seen. The lower ribs are not adequately imaged to rule out pathology on the chest film                                                                                             1\n",
       "COPD and probable right basilar scarring. No acute pulmonary disease identified.                                                                                                                                                                  1\n",
       "1. Cardiomegaly                                                                                                                                                                                                                                   1\n",
       "No focal infiltrate.                                                                                                                                                                                                                              1\n",
       "Name: IMPRESSION, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#displaying 10 datapoints with highest number of occurences\n",
    "df.IMPRESSION.value_counts()[-10:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 867
    },
    "id": "S66XLJL5v_sT",
    "outputId": "e06d10d1-745d-4836-e85e-b9a60e32eb39"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "df.IMPRESSION.value_counts(dropna = False)[:20].plot(figsize = (20,10),kind = 'bar',legend =True,grid =True,title='20-Most Occuring Impressions')\n",
    "plt.xlabel('impressions')\n",
    "plt.ylabel('count')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "id": "GxGp4jSWv-xU"
   },
   "outputs": [],
   "source": [
    "#filling NaN's with string 'no impressioni so that data does not get wasted'\n",
    "df.fillna('no impression', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "id": "j2G1Dz-qv_sY"
   },
   "outputs": [],
   "source": [
    "#getting word counts of each datapoint in impression column\n",
    "word_length = df.IMPRESSION.str.split().apply(lambda x:len(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 90
    },
    "id": "tPirwi-cv_sc",
    "outputId": "2a0e7e0e-c780-4105-aeb0-9e69c06f8a0f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean of words are in single impression are : 10.493378343287457\n",
      "median of words are in single impression are: 5.0\n",
      "maximum number of words are in single impression are : 130\n",
      "minimum number of words are in single impression are : 1\n"
     ]
    }
   ],
   "source": [
    "print('mean of words are in single impression are :',word_length.mean())\n",
    "print('median of words are in single impression are:',word_length.median())\n",
    "print('maximum number of words are in single impression are :',word_length.max())\n",
    "print('minimum number of words are in single impression are :',word_length.min())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 633
    },
    "id": "RkA276Wmr-Qm",
    "outputId": "f31abd75-6a6d-487e-822f-ffcbaf09dcd0"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "plt.figure(figsize = (20,10))\n",
    "plt.hist(word_length)\n",
    "plt.xlabel(\"count\")\n",
    "plt.ylabel(\"impression\")\n",
    "plt.title('number of words in each sentence')\n",
    "plt.tight_layout()\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7qDLWTiSv_sh"
   },
   "source": [
    "## Findings from EDA:\n",
    "1. I think i will only keep the impressions and image columns others have many null values.\n",
    "2. In impression column there were nan values too 30 around i dropped them as of now but i think i can replace nans with no impression and pass it to the model.\n",
    "3. Mostly the reports are normal only some are abnormal so data bias is there.\n",
    "4. Mainly there are 2 images per report so i will pass 2 images only, if less than that then i will repeat the 1st image itself twice. there fewer occurences of 3 and 4 images per image i will keep first two and discard the rest.there is only one report with 5 images so will perform same actions as for 2-4 images.,"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> 3.3 Data Cleaning </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 293
    },
    "id": "8MLJ6nVbHJ5r",
    "outputId": "875f8b52-eea1-4455-9ffd-99642b71c0f5"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>IMAGES</th>\n",
       "      <th>COMPARISON</th>\n",
       "      <th>INDICATION</th>\n",
       "      <th>FINDINGS</th>\n",
       "      <th>IMPRESSION</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[/content/images/CXR1_1_IM-0001-3001.png, /con...</td>\n",
       "      <td>None.</td>\n",
       "      <td>Positive TB test</td>\n",
       "      <td>The cardiac silhouette and mediastinum size ar...</td>\n",
       "      <td>Normal chest x-XXXX.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[/content/images/CXR2_IM-0652-1001.png, /conte...</td>\n",
       "      <td>None.</td>\n",
       "      <td>Preop bariatric surgery.</td>\n",
       "      <td>Borderline cardiomegaly. Midline sternotomy XX...</td>\n",
       "      <td>No acute pulmonary findings.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[/content/images/CXR3_IM-1384-1001.png, /conte...</td>\n",
       "      <td>no impression</td>\n",
       "      <td>rib pain after a XXXX, XXXX XXXX steps this XX...</td>\n",
       "      <td>no impression</td>\n",
       "      <td>No displaced rib fractures, pneumothorax, or p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[/content/images/CXR4_IM-2050-1001.png, /conte...</td>\n",
       "      <td>None available</td>\n",
       "      <td>XXXX-year-old XXXX with XXXX.</td>\n",
       "      <td>There are diffuse bilateral interstitial and a...</td>\n",
       "      <td>1. Bullous emphysema and interstitial fibrosis...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[/content/images/CXR5_IM-2117-1003002.png, /co...</td>\n",
       "      <td>no impression</td>\n",
       "      <td>Chest and nasal congestion.</td>\n",
       "      <td>The cardiomediastinal silhouette and pulmonary...</td>\n",
       "      <td>No acute cardiopulmonary abnormality.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              IMAGES  ...                                         IMPRESSION\n",
       "1  [/content/images/CXR1_1_IM-0001-3001.png, /con...  ...                               Normal chest x-XXXX.\n",
       "2  [/content/images/CXR2_IM-0652-1001.png, /conte...  ...                       No acute pulmonary findings.\n",
       "3  [/content/images/CXR3_IM-1384-1001.png, /conte...  ...  No displaced rib fractures, pneumothorax, or p...\n",
       "4  [/content/images/CXR4_IM-2050-1001.png, /conte...  ...  1. Bullous emphysema and interstitial fibrosis...\n",
       "5  [/content/images/CXR5_IM-2117-1003002.png, /co...  ...              No acute cardiopulmonary abnormality.\n",
       "\n",
       "[5 rows x 5 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.3.1 Dropping 'COMPARISON','INDICATION','FINDINGS' columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "MMWrY3_LG__E",
    "outputId": "bbb46a93-a651-45f0-d7fd-4ec4aff86871"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>IMAGES</th>\n",
       "      <th>IMPRESSION</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[/content/images/CXR1_1_IM-0001-3001.png, /con...</td>\n",
       "      <td>Normal chest x-XXXX.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[/content/images/CXR2_IM-0652-1001.png, /conte...</td>\n",
       "      <td>No acute pulmonary findings.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[/content/images/CXR3_IM-1384-1001.png, /conte...</td>\n",
       "      <td>No displaced rib fractures, pneumothorax, or p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[/content/images/CXR4_IM-2050-1001.png, /conte...</td>\n",
       "      <td>1. Bullous emphysema and interstitial fibrosis...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>[/content/images/CXR5_IM-2117-1003002.png, /co...</td>\n",
       "      <td>No acute cardiopulmonary abnormality.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              IMAGES                                         IMPRESSION\n",
       "1  [/content/images/CXR1_1_IM-0001-3001.png, /con...                               Normal chest x-XXXX.\n",
       "2  [/content/images/CXR2_IM-0652-1001.png, /conte...                       No acute pulmonary findings.\n",
       "3  [/content/images/CXR3_IM-1384-1001.png, /conte...  No displaced rib fractures, pneumothorax, or p...\n",
       "4  [/content/images/CXR4_IM-2050-1001.png, /conte...  1. Bullous emphysema and interstitial fibrosis...\n",
       "5  [/content/images/CXR5_IM-2117-1003002.png, /co...              No acute cardiopulmonary abnormality."
      ]
     },
     "execution_count": 42,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#dropping columns as it has many null values\n",
    "df.drop(labels = ['COMPARISON'],inplace = True,axis = 1)\n",
    "df.drop(labels = ['INDICATION'],inplace = True,axis = 1)\n",
    "df.drop(labels = ['FINDINGS'],inplace = True,axis = 1)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "id": "YawB155cDgDb",
    "outputId": "8c91e4d4-1da9-4d29-b702-1a97b2b92a60"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['/content/images/CXR1_1_IM-0001-3001.png',\n",
       " '/content/images/CXR1_1_IM-0001-4001.png']"
      ]
     },
     "execution_count": 43,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['IMAGES'][1]#sample datapoint "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.3.2 storing trimmed data into pickle file. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "id": "LLf2z6xvr__Z"
   },
   "outputs": [],
   "source": [
    "df.to_pickle('/content/drive/My Drive/case_study_2/extract')# storing dataframe to pickle file so as to use further"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "id": "zL-1-ftxLS1s"
   },
   "outputs": [],
   "source": [
    "df = pd.read_pickle('/content/drive/My Drive/case_study_2/extract')# reading  pickle file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "id": "4TfSAXXXv_sr"
   },
   "outputs": [],
   "source": [
    "df.reset_index(inplace=True,drop = True)# resettnig indeces"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "45EMdQRsC0T-",
    "outputId": "5b469cf2-6447-4778-b4d9-a713eae62d96"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['IMAGES', 'IMPRESSION'], dtype='object')"
      ]
     },
     "execution_count": 47,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#displaying columns\n",
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "H3g9VEltv_sy",
    "outputId": "d066a7cb-722c-486b-e9e6-2f30ec913b0f"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>IMAGES</th>\n",
       "      <th>IMPRESSION</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[/content/images/CXR1_1_IM-0001-3001.png, /con...</td>\n",
       "      <td>Normal chest x-XXXX.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[/content/images/CXR2_IM-0652-1001.png, /conte...</td>\n",
       "      <td>No acute pulmonary findings.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[/content/images/CXR3_IM-1384-1001.png, /conte...</td>\n",
       "      <td>No displaced rib fractures, pneumothorax, or p...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[/content/images/CXR4_IM-2050-1001.png, /conte...</td>\n",
       "      <td>1. Bullous emphysema and interstitial fibrosis...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[/content/images/CXR5_IM-2117-1003002.png, /co...</td>\n",
       "      <td>No acute cardiopulmonary abnormality.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              IMAGES                                         IMPRESSION\n",
       "0  [/content/images/CXR1_1_IM-0001-3001.png, /con...                               Normal chest x-XXXX.\n",
       "1  [/content/images/CXR2_IM-0652-1001.png, /conte...                       No acute pulmonary findings.\n",
       "2  [/content/images/CXR3_IM-1384-1001.png, /conte...  No displaced rib fractures, pneumothorax, or p...\n",
       "3  [/content/images/CXR4_IM-2050-1001.png, /conte...  1. Bullous emphysema and interstitial fibrosis...\n",
       "4  [/content/images/CXR5_IM-2117-1003002.png, /co...              No acute cardiopulmonary abnormality."
      ]
     },
     "execution_count": 48,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "id": "uIUC-cvQPBvP"
   },
   "outputs": [],
   "source": [
    "#filling nan's with no impression so that these columns do the image related to it doesent go to waste.\n",
    "df.fillna('no impression', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 182
    },
    "id": "k6Anf198OA9M",
    "outputId": "155a4a6f-d777-483e-b1f2-4f52fd5a8789"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 3851 entries, 0 to 3850\n",
      "Data columns (total 2 columns):\n",
      " #   Column      Non-Null Count  Dtype \n",
      "---  ------      --------------  ----- \n",
      " 0   IMAGES      3851 non-null   object\n",
      " 1   IMPRESSION  3851 non-null   object\n",
      "dtypes: object(2)\n",
      "memory usage: 60.3+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "id": "tjpkQv5esOHV",
    "outputId": "16530d50-438d-424e-b5d2-e149ef82927f"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 51,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from wordcloud import WordCloud, STOPWORDS\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem import SnowballStemmer\n",
    "from sklearn.manifold import TSNE\n",
    "import nltk\n",
    "import matplotlib.pyplot as plt\n",
    "nltk.download('stopwords')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3> 3.3.3 most common words </h3>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "id": "LjLap2yxsOFU"
   },
   "outputs": [],
   "source": [
    "stopwords1 = set(STOPWORDS)#setting stopwords"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 589
    },
    "id": "mVBU1lTCMpsv",
    "outputId": "9b018013-1c26-4dfb-8df3-6c0500ee8e7c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Word Cloud for IMPRESSION \n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# generating word cloud of most occuring words ignoring stop words \n",
    "wc = WordCloud(background_color=\"black\", max_words=len(df),width=1600,\n",
    "                          height=800, stopwords=stopwords1)\n",
    "wc.generate(str(df[\"IMPRESSION\"]))\n",
    "print(\"Word Cloud for IMPRESSION \")\n",
    "fig = plt.figure(figsize=(15,10))\n",
    "plt.imshow(wc, interpolation='bilinear')\n",
    "plt.axis(\"off\")\n",
    "plt.tight_layout(pad=0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2> 3.4 Data Preprocessing </h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "id": "V5gsb_8YTxUJ"
   },
   "outputs": [],
   "source": [
    "# replacing short words with their full words eg: won't to will not etc\n",
    "def decontracted(phrase):\n",
    "    # specific\n",
    "    phrase = re.sub(r\"won't\", \"will not\", phrase)\n",
    "    phrase = re.sub(r\"can\\'t\", \"can not\", phrase)\n",
    "\n",
    "    # general\n",
    "    phrase = re.sub(r\"n\\'t\", \" not\", phrase)\n",
    "    phrase = re.sub(r\"\\'re\", \" are\", phrase)\n",
    "    phrase = re.sub(r\"\\'s\", \" is\", phrase)\n",
    "    phrase = re.sub(r\"\\'d\", \" would\", phrase)\n",
    "    phrase = re.sub(r\"\\'ll\", \" will\", phrase)\n",
    "    phrase = re.sub(r\"\\'t\", \" not\", phrase)\n",
    "    phrase = re.sub(r\"\\'ve\", \" have\", phrase)\n",
    "    phrase = re.sub(r\"\\'m\", \" am\", phrase)\n",
    "    return phrase"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "-nD2iX_pMpko",
    "outputId": "f7b5091a-dc71-4eb1-963b-764f7267456e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No radiographic evidence of acute cardiopulmonary disease\n"
     ]
    }
   ],
   "source": [
    "text = df['IMPRESSION'][1057]\n",
    "print(text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "id": "FN9crQ0jMpii"
   },
   "outputs": [],
   "source": [
    "import re\n",
    "\n",
    "def preprocess(text):\n",
    "  text = re.sub('XXXX','',text)#substituting XXXX string which occurs n most of the impressions with space as it has no semantic meaning\n",
    "  text=re.sub(r'[^A-Za-z]+',' ',text)#replacing anything other than words wtih space\n",
    "  text = decontracted(text)#decontracting the words\n",
    "  text = re.sub('\\\\r', ' ',text)#replacing new line with space\n",
    "  text = re.sub('\\\\n', ' ',text)#replacing tab with single space\n",
    "  \n",
    "  return text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 90
    },
    "id": "lL69v8aHP0Nz",
    "outputId": "fda0dd1b-6a0d-4085-9c48-529ce5b04a71"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "before preprocessing \n",
      " Congestive heart failure with basilar pulmonary interstitial edema and bilateral pleural fluid.\n",
      "after preprocessing \n",
      " Congestive heart failure with basilar pulmonary interstitial edema and bilateral pleural fluid \n"
     ]
    }
   ],
   "source": [
    "print('before preprocessing \\n',df['IMPRESSION'][1065])\n",
    "clean = preprocess(df['IMPRESSION'][1065])\n",
    "print('after preprocessing \\n',clean)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.4.1 Preprocessing  IMPRESSION column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "id": "w06DGy8ZMpek"
   },
   "outputs": [],
   "source": [
    "df['IMPRESSION'] = df['IMPRESSION'].map(preprocess)# mapping whole column to the preprocess function , it will apply this function to whole column line by line"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.4.2 Preprocessing  IMAGES column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "id": "1F_emW7ZMpcL"
   },
   "outputs": [],
   "source": [
    "#keeping only first two images per datapoints\n",
    "df['IMAGES'] = df['IMAGES'].apply(lambda x:x[:2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "Kafr7ADXsNGt",
    "outputId": "15718f37-7371-46c5-cd61-f242e1c2afc1"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>IMAGES</th>\n",
       "      <th>IMPRESSION</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>[/content/images/CXR1_1_IM-0001-3001.png, /con...</td>\n",
       "      <td>Normal chest x</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>[/content/images/CXR2_IM-0652-1001.png, /conte...</td>\n",
       "      <td>No acute pulmonary findings</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>[/content/images/CXR3_IM-1384-1001.png, /conte...</td>\n",
       "      <td>No displaced rib fractures pneumothorax or ple...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>[/content/images/CXR4_IM-2050-1001.png, /conte...</td>\n",
       "      <td>Bullous emphysema and interstitial fibrosis P...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>[/content/images/CXR5_IM-2117-1003002.png, /co...</td>\n",
       "      <td>No acute cardiopulmonary abnormality</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              IMAGES                                         IMPRESSION\n",
       "0  [/content/images/CXR1_1_IM-0001-3001.png, /con...                                    Normal chest x \n",
       "1  [/content/images/CXR2_IM-0652-1001.png, /conte...                       No acute pulmonary findings \n",
       "2  [/content/images/CXR3_IM-1384-1001.png, /conte...  No displaced rib fractures pneumothorax or ple...\n",
       "3  [/content/images/CXR4_IM-2050-1001.png, /conte...   Bullous emphysema and interstitial fibrosis P...\n",
       "4  [/content/images/CXR5_IM-2117-1003002.png, /co...              No acute cardiopulmonary abnormality "
      ]
     },
     "execution_count": 60,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "id": "YBbTQtg3DQnP"
   },
   "outputs": [],
   "source": [
    "# for rows having only one image we are copying that image twice\n",
    "for i in range(len(df['IMAGES'])):\n",
    "  if len(df['IMAGES'][i])<2:\n",
    "    df['IMAGES'][i] = [df['IMAGES'][i][0],df['IMAGES'][i][0]]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "id": "tPPDtarzsNEt"
   },
   "outputs": [],
   "source": [
    "#finding heights and widths of all the images\n",
    "heights = []\n",
    "widths = []\n",
    "import cv2\n",
    "for i in range(len(df['IMAGES'])):\n",
    "  for j in range(len(df['IMAGES'][i])):\n",
    "    img = cv2.imread(df['IMAGES'][i][j], cv2.IMREAD_UNCHANGED) #reading/loading image\n",
    "    height, width, channels = img.shape #getting width and height of image\n",
    "    heights.append(height)\n",
