Natural_Language_Processing/Multimodal_Entailment/multimodal_entailment_KYK.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e3c07a55",
   "metadata": {},
   "source": [
    "# Multimodal Entailment"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db069d82",
   "metadata": {},
   "source": [
    "[Original Keras Code](https://keras.io/examples/nlp/multimodal_entailment/)\n",
    "\n",
    "<br/>\n",
    "\n",
    "**Author**: Yookyung Kho\n",
    "\n",
    "**Date presented**: 2022/01/17, DSBA keras2torch Study\n",
    "\n",
    "**Task description**: Multi-Label Classification, Training a multimodal model for predicting entailment.\n",
    "\n",
    "> You can check a whole python script pipeline for multimodal entailment(pytorch) in [this github repository](https://github.com/yookyungkho/Multimodal-Entailment-pytorch).\n",
    "\n",
    "**References**:\n",
    "\n",
    "- https://github.com/google-research-datasets/recognizing-multimodal-entailment\n",
    "\n",
    "- https://multimodal-entailment.github.io/\n",
    "\n",
    "\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f6dde77",
   "metadata": {},
   "source": [
    "## About Multimodal Entailment\n",
    "\n",
    "- [Recognizing Multimodal Entailment](https://multimodal-entailment.github.io/): Tutorial presented at `ACL 2021`\n",
    "\n",
    "On social media platforms, to audit and moderate content we may want to find answers to the following questions in near real-time:\n",
    "\n",
    "    > Does a given piece of information \"contradict\" the other?\n",
    "    > Does a given piece of information \"imply\" the other?\n",
    "\n",
    "In NLP, this task is called analyzing **textual entailment**. However, that's only when the information comes from text content. In practice, it's often the case the information available comes not just from text content, but from a multimodal combination of text, images, audio, video, etc. **Multimodal entailment** is simply the extension of textual entailment to **a variety of new input modalities**.\n",
    "\n",
    "- **Input**: 2 text,image pairs (`text_1`, `image_1`), (`text_2`, `image_2`)\n",
    "- **Output**: 3 classes(`Implies`, `NoEntailment`, `Contradictory`)\n",
    "\n",
    "![image](multimodal.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b423efdf",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "13dc7693",
   "metadata": {},
   "source": [
    "## 0. Import"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "652aeb92",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import json\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from timeit import default_timer as timer\n",
    "from PIL import Image\n",
    "from tqdm import tqdm\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import DataLoader\n",
    "from torchinfo import summary\n",
    "\n",
    "import torchtext\n",
    "from torchtext.utils import download_from_url\n",
    "import torchvision\n",
    "\n",
    "from transformers import AutoTokenizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "dc1e1e5e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.10.2+cu102\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "print(torch.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2c7e1a94",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.11.0\n"
     ]
    }
   ],
   "source": [
    "import torchtext\n",
    "print(torchtext.__version__)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e1ee08b",
   "metadata": {},
   "source": [
    "## 1. Data Preprocessing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8f513f4",
   "metadata": {},
   "source": [
    "**Data description**\n",
    "\n",
    "- [multimodal entailment dataset](https://github.com/google-research-datasets/recognizing-multimodal-entailment) recently introduced by Google Research\n",
    "\n",
    "- The original dataset is available here. It comes with URLs of images which are hosted on Twitter's photo storage system called the Photo Blob Storage (PBS for short). We will be working with the downloaded images along with additional data that comes with the original dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43919e5a",
   "metadata": {},
   "source": [
    "### 1.1. Data Download"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b3b3438c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "path = 'data/' # 현재 디렉토리에 data 폴더 추가\n",
    "\n",
    "if not os.path.exists(path):\n",
    "    os.mkdir(path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "9854d67d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 344M/344M [00:50<00:00, 6.87MB/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'/workspace/codes/keras2torch_study/multimodal/data/tweet_images.tar.gz'"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from torchtext.utils import download_from_url\n",
    "\n",
    "url = \"https://github.com/sayakpaul/Multimodal-Entailment-Baseline/releases/download/v1.0.0/tweet_images.tar.gz\"\n",
    "download_from_url(url, path=None, root=path, overwrite=False, hash_value=None, hash_type='sha256')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "34c166ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tarfile\n",
    "\n",
    "tar = tarfile.open(path+\"tweet_images.tar.gz\", \"r:gz\")\n",
    "tar.extractall(path)\n",
    "tar.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b3132ecd",
   "metadata": {},
   "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>id_1</th>\n",
       "      <th>text_1</th>\n",
       "      <th>image_1</th>\n",
       "      <th>id_2</th>\n",
       "      <th>text_2</th>\n",
       "      <th>image_2</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>653</th>\n",
       "      <td>1376568655857602566</td>\n",
       "      <td>Barack Obama speaks out following the passing ...</td>\n",