    "    widths.append(width)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 90
    },
    "id": "TbzTeZ5fsNBl",
    "outputId": "dc9946bc-034b-4ccf-9582-bbbd9956bf30"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean of heights:  531.6648922357829\n",
      "median of heights:  512.0\n",
      "mean of widths:  512\n",
      "median of widths:  512.0\n"
     ]
    }
   ],
   "source": [
    "from statistics import mean,median\n",
    "print('mean of heights: ',mean(heights))\n",
    "print('median of heights: ',median(heights))\n",
    "\n",
    "print('mean of widths: ',mean(widths))\n",
    "print('median of widths: ',median(widths))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "-aGqBXGNPuMw",
    "outputId": "0babf961-2f14-414f-b735-fd4b0d670094"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 3851/3851 [00:00<00:00, 1324716.21it/s]\n"
     ]
    }
   ],
   "source": [
    "#checking if there are only two images per datapoint\n",
    "from tqdm import tqdm\n",
    "image_paths = []\n",
    "for lst_of_img in tqdm(df[\"IMAGES\"]):\n",
    "  for img in lst_of_img:\n",
    "    image_paths.append(img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "bMDhStSC2KJ_",
    "outputId": "f5861394-1c45-4581-8631-7f3814c87060"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7702"
      ]
     },
     "execution_count": 65,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#so we have 7702 image_paths in total and we have3 3851 datapoints in our dataframe which is half of that\n",
    "len(image_paths)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "id": "BXU6ysPBE3L_"
   },
   "outputs": [],
   "source": [
    "word_len = []\n",
    "for i in range(len(df)):\n",
    "  word_len.append(len(df['IMPRESSION'][i].split()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 678
    },
    "id": "xmC6D6cVve7H",
    "outputId": "b495384f-badf-4348-e18b-c4e047005add"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/seaborn/distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt#matplot lib is also a plotting library\n",
    "import seaborn as sns#seaborn is a plotting library\n",
    "plt.figure(figsize = (20,10))\n",
    "sns.distplot(word_len)\n",
    "plt.grid()\n",
    "plt.xlabel(\"count\")\n",
    "plt.ylabel(\"impression\")\n",
    "plt.title('number of words in each sentence')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "A6FqiFk3FLP2",
    "outputId": "6d9467f7-b847-47c0-b8ca-f96b6799e063"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "99 percent of impressions has word length less that or equal to :  123\n"
     ]
    }
   ],
   "source": [
    "#finding 99 percentile value of word_lengths of each ipression so as to see that how many words does 99 percent of them contain\n",
    "import numpy as np\n",
    "max_length= int(np.percentile(word_len,100))\n",
    "print('99 percent of impressions has word length less that or equal to : ',max_length)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.4.3 Adding special string to IMPRESSION column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "id": "qIcsO49cPmol"
   },
   "outputs": [],
   "source": [
    "# inserting starting <start> and <end> string in impression column\n",
    "imp_inp = []\n",
    "imp_out = []\n",
    "for i in range(len(df['IMPRESSION'])):\n",
    "  imp_inp.append('<start> '+df['IMPRESSION'][i])#we will passs <start> token follwed by sentence as input to model \n",
    "  imp_out.append(df['IMPRESSION'][i] + ' <end>')#we will expect whole sentence and <end> token as output\n",
    "  df['IMPRESSION'][i]= '<start> ' +df['IMPRESSION'][i] + ' <end>' #sentence with whole <start> and <end> token"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "id": "uJInBUxrQ10u",
    "outputId": "9b24c080-14da-4c17-af55-48a2d75f996b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ORIGINAL SEQUENCE:   <start> No acute pulmonary findings  <end>\n",
      "INPUT SEQUENCE:   <start> No acute pulmonary findings \n",
      "OUTPUT_SEQUENCE:   No acute pulmonary findings  <end>\n"
     ]
    }
   ],
   "source": [
    "print('ORIGINAL SEQUENCE:  ',df['IMPRESSION'][1])\n",
    "print('INPUT SEQUENCE:  ',imp_inp[1])\n",
    "print('OUTPUT_SEQUENCE:  ',imp_out[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.4.4 Tokenizinzation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Bp81Dalt8UoK"
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "#tf.random.set_seed(42)\n",
    "tokenizer = tf.keras.preprocessing.text.Tokenizer(split=' ',lower=True,filters='')#initializing tokenizer\n",
    "tokenizer.fit_on_texts(df['IMPRESSION'])#fitting of whole corpus\n",
    "target_vocab_size = len(tokenizer.word_index)+1#defining length of corpus which is 1486 words\n",
    "target_inp = tokenizer.texts_to_sequences(imp_inp)#converting texts into sequence of integers for passing them to model\n",
    "target_inp = tf.keras.preprocessing.sequence.pad_sequences(target_inp,maxlen=max_length,\n",
    "                                                           padding='post')#padding input sequences to 99percentile of max sequence length which we will pass as input to model \n",
    "\n",
    "target_out = tokenizer.texts_to_sequences(imp_out)#converting texts into sequence of integers \n",
    "\n",
    "target_out = tf.keras.preprocessing.sequence.pad_sequences(target_out,maxlen=max_length,\n",
    "                                                           padding='post')#padding output sequences to 99 percentile of max sequence length which we will expect as output  model "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "id": "q7ucfSUBT_y6",
    "outputId": "cbb856e8-266f-46b6-905a-dcf5abaf4162"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<start>\n",
      "<end>\n",
      "no\n"
     ]
    }
   ],
   "source": [
    "#displaying words at 1,2,3 tokens\n",
    "print(tokenizer.index_word[1])\n",
    "print(tokenizer.index_word[2])\n",
    "print(tokenizer.index_word[3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 237
    },
    "id": "8oKQofrHPt8e",
    "outputId": "776126e6-2c81-46e8-aa50-74fc8167e4bb"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sample input sequence [ 1 12 18 88  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0]\n",
      "sample output sequence [12 18 88  2  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0\n",
      "  0  0  0]\n"
     ]
    }
   ],
   "source": [
    "print('sample input sequence',target_inp[0])\n",
    "print('sample output sequence',target_out[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "nqOfE4T6Pt6g",
    "outputId": "e2f4ad24-1b79-41c5-bcd0-96e0b52a2e21"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "vocab_size:  1486\n"
     ]
    }
   ],
   "source": [
    "print('vocab_size: ',target_vocab_size)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1>4. Machine Learning Models </h1>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.1 Split the data into train,validation and test."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "id": "tDcksCz_Pt0L"
   },
   "outputs": [],
   "source": [
    "#dividing dataset into train(90%), validation(9%), test(1%) .\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "img_name_train, img_name_temp, target_train_inp, target_temp_inp,target_train_out,target_temp_out = train_test_split(df['IMAGES'].values,\n",
    "                                                                                                                  target_inp,\n",
    "                                                                                                                  target_out,\n",
    "                                                                                                                  test_size=0.10,\n",
    "                                                                                                                   random_state=0,\n",
    "                                                                                                                    shuffle=True)\n",
    "\n",
    "\n",
    "img_name_val, img_name_test, target_val_inp, target_test_inp,target_val_out,target_test_out = train_test_split(img_name_temp,\n",
    "                                                                                                                  target_temp_inp,\n",
    "                                                                                                                  target_temp_out,\n",
    "                                                                                                                  test_size=0.019,\n",
    "                                                                                                                   random_state=0,\n",
    "                                                                                                                    shuffle=True)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "id": "NvYPNUdvHJkw",
    "outputId": "fb39f519-14e6-4091-877e-1c999aed0498"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input_train_image shape (3465,)\n",
      "target_train_input shape (3465, 123)\n",
      "target_train_output shape (3465, 123)\n"
     ]
    }
   ],
   "source": [
    "print('input_train_image shape',img_name_train.shape)\n",
    "print('target_train_input shape', target_train_inp.shape)\n",
    "print('target_train_output shape',target_train_out.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "id": "jBxYiTIlHJjZ",
    "outputId": "357df09b-2bb9-41b8-a86c-3284f85bb6ff"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input__val_image shape (378,)\n",
      "target_val_input shape (378, 123)\n",
      "target_val_output shape (378, 123)\n"
     ]
    }
   ],
   "source": [
    "print('input__val_image shape',img_name_val.shape)\n",
    "print('target_val_input shape', target_val_inp.shape)\n",
    "print('target_val_output shape',target_val_out.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 72
    },
    "id": "uPd_88PKM6UO",
    "outputId": "5bf97ba3-a8a4-4d96-e29d-3a548d01fca2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input test image shape (8,)\n",
      "target_test_input shape (8, 123)\n",
      "target_test_output shape (8, 123)\n"
     ]
    }
   ],
   "source": [
    "print('input test image shape',img_name_test.shape)\n",
    "print('target_test_input shape', target_test_inp.shape)\n",
    "print('target_test_output shape',target_test_out.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.2 Defining hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "id": "8utsC1XDMUCW"
   },
   "outputs": [],
   "source": [
    "img_height = 224#int(median(heights)) # input image height\n",
    "img_width = 224#int(median(widths)) #input image width\n",
    "\n",
    "BATCH_SIZE = 7 #batch size\n",
    "embedding_dim = 256 #embedding diensions for embedding layer\n",
    "units = 512#256 #lstm units\n",
    "rate = 0.1#dropout rate\n",
    "input_len_dec = target_val_inp.shape[1]#length of input sequence \n",
    "attention_features_shape = 49 #feature shape after squeezing CheX-Net output feature, which will act as hidden dimension for decoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.3 data augmentation "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "id": "ecwOSM-K-4h2"
   },
   "outputs": [],
   "source": [
    "#augmentating input images because we dont have much data to train on.\n",
    "import imgaug.augmenters as iaa\n",
    "aug2 = iaa.Fliplr(1)# flip images left to right or horizontally\n",
    "aug3 = iaa.Flipud(1) # flip images up down or vertically\n",
    "aug4 = iaa.Emboss(alpha=(1), strength=1)#Augmenter that embosses images and overlays the result with the original image.\n",
    "aug5 = iaa.DirectedEdgeDetect(alpha=(0.8), direction=(1.0))#detects edges that have certain directions and marks them in a black and white image and then overlays the result with the original image.\n",
    "aug6 = iaa.Sharpen(alpha=(1.0), lightness=(1.5)) #sharpens images"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.4 loading and preprocessing images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "id": "8CtCXYw4-hJQ"
   },
   "outputs": [],
   "source": [
    "#https://stackoverflow.com/questions/38025838/normalizing-images-in-opencv\n",
    "# loading and preprocessing images\n",
    "def preprocess_img(image_path,training = True):\n",
    "    img = cv2.imread(image_path, cv2.IMREAD_UNCHANGED)# reading image\n",
    "    image = cv2.resize(img,(int(img_width),int(img_height)),interpolation = cv2.INTER_NEAREST)# resizing image as desired\n",
    "    if training == True: # if this function is used during training only them we will augment images\n",
    "      a = np.random.uniform() # selecting random number between 0-1 and applying different augmentations according to its output\n",
    "      if a<0.2:\n",
    "              image = aug2.augment_image(image)\n",
    "      elif a<0.4:\n",
    "              image = aug3.augment_image(image)\n",
    "      elif a<0.6:\n",
    "              image = aug4.augment_image(image)\n",
    "      elif a<0.8:\n",
    "              image = aug5.augment_image(image)\n",
    "      else:\n",
    "              image = aug6.augment_image(image)\n",
    "\n",
    "    image = cv2.normalize(image, None, alpha=0, beta=1, norm_type=cv2.NORM_MINMAX, dtype=cv2.CV_32F)# normalizing images to compress the pixel values between 0-1.\n",
    "    return image ,image_path\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {
    "id": "laIVZ5Kq0gXi"
   },
   "outputs": [],
   "source": [
    "def load_image1(image_path,training = True):\n",
    "    img = tf.io.read_file(image_path)# reading image\n",
    "    img = tf.image.decode_jpeg(img, channels=3)\n",
    "    image = tf.image.resize(img, (int(img_height),int(img_width)))# resizing image as desired\n",
    "    if training == True: #if this function is used during training only them we will augment images\n",
    "      \n",
    "      a = np.random.uniform() #selecting random number between 0-1 and applying different augmentations according to its output\n",
    "      if a<0.2:\n",
    "              image = aug2.augment_image(image)\n",
    "      elif a<0.4:\n",
    "              image = aug3.augment_image(image)\n",
    "      elif a<0.6:\n",
    "              image = aug4.augment_image(image)\n",
    "      elif a<0.8:\n",
    "              image = aug5.augment_image(image)\n",
    "      else:\n",
    "              image = aug6.augment_image(image)\n",
    "              \n",
    "    image = tf.keras.applications.densenet.preprocess_input(image)#normalize image\n",
    "    return image,image_path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "id": "6a0DwTcs-tKR",
    "outputId": "59d85ab0-d130-4cfa-ce0d-727481acd041"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['/content/images/CXR314_IM-1477-1001.png',\n",
       " '/content/images/CXR314_IM-1477-2001.png']"
      ]
     },
     "execution_count": 135,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "img_name_train[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.5 Dataloader function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "id": "SUMQOVNP0KX1"
   },
   "outputs": [],
   "source": [
    "#https://github.com/keras-team/keras/issues/8130 \n",
    "#dataloader will first load images and then passs whole data in batches \n",
    "class Dataloder(tf.keras.utils.Sequence):    \n",
    "    def __init__(self, image_names,capin,capout, batch_size=1, shuffle=False):\n",
    "\n",
    "        self.image_names = image_names\n",
    "        self.capin = capin\n",
    "        self.capout = capout\n",
    "        self.batch_size = batch_size\n",
    "        self.shuffle = shuffle\n",
    "        self.indexes = np.arange(len(image_names))\n",
    "        \n",
    "    def __getitem__(self, i):\n",
    "        # collect batch data\n",
    "        start = i * self.batch_size\n",
    "        stop = (i + 1) * self.batch_size\n",
    "        image1 =[]\n",
    "        image2 = []\n",
    "        caption_in = []\n",
    "        caption_out = []\n",
    "  \n",
    "        for j in range(start, stop):\n",
    "            img,img1 = self.__data_generation(self.image_names[j])#loading images from specific index\n",
    "            image1.append(img)#apppending that image to array\n",
    "            image2.append(img1)#appending second imm=age to another array\n",
    "            caption_in.append(self.capin[j])#appending input impression to model \n",
    "            caption_out.append(self.capout[j])#appending expected output to array\n",
    "\n",
    "        return tuple([(tf.convert_to_tensor(image1),tf.convert_to_tensor(image2),tf.convert_to_tensor(caption_in)),(tf.convert_to_tensor(caption_out))]) #returning tuple of (x,y)\n",
    "    \n",
    "    def __len__(self):\n",
    "        return len(self.indexes) // self.batch_size #definig size of dataloader function\n",
    "\n",
    "    def on_epoch_end(self):\n",
    "        if self.shuffle:\n",
    "            self.indexes = np.random.permutation(self.indexes)#shuffle indeces after every epoch\n",
    "\n",
    "    def __data_generation(self,image_name):\n",
    "        #preprocess and load images\n",
    "        img,path = load_image1(image_name[0])\n",
    "\n",
    "        img1,path1 = load_image1(image_name[1])\n",
    "        \n",
    "        return img,img1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "id": "oDysHECln-SH"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "metadata": {
    "id": "CupybUOd0q_A"
   },
   "outputs": [],
   "source": [
    "train_dataloader = Dataloder(img_name_train, target_train_inp,target_train_out, batch_size=BATCH_SIZE, shuffle=True)\n",
    "val_dataloader = Dataloder(img_name_val, target_val_inp,target_val_out, batch_size=BATCH_SIZE, shuffle=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 138,
   "metadata": {
    "id": "euciJO_PYLPM"
   },
   "outputs": [],
   "source": [
    "#import tensorflow as tf\n",
    "tf.compat.v1.enable_eager_execution()\n",
    "from tensorflow.keras.layers import TimeDistributed\n",
    "tf.keras.backend.clear_session()\n",
    "from tensorflow.keras.layers import Input, Softmax, RNN, Dense, Embedding, LSTM,Dropout\n",
    "from tensorflow.keras.models import Model\n",
    "import numpy as np\n",
    "#tf.keras.backend.set_floatx('float64')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.6 Making model architecture. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.6.1 Loading pretrained Chex-Net model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "89qLO2a5FJJV"
   },
   "outputs": [],
   "source": [
    "#weights_link: https://github.com/brucechou1983/CheXNet-Keras\n",
    "#loading pretrained Densenet121 on chest X-rays \n",
    "import tensorflow as tf\n",
    "image_model = tf.keras.applications.DenseNet121(weights='/content/drive/My Drive/case_study_2/brucechou1983_CheXNet_Keras_0.3.0_weights.h5',\n",
    "                                                classes = 14,input_shape=(int(img_height),int(img_width),3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "metadata": {
    "id": "yDPtiXrjNWJP"
   },
   "outputs": [],
   "source": [
    "#removing last dense layer and making different model without top dense layer for encoder.\n",
    "new_input = image_model.input\n",
    "hidden_layer = image_model.layers[-3].output\n",
    "image_features_extract_model = tf.keras.Model(new_input, hidden_layer)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "KtFtwjnTOQPY",
    "outputId": "960905a2-e529-4a67-f16c-52614c85c9f3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"functional_1\"\n",
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            [(None, 224, 224, 3) 0                                            \n",
      "__________________________________________________________________________________________________\n",
      "zero_padding2d (ZeroPadding2D)  (None, 230, 230, 3)  0           input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv1/conv (Conv2D)             (None, 112, 112, 64) 9408        zero_padding2d[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv1/bn (BatchNormalization)   (None, 112, 112, 64) 256         conv1/conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv1/relu (Activation)         (None, 112, 112, 64) 0           conv1/bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "zero_padding2d_1 (ZeroPadding2D (None, 114, 114, 64) 0           conv1/relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool1 (MaxPooling2D)            (None, 56, 56, 64)   0           zero_padding2d_1[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_0_bn (BatchNormali (None, 56, 56, 64)   256         pool1[0][0]                      \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_0_relu (Activation (None, 56, 56, 64)   0           conv2_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_conv (Conv2D)    (None, 56, 56, 128)  8192        conv2_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_bn (BatchNormali (None, 56, 56, 128)  512         conv2_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_1_relu (Activation (None, 56, 56, 128)  0           conv2_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_2_conv (Conv2D)    (None, 56, 56, 32)   36864       conv2_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block1_concat (Concatenat (None, 56, 56, 96)   0           pool1[0][0]                      \n",
      "                                                                 conv2_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_0_bn (BatchNormali (None, 56, 56, 96)   384         conv2_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_0_relu (Activation (None, 56, 56, 96)   0           conv2_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_conv (Conv2D)    (None, 56, 56, 128)  12288       conv2_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_bn (BatchNormali (None, 56, 56, 128)  512         conv2_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_1_relu (Activation (None, 56, 56, 128)  0           conv2_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_2_conv (Conv2D)    (None, 56, 56, 32)   36864       conv2_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block2_concat (Concatenat (None, 56, 56, 128)  0           conv2_block1_concat[0][0]        \n",
      "                                                                 conv2_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_0_bn (BatchNormali (None, 56, 56, 128)  512         conv2_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_0_relu (Activation (None, 56, 56, 128)  0           conv2_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_conv (Conv2D)    (None, 56, 56, 128)  16384       conv2_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_bn (BatchNormali (None, 56, 56, 128)  512         conv2_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_1_relu (Activation (None, 56, 56, 128)  0           conv2_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_2_conv (Conv2D)    (None, 56, 56, 32)   36864       conv2_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block3_concat (Concatenat (None, 56, 56, 160)  0           conv2_block2_concat[0][0]        \n",
      "                                                                 conv2_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_0_bn (BatchNormali (None, 56, 56, 160)  640         conv2_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_0_relu (Activation (None, 56, 56, 160)  0           conv2_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_conv (Conv2D)    (None, 56, 56, 128)  20480       conv2_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_bn (BatchNormali (None, 56, 56, 128)  512         conv2_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_1_relu (Activation (None, 56, 56, 128)  0           conv2_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_2_conv (Conv2D)    (None, 56, 56, 32)   36864       conv2_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block4_concat (Concatenat (None, 56, 56, 192)  0           conv2_block3_concat[0][0]        \n",
      "                                                                 conv2_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_0_bn (BatchNormali (None, 56, 56, 192)  768         conv2_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_0_relu (Activation (None, 56, 56, 192)  0           conv2_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_conv (Conv2D)    (None, 56, 56, 128)  24576       conv2_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_bn (BatchNormali (None, 56, 56, 128)  512         conv2_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_1_relu (Activation (None, 56, 56, 128)  0           conv2_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_2_conv (Conv2D)    (None, 56, 56, 32)   36864       conv2_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block5_concat (Concatenat (None, 56, 56, 224)  0           conv2_block4_concat[0][0]        \n",
      "                                                                 conv2_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_0_bn (BatchNormali (None, 56, 56, 224)  896         conv2_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_0_relu (Activation (None, 56, 56, 224)  0           conv2_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_conv (Conv2D)    (None, 56, 56, 128)  28672       conv2_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_bn (BatchNormali (None, 56, 56, 128)  512         conv2_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_1_relu (Activation (None, 56, 56, 128)  0           conv2_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_2_conv (Conv2D)    (None, 56, 56, 32)   36864       conv2_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv2_block6_concat (Concatenat (None, 56, 56, 256)  0           conv2_block5_concat[0][0]        \n",
      "                                                                 conv2_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "pool2_bn (BatchNormalization)   (None, 56, 56, 256)  1024        conv2_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "pool2_relu (Activation)         (None, 56, 56, 256)  0           pool2_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool2_conv (Conv2D)             (None, 56, 56, 128)  32768       pool2_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool2_pool (AveragePooling2D)   (None, 28, 28, 128)  0           pool2_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_0_bn (BatchNormali (None, 28, 28, 128)  512         pool2_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_0_relu (Activation (None, 28, 28, 128)  0           conv3_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_conv (Conv2D)    (None, 28, 28, 128)  16384       conv3_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_1_relu (Activation (None, 28, 28, 128)  0           conv3_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block1_concat (Concatenat (None, 28, 28, 160)  0           pool2_pool[0][0]                 \n",
      "                                                                 conv3_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_0_bn (BatchNormali (None, 28, 28, 160)  640         conv3_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_0_relu (Activation (None, 28, 28, 160)  0           conv3_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_conv (Conv2D)    (None, 28, 28, 128)  20480       conv3_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_1_relu (Activation (None, 28, 28, 128)  0           conv3_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block2_concat (Concatenat (None, 28, 28, 192)  0           conv3_block1_concat[0][0]        \n",
      "                                                                 conv3_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_0_bn (BatchNormali (None, 28, 28, 192)  768         conv3_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_0_relu (Activation (None, 28, 28, 192)  0           conv3_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_conv (Conv2D)    (None, 28, 28, 128)  24576       conv3_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_1_relu (Activation (None, 28, 28, 128)  0           conv3_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block3_concat (Concatenat (None, 28, 28, 224)  0           conv3_block2_concat[0][0]        \n",
      "                                                                 conv3_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_0_bn (BatchNormali (None, 28, 28, 224)  896         conv3_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_0_relu (Activation (None, 28, 28, 224)  0           conv3_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_conv (Conv2D)    (None, 28, 28, 128)  28672       conv3_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_1_relu (Activation (None, 28, 28, 128)  0           conv3_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block4_concat (Concatenat (None, 28, 28, 256)  0           conv3_block3_concat[0][0]        \n",
      "                                                                 conv3_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_0_bn (BatchNormali (None, 28, 28, 256)  1024        conv3_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_0_relu (Activation (None, 28, 28, 256)  0           conv3_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_conv (Conv2D)    (None, 28, 28, 128)  32768       conv3_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_1_relu (Activation (None, 28, 28, 128)  0           conv3_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block5_concat (Concatenat (None, 28, 28, 288)  0           conv3_block4_concat[0][0]        \n",
      "                                                                 conv3_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_0_bn (BatchNormali (None, 28, 28, 288)  1152        conv3_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_0_relu (Activation (None, 28, 28, 288)  0           conv3_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_conv (Conv2D)    (None, 28, 28, 128)  36864       conv3_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_1_relu (Activation (None, 28, 28, 128)  0           conv3_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block6_concat (Concatenat (None, 28, 28, 320)  0           conv3_block5_concat[0][0]        \n",
      "                                                                 conv3_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_0_bn (BatchNormali (None, 28, 28, 320)  1280        conv3_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_0_relu (Activation (None, 28, 28, 320)  0           conv3_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_conv (Conv2D)    (None, 28, 28, 128)  40960       conv3_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_1_relu (Activation (None, 28, 28, 128)  0           conv3_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block7_concat (Concatenat (None, 28, 28, 352)  0           conv3_block6_concat[0][0]        \n",
      "                                                                 conv3_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_0_bn (BatchNormali (None, 28, 28, 352)  1408        conv3_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_0_relu (Activation (None, 28, 28, 352)  0           conv3_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_conv (Conv2D)    (None, 28, 28, 128)  45056       conv3_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_1_relu (Activation (None, 28, 28, 128)  0           conv3_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block8_concat (Concatenat (None, 28, 28, 384)  0           conv3_block7_concat[0][0]        \n",
      "                                                                 conv3_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_0_bn (BatchNormali (None, 28, 28, 384)  1536        conv3_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_0_relu (Activation (None, 28, 28, 384)  0           conv3_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_conv (Conv2D)    (None, 28, 28, 128)  49152       conv3_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_bn (BatchNormali (None, 28, 28, 128)  512         conv3_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_1_relu (Activation (None, 28, 28, 128)  0           conv3_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_2_conv (Conv2D)    (None, 28, 28, 32)   36864       conv3_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block9_concat (Concatenat (None, 28, 28, 416)  0           conv3_block8_concat[0][0]        \n",
      "                                                                 conv3_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_0_bn (BatchNormal (None, 28, 28, 416)  1664        conv3_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_0_relu (Activatio (None, 28, 28, 416)  0           conv3_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_conv (Conv2D)   (None, 28, 28, 128)  53248       conv3_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_bn (BatchNormal (None, 28, 28, 128)  512         conv3_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_1_relu (Activatio (None, 28, 28, 128)  0           conv3_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_2_conv (Conv2D)   (None, 28, 28, 32)   36864       conv3_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block10_concat (Concatena (None, 28, 28, 448)  0           conv3_block9_concat[0][0]        \n",
      "                                                                 conv3_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_0_bn (BatchNormal (None, 28, 28, 448)  1792        conv3_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_0_relu (Activatio (None, 28, 28, 448)  0           conv3_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_conv (Conv2D)   (None, 28, 28, 128)  57344       conv3_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_bn (BatchNormal (None, 28, 28, 128)  512         conv3_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_1_relu (Activatio (None, 28, 28, 128)  0           conv3_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_2_conv (Conv2D)   (None, 28, 28, 32)   36864       conv3_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block11_concat (Concatena (None, 28, 28, 480)  0           conv3_block10_concat[0][0]       \n",
      "                                                                 conv3_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_0_bn (BatchNormal (None, 28, 28, 480)  1920        conv3_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_0_relu (Activatio (None, 28, 28, 480)  0           conv3_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_conv (Conv2D)   (None, 28, 28, 128)  61440       conv3_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_bn (BatchNormal (None, 28, 28, 128)  512         conv3_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_1_relu (Activatio (None, 28, 28, 128)  0           conv3_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_2_conv (Conv2D)   (None, 28, 28, 32)   36864       conv3_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv3_block12_concat (Concatena (None, 28, 28, 512)  0           conv3_block11_concat[0][0]       \n",
      "                                                                 conv3_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool3_bn (BatchNormalization)   (None, 28, 28, 512)  2048        conv3_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool3_relu (Activation)         (None, 28, 28, 512)  0           pool3_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool3_conv (Conv2D)             (None, 28, 28, 256)  131072      pool3_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool3_pool (AveragePooling2D)   (None, 14, 14, 256)  0           pool3_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_0_bn (BatchNormali (None, 14, 14, 256)  1024        pool3_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_0_relu (Activation (None, 14, 14, 256)  0           conv4_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_conv (Conv2D)    (None, 14, 14, 128)  32768       conv4_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_1_relu (Activation (None, 14, 14, 128)  0           conv4_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block1_concat (Concatenat (None, 14, 14, 288)  0           pool3_pool[0][0]                 \n",
      "                                                                 conv4_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_0_bn (BatchNormali (None, 14, 14, 288)  1152        conv4_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_0_relu (Activation (None, 14, 14, 288)  0           conv4_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_conv (Conv2D)    (None, 14, 14, 128)  36864       conv4_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_1_relu (Activation (None, 14, 14, 128)  0           conv4_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block2_concat (Concatenat (None, 14, 14, 320)  0           conv4_block1_concat[0][0]        \n",
      "                                                                 conv4_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_0_bn (BatchNormali (None, 14, 14, 320)  1280        conv4_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_0_relu (Activation (None, 14, 14, 320)  0           conv4_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_conv (Conv2D)    (None, 14, 14, 128)  40960       conv4_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_1_relu (Activation (None, 14, 14, 128)  0           conv4_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block3_concat (Concatenat (None, 14, 14, 352)  0           conv4_block2_concat[0][0]        \n",
      "                                                                 conv4_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_0_bn (BatchNormali (None, 14, 14, 352)  1408        conv4_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_0_relu (Activation (None, 14, 14, 352)  0           conv4_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_conv (Conv2D)    (None, 14, 14, 128)  45056       conv4_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_1_relu (Activation (None, 14, 14, 128)  0           conv4_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block4_concat (Concatenat (None, 14, 14, 384)  0           conv4_block3_concat[0][0]        \n",
      "                                                                 conv4_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_0_bn (BatchNormali (None, 14, 14, 384)  1536        conv4_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_0_relu (Activation (None, 14, 14, 384)  0           conv4_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_conv (Conv2D)    (None, 14, 14, 128)  49152       conv4_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_1_relu (Activation (None, 14, 14, 128)  0           conv4_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block5_concat (Concatenat (None, 14, 14, 416)  0           conv4_block4_concat[0][0]        \n",
      "                                                                 conv4_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_0_bn (BatchNormali (None, 14, 14, 416)  1664        conv4_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_0_relu (Activation (None, 14, 14, 416)  0           conv4_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_conv (Conv2D)    (None, 14, 14, 128)  53248       conv4_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_1_relu (Activation (None, 14, 14, 128)  0           conv4_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block6_concat (Concatenat (None, 14, 14, 448)  0           conv4_block5_concat[0][0]        \n",
      "                                                                 conv4_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_0_bn (BatchNormali (None, 14, 14, 448)  1792        conv4_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_0_relu (Activation (None, 14, 14, 448)  0           conv4_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_conv (Conv2D)    (None, 14, 14, 128)  57344       conv4_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_1_relu (Activation (None, 14, 14, 128)  0           conv4_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block7_concat (Concatenat (None, 14, 14, 480)  0           conv4_block6_concat[0][0]        \n",
      "                                                                 conv4_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_0_bn (BatchNormali (None, 14, 14, 480)  1920        conv4_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_0_relu (Activation (None, 14, 14, 480)  0           conv4_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_conv (Conv2D)    (None, 14, 14, 128)  61440       conv4_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_1_relu (Activation (None, 14, 14, 128)  0           conv4_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block8_concat (Concatenat (None, 14, 14, 512)  0           conv4_block7_concat[0][0]        \n",