       "      <td>http://pbs.twimg.com/media/ExqN5ZzUYAMEZ46.jpg</td>\n",
       "      <td>1376604080840175620</td>\n",
       "      <td>Barack Obama Mourning Death of Beloved Grandmo...</td>\n",
       "      <td>http://pbs.twimg.com/media/ExquHcjVgAELk3f.jpg</td>\n",
       "      <td>NoEntailment</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>647</th>\n",
       "      <td>1333893428367781890</td>\n",
       "      <td>There were no winners to tonight's club lotto....</td>\n",
       "      <td>http://pbs.twimg.com/media/EoLw_XIWMAIF0bg.jpg</td>\n",
       "      <td>1336431970373558273</td>\n",
       "      <td>There was no winner of this week's club lotto....</td>\n",
       "      <td>http://pbs.twimg.com/media/Eov1yTaXYAIgubh.jpg</td>\n",
       "      <td>NoEntailment</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1257</th>\n",
       "      <td>1339225227046350848</td>\n",
       "      <td>15:00:Temp: 7.5°C, Wind: S, 171°, 10mph, 18mph...</td>\n",
       "      <td>http://pbs.twimg.com/media/EpXiPtFXIAI65nk.jpg</td>\n",
       "      <td>1363183542398824453</td>\n",
       "      <td>17:47:Temp: 11.2°C, Wind: S, 176°, 20mph, 38mp...</td>\n",
       "      <td>http://pbs.twimg.com/media/EusANN3WgAMelTX.jpg</td>\n",
       "      <td>NoEntailment</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>1378117789722931201</td>\n",
       "      <td>BOOM! These are my picks I just made on @foxsu...</td>\n",
       "      <td>http://pbs.twimg.com/media/EyAO0ypUcAMwDU7.jpg</td>\n",
       "      <td>1381240499571453957</td>\n",
       "      <td>BOOM! These are my picks I just made on @foxsu...</td>\n",
       "      <td>http://pbs.twimg.com/media/Eysm6a1WgAA1VVE.jpg</td>\n",
       "      <td>Implies</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>614</th>\n",
       "      <td>1335081080357994499</td>\n",
       "      <td>UPSC &amp;amp; State PSC Mains Practice Questions ...</td>\n",
       "      <td>http://pbs.twimg.com/media/EocpAE5UUAAQbUg.jpg</td>\n",
       "      <td>1355012873248890880</td>\n",
       "      <td>UPSC &amp;amp; State PSC Mains Practice Questions\\...</td>\n",
       "      <td>http://pbs.twimg.com/media/Es345dOU0AADrQm.jpg</td>\n",
       "      <td>NoEntailment</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     id_1                                             text_1  \\\n",
       "653   1376568655857602566  Barack Obama speaks out following the passing ...   \n",
       "647   1333893428367781890  There were no winners to tonight's club lotto....   \n",
       "1257  1339225227046350848  15:00:Temp: 7.5°C, Wind: S, 171°, 10mph, 18mph...   \n",
       "21    1378117789722931201  BOOM! These are my picks I just made on @foxsu...   \n",
       "614   1335081080357994499  UPSC &amp; State PSC Mains Practice Questions ...   \n",
       "\n",
       "                                             image_1                 id_2  \\\n",
       "653   http://pbs.twimg.com/media/ExqN5ZzUYAMEZ46.jpg  1376604080840175620   \n",
       "647   http://pbs.twimg.com/media/EoLw_XIWMAIF0bg.jpg  1336431970373558273   \n",
       "1257  http://pbs.twimg.com/media/EpXiPtFXIAI65nk.jpg  1363183542398824453   \n",
       "21    http://pbs.twimg.com/media/EyAO0ypUcAMwDU7.jpg  1381240499571453957   \n",
       "614   http://pbs.twimg.com/media/EocpAE5UUAAQbUg.jpg  1355012873248890880   \n",
       "\n",
       "                                                 text_2  \\\n",
       "653   Barack Obama Mourning Death of Beloved Grandmo...   \n",
       "647   There was no winner of this week's club lotto....   \n",
       "1257  17:47:Temp: 11.2°C, Wind: S, 176°, 20mph, 38mp...   \n",
       "21    BOOM! These are my picks I just made on @foxsu...   \n",
       "614   UPSC &amp; State PSC Mains Practice Questions\\...   \n",
       "\n",
       "                                             image_2         label  \n",
       "653   http://pbs.twimg.com/media/ExquHcjVgAELk3f.jpg  NoEntailment  \n",
       "647   http://pbs.twimg.com/media/Eov1yTaXYAIgubh.jpg  NoEntailment  \n",
       "1257  http://pbs.twimg.com/media/EusANN3WgAMelTX.jpg  NoEntailment  \n",
       "21    http://pbs.twimg.com/media/Eysm6a1WgAA1VVE.jpg       Implies  \n",
       "614   http://pbs.twimg.com/media/Es345dOU0AADrQm.jpg  NoEntailment  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_csv(\"https://github.com/sayakpaul/Multimodal-Entailment-Baseline/raw/main/csvs/tweets.csv\")\n",
    "df.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "534fff2a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1400, 7)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8bd3fc9f",
   "metadata": {},
   "source": [
    "The entailment task is formulated as the following:\n",
    "\n",
    "> Given the pairs of (`text_1`, `image_1`) and (`text_2`, `image_2`), do they **entail (or not entail or contradict)** each other?\n",
    "\n",
    "We have the images already downloaded. `image_1` is downloaded as `id_1` as its filename and `image_2` is downloaded as `id_2` as its filename. In the next step, we will _add two more columns_ to df - _filepaths_ of `image_1`s and `image_2`s."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "29d68a49",
   "metadata": {},
   "outputs": [],
   "source": [
    "image_base_path = path + \"tweet_images/\"\n",
    "images_one_paths = []\n",
    "images_two_paths = []\n",
    "\n",
    "for idx in range(len(df)):\n",
    "    current_row = df.iloc[idx]\n",
    "    id_1 = current_row[\"id_1\"]\n",
    "    id_2 = current_row[\"id_2\"]\n",
    "    extentsion_one = current_row[\"image_1\"].split(\".\")[-1] #jpg\n",
    "    extentsion_two = current_row[\"image_2\"].split(\".\")[-1]\n",
    "\n",
    "    image_one_path = os.path.join(image_base_path, str(id_1) + f\".{extentsion_one}\") #ex) 'data/tweet_images/1375936088968200205.jpg'\n",
    "    image_two_path = os.path.join(image_base_path, str(id_2) + f\".{extentsion_two}\")\n",
    "\n",
    "    images_one_paths.append(image_one_path)\n",
    "    images_two_paths.append(image_two_path)\n",
    "\n",
    "df[\"image_1_path\"] = images_one_paths\n",
    "df[\"image_2_path\"] = images_two_paths\n",
    "\n",
    "# Defien label map\n",
    "label_map = {\"Contradictory\": 0, \"Implies\": 1, \"NoEntailment\": 2}\n",
    "# Create another column containing the integer ids of the string labels.\n",
    "df[\"label_idx\"] = df[\"label\"].apply(lambda x: label_map[x])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c63f45f7",
   "metadata": {},
   "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>id_1</th>\n",
       "      <th>text_1</th>\n",
       "      <th>image_1</th>\n",
       "      <th>id_2</th>\n",
       "      <th>text_2</th>\n",
       "      <th>image_2</th>\n",
       "      <th>label</th>\n",
       "      <th>image_1_path</th>\n",
       "      <th>image_2_path</th>\n",
       "      <th>label_idx</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1375936088968200205</td>\n",
       "      <td>150 years after its inauguration, the strategi...</td>\n",
       "      <td>http://pbs.twimg.com/media/ExeIVRBWYAAwizQ.jpg</td>\n",
       "      <td>1376653595144241155</td>\n",
       "      <td>Look how only 190 km #suezcanel carries 12% of...</td>\n",
       "      <td>http://pbs.twimg.com/media/ExrbJffWgAoBaxO.jpg</td>\n",
       "      <td>NoEntailment</td>\n",
       "      <td>data/tweet_images/1375936088968200205.jpg</td>\n",
       "      <td>data/tweet_images/1376653595144241155.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1360981123388436489</td>\n",
       "      <td>Fall Guys is celebrating Valentine's Day with ...</td>\n",
       "      <td>http://pbs.twimg.com/media/EuMowPiXYAEylR8.jpg</td>\n",
       "      <td>1361075903606910976</td>\n",
       "      <td>Fall Guys Red Panda costume available for limi...</td>\n",
       "      <td>http://pbs.twimg.com/media/EuODUixWYAMRukK.jpg</td>\n",
       "      <td>NoEntailment</td>\n",
       "      <td>data/tweet_images/1360981123388436489.jpg</td>\n",
       "      <td>data/tweet_images/1361075903606910976.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1336637206899945472</td>\n",
       "      <td>Impressive Batting Display From Shoaib Malik 🏏...</td>\n",
       "      <td>http://pbs.twimg.com/media/EoywcVGXMAE5FZK.jpg</td>\n",
       "      <td>1339243901799030785</td>\n",
       "      <td>A Half century of King\\n\\n    🔥\"Shoaib Malik\" ...</td>\n",
       "      <td>http://pbs.twimg.com/media/EpXzOwRXYAYLaCB.jpg</td>\n",
       "      <td>NoEntailment</td>\n",
       "      <td>data/tweet_images/1336637206899945472.jpg</td>\n",
       "      <td>data/tweet_images/1339243901799030785.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1336490796279742464</td>\n",
       "      <td>How #confident are you today? https://t.co/5ay...</td>\n",
       "      <td>http://pbs.twimg.com/media/EowrS9aVEAAMhsF.jpg</td>\n",
       "      <td>1341081069248868353</td>\n",
       "      <td>How #confident are you today? https://t.co/KqZ...</td>\n",
       "      <td>http://pbs.twimg.com/media/Epx6IA-UwAA8pqN.jpg</td>\n",
       "      <td>Implies</td>\n",
       "      <td>data/tweet_images/1336490796279742464.jpg</td>\n",
       "      <td>data/tweet_images/1341081069248868353.jpg</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1358934899256877059</td>\n",
       "      <td>The Peacocks finish out the first quarter trai...</td>\n",
       "      <td>http://pbs.twimg.com/media/EtvoCsuWQAALmyP.jpg</td>\n",
       "      <td>1361131487077228546</td>\n",
       "      <td>10 minutes gone here at home as the Peacocks l...</td>\n",
       "      <td>http://pbs.twimg.com/media/EuO1xf2XIAAOIGb.jpg</td>\n",
       "      <td>NoEntailment</td>\n",
       "      <td>data/tweet_images/1358934899256877059.jpg</td>\n",
       "      <td>data/tweet_images/1361131487077228546.jpg</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  id_1                                             text_1  \\\n",
       "0  1375936088968200205  150 years after its inauguration, the strategi...   \n",
       "1  1360981123388436489  Fall Guys is celebrating Valentine's Day with ...   \n",
       "2  1336637206899945472  Impressive Batting Display From Shoaib Malik 🏏...   \n",
       "3  1336490796279742464  How #confident are you today? https://t.co/5ay...   \n",
       "4  1358934899256877059  The Peacocks finish out the first quarter trai...   \n",
       "\n",
       "                                          image_1                 id_2  \\\n",
       "0  http://pbs.twimg.com/media/ExeIVRBWYAAwizQ.jpg  1376653595144241155   \n",
       "1  http://pbs.twimg.com/media/EuMowPiXYAEylR8.jpg  1361075903606910976   \n",
       "2  http://pbs.twimg.com/media/EoywcVGXMAE5FZK.jpg  1339243901799030785   \n",
       "3  http://pbs.twimg.com/media/EowrS9aVEAAMhsF.jpg  1341081069248868353   \n",
       "4  http://pbs.twimg.com/media/EtvoCsuWQAALmyP.jpg  1361131487077228546   \n",
       "\n",
       "                                              text_2  \\\n",
       "0  Look how only 190 km #suezcanel carries 12% of...   \n",
       "1  Fall Guys Red Panda costume available for limi...   \n",
       "2  A Half century of King\\n\\n    🔥\"Shoaib Malik\" ...   \n",
       "3  How #confident are you today? https://t.co/KqZ...   \n",
       "4  10 minutes gone here at home as the Peacocks l...   \n",
       "\n",
       "                                          image_2         label  \\\n",
       "0  http://pbs.twimg.com/media/ExrbJffWgAoBaxO.jpg  NoEntailment   \n",