      "                                                                 conv4_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_0_bn (BatchNormali (None, 14, 14, 512)  2048        conv4_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_0_relu (Activation (None, 14, 14, 512)  0           conv4_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_conv (Conv2D)    (None, 14, 14, 128)  65536       conv4_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_bn (BatchNormali (None, 14, 14, 128)  512         conv4_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_1_relu (Activation (None, 14, 14, 128)  0           conv4_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_2_conv (Conv2D)    (None, 14, 14, 32)   36864       conv4_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block9_concat (Concatenat (None, 14, 14, 544)  0           conv4_block8_concat[0][0]        \n",
      "                                                                 conv4_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_0_bn (BatchNormal (None, 14, 14, 544)  2176        conv4_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_0_relu (Activatio (None, 14, 14, 544)  0           conv4_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_conv (Conv2D)   (None, 14, 14, 128)  69632       conv4_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block10_concat (Concatena (None, 14, 14, 576)  0           conv4_block9_concat[0][0]        \n",
      "                                                                 conv4_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_0_bn (BatchNormal (None, 14, 14, 576)  2304        conv4_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_0_relu (Activatio (None, 14, 14, 576)  0           conv4_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_conv (Conv2D)   (None, 14, 14, 128)  73728       conv4_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block11_concat (Concatena (None, 14, 14, 608)  0           conv4_block10_concat[0][0]       \n",
      "                                                                 conv4_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_0_bn (BatchNormal (None, 14, 14, 608)  2432        conv4_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_0_relu (Activatio (None, 14, 14, 608)  0           conv4_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_conv (Conv2D)   (None, 14, 14, 128)  77824       conv4_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block12_concat (Concatena (None, 14, 14, 640)  0           conv4_block11_concat[0][0]       \n",
      "                                                                 conv4_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_0_bn (BatchNormal (None, 14, 14, 640)  2560        conv4_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_0_relu (Activatio (None, 14, 14, 640)  0           conv4_block13_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_conv (Conv2D)   (None, 14, 14, 128)  81920       conv4_block13_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block13_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block13_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block13_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block13_concat (Concatena (None, 14, 14, 672)  0           conv4_block12_concat[0][0]       \n",
      "                                                                 conv4_block13_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_0_bn (BatchNormal (None, 14, 14, 672)  2688        conv4_block13_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_0_relu (Activatio (None, 14, 14, 672)  0           conv4_block14_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_conv (Conv2D)   (None, 14, 14, 128)  86016       conv4_block14_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block14_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block14_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block14_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block14_concat (Concatena (None, 14, 14, 704)  0           conv4_block13_concat[0][0]       \n",
      "                                                                 conv4_block14_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_0_bn (BatchNormal (None, 14, 14, 704)  2816        conv4_block14_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_0_relu (Activatio (None, 14, 14, 704)  0           conv4_block15_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_conv (Conv2D)   (None, 14, 14, 128)  90112       conv4_block15_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block15_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block15_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block15_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block15_concat (Concatena (None, 14, 14, 736)  0           conv4_block14_concat[0][0]       \n",
      "                                                                 conv4_block15_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_0_bn (BatchNormal (None, 14, 14, 736)  2944        conv4_block15_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_0_relu (Activatio (None, 14, 14, 736)  0           conv4_block16_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_conv (Conv2D)   (None, 14, 14, 128)  94208       conv4_block16_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block16_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block16_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block16_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block16_concat (Concatena (None, 14, 14, 768)  0           conv4_block15_concat[0][0]       \n",
      "                                                                 conv4_block16_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_0_bn (BatchNormal (None, 14, 14, 768)  3072        conv4_block16_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_0_relu (Activatio (None, 14, 14, 768)  0           conv4_block17_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_conv (Conv2D)   (None, 14, 14, 128)  98304       conv4_block17_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block17_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block17_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block17_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block17_concat (Concatena (None, 14, 14, 800)  0           conv4_block16_concat[0][0]       \n",
      "                                                                 conv4_block17_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_0_bn (BatchNormal (None, 14, 14, 800)  3200        conv4_block17_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_0_relu (Activatio (None, 14, 14, 800)  0           conv4_block18_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_conv (Conv2D)   (None, 14, 14, 128)  102400      conv4_block18_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block18_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block18_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block18_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block18_concat (Concatena (None, 14, 14, 832)  0           conv4_block17_concat[0][0]       \n",
      "                                                                 conv4_block18_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_0_bn (BatchNormal (None, 14, 14, 832)  3328        conv4_block18_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_0_relu (Activatio (None, 14, 14, 832)  0           conv4_block19_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_conv (Conv2D)   (None, 14, 14, 128)  106496      conv4_block19_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block19_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block19_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block19_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block19_concat (Concatena (None, 14, 14, 864)  0           conv4_block18_concat[0][0]       \n",
      "                                                                 conv4_block19_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_0_bn (BatchNormal (None, 14, 14, 864)  3456        conv4_block19_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_0_relu (Activatio (None, 14, 14, 864)  0           conv4_block20_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_conv (Conv2D)   (None, 14, 14, 128)  110592      conv4_block20_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block20_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block20_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block20_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block20_concat (Concatena (None, 14, 14, 896)  0           conv4_block19_concat[0][0]       \n",
      "                                                                 conv4_block20_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_0_bn (BatchNormal (None, 14, 14, 896)  3584        conv4_block20_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_0_relu (Activatio (None, 14, 14, 896)  0           conv4_block21_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_conv (Conv2D)   (None, 14, 14, 128)  114688      conv4_block21_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block21_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block21_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block21_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block21_concat (Concatena (None, 14, 14, 928)  0           conv4_block20_concat[0][0]       \n",
      "                                                                 conv4_block21_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_0_bn (BatchNormal (None, 14, 14, 928)  3712        conv4_block21_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_0_relu (Activatio (None, 14, 14, 928)  0           conv4_block22_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_conv (Conv2D)   (None, 14, 14, 128)  118784      conv4_block22_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block22_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block22_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block22_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block22_concat (Concatena (None, 14, 14, 960)  0           conv4_block21_concat[0][0]       \n",
      "                                                                 conv4_block22_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_0_bn (BatchNormal (None, 14, 14, 960)  3840        conv4_block22_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_0_relu (Activatio (None, 14, 14, 960)  0           conv4_block23_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_conv (Conv2D)   (None, 14, 14, 128)  122880      conv4_block23_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block23_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block23_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block23_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block23_concat (Concatena (None, 14, 14, 992)  0           conv4_block22_concat[0][0]       \n",
      "                                                                 conv4_block23_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_0_bn (BatchNormal (None, 14, 14, 992)  3968        conv4_block23_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_0_relu (Activatio (None, 14, 14, 992)  0           conv4_block24_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_conv (Conv2D)   (None, 14, 14, 128)  126976      conv4_block24_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_bn (BatchNormal (None, 14, 14, 128)  512         conv4_block24_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_1_relu (Activatio (None, 14, 14, 128)  0           conv4_block24_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_2_conv (Conv2D)   (None, 14, 14, 32)   36864       conv4_block24_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv4_block24_concat (Concatena (None, 14, 14, 1024) 0           conv4_block23_concat[0][0]       \n",
      "                                                                 conv4_block24_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool4_bn (BatchNormalization)   (None, 14, 14, 1024) 4096        conv4_block24_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "pool4_relu (Activation)         (None, 14, 14, 1024) 0           pool4_bn[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "pool4_conv (Conv2D)             (None, 14, 14, 512)  524288      pool4_relu[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "pool4_pool (AveragePooling2D)   (None, 7, 7, 512)    0           pool4_conv[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_0_bn (BatchNormali (None, 7, 7, 512)    2048        pool4_pool[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_0_relu (Activation (None, 7, 7, 512)    0           conv5_block1_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_conv (Conv2D)    (None, 7, 7, 128)    65536       conv5_block1_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block1_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_1_relu (Activation (None, 7, 7, 128)    0           conv5_block1_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block1_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block1_concat (Concatenat (None, 7, 7, 544)    0           pool4_pool[0][0]                 \n",
      "                                                                 conv5_block1_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_0_bn (BatchNormali (None, 7, 7, 544)    2176        conv5_block1_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_0_relu (Activation (None, 7, 7, 544)    0           conv5_block2_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_conv (Conv2D)    (None, 7, 7, 128)    69632       conv5_block2_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block2_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_1_relu (Activation (None, 7, 7, 128)    0           conv5_block2_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block2_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block2_concat (Concatenat (None, 7, 7, 576)    0           conv5_block1_concat[0][0]        \n",
      "                                                                 conv5_block2_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_0_bn (BatchNormali (None, 7, 7, 576)    2304        conv5_block2_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_0_relu (Activation (None, 7, 7, 576)    0           conv5_block3_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_1_conv (Conv2D)    (None, 7, 7, 128)    73728       conv5_block3_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block3_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_1_relu (Activation (None, 7, 7, 128)    0           conv5_block3_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block3_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block3_concat (Concatenat (None, 7, 7, 608)    0           conv5_block2_concat[0][0]        \n",
      "                                                                 conv5_block3_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_0_bn (BatchNormali (None, 7, 7, 608)    2432        conv5_block3_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_0_relu (Activation (None, 7, 7, 608)    0           conv5_block4_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_conv (Conv2D)    (None, 7, 7, 128)    77824       conv5_block4_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block4_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_1_relu (Activation (None, 7, 7, 128)    0           conv5_block4_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block4_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block4_concat (Concatenat (None, 7, 7, 640)    0           conv5_block3_concat[0][0]        \n",
      "                                                                 conv5_block4_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_0_bn (BatchNormali (None, 7, 7, 640)    2560        conv5_block4_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_0_relu (Activation (None, 7, 7, 640)    0           conv5_block5_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_conv (Conv2D)    (None, 7, 7, 128)    81920       conv5_block5_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block5_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_1_relu (Activation (None, 7, 7, 128)    0           conv5_block5_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block5_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block5_concat (Concatenat (None, 7, 7, 672)    0           conv5_block4_concat[0][0]        \n",
      "                                                                 conv5_block5_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_0_bn (BatchNormali (None, 7, 7, 672)    2688        conv5_block5_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_0_relu (Activation (None, 7, 7, 672)    0           conv5_block6_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_conv (Conv2D)    (None, 7, 7, 128)    86016       conv5_block6_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block6_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_1_relu (Activation (None, 7, 7, 128)    0           conv5_block6_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block6_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block6_concat (Concatenat (None, 7, 7, 704)    0           conv5_block5_concat[0][0]        \n",
      "                                                                 conv5_block6_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_0_bn (BatchNormali (None, 7, 7, 704)    2816        conv5_block6_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_0_relu (Activation (None, 7, 7, 704)    0           conv5_block7_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_conv (Conv2D)    (None, 7, 7, 128)    90112       conv5_block7_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block7_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_1_relu (Activation (None, 7, 7, 128)    0           conv5_block7_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block7_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block7_concat (Concatenat (None, 7, 7, 736)    0           conv5_block6_concat[0][0]        \n",
      "                                                                 conv5_block7_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_0_bn (BatchNormali (None, 7, 7, 736)    2944        conv5_block7_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_0_relu (Activation (None, 7, 7, 736)    0           conv5_block8_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_conv (Conv2D)    (None, 7, 7, 128)    94208       conv5_block8_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block8_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_1_relu (Activation (None, 7, 7, 128)    0           conv5_block8_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block8_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block8_concat (Concatenat (None, 7, 7, 768)    0           conv5_block7_concat[0][0]        \n",
      "                                                                 conv5_block8_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_0_bn (BatchNormali (None, 7, 7, 768)    3072        conv5_block8_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_0_relu (Activation (None, 7, 7, 768)    0           conv5_block9_0_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_conv (Conv2D)    (None, 7, 7, 128)    98304       conv5_block9_0_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_bn (BatchNormali (None, 7, 7, 128)    512         conv5_block9_1_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_1_relu (Activation (None, 7, 7, 128)    0           conv5_block9_1_bn[0][0]          \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_2_conv (Conv2D)    (None, 7, 7, 32)     36864       conv5_block9_1_relu[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block9_concat (Concatenat (None, 7, 7, 800)    0           conv5_block8_concat[0][0]        \n",
      "                                                                 conv5_block9_2_conv[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_0_bn (BatchNormal (None, 7, 7, 800)    3200        conv5_block9_concat[0][0]        \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_0_relu (Activatio (None, 7, 7, 800)    0           conv5_block10_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_conv (Conv2D)   (None, 7, 7, 128)    102400      conv5_block10_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_bn (BatchNormal (None, 7, 7, 128)    512         conv5_block10_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_1_relu (Activatio (None, 7, 7, 128)    0           conv5_block10_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_2_conv (Conv2D)   (None, 7, 7, 32)     36864       conv5_block10_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block10_concat (Concatena (None, 7, 7, 832)    0           conv5_block9_concat[0][0]        \n",
      "                                                                 conv5_block10_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_0_bn (BatchNormal (None, 7, 7, 832)    3328        conv5_block10_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_0_relu (Activatio (None, 7, 7, 832)    0           conv5_block11_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_conv (Conv2D)   (None, 7, 7, 128)    106496      conv5_block11_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_bn (BatchNormal (None, 7, 7, 128)    512         conv5_block11_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_1_relu (Activatio (None, 7, 7, 128)    0           conv5_block11_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_2_conv (Conv2D)   (None, 7, 7, 32)     36864       conv5_block11_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block11_concat (Concatena (None, 7, 7, 864)    0           conv5_block10_concat[0][0]       \n",
      "                                                                 conv5_block11_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_0_bn (BatchNormal (None, 7, 7, 864)    3456        conv5_block11_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_0_relu (Activatio (None, 7, 7, 864)    0           conv5_block12_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_conv (Conv2D)   (None, 7, 7, 128)    110592      conv5_block12_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_bn (BatchNormal (None, 7, 7, 128)    512         conv5_block12_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_1_relu (Activatio (None, 7, 7, 128)    0           conv5_block12_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_2_conv (Conv2D)   (None, 7, 7, 32)     36864       conv5_block12_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block12_concat (Concatena (None, 7, 7, 896)    0           conv5_block11_concat[0][0]       \n",
      "                                                                 conv5_block12_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_0_bn (BatchNormal (None, 7, 7, 896)    3584        conv5_block12_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_0_relu (Activatio (None, 7, 7, 896)    0           conv5_block13_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_conv (Conv2D)   (None, 7, 7, 128)    114688      conv5_block13_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_bn (BatchNormal (None, 7, 7, 128)    512         conv5_block13_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_1_relu (Activatio (None, 7, 7, 128)    0           conv5_block13_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_2_conv (Conv2D)   (None, 7, 7, 32)     36864       conv5_block13_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block13_concat (Concatena (None, 7, 7, 928)    0           conv5_block12_concat[0][0]       \n",
      "                                                                 conv5_block13_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_0_bn (BatchNormal (None, 7, 7, 928)    3712        conv5_block13_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_0_relu (Activatio (None, 7, 7, 928)    0           conv5_block14_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_conv (Conv2D)   (None, 7, 7, 128)    118784      conv5_block14_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_bn (BatchNormal (None, 7, 7, 128)    512         conv5_block14_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_1_relu (Activatio (None, 7, 7, 128)    0           conv5_block14_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_2_conv (Conv2D)   (None, 7, 7, 32)     36864       conv5_block14_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block14_concat (Concatena (None, 7, 7, 960)    0           conv5_block13_concat[0][0]       \n",
      "                                                                 conv5_block14_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_0_bn (BatchNormal (None, 7, 7, 960)    3840        conv5_block14_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_0_relu (Activatio (None, 7, 7, 960)    0           conv5_block15_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_conv (Conv2D)   (None, 7, 7, 128)    122880      conv5_block15_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_bn (BatchNormal (None, 7, 7, 128)    512         conv5_block15_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_1_relu (Activatio (None, 7, 7, 128)    0           conv5_block15_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_2_conv (Conv2D)   (None, 7, 7, 32)     36864       conv5_block15_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block15_concat (Concatena (None, 7, 7, 992)    0           conv5_block14_concat[0][0]       \n",
      "                                                                 conv5_block15_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_0_bn (BatchNormal (None, 7, 7, 992)    3968        conv5_block15_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_0_relu (Activatio (None, 7, 7, 992)    0           conv5_block16_0_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_conv (Conv2D)   (None, 7, 7, 128)    126976      conv5_block16_0_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_bn (BatchNormal (None, 7, 7, 128)    512         conv5_block16_1_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_1_relu (Activatio (None, 7, 7, 128)    0           conv5_block16_1_bn[0][0]         \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_2_conv (Conv2D)   (None, 7, 7, 32)     36864       conv5_block16_1_relu[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "conv5_block16_concat (Concatena (None, 7, 7, 1024)   0           conv5_block15_concat[0][0]       \n",
      "                                                                 conv5_block16_2_conv[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "bn (BatchNormalization)         (None, 7, 7, 1024)   4096        conv5_block16_concat[0][0]       \n",
      "__________________________________________________________________________________________________\n",
      "relu (Activation)               (None, 7, 7, 1024)   0           bn[0][0]                         \n",
      "==================================================================================================\n",
      "Total params: 7,037,504\n",
      "Trainable params: 6,953,856\n",
      "Non-trainable params: 83,648\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "image_features_extract_model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.6.2 CNN_Encoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "metadata": {
    "id": "f3Q3veiFeo3g"
   },
   "outputs": [],
   "source": [
    "class CNN_Encoder(tf.keras.layers.Layer):\n",
    "    # This encoder passes the features through a image_features_extract_model\n",
    "    def __init__(self, embedding_dim,feature_extracter):\n",
    "        super(CNN_Encoder, self).__init__()\n",
    "        # shape after fc == (batch_size, 64, embedding_dim/2)\n",
    "        #self.fc = tf.keras.layers.Dense(embedding_dim)\n",
    "        self.image_features_extract_model = feature_extracter\n",
    "    def call(self, x):\n",
    "        x = image_features_extract_model(x)\n",
    "        x = tf.reshape(x,(x.shape[0], -1, x.shape[3]))#squashing the output of image_features_extract_model.\n",
    "        #x = self.fc(x)\n",
    "        #x = tf.nn.relu(x)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.6.3 Luong's Concat Attention: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {
    "id": "CWV8jgJ6Rsr8"
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras import backend as K\n",
    "\n",
    "#Luong's attention type1 :\n",
    "\n",
    "class attention_concat(tf.keras.layers.Layer):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.W1 = tf.keras.layers.Dense(units)\n",
    "        self.W2 = tf.keras.layers.Dense(units)\n",
    "        self.V = tf.keras.layers.Dense(1)\n",
    "    \n",
    "    def call(self,decoder_hidden_state,encoder_output_states):\n",
    "        print('='*50,'attention','='*50)\n",
    "        \n",
    "        decoder_hidden_state = tf.expand_dims(decoder_hidden_state, 1)\n",
    "        # features(encoder_output_states) shape == (batch_size, attention_features_shape, 2048)\n",
    "        #decoder_hidden_state shape == (batch_size,1,units)\n",
    "        print(\"attention shapes {},{} :\".format(encoder_output_states.shape,decoder_hidden_state.shape))\n",
    "        similarities = self.V(tf.nn.tanh(\n",
    "        self.W1(decoder_hidden_state) + self.W2(encoder_output_states)))\n",
    "        print(\"similarities shape:\",similarities.shape)\n",
    "        #similarities shape = (batch_size,attention_features_shape,1)\n",
    "        attention_weights = tf.nn.softmax(similarities, axis=1)#shrinking similarities values between 0-1.\n",
    "        print(\"attention_weights shape:\",attention_weights.shape)\n",
    "\n",
    "        #context_vector shape after sum == (batch_size, hidden_size)\n",
    "        context_vector = attention_weights * encoder_output_states#multiplying these attention_weights with encoder output states to generate featue map of attention reigons.\n",
    "        print(\"initial context shape\",context_vector.shape)\n",
    "        #initial context shape (batch_size, attention_features_shape, 2048)\n",
    "\n",
    "        context_vector = tf.reduce_sum(context_vector, axis=1)#summing them up for passing them to next input.\n",
    "        print(\"final context shape\",context_vector.shape)\n",
    "        #shape after summing over the first axis = (batch_size,2048)\n",
    "        return context_vector, attention_weights"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "0TAd2KtdAy5g"
   },
   "source": [
    "### 4.6.4 Luong's General Attention: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {
    "id": "CSWL2eBSpYEM"
   },
   "outputs": [],
   "source": [
    "from tensorflow.keras import backend as K\n",
    "#luong's attention type 2:\n",
    "\n",
    "class attention_general(tf.keras.layers.Layer):\n",
    "    def __init__(self):\n",
    "        super().__init__()\n",
    "        self.W = tf.keras.layers.Dense(units)\n",
    "\n",
    "    def call(self,decoder_hidden_state,encoder_output_states):\n",
    "        # features(encoder_output_states) shape == (batch_size, attention_features_shape, 2048)\n",
    "        #decoder_hidden_state shape == (batch_size,units)\n",
    "\n",
    "        dense_out = self.W(encoder_output_states)\n",
    "        #dense out shape  (batch_size, attention_features_shape, units)\n",
    "        print(\"attention shapes {},{} :\".format(decoder_hidden_state.shape,dense_out.shape))\n",
    "        similarities = tf.keras.layers.Dot(axes=-1)([decoder_hidden_state,dense_out])\n",
    "        print(\"similarities shape:\",similarities.shape)\n",
    "        #similarities shape = (batch_size,attention_features_shape)\n",
    "\n",
    "        attention_weights = tf.nn.softmax(similarities, axis=-1)\n",
    "        attention_weights = tf.expand_dims(attention_weights, -1)\n",
    "        print(\"attention_weights shape:\",attention_weights.shape)\n",
    "        #attention_weighhts shape = (batch_size,attention_features_shape,1)\n",
    "\n",
    "        # context_vector shape after sum == (batch_size, 2048)\n",
    "        context_vector = attention_weights * encoder_output_states\n",
    "        print(\"context shape\",context_vector.shape)\n",
    "\n",
    "        return context_vector, attention_weights"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.6.5 Decoder "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {
    "id": "KJll8fU-nMgl"
   },
   "outputs": [],
   "source": [
    "#decoder is the main part in this model it will take image features and generate prediction for each timestep.\n",
    "class decoder(tf.keras.Model):\n",
    "    def __init__(self,vocab_size,embedding_dim,input_length,dec_units,atten_layer):\n",
    "        super().__init__()\n",
    "        self.vocab_size = vocab_size #1486\n",
    "        self.embedding_dim = embedding_dim #256\n",
    "        self.input_length = input_length #max_length\n",
    "        self.dec_units =dec_units #units\n",
    "        self.dense= tf.keras.layers.Dense(self.vocab_size) #1486\n",
    "        self.attention= atten_layer \n",
    "\n",
    "        self.embedding = Embedding(input_dim = self.vocab_size,output_dim =self.embedding_dim,\n",
    "                             input_length = self.input_length,name = 'decoder_embedding_layer',mask_zero=True)\n",
    "\n",
    "        self.lstm = LSTM(units = self.dec_units,return_state = True,return_sequences = True,name = 'train_decoder_LSTM',dropout=rate,recurrent_dropout=rate)\n",
    "\n",
    "    @tf.function\n",
    "\n",
    "    def one_step_decoder(self,input_to_decoder,decoder_hidden_state,encoder_output_states,decoder_cell_state):\n",
    "\n",
    "        # this function is called at every timestep , it calculates attention,updates hidden and cell state of LSTM .\n",
    "\n",
    "        embed = self.embedding(input_to_decoder)\n",
    "        #embedding shape = (batch_size,vocab_size,embedding_dim)\n",
    "        attn_context,attn_weights = self.attention(decoder_hidden_state,encoder_output_states)\n",
    "        print(\"decoder_embedding shape:\",tf.shape(embed))\n",
    "        attn_context = tf.expand_dims(attn_context, 1)\n",
    "\n",
    "        print(\"attention_context shape:\",attn_context.shape)\n",
    "        concat =tf.keras.layers.Concatenate(axis=-1)([embed,attn_context])\n",
    "        print('shape after concat',concat.shape)\n",
    "        #shape after concat == (batch_size,1,embedding_dim + 2048)\n",
    "        concat = tf.nn.dropout(concat, rate)# using dropout to reduce overfitting\n",
    "\n",
    "        decoder_outputs,decoder_hidden_state,decoder_cell_state = self.lstm(concat,initial_state=[decoder_hidden_state, decoder_cell_state])\n",
    "\n",
    "        decoder_outputs = tf.reshape(decoder_outputs, (-1, decoder_outputs.shape[2]))\n",
    "        output = self.dense(decoder_outputs)\n",
    "        #output shaoe == (batch_size,1,vocab_size)\n",
    "        return output,decoder_hidden_state,attn_weights,decoder_cell_state\n",
    "\n",
    "    def reset_hidden_state(self, batch_size):\n",
    "      #randomly initializing decoder hidden state\n",
    "      return tf.random.normal((batch_size, self.dec_units))\n",
    "\n",
    "    def reset_cell_state(self,batch_size):\n",
    "      #randomly initializing decoder cell state\n",
    "      return tf.random.normal((batch_size,self.dec_units))\n",
    "\n",
    "    def call(self,input_to_decoder,decoder_hidden_state,encoder_output_states, decoder_cell_state):\n",
    "        \n",
    "        all_outputs = tf.TensorArray(tf.float32,size = input_to_decoder.shape[1],name = \"output_arrays\")#storing outputs at every timestep.\n",
    "        for timestep in range(input_to_decoder.shape[1]):\n",
    "            #this loop will update the cell and hidden states of the LSTM the 'decoder_hidden_state' and 'decoder_cell_state' on the right side of '=' are states of lstm at current time step and \n",
    "            #the outputs of these of are passed as inputs in next iteration.\n",
    "            output,decoder_hidden_state,attn_weights,decoder_cell_state=self.one_step_decoder(input_to_decoder[:,timestep:timestep+1],decoder_hidden_state,encoder_output_states, decoder_cell_state)\n",
    "            all_outputs = all_outputs.write(timestep,output)#writing output at every time step\n",
    "        all_outputs = tf.transpose(all_outputs.stack(),[1,0,2])#reshapeing intto (batch_size,max_length,vocab_size)\n",
    "\n",
    "        return all_outputs\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.7 Loss Function and Optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {
    "id": "3S0e8Z6moOQS"
   },
   "outputs": [],
   "source": [
    "import os,datetime\n",
    "\n",
    "optimizer = tf.keras.optimizers.Adam(0.00001)\n",
    "\n",
    "def loss_function(targets, logits):\n",
    "    crossentropy = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)#loss function\n",
    "\n",
    "    mask = tf.math.logical_not(tf.math.equal(targets, 0))#detecting mask tokens so that we can ignore them while calculating loss.\n",
    "    mask = tf.cast(mask, dtype=tf.int64)#above output will give us True ,False array so we will convert it into 1's and 0's using this command\n",
    "    loss = crossentropy(targets, logits, sample_weight=mask)#while calculating loss we will pass this mask as weights because\n",
    "                                                            #the mask will contain 1's for only those values which are not padded.\n",
    "    return loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "id": "LHbBMuAQQaIb",
    "outputId": "b37c48cf-9f0d-41de-d318-e57fecfdfcf5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "length of train dataloader :  495\n",
      "length of validation dataloader :  54\n"
     ]
    }
   ],
   "source": [
    "#length of train and validation dataloader\n",
    "print('length of train dataloader : ',len(train_dataloader))\n",
    "print('length of validation dataloader : ',len(val_dataloader))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 54
    },
    "id": "O0GCchRdNglN",
    "outputId": "a2913142-27d0-44ee-ce3b-811cc3e93363"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "steps per epoch for train dataloader :  70\n",
      "steps per epoch for validation dataloader :  7\n"
     ]
    }
   ],
   "source": [
    "steps_per_epoch = int(len(train_dataloader)/BATCH_SIZE)\n",
    "print('steps per epoch for train dataloader : ',steps_per_epoch)\n",
    "steps_per_epoch_val = int(len(val_dataloader)/BATCH_SIZE)\n",
    "print('steps per epoch for validation dataloader : ',steps_per_epoch_val)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 55
    },
    "id": "3qNnoRnlcziX",
    "outputId": "4334beab-8e79-410a-eb39-5af24d7e4c41"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:Layer train_decoder_LSTM will not use cuDNN kernel since it doesn't meet the cuDNN kernel criteria. It will use generic GPU kernel as fallback when running on GPU\n"
     ]
    }
   ],
   "source": [
    "#initializing encoder and decoder.\n",
    "CNN_Encoder = CNN_Encoder(embedding_dim,image_features_extract_model)\n",
    "decoder = decoder(target_vocab_size, embedding_dim, input_len_dec,units,attention_concat())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {
    "id": "GhcvT7Hdwq1R"
   },
   "outputs": [],
   "source": [
    "#!rm -rf -r /content/Checkpoint"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.8 Model Checkpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {
    "id": "-tDwc6OYqEhe"
   },
   "outputs": [],
   "source": [
    "#creating model checkpoint because we are using custom training.\n",
    "checkpoint_path = \"/content/drive/My Drive/case_study_2/checkpoint\"\n",
    "ckpt = tf.train.Checkpoint(encoder=CNN_Encoder,\n",
    "                           decoder=decoder,\n",
    "                           optimizer = optimizer)\n",
    "ckpt_manager = tf.train.CheckpointManager(ckpt, checkpoint_path, max_to_keep=5)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {
    "id": "wIcr9wJbqF6M"
   },
   "outputs": [],
   "source": [
    "# determining start epoch based on latest checkpoint saved.\n",
    "start_epoch = 0\n",
    "if ckpt_manager.latest_checkpoint:\n",
    "  start_epoch = int(ckpt_manager.latest_checkpoint.split('-')[-1])\n",
    "  print(start_epoch)\n",
    "  # restoring the latest checkpoint in checkpoint_path\n",
    "  ckpt.restore(ckpt_manager.latest_checkpoint)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {
    "id": "8Ax8NXUVKJun"
   },
   "outputs": [],
   "source": [
    "#storing loss values for plotting later\n",
    "tr_loss_plot = []\n",