       "1  http://pbs.twimg.com/media/EuODUixWYAMRukK.jpg  NoEntailment   \n",
       "2  http://pbs.twimg.com/media/EpXzOwRXYAYLaCB.jpg  NoEntailment   \n",
       "3  http://pbs.twimg.com/media/Epx6IA-UwAA8pqN.jpg       Implies   \n",
       "4  http://pbs.twimg.com/media/EuO1xf2XIAAOIGb.jpg  NoEntailment   \n",
       "\n",
       "                                image_1_path  \\\n",
       "0  data/tweet_images/1375936088968200205.jpg   \n",
       "1  data/tweet_images/1360981123388436489.jpg   \n",
       "2  data/tweet_images/1336637206899945472.jpg   \n",
       "3  data/tweet_images/1336490796279742464.jpg   \n",
       "4  data/tweet_images/1358934899256877059.jpg   \n",
       "\n",
       "                                image_2_path  label_idx  \n",
       "0  data/tweet_images/1376653595144241155.jpg          2  \n",
       "1  data/tweet_images/1361075903606910976.jpg          2  \n",
       "2  data/tweet_images/1339243901799030785.jpg          2  \n",
       "3  data/tweet_images/1341081069248868353.jpg          1  \n",
       "4  data/tweet_images/1361131487077228546.jpg          2  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3164ed63",
   "metadata": {},
   "source": [
    "### 1.2. Data Visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0b5232af",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "def visualize(idx):\n",
    "    current_row = df.iloc[idx]\n",
    "    image_1 = plt.imread(current_row[\"image_1_path\"])\n",
    "    image_2 = plt.imread(current_row[\"image_2_path\"])\n",
    "    text_1 = current_row[\"text_1\"]\n",
    "    text_2 = current_row[\"text_2\"]\n",
    "    label = current_row[\"label\"]\n",
    "    \n",
    "    # image1\n",
    "    plt.figure(figsize=(8, 8))\n",
    "    plt.subplot(1, 2, 1)\n",
    "    plt.imshow(image_1)\n",
    "    plt.axis(\"off\")\n",
    "    plt.title(\"Image One\")\n",
    "    plt.show()\n",
    "    # text1\n",
    "    print(\"=========\"*10)\n",
    "    print(f\"Text1: {text_1}\")\n",
    "    print(\"=========\"*10)\n",
    "    \n",
    "    # image2\n",
    "    plt.figure(figsize=(8, 8))\n",
    "    plt.subplot(1, 2, 2)\n",
    "    plt.imshow(image_2)\n",
    "    plt.axis(\"off\")\n",
    "    plt.title(\"Image Two\")\n",
    "    plt.show()\n",
    "    # text2\n",
    "    print(\"=========\"*10)\n",
    "    print(f\"Text2: {text_2}\")\n",
    "    print(\"=========\"*10)\n",
    "    \n",
    "    # label\n",
    "    print(f\">>>>> Label: {label}\")\n",
    "    print(\"=========\"*10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "2f3ff688-a642-4930-86e8-d2c8457706bd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==========================================================================================\n",
      "Text1: #HyderabadFC's two-goal cushion is restored as Joao Victor makes it 3-1 with a peach of a finish. \n",
      "\n",
      " Follow #CFCHFC live: https://t.co/FDFVZy9cNv #ISL https://t.co/bH5nazqeEc\n",
      "==========================================================================================\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==========================================================================================\n",
      "Text2: #HyderabadFC is making merry here! Halicharan Narzary bags his second goal of the evening to hand his side a 4-1 lead. \n",
      "\n",
      "Follow #CFCHFC live: https://t.co/FDFVZy9cNv #ISL https://t.co/0iBUxjUk9D\n",
      "==========================================================================================\n",
      ">>>>> Label: NoEntailment\n",
      "==========================================================================================\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==========================================================================================\n",
      "Text1: 🚨 🏡 RENTAL HOUSE! LIKE NEW 3/2/1 IN THE HEART OF PALM COAST, FLORIDA\n",
      "\n",
      "👉 https://t.co/p77rBODlba\n",
      "\n",
      "#realestate  #realestateinvestor #realestatebroker #landlords #rentalproperty #propertyforrent #propertymanagement #386rent #PalmCoastPropertyManagement #EverestRealty https://t.co/kYv83AkCRu\n",
      "==========================================================================================\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==========================================================================================\n",
      "Text2: 🚨 🏡 RENTAL HOUSE! LIKE NEW 3/2/1 IN THE HEART OF PALM COAST FLORIDA\n",
      "\n",
      "👉 https://t.co/XMbG0zFzlF\n",
      "\n",
      "#realestate  #RealEstateAgent #realestateinvestor  #rentalproperty  #propertydevelopment #propertyforrent #propertymanagement #386rent #PalmCoastPropertyManagement #EverestRealty https://t.co/EBRbI6Ewza\n",
      "==========================================================================================\n",
      ">>>>> Label: Implies\n",
      "==========================================================================================\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAANkAAADpCAYAAACpxJLbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAACqy0lEQVR4nOy9d4AdV3n3/zkzc/v23nelVe/NktxkywV3bMCmOkAIEEJCSH/TQxLS3jeFEEJCfiHgUE2zccEVd0u2bKt3aaXtvd3d22fmnN8fM3Pv3dWqS5Zk9LWv9t4p55w5c57z1PMcoZTiMi7jMs4ftAvdgMu4jHc6LhPZZVzGecZlIruMyzjPuExkl3EZ5xmXiewyLuM84zKRXcZlnGdcJrLLuIzzjMtEdgYQQrQLIW660O04GYQQJUKI/xBC9AshEkKIXUKIX77Q7fpFg3GhG3AZ5wdCCD/wLDAIXAl0AzcCDwghSpVS/3wh2/eLhMuc7CwhhPi4EOJVIcS/CCHGhRBHhBBXuce7hBCDQoiP5V1/hxBimxBiwj3/hWnlfVQI0SGEGBFC/Fk+1xRCaEKIPxRCtLnnfyCEKDtO034JaALuU0odVUqZSqkngd8E/koIUeSW2S6E+D0hxE4hRFQI8aAQIpjXnjuFENvdZ9skhFh2bnvwnY/LRHZusA7YCZQD3wW+D1wBzAHuB74ihChwr40DHwVKgDuAXxNC3AMghFgEfBX4CFALFAP1efV8DrgHuA6oA8aAfz9Om24GnlBKxacd/zEQxOFuHt4P3ArMApYBH3fbsxL4H+BX3Wf7GvCIECJwkv64jDxcJrJzg6NKqW8opWzgQaAR+CulVFop9TSQwSE4lFIvKKV2KaWkUmon8D0cogG4F3hUKfWKUioD/DmQH1z6GeBPlFLdSqk08AXgXiHETGJ/BdA3/aBSygKG3fMevqyU6lVKjQKPAivc458GvqaUel0pZSulHgDSwPrT655fbFwmsnODgbzvSQCl1PRjBQBCiHVCiOeFEENCiCgO4XgDvg7o8m5SSiWAkbxymoGHXNFtHNgH2ED1DG0axuGGU+ASZIV73kN/3veE11a3vt/16nPrbHTbeRmniMtE9vbju8AjQKNSqhj4T0C45/qABu9CIUQIR0zz0AXcppQqyfsElVI9M9TzLHCbECIy7fj7cLjRa6fQ1i7gb6bVF1ZKfe9UHvQyHFwmsrcfhcCoUiolhFgLfDjv3I+Au1zDiR9HHBR55/8T+BshRDOAEKJSCHH3cer5Fo5F8YdCiBYhhE8IcQvwZeALSqnoKbT1/wM+43JfIYSIuIabwtN54F90XCaytx+fxbHuTeLoXD/wTiil9uAYN76Pw9ViOCb4tHvJv+Jwwafd+1/DMbocA1dnuwmHG70OTAD/jKPT/b9TaahS6k3gU8BXcIwsh3GNIpdx6hCXF21evHAtkuPAXKXU0QvcnMs4Q1zmZBcZhBB3CSHCri71j8AuoP3CtuoyzgaXieziw91Ar/uZC3xQXRY3LmlcFhcv4zLOMy5zssu4jPOMEwYIL/ru8GU2dxkXDRQgEhN0/t29JI5s42IanEopcbxz54GTKRRkPyjnSPb3ZZwG1LTPLyiUN4Ig/uZjJDp2X1K9cR6WugiEAiVAQyGFwqFllT1/GacPlddvv3A9KAQgIB1l9OffQtnm9Avy/s5EfgKQCETeWe/a6ePy3JPveSAyhRICFLQyxKAoYhw/Am3KI/6i4dSmGDXDO1aAQAjlfsse4thBJabUdfL6Lg0oFAJB+vB24h27jzkvUBgBA8MnMFOKcLEfw6cxMZxEoSiqCCMtSSxqYgSLAYUZj6KkjREuRtN92OkEViZxXtp/zogsfxAJJOvL0nzE10efnuLro/X0p3G76p3B0VTevw5E7oncw0pMu0YpHGOuBKVAKgQKJV1xSLkzq3I+Qgj3OO5srlBoIJzalBAgBEII0DTnGm/WF257XFVBCTWlx9XUFl/kEChhM/nmEygzdcxZBVz7vrlceU8TB94aYumV9fgCsPXpHjSfYMX1DVimxc+/10e66tfxFVWw61t/wmTPIRZ/8M8orJ/Hvu//NcOH3jgvrT9nRCbcsSQFBAX82vJClhStRlOCnv023zgoc6/0Unm3J4Az5HP/erqocAkEKRFSIpUEW4KyUUqClM7MrJQjVrvl5ZPrMdNQHsOaiayzx4RwZzkNoekoTYBuIDQNNB2laU5hyp3sLpH3IFCQjBPbf/yY5kBYp7I5jN9fw95Nvay6uZ6r399Mz/4J9m3uZ+W7atnwvmoe+ck+wi0fomz+laQnxylqWU4mPk605+B5a/9ZE5kzsJyZEgVBTbG2ymZOWQR/QEcoxfraBM/3JuhNgKmcFy2yM7Q7gQsQeQYa76tQU2tTIjfAhfIECRf5nEMJZ/Z2/2bvys7szstT7o1CuWJu9ohbxvRyvefFIRKkjXI/2BIpJSjpEJTKa5bbdq/TlMvTxTSyyf47nQDElD8Os8N7Bu+oW6eSYFvZGyUCoQmEpqF0DaH7csTnccRjq7pooBDI4W4yQx3HvUbazkjY/NhRnn3gAMXVARZeVcMzDxzg0NZBmpaUUFjuY7LtOUoW307Z/LUkBjswwkX0vvYQZmLyvLX/rIgsOwsL0IRgXrFkTRVElESO9WLZMRAGatxgQ1UISzfY0q9oi7n3CofYvDGsplKUW7ZDhQKBEKALha0ApQEShEJ5lOcZXRyKzRKNoSmULbAFznHcwa0cAlVCgfAGvSeWOXqQUKBczoSUKNtCShslLYRUOctXlqBcQnWlNu8oSuRxLgVu2dnzcHqjW3gTULZGr1OnqPDCe1YpnYnAAikyzjWajjB8CJ9DdAgNjyNfbMJkuq8NO5086XWpuIVtK6yUIy2kUzZWRmKmbYQIkhhuZ/zg65QuvIr6q96HzCQY3PEcIM9b28+YyPJ1hYgu2VAnWRRKUnrgVeZ3v8DoaCf7gtXUxLpYWhCmqGA2B1qup2r2WnZMBtnULzBlTrNZXSn44JwQ3zqYIm3a/MqiCK/3p1lW5WffSIatQxa/sjRMRVDnxe4kz/eYfGZJhG2DKapCBoaAmKkoCWq81J3i7tYI/7UryY1NOhsawvTFTb69P8F9c8IcGLcAxcqKAD88FOPds8McnLD5zgHTaY10RDxpWyhpZQnMVaiyAzc3UYhpBDLDAJ1CR2dAVDMVJ6b9nlbg8X4JnIlJ2BbKNlFpDaXraIbfJTjdvfDiIDOBwBpsB2mf4JqZjGrqmGuUtOjZ8hhli6+juHUVYwffIDZwfmOvT0JkM6v3KjtFCooMxe3NgrqJLq5+7f+jNjOIVl7N0dEAdv0iEv0W5Q3VzE5MMnv7Vzjauwq18pcpbKrh6Q5BRjqcsCGicUtzkGe60sR1uKUlSCyT4abGID4sVlWHuLomwPbBJL+xrIjx5DjvagpSYpgsrgjx6NEUS4t8XFkXIJUyubEpyJv9KX59eTE7BpPc3BjCtGzWVPuoCjsPflV9iKOjk9zWEiK5fwwZnwCpQNkuh5qq9+RzCHWRzfSnB5FHlO7wtC2UbSMzKYRuIPx+hOFz9DhXUhD5nfD2NpfMWD8nMq+n45LJwTSZhI0AEhMm0aEMtulwqNiYhS+QxrYk0c5dJMY6CVfMZnDbk0grM3OlU7Tj6d9zMs/JcEIiU3nDSiDy5A/nS0RXvGe2wj94lBu3/jOlZSWI+iXQsY9Kn8lk1y4Cfhu7twOjsha58GpmH91GYNOXSKz7Pd49u4JHjkhSSpvS7OybdKUuTShKA4KkBfvHJFfUCIqCjri3tDJCyKfxck+aD8w3KPBpXFnrAwSFAfBrsH84zbyyEBFDMJK0qfILfLqGBrQU6Pg0jb5oBmWZU8aQyLOXH0dFunRxPC6oFMrKOBxO0xD+AMIfAPSsPvy2Qylk5lirYj42PXqErT/vIJUwURIe+eoufvbfGsmURFRU8OPHYhihNJmmVoqvb0arLSPd18X4gdeZXVJHc1EtCOicGKAz2kdhULC0MoStFFv74iRMZxRUFxjMLw+RsCTb++OYNpzMt3ZiIkvEELoBuobSdMdq5YqIQilWVSkmJuO8f88DlJUUoXx+0ge38VL1jfyseQMdqhQdxVwGuHPsedZ0bkXMWUV923bW7PwOz635LAtLDbaPOIEnUwNTvFnWOai5x57qSvNKn+n8FlARMhhN2fQmcoSwrCZC0lIo01H+bdNEKYVl2gxNSlZU+RGaIGVLGosDCAEDibyOukjEpAsB18UJKLBtVDKBymQgEETzBdx3JKYajd6Gdp2MX6STJumk56RWJCYz+BtrKP3MXQRWLkREwiCgVGoEE4X4j/rpf/lBPjXvZm5qWkeBEQYgaSV5oftNnm5/lP+8rZwiv8bHHunh2fYYQig+u6qcz64s5tEjST79swSnosudkMic2cMzCwuE0BC6DpqO39CoCRiUHXoLbaSb8aIIgUw/P1j4GxQvuxK9N83YsIahga+mmP7l83j4rZ9z95Hvo2UsmjpfIVh7Ey31i9gxKgFXD0BD4ehMjiFEIV3jgkKxvNxgY0OAJ48mUMBY0iTiN1gQTqMsH70TgtKQ4bx603aseMq1+KEYmMxQ0hwhZSt2D8SYXxFGScnApPOs0/1Jv8jw/JrKtiAZR5kmBENoupE36N8OGVKgGSfLQjeVDH2VJVT94SdhTiOyvZfkYy9iTyQoKmxGmBUc2vtN3muU897W6+mND/M/Ox/C0DTeP/8Wbm++FlOm2T24iZuaQtwwq4jnOmIU+HSubw6DpnjycAzLPjVjyQmJTOTN6EpKwLGuCQRFEYE/qTF36C24/n76tj/HtqqNqNmraBuME0tJGu04pjKIJcP0+jTkvI28NNFB1eHn8C2+kdr2F7Ea5xERCqUCCGBDLewYlEilKDMEuoBYyiKs6/iKJMuKbNbXhHjmwAhCKfYOTrKkuoir6gIgJV0TJsNJm9ayEGnL8VPVFhj4dEHSkvQmJCG/QTxhsqMvzhX1RcQykrG0zJudf7GR5U0i75cCZWbAspChIMIXciykb0uLFL6iCqbqRjPDsRZD0U1XwpwmZEc3Q3/x78jRKCISJi5eBwVVdogbb/g/WNLmqzt+yMtdW0FANB3nD9f9CtfVr+Xlnje4oVlyTUOIAp/G/PIArSU++mM2r3Sfusn/lK2LIidHIIGyoI7PSrBw8gBDz2xD+Qt5o/EKbg5rbOpMEotl+LDaAuEivtU3H5kKcW1LAU8VXsWH9Fcwu9qYi0V3IkEoI4lNWLSPhVlR7mdz+yRPHZ5gTX2Y3skMPzs4SUkgzuySKtY3FfH4wVH2DZu0j6fZPZghYcepCusMJ0yiGUn3eAZd09g1lOTZtiirGoroncjwyP4JioIGbaMpuibSvNGX5vrRFONJm9GsyH+Zj50QSjoipCXRgiFHb1N5fs3zUSXgq2oGTQdpnfR6TdPwLZgFQpF+bRfWwDDhdcsp+837kUJAIkHTV56lxB8hmomxf/ho1gWye7iNuBmnwBdm37CP0ZRkVonO3LIgG1sKCBmKn3alGIiZpzwhn54JP+sgdlypEXuScFAwp6qJ/fEAnTLMd3eNs6pU8a6yXiLjgBbnz2pGeDRRy0/2TmCqUqpLi6nKdJP2FVNIEo0A+4dSHBxJUBYyiGcseiZSlIUEjx0Y50NLyni1a4JvvDXIPQtL2T+SpiSoMRJPU18U4ImDo2xoKaIq4iNjK77RHuOptgk+tqKM4oDON94aRBcat8wOEtA1+iZSSEuxrNrPS0ejxE3Ju2aHeXDf9GS7l5FFPvUoBZkUUtlooUgukuQ8Vu2vnY3uC2CnT0xkWd+t7ur5UjolpNKIoTHEnEaEz0AzdIS3CEXlc2Tl/U/PRIY3+0xunRXixpYI1zVFkBJ+djiGdRqLnc9oqUvWmCkVOgJfQQFJwvRNSlIZyUfn+ZjfUEfDsvXUL1nN3Lpy7m8NMBC3GEppWIEI/oXrwRdEYCOEYnlNmNV1hUymbW5oLWFFbYS1DUVc2xTm9rnFLCgPsKy2gJW1hWycVUhzkZ9lNQXUF+q8d1E5HeNJ1jUUMZo0mchIllZHuG1OMWlTcs+iUuaW+VhfX0jfRJoV9YUkbZuhuM3GWYXc2lpAxnKeTFxeKX4KcMU2K4NMTjpuj/MsaBtVzfhKj8nVOjOUQnYNgILQ2mXo5cUkdh6g72+/hoolAEV/bJRJK0GBL8y88uZsZM/CilYivjATZpy+xBhPtMWRSnBHawELygyOTkpe6zm96JAzIjIB6JpCEyAkSJWTlcO6oKisipLGWZQ0tFDWMIvyhtmES0oIaXmhToEi1zyvI4RgKGZiaAIp4TvbhpAIbKm4or6QlGVjCJvqAh9tw1FKgwY+H/h0jeKgj54Jk6faUiRtxXMdSUZSgsKgwFLwVn+MvYNxFAJTCZ44miJlweaeBM93pPEbOsWhAFv7E5e05+vth6s+mFaW0LxRcM7JTYEIFxOes+LULpeKiWc2w+g4WmsjVX/6GUruu4WCG9chAn6UgL7YEJt6t+HXfHxmxX3cPed63jN3I59ceg9CCJ7teI2hRJSXOycZTtksqfRRGoQXOuKMpo7vFJ8JZxTx4VjsnEGrdB2l6SghiBhwb80EXa/tot/nR9c