    "val_loss_plot = []"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.9 Custom Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {
    "id": "036LwvNEQxxU"
   },
   "outputs": [],
   "source": [
    "@tf.function\n",
    "def train_step(inputs, target):\n",
    "        loss2 = 0\n",
    "        img1,img2,input_text = inputs[0],inputs[1],inputs[2] #seperating inputs\n",
    "\n",
    "        # initializing the hidden state and cell_state for each batch\n",
    "        # because the captions are not related from image to image\n",
    "        hidden = decoder.reset_hidden_state(batch_size=BATCH_SIZE)\n",
    "        cell_state = decoder.reset_cell_state(batch_size=BATCH_SIZE)\n",
    "        \n",
    "        print(\"=\"*20, \"ENCODER\", \"=\"*20)\n",
    "        with tf.GradientTape() as tape:#calculating gradient\n",
    "\n",
    "          encoder_output_1 = CNN_Encoder(img1)#passing 1st image in encoder\n",
    "          encoder_output_2 = CNN_Encoder(img2)#passing second image in encoder\n",
    "          encoder_output = tf.keras.layers.Concatenate(axis=-1)([encoder_output_1,encoder_output_2])#concatenating outputs of both encoders\n",
    "          encoder_output = Dropout(rate)(encoder_output)#dropout\n",
    "          print(\"ENCODER ==> OUTPUT SHAPE\",encoder_output.shape)\n",
    "          #encoder output shape == (batch_size, attention_features_shape,2048)\n",
    "          print(\"=\"*20, \"DECODER\", \"=\"*20)\n",
    "          decoder_output = decoder(input_text,hidden,encoder_output,cell_state)\n",
    "\n",
    "          for i in range(decoder_output.shape[1]):\n",
    "            #calculating batch loss at every tim step and summing it up.\n",
    "            loss2 += loss_function(target[:, i], decoder_output[:,i])\n",
    "            \n",
    "          print(\"FINAL OUTPUT SHAPE\",decoder_output.shape)\n",
    "  \n",
    "        total_loss = loss2/int(decoder_output.shape[1])#dividing the loss with total time steps to get total loss.\n",
    "        trainable_variables = CNN_Encoder.trainable_variables + decoder.trainable_variables #getting trainable variable of both the encoder and decoder.\n",
    "        gradients = tape.gradient(loss2,trainable_variables)#Computes the gradient using operations recorded in context of this tape.\n",
    "        optimizer.apply_gradients(zip(gradients,trainable_variables))# Apply gradients to variables.\n",
    "        return total_loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {
    "id": "-PeDEajl-hs4"
   },
   "outputs": [],
   "source": [
    "#validatoion step performs the same steps as train_Step but without dropout and without calculating gradients, it is used just to predict loss on unseen data.\n",
    "@tf.function\n",
    "def validation_step(inputs, target):\n",
    "        loss2 = 0\n",
    "        img1,img2,input_text = inputs[0],inputs[1],inputs[2]\n",
    "\n",
    "        hidden = decoder.reset_hidden_state(batch_size=BATCH_SIZE)\n",
    "        cell_state = decoder.reset_cell_state(batch_size=BATCH_SIZE)\n",
    "        \n",
    "        encoder_output_1 = CNN_Encoder(img1)\n",
    "        encoder_output_2 = CNN_Encoder(img2)\n",
    "        encoder_output = tf.keras.layers.Concatenate(axis=-1)([encoder_output_1,encoder_output_2])\n",
    "          \n",
    "        decoder_output = decoder(input_text,hidden,encoder_output,cell_state)\n",
    "        for i in range(decoder_output.shape[1]):\n",
    "            loss2 += loss_function(target[:, i], decoder_output[:,i])\n",
    "        total_loss = loss2/int(target.shape[1])\n",
    "        return  total_loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 1000
    },
    "id": "eXm3Sm3O_dvZ",
    "outputId": "279ce3e6-d469-4be8-f538-b879fc163f71"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==================== ENCODER ====================\n",
      "ENCODER ==> OUTPUT SHAPE (7, 49, 2048)\n",
      "==================== DECODER ====================\n",
      "================================================== attention ==================================================\n",
      "Tensor(\"attention_concat/ExpandDims:0\", shape=(7, 1, 512), dtype=float32)\n",
      "attention shapes (7, 49, 2048),(7, 1, 512) :\n",
      "similarities shape: (7, 49, 1)\n",
      "attention_weights shape: (7, 49, 1)\n",
      "initial context shape (7, 49, 2048)\n",
      "final context shape (7, 2048)\n",
      "decoder_embedding shape: Tensor(\"Shape:0\", shape=(3,), dtype=int32)\n",
      "attention_context shape: (7, 1, 2048)\n",
      "shape after concat (7, 1, 2304)\n",
      "(7, 512) (7, 512)\n",
      "================================================== attention ==================================================\n",
      "Tensor(\"attention_concat/ExpandDims:0\", shape=(7, 1, 512), dtype=float32)\n",
      "attention shapes (7, 49, 2048),(7, 1, 512) :\n",
      "similarities shape: (7, 49, 1)\n",
      "attention_weights shape: (7, 49, 1)\n",
      "initial context shape (7, 49, 2048)\n",
      "final context shape (7, 2048)\n",
      "decoder_embedding shape: Tensor(\"Shape:0\", shape=(3,), dtype=int32)\n",
      "attention_context shape: (7, 1, 2048)\n",
      "shape after concat (7, 1, 2304)\n",
      "(7, 512) (7, 512)\n",
      "FINAL OUTPUT SHAPE (7, 123, 1486)\n",
      "==================== ENCODER ====================\n",
      "ENCODER ==> OUTPUT SHAPE (7, 49, 2048)\n",
      "==================== DECODER ====================\n",
      "FINAL OUTPUT SHAPE (7, 123, 1486)\n",
      "Epoch 1 batch 0 loss 0.7399\n",
      "Epoch 1 batch 23 loss 0.6828\n",
      "Epoch 1 batch 46 loss 0.4002\n",
      "Epoch 1 batch 69 loss 0.4925\n",
      "Epoch 1 Loss 0.642032 val_Loss 0.557420\n",
      "Time taken for 1 epoch 217.89571070671082 sec\n",
      "\n",
      "Epoch 2 batch 0 loss 0.6824\n",
      "Epoch 2 batch 23 loss 0.6171\n",
      "Epoch 2 batch 46 loss 0.3478\n",
      "Epoch 2 batch 69 loss 0.4503\n",
      "Epoch 2 Loss 0.585244 val_Loss 0.511624\n",
      "Time taken for 1 epoch 112.79011392593384 sec\n",
      "\n",
      "Epoch 3 batch 0 loss 0.6404\n",
      "Epoch 3 batch 23 loss 0.5733\n",
      "Epoch 3 batch 46 loss 0.3197\n",
      "Epoch 3 batch 69 loss 0.4229\n",
      "Epoch 3 Loss 0.549486 val_Loss 0.482062\n",
      "Time taken for 1 epoch 113.42280673980713 sec\n",
      "\n",
      "Epoch 4 batch 0 loss 0.6142\n",
      "Epoch 4 batch 23 loss 0.5446\n",
      "Epoch 4 batch 46 loss 0.3003\n",
      "Epoch 4 batch 69 loss 0.4045\n",
      "Epoch 4 Loss 0.525508 val_Loss 0.462615\n",
      "Time taken for 1 epoch 112.53309226036072 sec\n",
      "\n",
      "Epoch 5 batch 0 loss 0.5959\n",
      "Epoch 5 batch 23 loss 0.5254\n",
      "Epoch 5 batch 46 loss 0.2880\n",
      "Epoch 5 batch 69 loss 0.3918\n",
      "Epoch 5 Loss 0.509468 val_Loss 0.449865\n",
      "Time taken for 1 epoch 112.92834377288818 sec\n",
      "\n",
      "Epoch 6 batch 0 loss 0.5816\n",
      "Epoch 6 batch 23 loss 0.5096\n",
      "Epoch 6 batch 46 loss 0.2785\n",
      "Epoch 6 batch 69 loss 0.3813\n",
      "Epoch 6 Loss 0.497455 val_Loss 0.439923\n",
      "Time taken for 1 epoch 113.26151394844055 sec\n",
      "\n",
      "Epoch 7 batch 0 loss 0.5750\n",
      "Epoch 7 batch 23 loss 0.4993\n",
      "Epoch 7 batch 46 loss 0.2721\n",
      "Epoch 7 batch 69 loss 0.3750\n",
      "Epoch 7 Loss 0.488211 val_Loss 0.431838\n",
      "Time taken for 1 epoch 112.52667427062988 sec\n",
      "\n",
      "Epoch 8 batch 0 loss 0.5687\n",
      "Epoch 8 batch 23 loss 0.4915\n",
      "Epoch 8 batch 46 loss 0.2696\n",
      "Epoch 8 batch 69 loss 0.3693\n",
      "Epoch 8 Loss 0.480725 val_Loss 0.424944\n",
      "Time taken for 1 epoch 112.16953420639038 sec\n",
      "\n",
      "Epoch 9 batch 0 loss 0.5646\n",
      "Epoch 9 batch 23 loss 0.4797\n",
      "Epoch 9 batch 46 loss 0.2629\n",
      "Epoch 9 batch 69 loss 0.3623\n",
      "Epoch 9 Loss 0.474615 val_Loss 0.420617\n",
      "Time taken for 1 epoch 112.71247291564941 sec\n",
      "\n",
      "Epoch 10 batch 0 loss 0.5578\n",
      "Epoch 10 batch 23 loss 0.4761\n",
      "Epoch 10 batch 46 loss 0.2582\n",
      "Epoch 10 batch 69 loss 0.3551\n",
      "Epoch 10 Loss 0.469332 val_Loss 0.415542\n",
      "Time taken for 1 epoch 112.3783802986145 sec\n",
      "\n",
      "Epoch 11 batch 0 loss 0.5495\n",
      "Epoch 11 batch 23 loss 0.4730\n",
      "Epoch 11 batch 46 loss 0.2574\n",
      "Epoch 11 batch 69 loss 0.3538\n",
      "Epoch 11 Loss 0.464919 val_Loss 0.411469\n",
      "Time taken for 1 epoch 113.27892112731934 sec\n",
      "\n",
      "Epoch 12 batch 0 loss 0.5478\n",
      "Epoch 12 batch 23 loss 0.4673\n",
      "Epoch 12 batch 46 loss 0.2544\n",
      "Epoch 12 batch 69 loss 0.3494\n",
      "Epoch 12 Loss 0.461339 val_Loss 0.407857\n",
      "Time taken for 1 epoch 112.26357936859131 sec\n",
      "\n",
      "Epoch 13 batch 0 loss 0.5451\n",
      "Epoch 13 batch 23 loss 0.4636\n",
      "Epoch 13 batch 46 loss 0.2499\n",
      "Epoch 13 batch 69 loss 0.3467\n",
      "Epoch 13 Loss 0.457947 val_Loss 0.405693\n",
      "Time taken for 1 epoch 112.62867665290833 sec\n",
      "\n",
      "Epoch 14 batch 0 loss 0.5440\n",
      "Epoch 14 batch 23 loss 0.4577\n",
      "Epoch 14 batch 46 loss 0.2489\n",
      "Epoch 14 batch 69 loss 0.3445\n",
      "Epoch 14 Loss 0.455325 val_Loss 0.403787\n",
      "Time taken for 1 epoch 113.71763110160828 sec\n",
      "\n",
      "Epoch 15 batch 0 loss 0.5405\n",
      "Epoch 15 batch 23 loss 0.4551\n",
      "Epoch 15 batch 46 loss 0.2468\n",
      "Epoch 15 batch 69 loss 0.3415\n",
      "Epoch 15 Loss 0.452913 val_Loss 0.400609\n",
      "Time taken for 1 epoch 114.04616856575012 sec\n",
      "\n",
      "Epoch 16 batch 0 loss 0.5388\n",
      "Epoch 16 batch 23 loss 0.4513\n",
      "Epoch 16 batch 46 loss 0.2446\n",
      "Epoch 16 batch 69 loss 0.3414\n",
      "Epoch 16 Loss 0.450415 val_Loss 0.399131\n",
      "Time taken for 1 epoch 113.89366817474365 sec\n",
      "\n",
      "Epoch 17 batch 0 loss 0.5344\n",
      "Epoch 17 batch 23 loss 0.4518\n",
      "Epoch 17 batch 46 loss 0.2410\n",
      "Epoch 17 batch 69 loss 0.3393\n",
      "Epoch 17 Loss 0.449091 val_Loss 0.397370\n",
      "Time taken for 1 epoch 113.17099404335022 sec\n",
      "\n",
      "Epoch 18 batch 0 loss 0.5322\n",
      "Epoch 18 batch 23 loss 0.4488\n",
      "Epoch 18 batch 46 loss 0.2412\n",
      "Epoch 18 batch 69 loss 0.3369\n",
      "Epoch 18 Loss 0.447399 val_Loss 0.395111\n",
      "Time taken for 1 epoch 112.71352863311768 sec\n",
      "\n",
      "Epoch 19 batch 0 loss 0.5338\n",
      "Epoch 19 batch 23 loss 0.4466\n",
      "Epoch 19 batch 46 loss 0.2417\n",
      "Epoch 19 batch 69 loss 0.3328\n",
      "Epoch 19 Loss 0.445790 val_Loss 0.394529\n",
      "Time taken for 1 epoch 112.86571311950684 sec\n",
      "\n",
      "Epoch 20 batch 0 loss 0.5291\n",
      "Epoch 20 batch 23 loss 0.4464\n",
      "Epoch 20 batch 46 loss 0.2392\n",
      "Epoch 20 batch 69 loss 0.3349\n",
      "Epoch 20 Loss 0.444172 val_Loss 0.393788\n",
      "Time taken for 1 epoch 113.155282497406 sec\n",
      "\n",
      "Epoch 21 batch 0 loss 0.5292\n",
      "Epoch 21 batch 23 loss 0.4459\n",
      "Epoch 21 batch 46 loss 0.2418\n",
      "Epoch 21 batch 69 loss 0.3319\n",
      "Epoch 21 Loss 0.443443 val_Loss 0.391837\n",
      "Time taken for 1 epoch 113.61511325836182 sec\n",
      "\n",
      "Epoch 22 batch 0 loss 0.5273\n",
      "Epoch 22 batch 23 loss 0.4438\n",
      "Epoch 22 batch 46 loss 0.2385\n",
      "Epoch 22 batch 69 loss 0.3333\n",
      "Epoch 22 Loss 0.442173 val_Loss 0.391354\n",
      "Time taken for 1 epoch 112.47574853897095 sec\n",
      "\n",
      "Epoch 23 batch 0 loss 0.5255\n",
      "Epoch 23 batch 23 loss 0.4407\n",
      "Epoch 23 batch 46 loss 0.2368\n",
      "Epoch 23 batch 69 loss 0.3294\n",
      "Epoch 23 Loss 0.440721 val_Loss 0.390699\n",
      "Time taken for 1 epoch 112.36650800704956 sec\n",
      "\n",
      "Epoch 24 batch 0 loss 0.5240\n",
      "Epoch 24 batch 23 loss 0.4429\n",
      "Epoch 24 batch 46 loss 0.2389\n",
      "Epoch 24 batch 69 loss 0.3308\n",
      "Epoch 24 Loss 0.439941 val_Loss 0.390132\n",
      "Time taken for 1 epoch 113.27413249015808 sec\n",
      "\n",
      "Epoch 25 batch 0 loss 0.5220\n",
      "Epoch 25 batch 23 loss 0.4412\n",
      "Epoch 25 batch 46 loss 0.2351\n",
      "Epoch 25 batch 69 loss 0.3292\n",
      "Epoch 25 Loss 0.438844 val_Loss 0.389193\n",
      "Time taken for 1 epoch 112.00620913505554 sec\n",
      "\n",
      "Epoch 26 batch 0 loss 0.5193\n",
      "Epoch 26 batch 23 loss 0.4405\n",
      "Epoch 26 batch 46 loss 0.2377\n",
      "Epoch 26 batch 69 loss 0.3280\n",
      "Epoch 26 Loss 0.438063 val_Loss 0.389090\n",
      "Time taken for 1 epoch 112.44405317306519 sec\n",
      "\n",
      "Epoch 27 batch 0 loss 0.5208\n",
      "Epoch 27 batch 23 loss 0.4388\n",
      "Epoch 27 batch 46 loss 0.2365\n",
      "Epoch 27 batch 69 loss 0.3271\n",
      "Epoch 27 Loss 0.437518 val_Loss 0.388400\n",
      "Time taken for 1 epoch 112.49250316619873 sec\n",
      "\n",
      "Epoch 28 batch 0 loss 0.5201\n",
      "Epoch 28 batch 23 loss 0.4395\n",
      "Epoch 28 batch 46 loss 0.2347\n",
      "Epoch 28 batch 69 loss 0.3283\n",
      "Epoch 28 Loss 0.436361 val_Loss 0.387428\n",
      "Time taken for 1 epoch 112.86984848976135 sec\n",
      "\n",
      "Epoch 29 batch 0 loss 0.5192\n",
      "Epoch 29 batch 23 loss 0.4360\n",
      "Epoch 29 batch 46 loss 0.2365\n",
      "Epoch 29 batch 69 loss 0.3262\n",
      "Epoch 29 Loss 0.435995 val_Loss 0.386646\n",
      "Time taken for 1 epoch 113.07509803771973 sec\n",
      "\n",
      "Epoch 30 batch 0 loss 0.5181\n",
      "Epoch 30 batch 23 loss 0.4369\n",
      "Epoch 30 batch 46 loss 0.2357\n",
      "Epoch 30 batch 69 loss 0.3253\n",
      "Epoch 30 Loss 0.434778 val_Loss 0.386960\n",
      "Time taken for 1 epoch 114.47582411766052 sec\n",
      "\n",
      "Epoch 31 batch 0 loss 0.5165\n",
      "Epoch 31 batch 23 loss 0.4344\n",
      "Epoch 31 batch 46 loss 0.2318\n",
      "Epoch 31 batch 69 loss 0.3250\n",
      "Epoch 31 Loss 0.434164 val_Loss 0.386817\n",
      "Time taken for 1 epoch 114.17380809783936 sec\n",
      "\n",
      "decreasing learning rate from 0.000010 to 0.000001 :\n",
      "Epoch 32 batch 0 loss 0.5157\n",
      "Epoch 32 batch 23 loss 0.4368\n",
      "Epoch 32 batch 46 loss 0.2318\n",
      "Epoch 32 batch 69 loss 0.3250\n",
      "Epoch 32 Loss 0.433464 val_Loss 0.386289\n",
      "Time taken for 1 epoch 113.47918653488159 sec\n",
      "\n",
      "Epoch 33 batch 0 loss 0.5174\n",
      "Epoch 33 batch 23 loss 0.4367\n",
      "Epoch 33 batch 46 loss 0.2306\n",
      "Epoch 33 batch 69 loss 0.3262\n",
      "Epoch 33 Loss 0.433450 val_Loss 0.386834\n",
      "Time taken for 1 epoch 112.03699803352356 sec\n",
      "\n",
      "decreasing learning rate from 0.000001 to 0.000000 :\n",
      "Epoch 34 batch 0 loss 0.5144\n",
      "Epoch 34 batch 23 loss 0.4365\n",
      "Epoch 34 batch 46 loss 0.2318\n",
      "Epoch 34 batch 69 loss 0.3272\n",
      "Epoch 34 Loss 0.433114 val_Loss 0.387015\n",
      "Time taken for 1 epoch 112.35937666893005 sec\n",
      "\n",
      "decreasing learning rate from 0.000000 to 0.000000 :\n",
      "stopping due to no change in validation loss\n"
     ]
    }
   ],
   "source": [
    "#custom training\n",
    "import os,time\n",
    "EPOCHS = 40\n",
    "\n",
    "for epoch in range(start_epoch,EPOCHS):\n",
    "  start = time.time()\n",
    "  total_loss = 0\n",
    "  total_loss_val = 0\n",
    "  for (batch,(inputs,target)) in (enumerate(train_dataloader)):\n",
    "      t_loss = train_step(inputs,target)#calculating loss per batch\n",
    "      total_loss+=t_loss#adding batch loss to total loss variable\n",
    "      if batch % int(steps_per_epoch/3) ==0:\n",
    "        print('Epoch {} batch {} loss {:.4f}'.format(epoch+1,batch,t_loss.numpy()))\n",
    "      if batch >= steps_per_epoch:  # manually detect the end of the epoch\n",
    "           break  \n",
    "\n",
    "  for (batch,(inputs,target)) in (enumerate(val_dataloader)):\n",
    "    t_loss = validation_step(inputs,target)# calculating validation loss\n",
    "    total_loss_val+=t_loss\n",
    "     \n",
    "    if batch >= steps_per_epoch_val:  # manually detect the end of the epoch\n",
    "           break  \n",
    "  #appending total loss value in the list for plotting\n",
    "  tr_loss_plot.append((total_loss/steps_per_epoch).numpy())\n",
    "  val_loss_plot.append((total_loss_val/steps_per_epoch_val).numpy())\n",
    "  if epoch % 1 == 0:\n",
    "    ckpt_manager.save()#saving model after every epoch\n",
    "    \n",
    "  train_dataloader.on_epoch_end()#shuffling train indeces\n",
    "  val_dataloader.on_epoch_end()#shuffling validation indeces\n",
    "\n",
    "  print ('Epoch {} Loss {:.6f} val_Loss {:.6f}'.format(epoch + 1,\n",
    "                                         total_loss/steps_per_epoch,total_loss_val/steps_per_epoch_val))\n",
    "  print ('Time taken for 1 epoch {} sec\\n'.format(time.time() - start))\n",
    "  \n",
    "  if (len(val_loss_plot)>3) and val_loss_plot[-1] > np.asarray(val_loss_plot[-3]):\n",
    "    #custom learning rate reducer, reduces lr if loss doesent change for last 3 epochs\n",
    "    new_learning_rate = 0.1*optimizer.lr.numpy()\n",
    "    print('decreasing learning rate from {:.7f} to {:.7f} :'.format(optimizer.lr.numpy(),new_learning_rate))\n",
    "    optimizer.lr.assign(new_learning_rate)\n",
    "\n",
    "  if (len(val_loss_plot)>6) and val_loss_plot[-1] > np.asarray(val_loss_plot[-6]):\n",
    "    #stopping if there is no chamge in validation loss for 6 epochs.\n",
    "    print('stopping due to no change in validation loss')\n",
    "    break\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.10 Plotting losses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 350
    },
    "id": "zNkV0vrU-NwP",
    "outputId": "057fcf68-164e-45f6-e41e-ac9511028948"