1pCVRmsBWgvfU1vL9Th8AQmVQuoHp+sIORU2+s32I9y0p5eMry9FQdIzEaSwNMxjLENA1qsMGvdEYC2rDlAZ0pFTEMxaa5oVEAUKguTF8CuiKWgwnFatqQniRTs5HYzQpaRvLENQFfTHpugsuk9pJMSVcTIFpoUQSQiHHzXOOLY6OvqdRtPo2xjb/dIb1ZMcifbSLsX/4BiWfuAd9diORhbPRlEJaEn14DDOT5r92PITf8HFVzQp+a/VHECiSZppHj7zIt/c8jlKSnknJ671pbmsJklIajx2cRJ6mtHMWsYsKWyqEmUYkJoFCioJQqaVIj49g+wNIy0SgsGyJrmmUlZYR9vmdCdDMoKQTBwiwqibIuxeWsat/ksaiALaCvrhNUdhmKGER8QtKQgbFgWIKfRoVIUHKVnRFM9QW+HOuLTd2cTJtoyO5rbWQ2ZUh9vTFj5lhTQXDCZuAoWGpyyaP00YuehplphC6jvAHz5vTOrDoagLVs0n1HjjmnKZp+Hw+TNN0g7QFiR37sf70K2jV5fiqylAamCPj2IOjWCMTjCrF32/+BrOK65lVXI8lJYfHO+ieHMCUTuSIKQV//uIgX9uqY0rYMXDy+MnpOCMiE4AuBIYuaJuwSZsxJoNFDMUVCdMmtm0Ti297P53bN+MPRSira+LQa88zduUyxlOuC9NOI4SGoWsIodg9lOLAaIrawhA/PRClpdTPSDzDUMoinrbxYXFkNMXjh8e5rbWEuKXYNZigI5ohnJJkLMWugQSxjI1AsG0gxc6BNA0lIR7cNUbQEAgtSUbC7oE4IymHCDsn0vj1fMX9QsUOXbpQOMYwmU4iDMMJYDiH8FYaaAVllG24j74H/87xfeZNm/X19Xzyk5+kra2N73znO9i2zYoVK7nzzjt54YUXeOWVV5gpM1vaNtk/2s7+0fYZnwsU7dEU7dGzaP+JUsLN/48DM55UCpZW6Kwx+rih46cYdXM41DnA3xXex68knqJm09cpvfYefP4AnXu2UVlWwtDuNzj0rj/icbWcb078K01NVUx2tvOtZb/Dc8MFpGxYXK7jE4qkraFrCokTaa+5416gkEoDIVFKQxPOIkghFKbS8OFsceLcpZyJQCgy0gm80wRYUsPQbGzpRJwbwrl2OCnZOyy9ii7jdOGNI38ALVQwxcd61kXjrZsQMNZLxxffS6r/cPa8EIKmpibKysq47rrr6OvrY9euXVx55ZVkMhmKi4vZuXMnmzZtwrIsNE2jpaWFYDB4/EpPAR0dHcTjjkX6RIl0znDKUeiaRtDvo6Z3C+aRlyhtvAGFoEBXNJfojFsmwcoaggUFhH3QVKDRJp0lHgrQOvejZRQhfwBNU/iVYnltEUOTafy6TkFQ8ExbjKaiALNLdEYyNiU+jeGESXNpmAf3RLlrfgn9kxlW1Qb56ptj3DGngLGkyZKqELGMZDwl6ZtMcd2sQrb3xmksDvDd3VE+u6aU3QNJDE1jVlmQrokU5SHB3uEklznZmUG5AcTKzIDfBMOXd/Zs+9MLrZOo0lqq7vgMXQ/8McpydDMhBAsWLOBDH/oQL7/8Mrfffju2bVNWVkZDQwPj4+PcfPPNbNmyJUtkra2tfPzjH+eJJ55g48aNvPzyy5imyXvf+15+8IMfUF9fT1lZGaFQiC1btrBmzRoGBwc5evRoljsODAxmiexEOMNsVQKUxLQkIX+Q8kgQpesIFCk0hjJBlG4wOTyCbYOFD8sfQXfFCEsLYI8OYClBxrJACZqKA6yvCzK/1E9jiR/TtBlJQnOxD8uWNBf5sSxJ17hFechgLCXI2BK/LlhYZlASNCjya7SWGKxtiLCqNsKq2gAra0MsqQwgbYWlFD7NoDRkUBTQCeoaKUsRd1MjqMtc7IzhJZZwxMZUjrOdk7Jzr0WgCF9zH8XLNyKE4+mSUjEwMEAg4IReHTlyhE2bNrF48WKeffZZNm3axK5du0ilnIgDy7IwDIPS0lIsy8Ln8/HUU09RV1dHZ2cnO3bsoLy8nN7eXhobG6mtreXw4cO8/PLLpNNp/H4/Dz/8MCMjwzO2dzrOmJONZxTb7FJ2NvweupJEtWIySuPl8uspvH4l/kghmVQSreQmNN2HVZ9kQC8lGTf4+6L7CIfvwNJ8aPEwcUvRH7f47u4RokmwECRMScJSvN6Xpq5AsHM0QTItGYgrfnpwkoyEt/qSJDISXYO+mM22wQwaNntHh3GXqpG2bHYOpTkctelJpohmJFu6E7SNpUnZgtC4yVDCEUwFuOLnTE+cg7fY0tP7j7n8TMfXtBUVXhRC9rC4GJdT5pCNCrIsZ4GobpzTlmZX8/nDVH3oz0n1HCLVfxSEIh5PsHnzZnp6eqitrWV0dJS2tja2bdvG4sWLjymrt7eXF154gW3btrFs2TJs2+bgwYPs27ePgYEBuru7s0Tq9/spK3O25m5ra2PLli1MTp66r+yMdDLPPD5FtPJWER9bxdTrZlyjM3NUWn6s97HXT400dM6eYmzbTG3LX6k8I5V5T3zsEJ8qYOa399jrprdl6rm8yAPhGRSOvepiXiWg3JlBC4YQgfA51c2m9pIitfMZuv/9N7Fjo45pRAjKy8u57777+NrXvkZFRQV33XUXXV1dBINBHnnkkWxJQgjuvPNOenp6WLt2LT/60Y+IRqPce++9vPXWWxw+fJglS5awfPlyTNMknU7T2dnJ1q1b2bhxIwDPPfdc3nMfXyc7MyL7RYTK/pP9KdzsUk7uHIkmlWPBcj/Kyz6FF94zFU5KBYEUCoHmDEghQHOOO9mo8o67d3mUeHxyvnDwuknTdURBkes3Oz8VCaWY3PQj+h74I+xE1HknQhAOh7O6UlFREZlMBiEEyeRU87thGIRCISzLIp1OI6UkEAhgGAbxeBxd1/H7/SSTSXw+R8c0TRNN0ygsLCQazZkcLxNZPtSMX6ewOJF/1iMWL8dH/l8l84hpKjfLcrdpouV0ZLl/XoDtVKHQITAn45QAXUdoOugaQrhp4Ka5H7LPdQFUTOWKt0KAiBSB4TtvQcOgEFKSePVBer79BezYeF7Pv71D9zxYFy9NqLx/YdrQlAqUQzzKdpLOKNsjIuevl6Uqe78QU8ohr+R8sW/ma3Lnp5+b+luBVFmn/ZSCNWd9n9ANhOESX7Y8T948XsXnB94EJZVAsy00wzgvjfCeUWka4Ws+SGNhJb0P/BHpwS4utiVLlzCRTWURWSPEFMPBVILyzgnpEI6ybecjbZczOWKel2hmutLkHLoAwllevbmJwsnriLRRprvswiM6nw8Mw+F0WatffoHq/D+LUI6J3e/4opzFsOenPiUEgRU301zRwMB3vsDE7pdOaUnM24VLWFw8Vg7LflXe0HJFuSxnshG2Q1BZEU/lypuef9DDlAF6oRWg7CQyTSPLe4/ZzMI+H8LnR9ONqRzwPBNZNjGB0NEKixFo543IsgKyo5BBaoLo899i+Gf/H+ZYb57BCs6tGJmTYvRQIWYi+s4RF3MSt/dvnhYjPXHPRrocKj+VG5AjLo+g8gbfiUS6iwbT5dDsTzHlEqUkKpNGZTJuQhw/wud3Epye5+dxMpq5Rh8pQT9/Feb0VncUBIsovvXXKVh5K+NP/zdjmx/BnhzGS4F+bkjM8c9p4SKKVtxI6W2fPvHVF5KTndBf6VkNvBmKnIldKYXwxD2XQ3n6E3kWPaeYi4lC3l5MdX8AQqD7/Qh/EDQtxwXcc+e+ftDDBQhfgLdDQZxKQhpCWci+NiY2P8z4G4+R7j+CNNNnWLo7mWsGwfJ6ClfdRPE178fXvAR0H3s/XHlm1sV5X92vOF9ihUc/x/jPnM5yaMwzRig3DbbtGiUcy54nJCqXEH9xyelU4YpOmobmD6L5/Sjh5gE7H69YKbRQBM0fBuHGhZ5XzOxVFQDxKOmj24nteonEwS2kB45iJSZQVsaxEmchpnwXhoEeLCBQ2USodRUFS68lMO8KRFEVjiTlxMCeiMhOKi6KfP/QdHo43T7LL8YtV7nWM8+yh0dQnt4k5ZRq1LTveWnuc/Fz3sTt5so/RrI6zWafS2T1lWyufrK6YH5e+5xpX+UlFT3Llnv7AkiJTCVRpokWDCIMY4rjOys8HP8h3L7OGtKnvAiRP+nl+Qez0TTnjamJaSTiVKUUECnGv+R6ypdsoMzKoKJDmCO9WIPtWOODWBMjWJkUQil0fwA9UoJRUoNR2Yi/og5RWgP+MMfutXdyYe/EKeFik46FynWOkvXLkP0r3HNSKndTEScFW5ZBei9Cydwydelwp6xhIo+bZok6r7dyjySmXKeyAy9f6Xc5ofJ0j/wNm7ImwgsGj7g8w0OO0HKThpObH3JGCuEQwNm2WziW0+rIAJXhUQ6OziadsND8AUQwSH4FJ6YDBcpZnCm9t5Iv2ecqdMR5oSg0xmkpaKMr3kI0U3GWD3I6yPWl8NqtBxBljfjLGwnMWz/l2jwN3/2Wt6GHIo8jn3qmrhNzMttGKuv4yvZpYZp9Z4oqkFdu3u+Zapt6zBmsy6r2sLxmL2/0rKJjvI475j6PrXR+dvh6fFqGyvAwaStIX7w2b4eXCwMlFIX+CSJGjLhZQMIsoKa4m4CwGIjXELfCKAGl/nFKg8PErUIG4jXnZG5QCIJ6kn+69e9ZXLWfP//57/DIwVtQ6bSzvVPYWdl8spo0YdMQPoqhWSjAUgYjqUoSVgRnIxCvPkd3BotPzv1PPjL7azzS/QH+evvfcV52Up4Bx/C2Y8bXVE6kjvk23ecp8n6cmsni5NZFMX1+y68xf9sej4imvSCVd73IEdpZKwFCURIYI+yL8omVP+TWOS/xZ899lvLAAH9y/VfoGKvjrZ75LKpp44sbv8yW3hV8/vE/xcRwxSbI27NpmgQgsqKnNxNmt2BSuWumxGuK/K2cpm6TlKvDOfjR5T/lI8se5fu77uCrb3yQP772G6xveJN/evUTfHfXe9Gx+O2r/oc75z3PN7ffx7++/tG8Jrovd4rc5e2Nlmtbtg3uMe+UpQx29C1EQ9EZrcN7h8oyEXGFFg47Fkiv3Rz71gv9E3z1pt+nKjgAQmFLna54PQ/s+RBPdtyOrYzs/Y6kIjgQncuhiSXsG1vojBm374W3K4/XwtwwwVkWlbfPjvdezm7kvO04uU52wkc6XWPDOeweJfns2u9z97ynKPRPoKsMf3jNN0EoIvok88ra+Pd3/z9+uPNqCv1jhI2YOwA1lJAsLD/AXQueQ0qNnf3zWFm3j86Jep46dA0fW/4wpvLx8tGlXNF4kB/svp0llQe4oXUzY4li2sdqWFpziO2Di3ji0EbmlLRx+/wXqAyNcWisiccP3MRoqpymog7es+DnlIdHOTrWyE8P3EhFeIzrWl6hMtTLxpYX2T/czGudC7m55Vmun7WZH+65nUJ/jKsa38IQaTZ3LiKsx7lrwbMsqjjCroFWioJxKiPj/GjPrXSMN3BF3Q42znodv2axuXcpLxy9iowM0VDQxR0LnqO+YJjO8RoeOXATI8ky+iYr8BuNJMwAN81+iTV1O9jat4jqyChNpYO8NnQ1r/RuQCqNhWW7uKX55xT4Y+wbWcCT7beAEBQY44R8E7w1uIKwkWF+6T6+cOXfMZkpoH2yhfvm/oShRDV7R+eyoK6Xnngje8cWMJYp466GHzGvZBdbBjfw0tCNrCt/hQ01T3MkNo+fdt7H0tKt3FD7DCEjwe6xZTzddwcpqySbUfpSI7Nz6Cc7zoOf4/7I553jqWJGU8WUBEaZtAronGikoagDUHTF6umPVWKrvFg+Z86mvqCHL93+RWYXtWOqACnLR4F/ki09a3mtYxEfXv