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmcAAAFNCAYAAABFbcjcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXhU1f3H8feZmUx2kpANSNhJQBbZQUUguCIqWBW1WivWpVqt2lZa7Ka1tnVpbbWlrlX7s1WqUFtU1KoQEFcWQfYt7CAESMi+n98fdwgBQTIhk5lMPq/nuc/M3GXud+6T0o/n3HuOsdYiIiIiIqHBFewCREREROQwhTMRERGREKJwJiIiIhJCFM5EREREQojCmYiIiEgIUTgTERERCSEKZyIizcgYc58x5h/BrkNEWi+FMxFptYwxW4wx5wThvC8YY6qMMSXGmAPGmHeNMX2a8D1BqV9EQpvCmYhI0zxsrY0DMoG9wAvBLUdEwoXCmYiEHWNMpDHmT8aYXb7lT8aYSN+2FGPMG8aYQl+r1wfGGJdv20+MMTuNMcXGmHXGmLNPdC5rbRnwEtD/OLVMNMas8p0v1xhzim/9i0AX4HVfC9yPm+v3i0jrpnAmIuHoZ8BpwCBgIDAC+Llv24+AHUAqkA78FLDGmN7A7cBwa208cD6w5UQnMsbEAdcAnx9jWzbwMnCX73xzcMKY11p7LbANuNhaG2etfbjJv1ZEworCmYiEo2uA+621e621+cCvgGt926qBjkBXa221tfYD60wyXAtEAn2NMRHW2i3W2k1fc467jTGFwEYgDphyjH2uBN601r5rra0Gfg9EA2c0w28UkTClcCYi4agTsLXB562+dQCP4ASq/xlj8owx0wCstRtxWrjuA/YaY2YYYzpxfL+31iZaaztYayceJ8gdUYe1tg7YDmQ08XeJSBugcCYi4WgX0LXB5y6+dVhri621P7LW9gAmAj88dG+ZtfYla+2ZvmMt8FBz1mGMMUBnYKdvlT3J7xeRMKRwJiKtXYQxJqrB4sG5z+vnxphUY0wK8EvgHwDGmIuMMb18QekgTndmnTGmtzHmLN+DAxVAOVB3krW9AlxojDnbGBOBc79bJfCRb/seoMdJnkNEwozCmYi0dnNwgtSh5T7gAWAx8AWwAljqWweQBbwHlAAfA3+11s7Dud/sQWAf8CWQBtxzMoVZa9cB3wL+7Pvei3EeAKjy7fI7nBBZaIy5+2TOJSLhwzj3wYqIiIhIKFDLmYiIiEgIUTgTERERCSEKZyIiIiIhROFMREREJIQonImIiIiEEE+wC2guKSkptlu3bgE/T2lpKbGxsQE/TzjRNfOfrpl/dL38p2vmP10z/+maHd+SJUv2WWtTj7UtoOHMGDMeeAxwA89aax88xj5X4IxLZIHl1tqrfetrccYnAthmrZ34defq1q0bixcvbsbqjy03N5ecnJyAnyec6Jr5T9fMP7pe/tM185+umf90zY7PGLP1eNsCFs6MMW5gOnAusANYZIyZba1d3WCfLJxBHkdZawuMMWkNvqLcWjsoUPWJiIiIhKJA3nM2Athorc3zjYY9A5h01D43AdOttQUA1tq9AaxHREREJOQFMpxlANsbfN7hW9dQNpBtjPnQGPOJrxv0kChjzGLf+ksCWKeIiIhIyAjY9E3GmMuB8dbaG32frwVGWmtvb7DPG0A1cAWQCSwABlhrC40xGdbancaYHsBc4Gxr7aajznEzcDNAenr60BkzZgTktzRUUlJCXFxcwM8TTnTN/Kdr5h9dL//pmvmvrV0zYwyxsbG43e4mf4e1FmNMM1bV+tTW1lJaWsrReWvcuHFLrLXDjnVMIB8I2Al0bvA507euoR3Ap9baamCzMWY9zqTEi6y1OwGstXnGmFxgMHBEOLPWPg08DTBs2DDbEjcd6uZG/+ma+U/XzD+6Xv7TNfNfW7tmmzdvJj4+nuTk5CYHrOLiYuLj45u5stbDWsv+/fspLi6me/fujT4ukN2ai4AsY0x3Y4wXuAqYfdQ+/wFyAIwxKTjdnHnGmCRjTGSD9aOA1YiIiEiLqKioOKlgJk7rY3JyMhUVFX4dF7CWM2ttjTHmduAdnKE0nrPWrjLG3A8sttbO9m07zxizGqgFplpr9xtjzgCeMsbU4QTIBxs+5SkiIiKBp2B28ppyDQM6zpm1dg4w56h1v2zw3gI/9C0N9/kIGBDI2kRERERCkaZvEhERkZBTWFjIX//6V7+PmzBhAoWFhX4fN2XKFGbOnOn3cYGgcNZI1bV1vL58FxsLaoNdioiISNg7Xjirqan52uPmzJlDYmJioMpqEQpnjeQ2hntnr2Le9q//oxAREZGTN23aNDZt2sSgQYMYPnw4o0ePZuLEifTt2xeASy65hKFDh9KvXz+efvrp+uO6devGvn372LJlC6eccgo33XQT/fr147zzzqO8vLxR537//fcZPHgwAwYM4Dvf+Q6VlZX1NfXt25dTTz2Vu+++G4BXX32V/v37M3DgQMaMGdMsvz1sJj4PNJfLMDorhXmrd1NXZ3G5dJOkiIi0Db96fRWrdxX5fVxtbe1xx0nr26kd917c77jHPvjgg6xcuZJly5aRm5vLhRdeyMqVK+uHpHjuuedo37495eXlDB8+nMsuu4zk5OQjvmPDhg28/PLLPPPMM1xxxRXMmjWLb33rW19bc0VFBVOmTOH9998nOzubb3/72zzxxBNce+21vPbaa6xduxZjTH3X6f33388777xDRkZGk7pTj0UtZ34Yk5VKUZVl9W7//0BFRESk6UaMGHHEWGGPP/44AwcO5LTTTmP79u1s2LDhK8d0796dQYOcabqHDh3Kli1bTniedevW0b17d7KzswG47rrrWLBgAQkJCURFRXHDDTfw73//m5iYGABGjRrFlClTeOaZZ6itbZ5bn9Ry5ofR2SkAzF+fT/+MhCBXIyIi0jK+roXr6zTnILSxsbH173Nzc3nvvff4+OOPiYmJIScn55hjiUVGRta/d7vdje7WPBaPx8Nnn33G+++/z8yZM/nLX/7C3LlzefLJJ/n000958803GTp0KEuWLPlKC56/1HLmh7T4KLq2c7FgfX6wSxEREQlr8fHxFBcXH3PbwYMHSUpKIiYmhrVr1/LJJ58023l79+7Nli1b2LhxIwAvvvgiY8eOpaSkhIMHDzJhwgT++Mc/snz5cgA2bdrEyJEjuf/++0lNTWX79u1f9/WNopYzPw1IcfP2lgKKK6qJj4oIdjkiIiJhKTk5mVGjRtG/f3+io6NJT0+v3zZ+/HiefPJJTjnlFHr37s1pp53WbOeNiori+eefZ/LkydTU1DB8+HBuueUWDhw4wKRJk6ioqMBay6OPPgrA1KlT2bBhA9Zazj77bAYOHHjSNSic+al/ips38qr5aNN+zu/XIdjliIiIhK2XXnrpmOsjIyN56623jrnt0H1lKSkprFy5sn79oacrj+eFF16of3/22Wfz+eefH7G9Y8eOfPbZZ1857t///vfXfm9TqFvTT70SXcRFepivrk0REREJALWc+cnjMpzRM5kF6/Ox1mreMRERkVbktttu48MPPzxi3Z133sn1118fpIq+SuGsCcZkp/K/1XvI21dKz9S4YJcjIiIijTR9+vRgl3BC6tZsgrHZqQDMX6euTREREWleCmdN0Ll9DD1SY3XfmYiIiDQ7hbMmGpudyqeb91NRrYnQRUREpPkonDXRmOxUKqrr+GzzgWCXIiIiImFE4ayJTuuejNfjUtemiIhICIiLO/4Delu2bKF///4tWM3JUThromivm5Hd22sqJxEREWlWGkrjJIzNTuWBN9ews7CcjMToYJcjIiISGG9Ngy9X+H1YdG0NuI8TNToMgAsePO6x06ZNo3Pnztx2220A3HfffXg8HubNm0dBQQHV1dU88MADTJo0ya+aKioquPXWW1m8eDEej4dHH32UcePGsWrVKq6//nqqqqqoq6tj1qxZdOrUiSuuuIIdO3ZQW1vLL37xC6688kq/ztcUajk7CYeG1FDrmYiISPO68soreeWVV+o/v/LKK1x33XW89tprLF26lHnz5vGjH/0Ia61f3zt9+nSMMaxYsYKXX36Z6667joqKCp588knuvPNOli1bxuLFi8nMzOTtt9+mU6dOLF++nJUrVzJ+/Pjm/pnHpJazk9ArLY5OCVHMX5fPN0d0CXY5IiIigfE1LVxfp7y4mPj4+CYdO3jwYPbu3cuuXbvIz88nKSmJDh068IMf/IAFCxbgcrnYuXMne/bsoUOHxs91vXDhQr7//e8D0KdPH7p27cr69es5/fTT+c1vfsOOHTu49NJLycrKYsCAAfzoRz/iJz/5CRdddBGjR49u0m/xl1rOToIxhrG9U/lw4z6qa+uCXY6IiEhYmTx5MjNnzuRf//oXV155Jf/85z/Jz89nyZIlLFu2jPT0dCoqKprlXFdffTWzZ88mOjqaCRMmMHfuXLKzs1m6dCkDBgzg5z//Offff3+znOtEFM5O0pisVIora1i2vTDYpYiIiISVK6+8khkzZjBz5kwmT57MwYMHSUtLIyIignnz5rF161a/v3P06NH885//BGD9+vVs27aN3r17k5eXR48ePbjjjjuYNGkSX3zxBbt27SImJoZvfetbTJ06laVLlzb3TzwmdWuepDN6peB2Geavy2d4t/bBLkdERCRs9OvXj+LiYjIyMujYsSPXXHMNF198MQMGDGDYsGH06dPH7+/83ve+x6233sqAAQPweDy88MILREZG8sorr/Diiy8SERFBhw4d+OlPf8qiRYuYOnUqLpeLiIgInnjiiQD8yq9SODtJCdERDOmSyPz1+dx9fu9glyMiIhJWVqw4/JRoSkoKH3/88TH3KykpOe53dOvWjZUrVwIQFRXF888//5V9pk2bxrRp045Yd/7553P++ec3peyTom7NZjA2O5UVOw+yr6Qy2KWIiIhIK6dw1gzG+IbUWLhhX5ArERERabtWrFjBoEGDjlhGjhwZ7LL8pm7NZtC/UwLtY73MX5/PJYMzgl2OiIhImzRgwACWLVsW7DJOmlrOmoHLZRiTlcKC9fnU1fk3GJ6IiEio8neAV/mqplxDhbNmMiY7lf2lVazeXRTsUkRERE5aVFQU+/fvV0A7CdZa9u/fT1RUlF/HqVuzmYzOcu47m78+n/4ZCUGuRkRE5ORkZmayY8cO8vObPkVhRUWF38Ek3ERFRZGZmenXMQpnzSQ1PpL+Ge2Yvy6f28b1CnY5IiIiJyUiIoLu3buf1Hfk5uYyePDgZqqo7VC3ZjMam53Kkm0FFFVUB7sUERERaaUUzprRmKxUaussH23cH+xSREREpJVSOGtGQ7omERfpYf76pvfPi4iISNumcNaMItwuRvVKZsH6fD3dIiIiIk2icNbMxmansbOwnE35x5/jS0REROR4FM6a2ZjsFADmr9dUTiIiIuI/hbNmlpkUQ8/UWN13JiIiIk0S0HBmjBlvjFlnjNlojJl2nH2uMMasNsasMsa81GD9dcaYDb7lukDW2dzGZqfxad5+Kqprg12KiIiItDIBC2fGGDcwHbgA6At80xjT96h9soB7gFHW2n7AXb717YF7gZHACOBeY0xSoGptbmOyU6isqePTzQeCXYqIiIi0MoFsORsBbLTW5llrq4AZwKSj9rkJmG6tLQCw1u71rT8feNdae8C37V1gfABrbVan9Ugm0uNi/jp1bYqIiIh/AhnOMoDtDT7v8K1rKBvINsZ8aIz5xBgz3o9jQ1ZUhJuRPZKZv37viXcWERERaSDYc2t6gCwgB8gEFhhjBjT2YGPMzcDNAOnp6eTm5gagxCOVlJQ06jwZrmoW5Fcx8625pES37ecuGnvN5DBdM//oevlP18x/umb+0zVrmkCGs51A5wafM33rGtoBfGqtrQY2G2PW44S1nTiBreGxuUefwFr7NPA0wLBhw2xOTs7RuzS73NxcGnOezL3FvLx2AVXte5EzskvA6wpljb1mcpiumX90vfyna+Y/XTP/6Zo1TSCbdBYBWcaY7sYYL3AVMPuoff6DL4QZY1JwujnzgHeA84wxSb4HAc7zrWs1eqbGkZEYra5NERER8UvAWs6stTXGmNtxQpUbeM5au8oYcz+w2Fo7m8MhbDVQC0y11u4HMMb8GifgAdxvrW1Vjz4aYxiTncrry3dRXVtHhLttd22KiIhI4wT0njNr7RxgzlHrftngvQV+6FuOPvY54LlA1hdoY7NTePmzbSzdWsDIHsnBLkdERERaATXnBNAZvVJwuwwLNmhIDREREWkchbMAahcVwdAuSZrKSURERBpN4SzAxvZOZeXOIvKLK4NdioiIiLQCCmcBNiYrFYAP1LUpIiIijaBwFmD9OrUjOdbLAnVtioiISCMonAWYy+UMqbFgwz7q6mywyxEREZEQp3DWAsZmp3KgtIqVuw4GuxQREREJcQpnLeDMrBQA5q9T16aIiIh8PYWzFpASF8mAjAQNqSEiIiInpHDWQs4+JY0l2wrYsq802KWIiIhICFM4ayFXj+iCx2V4/sPNwS5FREREQpjCWQtJaxfFxIEZvLJ4BwfLqoNdjoiIiIQohbMWdMOZ3SmvruWlz7YFuxQREREJUQpnLahvp3aM6pXMCx9tpqqmLtjliIiISAhSOGthN57Zgz1FlcxZsTvYpYiIiEgIUjhrYWOzU+mZGsuzC/OwVjMGiIiIyJEUzlqYy2W44cwerNxZxKebDwS7HBEREQkxCmdBcOmQDJJiInj2Aw2rISIiIkdSOAuCqAg33zqtK++v3UNefkmwyxEREZEQonAWJNee3pUIl4vnP9wS7FJEREQkhCicBUlafBQTB3Vi5pIdFJZVBbscERERCREKZ0GkQWlFRETkaApnQXRKx3ac2SuFv3+0RYPSioiICKBwFnQ3jO7OnqJK3lyxK9iliIiISAhQOAuysVmp9EqL49kPNmtQWhEREVE4CzZnUNrurNpVxCd5GpRWRESkrVM4CwHfGJxB+1gvf1uYF+xSREREJMgUzkLAoUFp31uzV4PSioiItHEKZyHi2tO64nW7eO5DTekkIiLSlimchYjU+Egm+QalLSjVoLQiIiJtlcJZCLlhdHcqqus0KK2IiEgbpnAWQvp0aMfoLA1KKyIi0pYpnIWYG87szt7iSt74QoPSioiItEUKZyFmbHYqWRqUVkREpM1SOAsxxjiD0q7eXcTHefuDXY6IiIi0MIWzxqouhzd+SNqe+QE/1SWDM0iO9fK3DzSshoiISFujcNZYnijYPJ+Ou/8X8FMdGpT2/bV72aRBaUVERNoUhbPGMgb6X05i4SooCvzN+t86rStej4vnFqr1TEREpC1ROPPHgMsxWFj1WsBPlRofyTcGZTBr6Q4OaFBaERGRNiOg4cwYM94Ys84Ys9EYM+0Y26cYY/KNMct8y40NttU2WD87kHU2WkoWxXE9YMXMFjndd870DUr76dYWOZ+IiIgEX8DCmTHGDUwHLgD6At80xvQ9xq7/stYO8i3PNlhf3mD9xEDV6a+9aaNh11LYvyng5+rdId4ZlPbjrVTW1Ab8fCIiIhJ8gWw5GwFstNbmWWurgBnApACer0XsTRvtvFn57xY5342je5BfXMkby3e3yPlEREQkuAIZzjKA7Q0+7/CtO9plxpgvjDEzjTGdG6yPMsYsNsZ8Yoy5JIB1+qUyKhW6nA4rZ0ILDBI7JivFGZR2oQalFRERaQtMoP4P3xhzOTDeWnuj7/O1wEhr7e0N9kkGSqy1lcaY7wJXWmvP8m3LsNbuNMb0AOYCZ1trNx11jpuBmwHS09OHzpgxIyC/paGSkhKyDy4ge8NTLBr2GKVx3QJ+zvk7qnl+ZRV3DolkcJon4OdrbiUlJcTFxQW7jFZF18w/ul7+0zXzn66Z/3TNjm/cuHFLrLXDjrUtkOHsdOA+a+35vs/3AFhrf3ec/d3AAWttwjG2vQC8Ya097p34w4YNs4sXL26O0r9Wbm4uOcP7w++zYdQdcM59AT9nZU0tFz2+kJLKGt75wRjaRUUE/JzNKTc3l5ycnGCX0aromvlH18t/umb+0zXzn67Z8RljjhvOAtmtuQjIMsZ0N8Z4gauAI566NMZ0bPBxIrDGtz7JGBPpe58CjAJWB7BW/8SmQI8cWDmrRbo2Iz1uHpk8kD1FFfz2zTUBP5+IiIgET8DCmbW2BrgdeAcndL1irV1ljLnfGHPo6cs7jDGrjDHLgTuAKb71pwCLfevnAQ9aa0MnnAEMuBwKt8GORS1yukGdE7l5TE9mLNrOgvX5LXJOERERaXkBHefMWjvHWpttre1prf2Nb90vrbWzfe/vsdb2s9YOtNaOs9au9a3/yFo7wLd+gLX2b4Gss0n6XATuSKf1rIXcdU4WPVNjmTbrC4orqlvsvCIiItJyNENAU0W1g+zznNkC6lpmDLKoCKd788uiCn47Z22LnFNERERalsLZyeh/OZTsgS0ftNgph3RJ4sbRPXj5s20s3LCvxc4rIiIiLUPh7GRknw/euBabzumQH56bTY+UWH4y6wtKKmta9NwiIiISWApnJyMiGvpcCGtmQ01li53W6d48lV0Hy3nwLT29KSIiEk4Uzk5W/8uh4iBsfL9FTzu0a3tuGNWdf3yyjY82qntTREQkXCicnaye4yC6vTOdUwv70Xm96Z4Sy49nfUGpujdFRETCgsLZyXJHQN9JsO4tqCpt0VNHe908fPmp7Cws5+G39fSmiIhIOFA4aw4DLofqMiegtbDh3doz5Yxu/P3jrXySt7/Fzy8iIiLNS+GsOXQ5A+I7tfhTm4dMPb83XZNj+PHMLyirUvemiIhIa6Zw1hxcLuh/KWx8D8oOtPjpY7weHr7sVLYdKOPht9e1+PlFRESk+SicNZf+l0FdNax5PSinH9kjmSlndOOFj7bwqbo3RUREWi2Fs+bSaTC07xGUpzYP+fH43nRpH8OPZ31BeVXLTCklIiIizUvhrLkY44x5tvkDKP4yKCXEeD08dNmpbN1fxu//p+5NERGR1kjhrDkNuBywzmToQXJ6z2S+fXpXnvtwM4u3tPz9byIiInJyFM6aU2pvSB8AK2cFtYyfjO9DRmI0U2d+QUW1ujdFRERaE4Wz5jbgMtixCAq2BK2E2Ejn6c3N+0r5g7o3RUREWhWFs+bW71LnNcitZ2f0SuGakV14duFmlmwtCGotIiIi0ngKZ80tqStkjoAVwQ1nAPdMOIVOCdFMnblc3ZsiIiKthMJZIAy4HPaugr1rglpGXKTz9GZefikPvLkaa21Q6xEREZETUzgLhH7fAOMK2nRODZ2ZlcLNY3rwj0+28bu31iqgiYiIhDiFs0CIS4PuY5wBaUMgDN1zQR++fXpXnl6Qx4NvK6CJiIiEskaFM2NMrDHG5XufbYyZaIyJCGxprVz/y50nNncuDXYlGGP41cR+fOu0Ljw1P4+H31mngCYiIhKiGttytgCIMsZkAP8DrgVeCFRRYeGUi8HtDep0Tg0ZY7h/Yn+uGdmFJ3I38fv/KaCJiIiEosaGM2OtLQMuBf5qrZ0M9AtcWWEgOhF6nQsr/w11ofGkpMtl+PWk/nxzRBemz9vEo++uV0ATEREJMY0OZ8aY04FrgDd969yBKSmMDLgMSr6ErR8Gu5J6LpfhN5f056rhnfnz3I386b0NwS5JREREGvA0cr+7gHuA16y1q4wxPYB5gSsrTGRfABGxzlOb3ccEu5p6Lpfht98YQJ21PPb+BoyBu87JDnZZIiIiQiPDmbV2PjAfwPdgwD5r7R2BLCwseGOgzwRY/V+Y8HvweINdUT2Xy/DgpadSZ+FP723AZQx3nJ0V7LJERETavMY+rfmSMaadMSYWWAmsNsZMDWxpYaL/5VBRCJvmBruSr3C5DA9ddiqXDcnk0XfX85e56uIUEREJtsbec9bXWlsEXAK8BXTHeWJTTqTnWRCVGDJPbR7N7TI8fPmpXDo4g9//bz3T520MdkkiIiJtWmPvOYvwjWt2CfAXa221MUaP+TWGxwt9Jzn3nVWVOV2dIcbtMjwyeSB11vLIO+twGcOtOT2DXZaIiEib1NiWs6eALUAssMAY0xUoClRRYaf/ZVBdCuvfDnYlx+V2Gf5wxSAmDerEQ2+v5an5m4JdkoiISJvU2AcCHgceb7BqqzFmXGBKCkPdzoS4DrByFvS/NNjVHJfbZfjD5IHUWfjdW2txGcNNY3oEuywREZE2pbEPBCQYYx41xiz2LX/AaUWTxnC5YcDlTsvZ/tBukfK4XfzxioFceGpHfjNnDc9+kBfskkRERNqUxnZrPgcUA1f4liLg+UAVFZbO+L4zndPcB4JdyQl53C4eu3IQEwZ04IE31/C3hZuDXZKIiEib0dhw1tNae6+1Ns+3/ApQf5c/4jvAad+DVf+GXZ8Hu5oT8rhdPHbVYC7o34Ffv7GaabO+oKI6NKahEhERCWeNDWflxpgzD30wxowCygNTUhgbdQdEt4f37gt2JY0S4Xbx528O5ns5PZmxaDuXTP+QvPySYJclIiIS1hobzm4BphtjthhjtgB/Ab4bsKrCVVQCjJkKebkhOSjtsXjcLn48vg/PTxnOl0UVXPznhby+fFewyxIREQlbjQpn1trl1tqBwKnAqdbawcBZAa0sXA2/ARK6wLv3Ql1dsKtptHF90phzx2h6d4jn+y9/zs//s0LdnCIiIgHQ2JYzAKy1Rb6ZAgB+GIB6wp8nEs76OXz5hXP/WSvSKTGaf333dG4a3Z1/fLKNy5/8iK37S4NdloiISFjxK5wdxTRbFW3NgMmQ3h/m/hpqqoJdjV8i3C5+dmFfnvn2MLbtL+Oixxfy9srdwS5LREQkbJxMODvh9E3GmPHGmHXGmI3GmGnH2D7FGJNvjFnmW25ssO06Y8wG33LdSdQZelwuOOc+KNgCS14Ibi1NdG7fdN68YzQ9UmO55R9L+dXrq6iqaT3dtCIiIqHqa8OZMabYGFN0jKUY6HSCY93AdOACoC/wTWNM32Ps+i9r7SDf8qzv2PbAvcBIYARwrzEmyf+fF8J6nQPdRsP8h6CyONjVNEnn9jG8essZTDmjG89/uIXJT33M9udhM5wAACAASURBVANlwS5LRESkVfvacGatjbfWtjvGEm+tPdHUTyOAjb5x0aqAGcCkRtZ1PvCutfaAtbYAeBcY38hjWwdj4JxfQdk++Ogvwa6mybweF/dN7McT1wwhb28JFz7+Ae+u3hPsskRERFqtk+nWPJEMYHuDzzt86452mTHmC2PMTGNMZz+Pbd0yh0LfSfDRn6Fkb7CrOSkXDOjIG3ecSef2Mdz0f4v57Zw1VNeqm1NERMRfxtoT3jrWtC825nJgvLX2Rt/na4GR1trbG+yTDJRYayuNMd8FrrTWnmWMuRuIstY+4NvvF0C5tfb3R53jZuBmgPT09KEzZswIyG9pqKSkhLi4uGb7vuiynYz47HZ2ZlzAxqybm+17g6Wq1jJjbRVzt9fQK9HFrQMjiawta9Zr1hY0999ZuNP18p+umf90zfyna3Z848aNW2KtHXasbSfqmjwZO4HODT5n+tbVs9bub/DxWeDhBsfmHHVs7tEnsNY+DTwNMGzYMJuTk3P0Ls0uNzeXZj9P7SIyP3+RzMsegPatf1as886G2ct3cc+sL3hgUQ1X9Ipi2oVjMUYP+DZWQP7Owpiul/90zfyna+Y/XbOmCWS35iIgyxjT3RjjBa4CZjfcwRjTscHHicAa3/t3gPOMMUm+BwHO860LTznTWs2k6I01cWAnZn/f6eZ86otKJj/5MSt3Hgx2WSIiIiEvYOHMWlsD3I4TqtYAr1hrVxlj7jfGTPTtdocxZpUxZjlwBzDFd+wB4Nc4AW8RcL9vXXg6NCn6ylmtYlL0xuqZGsdr3xvF9f29bNlfysV/Wci0WV+wr6Qy2KWJiIiErEC2nGGtnWOtzbbW9rTW/sa37pfW2tm+9/dYa/tZawdaa8dZa9c2OPY5a20v3/J8IOsMCfWTov8q2JU0K7fLMDYzgrl353DDqO7MXLKDcY/k8uwHeRoXTURE5BgCGs7ED/WTos9rNZOi+6NdVAQ/v6gv7/xgDEO7JfHAm2sY/9gC5q1r3U+pioiINDeFs1ByaFL09+5rVZOi+6NnahwvXD+C56YMw1q4/vlFfOeFRWzepzk6RUREQOEstHgi4ayfwe7lrW5SdH+d1Sedd+4aw08n9OGzzQc474/z+d2cNRRXVAe7NBERkaBSOAs1rXhSdH95PS5uHtOTuXeP5ZJBGTy1II9xv5/PK4u3U1cXmPH3REREQp3CWahxuQ9Pir7070EupmWkxUfxyOSB/Pe2UXRuH82PZ37BJX/9kCVbC4JdmoiISItTOAtFYTApelMM7JzIrFvO4I9XDuTLgxVc9sRH3P7SUlbvKgp2aSIiIi1G4SwUGeO0npXmw8fTg11Ni3K5DN8YnMm8u3O4bVxP5q3dy4THP+D65z9j0ZbwHepORETkEIWzUJU5DE6Z6JsUPT/Y1bS42EgPU8/vw4fTzuJH52azfMdBJj/5MZc/8RFz1+4hUHPCioiIBJvCWSg7+5dQXQ4LHgl2JUGTGOPl+2dnsfAn47jv4r7sPljBd15YzAWPfcB/l+2kpjY8hxwREZG2S+EslKVkwZBvw+Ln4EBesKsJqhivhymjupM7NYc/TB5ITZ3lzhnLGPeHXF78ZCsV1bXBLlFERKRZKJyFupxp4I6Aub8JdiUhIcLt4rKhmfzvrjE8fe1QkmMj+cV/VnLmQ/P4a+5GijROmoiItHIKZ6EuvgOcdiusnAnbPgl2NSHD5TKc168Dr33vDF6+6TT6dmrHw2+vY9Tv5vLQ22vJL9bk6iIi0jopnLUGo+6EpG7wyrehaFewqwkpxhhO75nM/31nBG98/0zG9E7lyfmbGPXQXH722goNwyEiIq2OwllrEJUA35wBVaUw42rnIQH5iv4ZCUy/eghzf5TDZUMyeHXxDiY8/gEX/3khL368hYNl6vIUEZHQp3DWWqSdApc+A7uWwezvg4aSOK7uKbH87tJT+exnZ/Orif2orbP84r+rGP7b97jj5c/5cOM+TQ8lIiIhyxPsAsQPfSY4E6PPfcCZf/PMu4JdUUhLjPFy3RnduO6MbqzceZBXF2/nP8t2MXv5LjISo5k8LJPLh2aSmRQT7FJFRETqKZy1NqPvhj2r4b37nNa07PODXVGr0D8jgf4ZCdwz4RT+t3oPry7ezmPvb+Cx9zdwZq8UJg/rzHl904mKcAe7VBERaeMUzlobY2DSdNi/EWbdCDe+B6m9g11VqxEV4WbiwE5MHNiJHQVlzFyyg1cX7+COlz8nITqCSwZ1YvKwzvTPSAh2qSIi0kbpnrPWyBsDV70Enkh4+SooLwh2Ra1SZlIMd52TzQc/Hsc/bxxJTu9UXl60nYv+vJAJj33AMwvy2FFQFuwyRUSkjVE4a60SO8OV/4DC7fDq9VBbE+yKWi2XyzCqVwqPXTWYRT89h19P6ofHbfjNnDWc+dA8Jk3/kKcXbGL7AQU1EREJPIWz1qzLaXDhHyBvHrz7y2BXExYSYiK49vRuzL79TOZPzeEn4/tQV2f57Zy1jH54HpP+spCn5iuoiYhI4Oies9Zu6HWwZxV8Mh069IdBVwe7orDRNTmWW3N6cmtOT7btL2POyt3MWbGb3721lt+9tZZTMxOYMKAjFw7oSOf2euJTRESah8JZODj/N5C/Bl6/E5KzoPPwYFcUdrokx3DL2J7cMvbIoPbgW2t58K21DMg4HNS6JCuoiYhI0ymchQN3BEz+OzwzDv51DdycC+06BbuqsNUwqG0/UMacFU5Qe+jttTz09lr6Z7Tjgv4dGZudSt+O7XC5TLBLFhGRVkThLFzEtHemeHr2HGeKp+vfgojoYFcV9jq3j+G7Y3vyXV9Qe2vlbt5c8SWPvLOOR95ZR3Ksl9FZKYzJTuXMrBTS4qOCXbKIiIQ4hbNwcmiKpxlXO1M8XfqMMy6atIjO7WO4eUxPbh7Tk73FFSzcsI8F6/P5YMM+/rPMmbD+lI7tGJOdwtisVIZ2SyLSo0FvRUTkSApn4UZTPIWEtPgoLh2SyaVDMqmrs6zeXcT89fksWJ/P3z7YzFPz84jxujmtRzJjfC1r3VNiMQrTIiJtnsJZOBp9t/MEp6Z4Cgkul6mfPuq2cb0oqazh4037WbA+nwUb8pm7di8AmUnRjMlOZUxWKlVVmphdRKStUjgLR/VTPG3SFE8hKC7Sw7l90zm3bzoAW/eXsmB9PvPX7+O/n+/kpU+3AfDo8nkM6ZrE0K5JDOmSRHZ6PG49XCAiEvYUzsKVN9aZ4umZcc4UTzfNheikYFclx9A1OZZrT4/l2tO7UVVTx/Idhbw6dzEFnngWrM/n30t3Ak6oG9wlkcFdnMA2qHMiCdERQa5eRESam8JZOEvsDFe8CH+/GP5xGVz5T2jXMdhVydfwelwM79ae0h5ecnKGYa1l24Eylm4rYMnWApZsLeQvczdQZ50G0qy0OIZ2TaoPbD1035qISKuncBbuup4OV/wdZt0ET491wlqXkcGuShrJGEPX5Fi6JsfyjcGZAJRU1rB8eyFLthawdFsBb36xm5c/2w5AUkwEgzonMqhzEgM7JzCocyKJMd5g/gQREfGTwllb0OdC576zGVfDCxfChIdh2HeCXZU0UVykh1G9UhjVKwWAujrLpvyS+ta1ZdsLyV2fj/U9U9AtOcYX2BIZ2DmRvp3aaQgPEZEQpnDWVqT3hZvnOQ8IvPED2LUMJjwCnshgVyYnyeUyZKXHk5Uez5XDuwBQXFHNip0HWba9kGXbCvlo0/76sdYi3Ia+Hds5ga1LIgMzEzWMh4hICFE4a0uik+DqV5wx0BY+CnvXwBX/p/vQwlB8VARn9EzhjJ4p9et2Hyxn+fZCPt9eyPLthby6ZAd//3grAAnREZyamcDAzESy0uPISounR2osURFqYRMRaWkKZ22Nyw3n3AsdT4X/fA+ezoErX4TOI4JdmQRYx4RoOiZEM76/E8Zr6ywb95awbLvTFbps+0GemL+J2jqnP9RlnCdJe6XFkZUWVx/aeqbGEe1VaBMRCRSFs7aq3zcgJdu5D+35CXDh72HolGBXJS3I7TL07hBP7w6Hu0Mra2rZsq+MDXuL2bCnpP513tq91PhCmzHQOSmGrLQ4evkCW1ZaHL3S4oiN1D8pIiInS/+StmXp/eCmeTDrBnj9Tuc+tAseBo+e7murIj3u+sDWUHVtHVv3l/oCm2/ZU8wHG/ZRVVtXv1/HhCi6p8TSIzWWHilx9EiNpWdqHJ0SozWArohIIymctXUx7eGamfD+/fDhn2Dvauc+tPgOwa5MQkiE20WvtHh6pcVzQYP1NbV1bDtQVh/W8vJLydtXyuxluyiqqKnfz+tx0T059nBwS/UFt5Q4EmI0kK6ISEMBDWfGmPHAY4AbeNZa++Bx9rsMmAkMt9YuNsZ0A9YA63y7fGKtvSWQtbZpLjec+yvoOBD+e5tzH9oVL0Ln4cGuTEKcx+3yBa04zu93ONBba9lfWuWEtfwS8vaVkpdfyvq9xby3Zk99FylAcqy3vqWtZ5pzj1vP1Dgyk2LU2iYibVLAwpkxxg1MB84FdgCLjDGzrbWrj9ovHrgT+PSor9hkrR0UqPrkGPpfevg+tBcmwIV/gCHfDnZV0goZY0iJiyQlLpIR3dsfsa26to7tB8rIyy9l875S8vaVsCm/lPfX7uFfi6vq9zvU2tYzLZZeqXH09IW2HqmxxHjV6C8i4SuQ/8KNADZaa/MAjDEzgEnA6qP2+zXwEDA1gLVIY3XoDzfnwszvwOzvO/ehjX9Q96FJs4lo0Np2tMKyKjbll7Ipv8RZ9pawZncxb6/8kgaNbXRKiKoPaz3T4ijeX0vHL4tpH+slKSYCj9vVgr9IRKR5BTKcZQDbG3zeARwxb5AxZgjQ2Vr7pjHm6HDW3RjzOVAE/Nxa+0EAa5WG6u9D+xV89DjsWQmT/gopvYJdmYS5xBgvQ7t6Gdo16Yj1lTW1bNtfxqb8EjbuLakPcK8u3k5pVS0ADy9aADhPkyZGR9A+1ktybCTJcV7fey/JcZH179vHOdsV5kQk1Bhr7Yn3asoXG3M5MN5ae6Pv87XASGvt7b7PLmAuMMVau8UYkwvc7bvnLBKIs9buN8YMBf4D9LPWFh11jpuBmwHS09OHzpgxIyC/paGSkhLi4r76X/zhKm3PArLXP4mrrpKtXSezrctlWJd/N3C3tWvWHHTNGsdaS2GlZcv+MqrdURRVWoqrLMXV9oj3xZWWkmo41r92Boj3QlKUi6RIQ1KUbzn0PtJFUpQh2kNYzaKgvzH/6Zr5T9fs+MaNG7fEWjvsWNsCGc5OB+6z1p7v+3wPgLX2d77PCcAmoMR3SAfgADDRWrv4qO/KxRfcjne+YcOG2cWLj7u52eTm5pKTkxPw84SU4i/h7Xtg1b8hOQsu/hN0O7PRh7fJa3aSdM3805jrVVtnKSyrYn9pFftLqjhQWsWB0kryS6rIL67gy4MVfFlUyZ6iCg6UVn3l+Bivmw7tokhvF0WHBOe1o+81vV0k6e2iSI2PJKKVtMLpb8x/umb+0zU7PmPMccNZILs1FwFZxpjuwE7gKuDqQxuttQeB+rlljmo5SwUOWGtrjTE9gCwgL4C1yteJ7wCTn4dB18CbP3QmTx/0LTjv104XqEgr4HYZkuMiSY6LhPSv37eiupa9RZV8WVThLAfL+fKgE9y+LKrgs80H2FNUccRTp4ekxHlJjfcFNt9rWrsjQ1xyrFddqSJyXAELZ9baGmPM7cA7OENpPGetXWWMuR9YbK2d/TWHjwHuN8ZUA3XALdbaA4GqVRop6xz43iew4GH46M+w/i047zcw8CrnRh+RMBEV4aZLcgxdkmOOu09dnTNcyJ6iCvYWV7DH1+q2p6iSfN/n1buK2FdSydEZzmUgOS6yPsCltYvytco54S3N99o+xotLw4mItDkBfR7dWjsHmHPUul8eZ9+cBu9nAbMCWZs0kTcGzrkPBkyG1++C/9wCy1+CC/+oBwakTXG5DKnxkaTGRwIJx92vprauPsQdCnB7iyrYW+y0zO0+WMHyHYXsK/lqV6rHZUiLP9TyFulrfYsiLd55nxgTQYzXQ4zXTazXQ0yku9V0q4rI8WmwIGma9H7wnXdg6Qvw7n3wxOkw+m448y7wRAa7OpGQ4XG76kPV16mqqSO/5HB4a9gSt7e4gs37Svkk7wAHy6u/9nu8bhfRXjexXjcxkU5wOxzePMREuImJdLP/yyo2R2ymfazXNwSJt/59VIQmthcJJoUzaTqXC4Z9B3pfCO/cA7m/hRWv+v3AgIg4g+5mJEaTkRj9tfs1vB+uqLyasupayiprKK2qpbzKeS2rrKGsqpayqlpKq2ooq6xlT3EFZft8n6tqKamoYfamo4eddMR43UeEtuRYL0mx3iPWpcZ7SY2LIiXeq0GBRZqZ/hclJy8+HS5/DgZerQcGRAKsMffDNcbcefMYPGIU+0urKChznmAtKDv0FGsVBaVV9ds27i2hoKyKMt+YckeL8bpJjY/0zQrhrX9/eF0kab730V61yomciMKZNJ/6BwYecQavPfTAgNUk6iKhxmUMSb4WscaqqK6tD2/5JZXsK670vVaxr6SS/OJK8vJL+XTzAQrLjt39Gut1kxjjJTbSTWykhzhf12tspIdYr8f32mBbZMNtbuIiPSRGe4mP8uhhCQlbCmfSvLwxcM69zgMDbzgPDAyJ7wUZv4Osc/VUp0grFhXhplNiNJ1O0PUKzj10B0qryC+urA9u+b7XovJqSqtqKK2spaSyhj1FFZRW1vrW1VBde+LxN42BdlERJMZEkBDtLIkxXhKinfB29PpDnw+FwXAaUFjCj8KZBEZ6X7j+bfhiBhFv3wcvTYaMYTDup9DzLIU0kTDn9bjokOAM2Ouvqpo6Sitr6gPcodB2KMwdLK/mYFkVheXVHCyvprCsmsLyanYUlFNYVsXB8uqvDF/SkDEQE+FrkYt0WuRivIdb8ZxXD3GRzkMVh1rz8r6swa7bS5THTVSEi0jfa1SEm6gIN5Ee571bLXpykhTOJHBcLhh0NZ8dSGNswg5Y8Hv4x6XQ+TQnpPUYG+wKRSQEeT0uvB7/ulwbqquzlFTVcLCsYXhzQltJhfPQhBP2jny/t7iCMl8APPQwxVcm0Vm26ITnj3AbojxuIusDm/ME7aFWvKSYiPrWvfrPvta9pBgvCdERCnhtnMKZBJx1eWDoFBj4Tfj8RVjwB/i/idBttBPSup4R7BJFJIy4XIZ2URG0i4qg80l8T12dpaLGCWullbUs+OgTBgwaQkV1LZXVdVRU11JR0/C981rpe62orqPSt095VS2F5dXsLiyioBGte+2iPCTFekn0BboYrzOGXYTbhddj6t87iznivddz5LaoCLdvqJTD4U8zVIQ2hTNpOZ5IGH6j8yTn0r/DB3+A5y+AHjkw7mfQeUSwKxQRqedyGd8gvx6Ih63t3AzpktQs311XZymuqKHA1z1bUFbFwTLntbCsmsKyKgp83bUFZVXsKqylps5SVVNHde2hxVJVW0dVTZ3f50+IjiApJsIZIiXGS6IvvCXGHBouxQlySbFOMIyOcBPtdRPlcetBjBagcCYtLyIKRn4XhnwbFj8HC/8IfzsXep0DOT+FzKHBrlBEJKBcLkNCTAQJMREn/V3WWmrrbH1Yqw9vNZbqujrKq2rrh0kpKHVCX4Ev/BWUVvFlUQVrdhdxoKyKiuoTB71D99bVB7YIN9G+rtvoCPcR2/K/rGJZzXrfYMiHB0WO9jr38UUftT7G61GXLgpnEkwR0XD6bU6X52fPwIePwbNnQfZ4yLkHOg0KdoUiIiHPGIPHbfC4IZqTG0fuUJArKKuioNQJceVVtZRX11JR7byWV9dSUXXo/aEuXF/XbVk15b5u37KqGkorqnlr8wa/avB6XPWtdceKaY190tbj6+71uJyuXo/L4HG78Lpd9dsi3AaP68juYY/bkN4uilvG9vSr7uakcCbB5411pn0afgN8+pQzqfrTY6HPRTD2x9BxYLArFBFpE6K9bqK9jRsupTFyc3MZPWYs5dW1lFXVUF5VS2llLeXVvocuGrwvq3RmtSirdma1qKj+6qDHx7pN7ysPbQAWpzWxxteaWOPrBq6uraO8upbqisOfG26rqbNU19TRLSVW4UwEgMh4GHM3jLgJPnkSPv4LrH0Duo6CkbdAnwvBpdHFRURaE7fLEOcbVFgaR49rSOiJSoCcn8BdK+DcX0PhdnjlWnhsEHz4OJQXBLtCERGRgFE4k9AVnQij7oA7PocrXoTEzvDuL+DRvvDGDyF/fbArFBERaXZqY5TQ5/ZA34nOsvsL5760z/8Bi/8GPc+G0251Xl36bw0REWn99P9m0rp0PBUumQ4/WAXjfg57VsE/L4fpw+HTp6GyJNgVioiInBSFM2md4lJh7FTnvrRLn3XuU3trKjx6Crz9UziwOdgVioiINIm6NaV183jh1MnOsn0RfPoEfPYUfPJXZ7y0PhOgxzjnfjUREZFWQOFMwkfn4c5StAsWPQvLXoL1bznbkrOg51nQcxx0O9MZtkNERCQEKZxJ+GnXCc7+JZz1C8hfC5vmwqZ5sPT/nFY1lwcyR/jC2lnOTAQaP01EREKEwpmEL2Mg7RRnOf02qKmEbZ9A3jwnsM17wFmiEqHHWCeo9RgHSV2DXbmIiLRhCmfSdnginRDWYyyccx+U7oO8XKdVbdNcWP1fZ7/2PZ2gln0+dB/jHCciItJCFM6k7YpNgQGXO4u1sG+9rwt0Liz7Jyx6BiLbOSGtz0XQ6xyIjAt21SIiEuYUzkTA6QJN7e0sp90K1RWweT6smQ1r58CKV8ET5Qx2e8pFzpOgMe2DXbWIiIQhhTORY4mIclrMss+Hi2pg28ew5nVnIvZ1b4JxQ/fRTotan4ugXcdgVywiImFC4UzkRNweJ4h1Hw0XPAS7ljpBbc3rMOduZ8kc4bSo9bkIknsGu2IREWnFFM5E/GEMZAx1lrPvhfx1sNYX1N79pbOk94feF0D3sZA53GmFExERaSSFM5GmMgbS+jjLmKlQsBXWvukEtQ/+AAseAXckdBkJ3cY4T35mDAF3RLArFxGREKZwJtJckrrC6d9zloqDsPVj2LwAtizwjakGRMRC19Oh22gnrHUcqAFwRUTkCApnIoEQlQC9xzsLQNkB2LLQCWubF8B79zrrIxOg26jDYS2tL7hcwatbRESCTuFMpCXEtIe+E50FoPjLI8PaujnO+uj20O1MMqrTYGc8dDhV3aAiIm2MwplIMMR3ODwALkDhdtjyga8bdCFZB7fDxmfBE+3cp9Z5BHQe6TwVGpsc3NpFRCSgFM5EQkFiZxh0tbMAH789k9M7u2H7Z858oB/9Ger+6Oyb3MsJaoeWlGx1hYqIhBGFM5EQVBmVAv1yoN83nBVVZbDrc9j+qRPY1r3lTDEFzv1tmb6Wtc4joMMAiE5yniYVEZFWR+FMpDXwxvgeHBjlfLYW9m/yhbVPnMC28d3D+0cmQPtukNQd2neHpAbv22XoCVERkRCmcCbSGhkDKb2cZfA1zrryAtix2BkYt2AzHNgMX37hTDlVV3P4WLcXErscGdgOvSZ2dYKgiIgEjcKZSLiIToKsc52lodoaKNp5OLAVbIaCLc777Z9BZdGR+8ekQEKmcx9cQpcG731LTHt1mYqIBJDCmUi4c3ucAXKTukKPnCO3WeuMwVawxRfaNsPBHc7To/nrYMN7UFN+5DERsU5gOzq0JXaBtFMgOrGFfpiISHgKaDgzxowHHgPcwLPW2gePs99lwExguLV2sW/dPcANQC1wh7X2nUDWKtImGeMMzRGbDJlDv7r9UHg7uO1waDu4HQp9n3cvg7L9Rx6T1B06DYKOg3yvA51WPRERaZSAhTNjjBuYDpwL7AAWGWNmW2tXH7VfPHAn8GmDdX2Bq4B+QCfgPWNMtrW2NlD1isgxNAxvnQYfe5+qUji4Ewq3Ove47focdiyBVa8d3keBTUSk0QLZcjYC2GitzQMwxswAJgGrj9rv18BDwNQG6yYBM6y1lcBmY8xG3/d9HMB6RaQpvLGQmu0sDe93K93vtKztXga7lsHOowNbNyfwHQpsKb0hNtXphhURacMC+a9gBrC9wecdwMiGOxhjhgCdrbVvGmOmHnXsJ0cdmxGoQkUkAGKTodfZznJI2YHDYW3X518NbBiISYa4dIhL9b2m+V7TnfB26H10kgbfFZGwFLT/RDXGuIBHgSkn8R03AzcDpKenk5ub2yy1fZ2SkpIWOU840TXzX3hfMxcwBNKHQDp4qouIL84junw33qpCIqoL8VYV4N23G++u1XirCnHXVX3lW+qMm+qIRKq8iZzijmfXunQqI1OojEymIsp5rYxMoc4d1fI/sRUI77+xwNA185+uWdMEMpztBDo3+JzpW3dIPNAfyDXOY/kdgNnGmImNOBYAa+3TwNMAw4YNszk5Oc1Y/rHl5ubSEucJJ7pm/tM1a8BaqCyGkr1Qsse37MVVsofI0r1EluyF3RuIP7gDyvZ99fioRGfg3XadICHj8Pt2Dd5HxrX87woy/Y35T9fMf7pmTRPIcLYIyDLm/9u709i4rjKM4//H43FsJ27S4FCVLqQtFfteIYEQqpBALB8AgaAFJEBILCpQhECt+MAmKiEECBUQCERREUuoWPsBlVbQsqioG6RLWgFt5NAl3UizOI7tseflwzm3c23PODOpnbl2np90dc89dzJz5uRE8+Ys9+gsUmB1AfDu4mZEHADGi2tJNwCfjohbJR0BfibpG6QFAecCN69iWc2sqiQYPikd489q+5Lbih+AxjQceggOPpQWKRx8MKUP5vTenXD4saVvMDoOW89Ox9POyemzYOs5fjSImR13qxacRcScpI8BfyA9SuOKiNgl6UvArRFx9TJ/dpekq0iLB+aAi7xS08yOqj7cCrI6aUzDob05aHsoPRrkiQnYtxsm/gZ3Q2x6EgAACk1JREFU7Fj4+pGtpYDt7BSwFcHb6NZV/TpmdmJa1TlnEfF74PeL8j7X4bXnL7q+DLhs1QpnZiem+nDuFTur/f3GkVaw9r/70nnfbthzI9xxFRCt146cDGPPgPpI6xgchvpo+pzBxflt0kOb0orXoY2t9OCwd2EwO4F5zbqZWVl9JO108PTnLr3XmE7PcysHbYceTrsoNKbTvLi56RTgNY600s1Gb2XQwMKgrT7aPojbtC1tsbUl79Awdqo3tTdbBxycmZl1qz4M256djl7Mz+UArhy0TaX07BTMTqaH+c4ehsbhVrqcP3s4LXjYvyelZyZh5sDCzxkYTIsctpzZ2lJrS2l7rZNOg8GhlasPM1sVDs7MzFZbbRBqY7BhbGXfd3YqbaN14L9pa639/83ba90Pu29Ic+vKw7Ao9a5tOZPnTdfgiR1QG0rDqIMb8rm4znm1DaV7G1pHfbR1DI2mIVw/d85sRTg4MzNbq4ZGW7sztDM3m1aploO2nN40uRv23A9zM6knrzg/FcUcu2Io9sn0yNJArj5SmpdXOtdHW3PyFpxHW/c9dGvrnIMzM7P1anCo4+KHm9s9fyoC5hutYG1+phS8Tadgr0g3jpSGZg/n68OpN69IF8O2U/ug8UB6fXF/7ghE89i+V31jerTKhpM6nDfD8Ob292pDgNK8PhXn3ONXpIt8tDAvYrlSma0YB2dmZpZIeVjzOMxLezIQzIspinNjqhX8tTvPTsHMQZg+kM8HYXp/6hEs8p5qD2AH5wPcWPQAlhdrjKaAcah0r8gf2tTKG96cg8QcOA5vTve8MtcWcXBmZmbHXzkQHN68su89N9smgMvp5lzqsYto9dxFc2FeNIFYlBdM7P4325/x9NxTOFU6T8GRJ/LCjanWvW4ezzkwuDRgG95c6v0regDHckA4lna0KFbtFvl+/MrKKf7O+zh87uDMzMzWl8EhGByHjeNHf20PJuIGtne7FVEEzM+2AriZybQN2fSBtMp2ujgOttJFQPn4va28xuHuPk+1HLTl48kAbiwt4Gg20qrh+dlWutlIvZfNuXxenJ+vpfT+xXDvQK003Duw8N6i+y+fmoH7xluLTOrDpQUnna5H8mKUehpWL3pN56YX9rLOHVl4f/HrmnMpQG7Op2CrOV+6nodmM50X3yNg23PgopuOtak8ZQ7OzMzMVprUWtnKU9hJYn4uBW0zh9KjVWYm8yNWJluPVJk9VEofTtdFempPClZqQ6mXrlaHgXo6FwHQQD2tKC7yF7+u3KNYBDPF0WwuvC7fb84zM/8IY4PDKYia3l8KpmZaQdb8TI91W2stFCkHdEWgNzre+m6q5WCxllYTL7iutQLLxXmbth3739kKcHBmZmZWVbXBtE3YGt0q7K5uNj5vNvPik6JnLB/zjfbBV61+XMreTw7OzMzMrH8GBmAgP15lpN+FqQY/MdDMzMysQhycmZmZmVWIgzMzMzOzCnFwZmZmZlYhDs7MzMzMKsTBmZmZmVmFODgzMzMzqxAHZ2ZmZmYV4uDMzMzMrEIcnJmZmZlViCKi32VYEZIeA/Ych48aBx4/Dp+znrjOeuc6643rq3eus965znrnOuvsmRHRdof1dROcHS+Sbo2I8/pdjrXEddY711lvXF+9c531znXWO9fZsfGwppmZmVmFODgzMzMzqxAHZ737fr8LsAa5znrnOuuN66t3rrPeuc565zo7Bp5zZmZmZlYh7jkzMzMzqxAHZ12S9AZJ/5J0r6RL+12etUDShKQ7Je2UdGu/y1NFkq6Q9Kiku0p5WyVdJ+k/+XxyP8tYNR3q7AuSHsxtbaekN/WzjFUj6QxJ10u6W9IuSRfnfLe1NpapL7ezDiQNS7pZ0u25zr6Y88+SdFP+7fyFpKF+l3Ut8LBmFyTVgH8DrwMeAG4BLoyIu/tasIqTNAGcFxF+xk0Hkl4DTAI/jogX5LyvAvsi4iv5PwInR8Ql/SxnlXSosy8AkxHxtX6WraoknQqcGhH/kDQG3Aa8FXg/bmtLLFNf78TtrC1JAjZGxKSkOvA34GLgU8CvI2KHpO8Bt0fEd/tZ1rXAPWfdeQVwb0TsjohZYAfwlj6XydaBiPgLsG9R9luAK3P6StKPgmUd6syWERF7I+IfOX0IuAc4Dbe1tpapL+sgksl8Wc9HAK8Ffpnz3ca65OCsO6cB95euH8D/ULsRwLWSbpP0oX4XZg05JSL25vTDwCn9LMwa8jFJd+RhTw/PdSBpO/BS4Cbc1o5qUX2B21lHkmqSdgKPAtcB9wH7I2Iuv8S/nV1ycGar6dUR8TLgjcBFeTjKehBp3oHnHhzdd4FzgJcAe4Gv97c41SRpE/Ar4JMRcbB8z21tqTb15Xa2jIiYj4iXAKeTRpye0+cirVkOzrrzIHBG6fr0nGfLiIgH8/lR4Dekf6x2dI/kOS/F3JdH+1yeyouIR/IPQxP4AW5rS+R5QL8CfhoRv87ZbmsdtKsvt7PuRMR+4HrglcAWSYP5ln87u+TgrDu3AOfmVSdDwAXA1X0uU6VJ2pgn0iJpI/B64K7l/5RlVwPvy+n3Ab/rY1nWhCLAyN6G29oCebL2D4F7IuIbpVtua210qi+3s84kbZO0JadHSAvo7iEFae/IL3Mb65JXa3YpL5n+JlADroiIy/pcpEqTdDaptwxgEPiZ62wpST8HzgfGgUeAzwO/Ba4CzgT2AO+MCE+AzzrU2fmkoaYAJoAPl+ZSnfAkvRr4K3An0MzZnyXNo3JbW2SZ+roQt7O2JL2INOG/Rur4uSoivpR/C3YAW4F/Au+NiJn+lXRtcHBmZmZmViEe1jQzMzOrEAdnZmZmZhXi4MzMzMysQhycmZmZmVWIgzMzMzOzCnFwZmbrmqR5STtLx6Ur+N7bJflZV2a2ogaP/hIzszXtSN5SxsxsTXDPmZmdkCRNSPqqpDsl3SzpWTl/u6Q/5c2t/yjpzJx/iqTfSLo9H6/Kb1WT9ANJuyRdm5+OjqRPSLo7v8+OPn1NM1uDHJyZ2Xo3smhY812lewci4oXAt0k7gAB8C7gyIl4E/BS4POdfDvw5Il4MvAzYlfPPBb4TEc8H9gNvz/mXAi/N7/OR1fpyZrb+eIcAM1vXJE1GxKY2+RPAayNid97k+uGIeJqkx4FTI6KR8/dGxLikx4DTy1vPSNoOXBcR5+brS4B6RHxZ0jXAJGk7rt9GxOQqf1UzWyfcc2ZmJ7LokO5FeZ/AeVpzed8MfIfUy3aLJM/xNbOuODgzsxPZu0rnv+f0jcAFOf0e0gbYAH8EPgogqSZpc6c3lTQAnBER1wOXAJuBJb13Zmbt+H9yZrbejUjaWbq+JiKKx2mcLOkOUu/XhTnv48CPJH0GeAz4QM6/GPi+pA+Sesg+Cuzt8Jk14Cc5gBNweUTsX7FvZGbrmuecmdkJKc85Oy8iHu93WczMyjysaWZmZlYh7jkzMzMzqxD3nJmZmZlViIMzMzMzswpxcGZmZmZWIQ7OzMzMzCrEwZmZmZlZhTg4MzMzM6uQ/wMHVoDgy4al1gAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#plotting losses\n",
    "plt.figure(figsize = (10,5))\n",
    "plt.plot(tr_loss_plot,label='train_loss')\n",
    "plt.plot(val_loss_plot,label = 'val_loss')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.xlabel('Epochs')\n",
    "plt.ylabel('Loss')\n",
    "plt.title('Loss Plot')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.11 model Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {
    "id": "Hf_VCS0JqfUc"
   },
   "outputs": [],
   "source": [
    "#The evaluate function is similar to the training loop, except you don't use teacher forcing here. \n",
    "#The input to the decoder at each time step is its previous predictions along with the hidden state and the encoder output.\n",
    "def evaluate(image_path,capin):\n",
    "    attention_plot = np.zeros((input_len_dec, attention_features_shape))\n",
    "    encoder_output_1 = CNN_Encoder(image_path[0])\n",
    "    encoder_output_2 = CNN_Encoder(image_path[1])\n",
    "    encoder_output = tf.keras.layers.Concatenate(axis=-1)([encoder_output_1,encoder_output_2])\n",
    "    hidden = decoder.reset_hidden_state(batch_size=1)\n",
    "    cell_state = decoder.reset_cell_state(batch_size=1)\n",
    "    dec_input = tf.expand_dims([tokenizer.word_index['<start>']], 0)\n",
    "    result = []\n",
    "    for i in range(input_len_dec):\n",
    "\n",
    "        predictions,hidden,attention_weights,cell_state=decoder.one_step_decoder(dec_input,hidden,encoder_output, cell_state)#calling one step decoder function inside decoder.\n",
    "        attention_plot[i] = tf.reshape(attention_weights, (-1, )).numpy()\n",
    "\n",
    "        predicted_id = tf.random.categorical(predictions, 1)[0][0].numpy()# Draws samples from a categorical distribution.\n",
    "        result.append(tokenizer.index_word[predicted_id])\n",
    "\n",
    "        if tokenizer.index_word[predicted_id] == '<end>':#if prediction is <end> token then breaking the loop\n",
    "            attention_plot = attention_plot[:len(result), :]\n",
    "\n",
    "            return result, attention_plot\n",
    "\n",
    "        dec_input = tf.expand_dims([predicted_id], 0)#reshaping input to decoder for feeding it again\n",
    "    attention_plot = attention_plot[:len(result), :]#slicing\n",
    "\n",
    "    return result, attention_plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {
    "id": "VagdqPgoEEnA"
   },
   "outputs": [],
   "source": [
    "from google.colab.patches import cv2_imshow\n",
    "from PIL import Image\n",
    "\n",
    "def get_concat_h(im1, im2):\n",
    "    #function for concatenating two images side by side.\n",
    "    dst = Image.new('RGB', (im1.width + im2.width, im1.height))\n",
    "    dst.paste(im1, (0, 0))\n",
    "    dst.paste(im2, (im1.width, 0))\n",
    "    return dst\n",
    "\n",
    "def plot_attention(image, result, attention_plot):\n",
    "    #plotting attention\n",
    "    temp_image1 = Image.open(image[0])\n",
    "    temp_image2 = Image.open(image[1])\n",
    "\n",
    "    img = get_concat_h(temp_image1, temp_image2)#concatening images side by side.\n",
    "    \n",
    "    fig = plt.figure(figsize=(20, 20))\n",
    "\n",
    "    len_result = len(result)\n",
    "    for l in range(len_result):\n",
    "        temp_att = np.resize(attention_plot[l], (7, 7))#resizing attention plot to 7,7 because our attention weight shape was 49.\n",
    "        ax = fig.add_subplot(len_result//2+1, len_result//2+1, l+1)#adding subplots\n",
    "        ax.set_title(result[l])#setting predicted word as the title of subplot.\n",
    "        img1 = ax.imshow(img)#showing concatenated images.\n",
    "        ax.imshow(temp_att, cmap='Greys', alpha=0.6, extent=img1.get_extent())#plotting attention weights over that image\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 186,
   "metadata": {
    "id": "u-BIkUqvpcFy"
   },
   "outputs": [],
   "source": [
    "#removing <start> and <end> tokens from sentence\n",
    "def remove_tok(sentence):\n",
    "  clean_sent = ''\n",
    "  for i in sentence.split():\n",
    "      if i == \"<start>\" or i == \"<end>\":\n",
    "        continue\n",
    "      else:\n",
    "        clean_sent+= \"\".join(i+\" \")\n",
    "  return clean_sent.strip()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 187,
   "metadata": {
    "id": "caqgXi6tpMR_"
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import nltk.translate.bleu_score as bleu\n",
    "def bleau_score(sentence,result):\n",
    "\n",
    "  reference = remove_tok(sentence) # the original\n",
    "  translation = remove_tok(result) # trasilated using model\n",
    "  print('BLEU score: {}'.format(bleu.sentence_bleu([reference.split()], translation.split())))\n",
    "  # cumulative BLEU scores\n",
    "\n",
    "  # NOTE: \n",
    "  #sentence_bleu takes list of list of tokens for reference sentence and list of token as translation sentence\n",
    "\n",
    "  print('Cumulative 1-gram: %f' % bleu.sentence_bleu([reference.split()], translation.split(), weights=(1, 0, 0, 0)))\n",
    "  print('Cumulative 2-gram: %f' % bleu.sentence_bleu([reference.split()], translation.split(), weights=(0.5, 0.5, 0, 0)))\n",
    "  print('Cumulative 3-gram: %f' % bleu.sentence_bleu([reference.split()], translation.split(), weights=(0.33, 0.33, 0.33, 0)))\n",
    "  print('Cumulative 4-gram: %f' % bleu.sentence_bleu([reference.split()], translation.split(), weights=(0.25, 0.25, 0.25, 0.25)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 188,
   "metadata": {
    "id": "wgrbonjcL3qq"
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import nltk.translate.bleu_score as bleu\n",
    "def bleau_score_wrong(sentence,result):\n",
    "  #NOTE:\n",
    "  #this is the wrong implementation of bleu score that i was using earlier.\n",
    "\n",
    "  reference = remove_tok(sentence) # the original\n",
    "  translation = remove_tok(result) # trasilated using model\n",
    "  print('BLEU score: {}'.format(bleu.sentence_bleu(reference, translation)))\n",
    "  # cumulative BLEU scores\n",
    "\n",
    "  print('Cumulative 1-gram: %f' % bleu.sentence_bleu(reference, translation ,weights=(1, 0, 0, 0)))\n",
    "  print('Cumulative 2-gram: %f' % bleu.sentence_bleu(reference, translation ,weights=(0.5, 0.5, 0, 0)))\n",
    "  print('Cumulative 3-gram: %f' % bleu.sentence_bleu(reference, translation ,weights=(0.33, 0.33, 0.33, 0)))\n",
    "  print('Cumulative 4-gram: %f' % bleu.sentence_bleu(reference, translation ,weights=(0.25, 0.25, 0.25, 0.25)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 189,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 109
    },
    "id": "F9YEK_7WK5Bv",
    "outputId": "484f6c75-d2e3-4579-c147-eb00825fd3f2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BLEU score: 0.7598356856515925\n",
      "Cumulative 1-gram: 0.666667\n",
      "Cumulative 2-gram: 0.577350\n",
      "Cumulative 3-gram: 0.695905\n",
      "Cumulative 4-gram: 0.759836\n"
     ]
    }
   ],
   "source": [
    "bleau_score('i am sezaz','i am ramana')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 192,
   "metadata": {
    "id": "HvUdu8YvjMm7"
   },
   "outputs": [],
   "source": [
    "#this function loads the images and preprocesses them and then add batch dimension at 0 axis for evalueate unction.\n",
    "def load_img(image_path):\n",
    "    img1,path1 = preprocess_img(image_path[0],training = False)\n",
    "    img2,path2 = preprocess_img(image_path[1],training = False)\n",
    "    img1 = tf.expand_dims(img1, axis=0)\n",
    "    img2 = tf.expand_dims(img2, axis=0)\n",
    "    \n",
    "    return (img1,img2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.12 Model Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 193,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 419
    },
    "id": "lUD22wdtEFd2",
    "outputId": "56f16180-b8c4-4834-a710-ffb8d71b3aff"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Real Caption: no acute cardiopulmonary findings chronic changes of emphysema and left basilar scarring\n",
      "Prediction Caption: no hypoinflation minimal acute recommended lungs a left are identified no cannot <end>\n",
      "BLEU score: 0.7071067811865476\n",
      "Cumulative 1-gram: 0.250000\n",
      "Cumulative 2-gram: 0.500000\n",
      "Cumulative 3-gram: 0.632878\n",
      "Cumulative 4-gram: 0.707107\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 13 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# captions on the validation set\n",
    "rid = np.random.randint(0, len(img_name_test))#generating random datapoint from test dataset.\n",
    "image = load_img( img_name_test[rid])#loading and preprocessing image\n",
    "\n",
    "result, attention_plot = evaluate(image,target_test_inp[rid])#generating caption\n",
    "rc = []\n",
    "\n",
    "for i in target_test_out[rid]:\n",
    "  if tokenizer.index_word[i] == '<end>':\n",
    "    break\n",
    "  else:\n",
    "    rc.append(tokenizer.index_word[i])\n",
    "\n",
    "\n",
    "print ('Real Caption:', ' '.join(rc))\n",
    "print ('Prediction Caption:', ' '.join(result))\n",
    "bleau_score(' '.join(rc), ' '.join(result))#calculating cumulative bleu score.\n",
    "plot_attention(img_name_test[rid], result, attention_plot)#plotting attention\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 194,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 364
    },
    "id": "JIETTmqIU4ed",
    "outputId": "1fed8394-a3bf-453f-9f39-66acb98441d0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Real Caption: status post mediastinal surgery cardiomegaly atherosclerotic disease of the thoracic aorta lung volumes are reduced suspicion for at bilateral pleural effusions no acute airspace disease no pulmonary edema\n",
      "Prediction Caption: evidence volumes pattern left cardiopulmonary these obliquely difficult right further recent silhouette visualized interstitial radiation may abnormalities <end>\n",
      "BLEU score: 0.2578538932915104\n",
      "Cumulative 1-gram: 0.030799\n",
      "Cumulative 2-gram: 0.126988\n",
      "Cumulative 3-gram: 0.205560\n",
      "Cumulative 4-gram: 0.257854\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 18 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# captions on the validation set\n",
    "rid = np.random.randint(0, len(img_name_test))#generating random datapoint from test dataset.\n",
    "image = load_img( img_name_test[rid])#loading and preprocessing image\n",
    "\n",
    "result, attention_plot = evaluate(image,target_test_inp[rid])#generating caption\n",
    "rc = []\n",
    "\n",
    "for i in target_test_out[rid]:\n",
    "  if tokenizer.index_word[i] == '<end>':\n",
    "    break\n",
    "  else:\n",
    "    rc.append(tokenizer.index_word[i])\n",
    "\n",
    "\n",
    "print ('Real Caption:', ' '.join(rc))\n",
    "print ('Prediction Caption:', ' '.join(result))\n",
    "bleau_score(' '.join(rc), ' '.join(result))#calculating cumulative bleu score.\n",
    "plot_attention(img_name_test[rid], result, attention_plot)#plotting attention\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 520
    },
    "id": "jX6Uglm6pPbv",
    "outputId": "f6c3e2a4-722e-4347-d9c7-48ca53447eb6"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Real Caption: no acute findings\n",
      "Prediction Caption: no evidence have right nodules of chest <end>\n",
      "BLEU score: 0.6147881529512643\n",
      "Cumulative 1-gram: 0.142857\n",
      "Cumulative 2-gram: 0.377964\n",
      "Cumulative 3-gram: 0.526160\n",
      "Cumulative 4-gram: 0.614788\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# captions on the validation set\n",
    "rid = np.random.randint(0, len(img_name_test))#generating random datapoint from test dataset.\n",
    "image = load_img( img_name_test[rid])#loading and preprocessing image\n",
    "\n",
    "result, attention_plot = evaluate(image,target_test_inp[rid])#generating caption\n",
    "rc = []\n",
    "\n",
    "for i in target_test_out[rid]:\n",
    "  if tokenizer.index_word[i] == '<end>':\n",
    "    break\n",
    "  else:\n",
    "    rc.append(tokenizer.index_word[i])\n",
    "\n",
    "\n",
    "print ('Real Caption:', ' '.join(rc))\n",
    "print ('Prediction Caption:', ' '.join(result))\n",
    "bleau_score(' '.join(rc), ' '.join(result))#calculating cumulative bleu score.\n",
    "plot_attention(img_name_test[rid], result, attention_plot)#plotting attention\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 196,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 446
    },
    "id": "sJ0fz3DoI4yE",
    "outputId": "0d9e35b4-734c-4c86-9bf8-b546a95ddbf2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Real Caption: bilateral opacities most prominent in the lower lobes representing airspace disease or edema\n",
      "Prediction Caption: retained volumes curvature normal lung prominent normal volumes fractures disease disease <end>\n",
      "BLEU score: 0.5444358245099123\n",
      "Cumulative 1-gram: 0.151591\n",
      "Cumulative 2-gram: 0.355513\n",
      "Cumulative 3-gram: 0.475026\n",
      "Cumulative 4-gram: 0.544436\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# captions on the validation set\n",
    "rid = np.random.randint(0, len(img_name_test))#generating random datapoint from test dataset.\n",
    "image = load_img( img_name_test[rid])#loading and preprocessing image\n",
    "\n",
    "result, attention_plot = evaluate(image,target_test_inp[rid])#generating caption\n",
    "rc = []\n",
    "\n",
    "for i in target_test_out[rid]:\n",
    "  if tokenizer.index_word[i] == '<end>':\n",
    "    break\n",
    "  else:\n",
    "    rc.append(tokenizer.index_word[i])\n",
    "\n",
    "\n",
    "print ('Real Caption:', ' '.join(rc))\n",
    "print ('Prediction Caption:', ' '.join(result))\n",
    "bleau_score(' '.join(rc), ' '.join(result))#calculating cumulative bleu score.\n",
    "plot_attention(img_name_test[rid], result, attention_plot)#plotting attention\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.13 Whole corpus BLEU score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 197,
   "metadata": {
    "id": "l-WwyEebzGSi"
   },
   "outputs": [],
   "source": [
    "#http://www.nltk.org/api/nltk.translate.html#nltk.translate.bleu_score.sentence_bleu\n",
    "\n",
    "import nltk.translate.bleu_score as bleu\n",
    "\n",
    "def bleau_score1(sentence,result):\n",
    "\n",
    "  reference = remove_tok(sentence) # the original\n",
    "  translation = remove_tok(result) # trasilated using model\n",
    "\n",
    "  bleau1 =  bleu.sentence_bleu([reference.split()], translation.split(), weights=(1, 0, 0, 0))\n",
    "  bleau2 = bleu.sentence_bleu([reference.split()], translation.split(), weights=(0.5, 0.5, 0, 0))\n",
    "  bleau3 = bleu.sentence_bleu([reference.split()], translation.split(), weights=(0.33, 0.33, 0.33, 0))\n",
    "  bleau4 =  bleu.sentence_bleu([reference.split()], translation.split(), weights=(0.25, 0.25, 0.25, 0.25))\n",
    "\n",
    "  return bleau1,bleau2,bleau3,bleau4\n",
    "\n",
    "def calc_bleau(image,sentence_in):\n",
    "  #this function will return bleu score\n",
    "  result, _ = evaluate(image,sentence_in)\n",
    "  rc = []\n",
    "  for i in sentence_in.numpy()[0]:\n",
    "    if tokenizer.index_word[i] == '<end>':#if the token number corresponds to end value then it will break\n",
    "      break\n",
    "    else:\n",
    "      rc.append(tokenizer.index_word[i])\n",
    "\n",
    "  score = bleau_score1(' '.join(rc), ' '.join(result))\n",
    "  return score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 198,
   "metadata": {
    "id": "6psvKJcDzGQ6"
   },
   "outputs": [],
   "source": [
    "#making dataloader for whole data available.\n",
    "evaluate_dataloader = Dataloder(df['IMAGES'].values,target_inp,target_out, batch_size=1, shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 200,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "4AqYyI67zGNc",
    "outputId": "64560199-1456-44e9-85dd-8859b7117fcd"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "3851it [20:32,  3.12it/s]\n"
     ]
    }
   ],
   "source": [
    "from tqdm import tqdm\n",
    "bleau1_final = 0\n",
    "bleau2_final = 0\n",
    "bleau3_final = 0\n",
    "bleau4_final = 0\n",
    "\n",
    "for (batch,(inputs,target)) in tqdm(enumerate(evaluate_dataloader)):\n",
    "  score1,score2,score3,score4 = calc_bleau(inputs,target)\n",
    "  bleau1_final+=score1\n",
    "  bleau2_final+=score2\n",
    "  bleau3_final+=score3\n",
    "  bleau4_final+=score4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 201,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 35
    },
    "id": "gS_6ARemdOao",
    "outputId": "f5a010c2-82d3-4626-bbee-c2f8d69f79f6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3851"
      ]
     },
     "execution_count": 201,
     "metadata": {
      "tags": []
     },
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Tolal number of datapoints that we are considering\n",
    "len(evaluate_dataloader)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 202,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 90
    },
    "id": "V-Dxd4yWzGLC",
    "outputId": "b67639c4-cf9f-4e68-90e6-6a4223e8b985"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "final_bleau1_score 0.11377335230141303\n",
      "final_bleau2_score 0.20860306243858598\n",
      "final_bleau3_score 0.27927417083079303\n",
      "final_bleau4_score 0.3235168641768014\n"
     ]
    }
   ],
   "source": [
    "#finding final i,2,3,4 gram BLEU score for entire corpus.\n",
    "final_bleau1 =bleau1_final/ len(evaluate_dataloader)\n",
    "print('final_bleau1_score',final_bleau1)\n",
    "final_bleau2 =bleau2_final/ len(evaluate_dataloader)\n",
    "print('final_bleau2_score',final_bleau2)\n",
    "final_bleau3 =bleau3_final/ len(evaluate_dataloader)\n",
    "print('final_bleau3_score',final_bleau3)\n",
    "final_bleau4 =bleau4_final/ len(evaluate_dataloader)\n",
    "print('final_bleau4_score',final_bleau4)"
   ]
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "collapsed_sections": [],
   "name": "Copy of case_study_2.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}