4QIT3Fh5YUABo9Y6V84ab/oNQ/jKmCpC0fhf4JHtx7D4cGm/iPu/6KhoJuYlaY+/xxrm3azh8/89v87U1fYkX1Tronannf4idYWbePJw6tZ1HlUVCwsPIoaxt28d0ddzCWKWVJ1WGqI0PMKu2krmCQQ6Mt7BuewydW/pjPrf0GApv3LgohlIluwKbO5cwvO8IXb/5nIrqTtvzepY/yL5s+yc8ObeDf7vwrFpUfwFI+DC3Dhllv8Ts/+31um/ciq2t28HrXKq5s3MkvLf0J71lQiM8QFOgTvNd6is88+yUQki9d90cUB8ZJWgHeP/fHLK/aw5e3/ioASTvC37z+B3THmvjd1V/iw/N+wL1zH+HBg3fzsYXfIWn5kBh0JmaxZXA9H5z9v2QI0D3ZzMdn/3+0Fhzh9eErubflO9xe9zD/sv//sLbiJf7v6t8mZMQxbT/3NX+XBcX7+Me9f4qt/OdGN32b8fYIxucaCnyGycN7NvBfb9yDROP17qX8zuO/xeHRJtIyxP998WP81hOfJ5ounyJKK6Wxvmkbs4o6OTjWymcf+zOeO3qV+/I82UrhFyY9E3U8dvAGNrRuodQ/zKtdV/CZR/6CfUNz3G1r4fZ5L9BSfJRnj6zlD574HXomK7mq4U1W1W6nOjJMyvLz0J5r+d7OO9jRP5/d/Yt58uBVKKF44tAGvrntHronGtg1MJ+y0Ahr6nZyTfNWfFqKlzrWoCG4Y/6zgOSrb32YP376d4hbBQgJfpHk/cuepNCY4Gtb7+MvX/gcW/tW0Frey21zXmRx+QF2Di7ks4/+KT8/egNSGTSX9uQsuwKQTrKGHX2L+fiP/5rXe1ZQ7BtjbeUm3tP6KJWhfn5y8A7+/JU/YiJdxM1NP6epoDMrQ0oMEnYRL3Zdi610KiND+EUGhCRspNk9vJjnem/CRkcoR+x+dehaomY5C4v2MKdgN0tLdzBpF7J54Gre1/xDSvzDfK/tw3xx+5+RtELc0fAQ1YG+nFHoEsMlF/HhWeTW1m3l727+F8J6DF1YXNW0je++/w8oDkTRNJu/vOlr/PYTlc6eG67pTiGQQlIUnEQIye7B2bzYvoGQP81d857F275WoBjLlPAHT/8e7RMt/MstfwUabO5awUtd17Kidj/r6rcisKmMjCGUYlZZP/eueIHeyToGYjWMpUv519c+widW/phfXvszNGyea7ua76XuZDBRjaYEQ/FyuiYbAPh521Vc27iFO+Y9Q23hKEkzxHNH1uHTLYr8MVIyyDOHNtA2PptPrHmQ4vI4um5RGpjEUjqvdq5mR/8KXmpfjaV03r/4GTQk+4Zm8VLXNewbmoehm8TMgpyuBtmJZevAYnYMrmDH0DyubNhKkBilxhAoxcLKg5QXRmmLtmLjc3SlrPFNYmCysOwAQrOZSBdiKQOhBEcnZvE7L/49E/5mfmPJvzgTnVD0JJrZObqSqyuf556GH1IT6GX7+Go647MoDYwjlGJ5xU6ai/o4MLkQUzopKpSr915qnOySIzKFRKFh2T6iyQJKi0eQSjA4WYnfSOPTTCbSxYwkijGl7toBFEX+CdbUvkXaDjOeKMBWOlc3buczV3yTa1veQgg7T9oXmNJP0nL8Ih3RelBw94LnQMCd81/AEVw1jow2IIVBf7ScF48s5c75LxGXQaLJIm6Zs4WuaC3f2nYrn1rzE26f+zz/u+MeUpYfBKyo2cc1jVvY1LWOTd0rGUuVcnXjNoSA/cPzOTgyG4VOf7KCZeEhfnXd9+gar6G1pBslFKYdoHuymsUVe7h/2aPcMfclbpj9BruHF/Dzg6swlY9rGrfxa6v/l3cvfBGfpvjDpz/vdKTnf/OsvEqBMtxJzPG2tEfrAI32sSZ2j8znXS3PM5ouYzhRCQoCWpqPzP8BAV+Gm5qeR0ofT3XchCWdNAApO0hGBRFe4IBbV0YF+Xn/TWyo+jnvbfkRhpbmhf6NxGUJ3fFG1pUJDo7NpTfRyHW1LzCcqmQ0XZ4r4BLDJSguOpazN/pW8omH/5aeeCND6Wp+5ZG/4+vbPoAtAnxv71184If/zq6BJQg0FD4WVbXx3/d8gS/f+X85ONzE672rKY+M85vrv0N9wRBS+VHoLrfzoYTuEp3g0X030TY+lzll7Xxu/XcI6smsk/vxgxt5rXs1V83awV/d+J+sqD3MgeE5dESbiKaCXD/rDf504zdpKh3k0NhsBuIVvNKxht5EHctr9nP7vBcBRe9EHdsGFmPoTg7B5zrWEzcLSVphvrPt3UxaRdzW+iIfWPIEpu0D5cNSfv7nrffSMdnELXNe4YNLH0PXLB7acz1PHdnIs0evpbpwmM9f+QANhX1s6lrGfpdwFX4QAik0JIbzEQqldBQ+wOD7O+9g7/AC7mx9lj9c+xXmlx1l+8ByJqxCFAaGbvGBBQ9x9+zHsaWPr+/5GI8dvQ0lBBIfEt2hCymRwsAWPpSb6GHL4NUMmTUEdZOoXcGmgetA+fjukY9yKLaQe2f9mM8v+Rdaiw7zxvA6knYBkAs0uJRwwrCqOX/32sU7dSgI+Sa5pvktTMvHKx1XMKfiKK2l7ewZWED7RBMSaCjoZUXtrqw72rSDvNx5BWEjyfqGbdgI2kaamF9+lOFUBTsH5nJN01soqfFS5xWk7AgVoWHW1LzFnPIO9gzN44q6HXx69Q/49u738pfP/xaFgUmuqNtOWWico2NN7Bhcgil9+LUMS6p301rSQ9IM8GbvMvoTNWgoWoo7WVBxiN5YNdsHlgKwoKyNeeWHkErnzb6lDMQcv54ubJZX7WFWWTuHR5opDCQpDk7wRs9KBhOV1Bd0s7J2L37NYtfgfA6PzUYhKDBirK7bQXXBCN0T1WztW0pG+llfv42S8CjbelZQFh5jdkkH+0bm0DbSysLK/bSWdnJodDb7R1qpDA2xpuUghaE0B8bmsW9sEYaQXFP/CmE9BijSdoi2aCtHY00o6ac8NMj6qjcYN4t4rW89thFhfk0H8wr3cjQ+lz3R5fiwWV/xEkX+EeJmEa8O34ClnNXHVYFuVpW/TlDLsG9iEYcnlpLLwnIxIUceZxy7eDETWTY6AvIMFu7J6cpx/vSXf63nu3L/KqHQVJ7f3/U5XT/rFf7plr8jmixkU9dSrmt+i8rICH/63O/yo73vzt7viaZONcK1oQhAOn40lOfR8hqDAqdOL5IgG27l+eGcv/l+q9yzAEgnuiXPz5Qrmey6uJybLuevyx3M6TnCdXFMseIZBlqkMFdlnlNxquXWuTfrG3S9XMIwEJHiXJvz+j73rFO6PNsu79QFjiE4Dk6NyC5BcdGBZzH0QrkE7vfs77z/xPRzed/z/mpugc617pVC8VrXSn605xYigTjvWfA8kUCCxw7ezFNtG9xxLbJZALw6s4NWKLcsp8ECLa9lZLflzW+rltcu70GFmPYc3uBDc/tB5B3zeih3jOw5MbUv8p7Ve9789itAWRbYZq7wvAbk2pO7N78tgBOLisy1b1qf5x/PtlMo93yu/ZcqLllOdqqYHq0w/UT+BDmzediZUXXNpLGgh/LgGDEzwtGJRkwrPK2G49yerUdkOU42uPYiXz3gWVvxB9BD4dNrb15EjVZQDHou4enF/MynjlPjZJecdfF0kR8JBa50yPRV0MoZEG7QspeeQCicVQFKYgHtsSraVaW7INFCMDkjiXmZpkQ+29EEKj/zVDYBTr4AmRPGLq5BKJx9o1WePHcKDcxG6iuc/jzv7bw4cckS2UwDUeX96w1a5Qr6yjsl7ezSGqQE29lXWnjp3FDZRQG5SETv10z15X0R+T/FtPtz6ojKIzyh6QhdQ+k6zioHhxDVtILFlNrfvuGarcmWYNsg3CFzSjNBdmoDy0IY/rx4nV8cXJJEdiz3yC3cBG/2dAnJdpLNkLfYUx1DHRzn5efvtDXzMM+3R2RLULmrvPNTrvfaJ0Fh513uEJ3SdTB0hEt4+Ut6LgypubVb1rQc9ye/x5tZlGU7k9hFLh6fD1wyRJbPTXKCFtmoA9wUbtlEOdncHio76I/NZnt6r3tmMjzxwdOqUSlngzvbhAzOqmVdRxgGGAa4my1OnRLeniHrCAG2s2Tx1KTFvHuFk1kLV1LgF4vQ3nYim56mbZqM5XydMkg9kzjg5faQNjKPoPJze+SX7g3Htz2rtcrT+bLKf/Zkzm2AwGOrXhr5fNYnlATLSa2mhHAIzudziE5zHOdZc3gWnnkll0RO5dXj6VTH8MXj5pD0blMoaTn97xmlT7FfBTiToGUjfPqp3fQOwgXlZGralJ8bK46O5KQikE4OPdtG2a4IiHetmvaicwPF5WEXBM4kISn0a8ypKsDvM+gcjtE/6eYCdM3TjaV+msrCTKZMDg7ESZiKopBOa2WEtqEEkyknn2LEL5hXXUDXaJKqYp2xhElcSuY2FCOUwFaS/ok0g3E3ZcOxcivVhQYNxX5MW7F3MIUGXNEQZnZZiJ6JNK91xklYJ1C0skYg283neFo94vxrZZyckL9QfOwCENlUFd4lGjeXh5KOcu1koFK5WX7KwBF5fpPjix0X9jUKFteE+MI9S2ipKEAqSJsWX3mujR+/2YOma3xmQxMfWt+CQuDTYV9vlD9/eC8oxZc/soKvPXeYb73eh4Zk4/xK/uzdi/n1/32Dv3jvMp7Y2cfW9hG+9oFFJNI2UilspXh0zxBf3dRH0iRvxnK42MY5Jfz+xmaiSYsPPLCbdy8t5dPrm+gYS1JfHOSxPYP8w/PdU8xG+cjqkrY8A++qy1NNEwIStIsxeuP84YREdjJO4HGMY/Js5r2q3HmVJabsx7ZRyrP2edfkiU6eUCPyedTFaOLOQQFhQ/FHdy1CE4JPfmMLE0mLj13dzG/fPIddXePMqgzz8Wtb+bdnDvDUngFqigL85XuW8Cd3LODz39/O3p4oNy2u4sE3erER3LK4irbBGIeHkvg0Dd01Tiql+KMfbqVtKMl18yv47XfNZzRu8j9vDuWpok7/PbFvhNoCH+9bVk3QUNy1qJofbu/ny6/0cOeCEn7zuhb++/U+hhLHsf8Jd0qbYeOMk/aJOxCElA4387IK54p+R+OERDZd6siG5XAsYYHrZ1LK4TyuqTzfCOHlS5xOvTmDRJ6ukj1y7Cu4qF+KUjSXh1hcW8wf/3gnO7qd3UX+8cnDPL1ngJGEyacXV7O/d5zvvd5N2hb0RS0eePkof3TXEqqLAjyxq48/vmMxjWV+UqZiZUsFX3/xMBnbIxvP9APDMZOucZPvvt7N6uYSbphbyv6hGKWhQLZJewfjHBk1GU1aKJzl/rZS+HXwOV4DioM+KiN+huLp41pzFMrJwuz6GU/9PeT8azKdwfAFpgX6XtRv9KxxYk6WyTCVIhwCUiq3XZAj8Umn4xXZSIZpJQFuMq3jWvo8HF8Bv1ReRTjgR9c0xuNuIk0hiGckrx0ZRxOKkrCf0YRJRgqXIwmG4yaGJogEfLzWNk7asrlqbiUTCROfrvHiwdETCMfOXtr9ExmaK+DWOUUsaSjLGjO+uaWboyMZpylA2lY8tHOAz17dxLqWEuIpE1NKTmYh8oxPpw/PzKJAWu6Wt8FpTpF3Lk5MZIlEtnscFch17k4xVE21XM1kO1TZUsiJHe9YCAbGE8TSJqtaSnn9aBSJYGFtAR+/poX/3dTO/r4J3rWkltpCH70TaTRNcGVrKWOJDAPjKUYTNluOjnDrkmqSaYud3WN0jibx+3LeNoEjLjpStqQ8bLBudhldo0n+5qcHnA0nQiGUrmMp5aTqzr1JdvbG+dLzhzkyZuPXJH9313wmkpkTPlku+dDp9oj7r3AkHZlOoRt+lKZln+WdjBOLi1kna343THWSiLwLT583vfMgBPROmjy0tYuPX9NKZVGQsbjJnctrmUia9I6l+OEbPdy8uIYv37+S5/YN0FIe4oZFtfzH84cYSdig4MldvfzLB1eBgC88vBtLgd+rA0ApfLrgvatquWGR5Nq5FVQXBfnbx/aSkkDGAiuOCEdA9xJy5nTaK5oL+eCKGr7yUic3zC3lyHCSwbg84WsSrhFF5cSR0+0dcEVOaaYRgeBp3n9p4pJxRl9KkErwleeO0h9Nc/PiGkL1Oi8dGOKbr3YwlrQYTUo++62t/PJVTdy4uIaJRIa/eXQPj+8cckhBCN7qmODVw6MYhuDVQ6NOudKxQvZNpJlIS3Z2TbCkoRzTluzrifLXP93L7r54TnqQEpWIo4XDoBsMxkx298ewJPxwxxDlIR/3X1HPwGSKf3i+g4ytjivGZ3mo55M8Q+djNmVeOuksgdF8Uyt4B+KEUfitf/jkL4bQfM6hstY9Q3eypZu2mhKtoZSzcaaha0gpsTwmInIDUXfXgNgyF+miu3GNSrkMCs+ynieOK7J1AaDpaJEIQtPRBY74iGPK8GsallLY8kR6cvapnGcqKHTCvs6AMhxrs1u/4UMLFzr7qp12SRcDcuTxCxOFn33k/IkjO3KnzhdCCKZPMN64zNczVV4R2WgKz9qaN6HPVB6AZeddlJ+OXAikUmQsma08p+86g9CaVl72mNtQaU+N9Mi5UPLrEmDb2IkEWjiCdAe0UAIpHCOI94wnmnCzfaBwrcRTI0bU9Gun/Z5qPnP/WiYynUALekto8pYCHbcllx7eUUTmjZaIX/Dxq5uIpWy+vaUbWyoiAY1fvqqRiZTFd1/vwZRw5ewirplTwf6+SR7bPQSuaTpkKG5aUMHK5jJsJdncNsrLh0YxXXpYXBviXYtrKA776RiO8diuQYYmM6xsLOKWxVU8+GYfzRUh1reU8OiOXvb0pSgOG3ziqkYODMT52e5BhIJ51SHuWlbDwESKB9/sJSMFG+eVcmVrBTkvl2AklmZ7V5QbF1Tw0PZeDgwkKQ0b3LW0kvm1xcTTFs/sHeCtzkkAPrKujooCP/+7uZuRuMWiugLevbyanx0YY+eow0kjfrh7cQWLqiNMpi0e2ztCx3iaX76ihrAb+uQN+Z/tH2NXX5yIofGRVZWEw0F+eCBOf8KmqdDH+xcVobtGLw2wleK59jgbmgvYPZTi2aMJinzwseUldERNHj8c56p6PxtbCtE0jTdHbJ7tVZhT0i6/c8jsHUVk3o6LaQuWN5WzsqmYFw4M0jGWYWl9AZ+9YS7f3tyBpSCgKz51/Ryun1dB22CMlw+PEk3Z+HT4kzvn895VTUymTHRd44PrW/jvFw7zlefauXZuKX///mWEfDqxtKQs4uO25XV87tvbmF9TwMevbWXT4RGuaCnhU9fPYWlTKb/6wFsUBjQ+clULT+8Z4IldAwB8YG0DH7u6lWgiyeajY7QNpKkpCbGiqZjW6kIKAj729IzTPhwnbUl+eUMrO7pGGI6ZfOmDy1ndXMpoPEMwYHDvFU38/eN7eXhrL3ctr2VlSwUBn8H/ffIQrZUhPrFhDocGt7JrIAGGwSfX1fCpKxvY2ROjuSzEuxaU83sPHWBRbZi6wiALqgron0jSHc2wudOxkK6sj/D565rxaRpRs4dv7YpT4FcsqQ5SFTaYXRrk6FiKwYTF4dEU9y8r56H9Yzx7NEHEBx9aVsrmziTdEyb/9K56RhM2KUtyz/wQgS3jPNSheCfaGy/Z9APHgxICS8LTu3spCvlYP7sMhWLjgkokgmf29IGCWRVhVjaWcGhgkrrSCGtbilAIVjQUcPeqRl482M97/n0TH/rPTWzrGOPmpbW0lPv51Y2tCDR+7YE3efe/vspXnjlAbUmImxZXZcUgmafbrGku40PrGsh59B1TeGlY4/r5VbQPTRDw+bhpQRUIyQ/f7OWj/72Ft9qHGU+m+c3vbuP//Gg3KSsnvr5/TR2rW8r512f28+5/e5X7v/Ya/eNJPnvDHCoL/IBAE4IPrmtm3axiN8+I44+T6TQ6kkW1BUgJ//lqB7//0/08sKWbwZTNb/74EF986gimrXh4Zz8f++4+XjkyiSbg1gVlJDMWAxNJbm4tIaAr9o/afPrRbv7z9T5sCf/1xjCferSHPcM5d0BurxTnb1Oxn0K/j5c6Jvi9p3v4p019DIxMIqwM+YHNJ485ujTwziKyvMG9uW2U0XiG6xdWUxIQXD23iiODk+ztj6OE5MaFlRi64MtP72cimeG2ZfXoSBbWFeHTBT96o5veaIZDQyk+++1tfOy/t5CR0FpZwPaOEV4/Os5IwuJrL3dy979t4uFtA1MMbgJImDbbO4f5xLWtLK6NAKBw1rNdMauM2pIQX3+pjQP9UW5ZUk3IryEVmFIg0RBoWFLkiVFOTowVTSWMJ9I8tLWXsYTNvv44P9/bT3VxmOaKMAroH48xEI3zO7fMpziYWxcnbBvbtHhy7xCmLfnHuxfygVW1HB5K0Rs1ySgN09XNLAUZBRJBedjgmtmlvNYe5aFdAyyqDDG/3IdUGhmlY7lDyUKQUbl1dM4fb1NH59eOgQRHxlK8f0k5f31DHbaEbf1pZCLmbOZ+Rg7vixfvLCIj55Prj2Z48+gIK5pK2TCvnOaKAp7bP0A8rSjwabxrSQ2D0SR90QxHhyZZ31pBbbHfsbIBfkPLWggL/BolYT86TuSYoWsooSNQGJpGeUgnEjhW3bel4ivPtRFPW/zWu+YRMBzlXhdw+7Jakhmb7pE4h/vGmVdTzJK6whnm7ulHJJYt0TWBbugoHCez36cjpY0lnaxXYwmTf3lyH61VBXxofbO7REUgFMhMhkf3jPEr39/DT3cNckVDEf/6voWsaYjM2J9KwVXNEaoKAuwfmODoSBKfgJtnFRw3akMpgVJguFm4hBBowjGatEcln3msgy+9NohfF/zxhho+varECcFLTDpr6t5BIuM7jsjAYWi2gmf29FMS9vEbN81FSslzewcAxeL6IuZUFVBVFOI/PraGxXXFlBf42TCvgp1dUVIZyceunsWqpkI2tJbwP7+8hv/4pVVIJdnbO86qljLuXVHBopowf3rnXB787FXcML98yoprb5ul3vE0//rMQZoqCgn5HFGuoTjAutZygj6df/zQam5a0oDfp3Hb0mo3LE1l/04Vn0AqjVcPjVAU8vPrG2exuDbAzYuquH15HUeG47QNJZz7BbzSNsZDW7uZX1vs5BfxQptsm/tXlvPL6+p4aFc/D7zZTaFfZ2F1gRNmqKaaHgwNbltYiSbgo2ub+MOb5iCE4PpZRRT7ppKD930kZTGeNFlVE2F1tcEtswqJGBod0TQrqn38wTU1xNIWX3ihl0RGsqw67OzfLiV2IoY001nTj/ffpYp3lOFjCoTgtSNjDE2mmVNVzNaOUQ4OphAIbl9Wg63g97+/lY7hJEVhg3/9yCpuW1bLw1t7+fqLB/mV6+byv59ahyYEibTFPz25j+4xky89c5j/9/5lfOG9y53dRTWNlw4M8sSuQW5bWuNWrk3Zgvap3QNcv6CCu1c2olBct6CC4pCfv318D6+1jeLXFF9833I2LKih9LkjjCWsPC1G5aknzn5eP90+wKrmEu5Z1cTdKxvRNcHgRJp/eGw/E8lcwhpbwX++0Mn61nLmVhfl+kZB18Akv3ZtCyvqi/DrGoOxDG90TiAAW9MBh/MpoLnEx+rGYn5+aJh/fakbELxvdQ33L69mZU2A57tS2aKlm/s7moIf7hnlM2uq+Pq7m9EEtI2lefTAOGlbUFvg5/9cU8d40sana7zcMenkMcKJ1peJOFpQQcB/yRv0L1pntMr7N99vkh8clOcJYsrl7g+hJDctKKOqMMjBvjFeOxJFR3LLkmp8huDJ7b2kTIku4OYl1RSG/Ty6rRfLsljWUMzS5nKkLXnryDAHhhLY7ori2pIgV82ppLjAT9doks3tURImzKuOsG52KT/fP0RNcYD5NUU8vmuIiWSG2iI/Nyysom0ojqFBQ1mYJ3YNMJG0UGhc2VLIrOpCnt4zxHDMZOP8MioL/Ty2Y4BkRjK3Ksza1lJePjhE+2iGkK5Y3VLKwtpCJpImm9pG6R430YTiXQsr8BkaT+wawJawtKGQZY0lbG4boW0wBSg0TbB4diWrmooRAl5vj7J/JIMAKsI6N88vZU9fjB19KZpL/Vw7u5itXZPsHUyDUDRWFbJxbhl7BlPsGEzTXGRwdWOYTd0J2qOOr8MQijW1ARZVBoimJC93xRlMOK+pOgRXNUQoD+scHEnyWq+JJSVSiOwKDyVA8wXRQiF3A3lXGXCNR1Pe/QXBqTmjL0Eiy0VMgLsPtXRSEjirqO28pTXOhXnbnXvRc+4vt0zPyYq7UiC7aybZ+p1IipyFbEq2EUE21ZvSNDTdTYKjGQhNcx3N+Q5xL/LDvdd1cDv157dwahB2tkbhRU5M7aFchmD3fPa1O45yh7HmeKQIR1CGcczx7P3ksm1lBVfhlKP5g6hQ2DkrvHvc5KbTXNM5j1/ut+eDy+vF/K7J+yKcDMbBsJPnxNUrLyUiu+jExanD0P0uJQrpZpyysd3sU0iZTeOGyu9wlRtkXibgPBz7m/yACXfgzBB3kLfu7Zhz0iVm20aZJlKAl95N6IbzMXTQdK9hec84NePu9Jo9Ijj22Ez6kDjmuHP7DJqTbaL5jNyIzZkDc2XhEUF+7zhBvhpy2j3TypneLpesRF6JM47MKQcVWCZ2YhItEEL4AzO+04sZ55nI8sNuxJTDuQlLTTnm5UOUUqJs21l/ZEsUdv6lgPvyZ9zmI5dMxqkif5GhN1u757NjZDq/O02IfP7kbUskna3RpA2ms7ZMaAbC8Dm5LnQdJ3/8hZiRFcq28va5Ps27pe1McppxGveebi05+UVIiUzFEZaJFgi5XG3qZHCx4m3hZLnB7v4WKhfNbbs5EW2XQ0npLAb1ZBZv0M7Qj1luNYNQ62UKzn33six5HERDyxPTlNCcVdt4hZ4LTGucwsnEa1vITAo0Hd3nQ/P5Qc9t1XS+kdVrbVcKOM0qs0tezijfx5lDKMDMYNs2mt/viKya5rbp4sVJiexESln2wTz9Zbqc7L5NJ+WAQ0hKSmegSRvlZqRybsppu9nQmimC+gz1HvcAWeueoQsK/RrjCQXCRiktS3RCg+qiEJZlMRIzc0a8KfqLI2oZmhNJr2sOGVhuu3XNia6wpUS647W8OEImkWLSlFmi0YRA1/PLUNjSQrcktplB6Tr+YBBb6Dkdbgr/z33z7vdiiwEMTWBJx6AhlMJWznVKKaT7XUqFoTv+O01zNr5QSmGd4ZYp0jTRfb4z4oSnhWyAtVexjUwnUaaJFggiXNdILvV6Tia5GHAOOJmrv3iKr5tTXtq2m3DUmpbjw71N5GZUJbyIgJxAd9YQTqTCu5ZW874r5/C5r2+mJByipbKAHZ1jxNM2H766hYX1pfh9Bj9+/SiTiQxBv040YXJ4IOYUo+l8/NalrKsL8/Dmdm64ogWRjPG3P9nFmCn41N1rWF4s+O+nd7G9P4lScNvV8+nbdZDnu+NIBZpu8Es3LeKqpgjfePogd26YR8hK8PD2Ee6+qoWO/R28Niq4d1kpf/3TPSS1IMLnQ9c0lFK01JVSaCbZMZiipKiAP7y9FV8qyRcfP8SYqVi1oJ6PLorwZ48c5n03LGAOcf72xT7++D2L2bH9CD9tS/Fr19dxZERy0/xiXts/yJJ5NTRGNH66rY+fHEmcURdL20ST0tUzz8VLOzGmaAYKsC3sZByRSTvEZvizhHZm08b5wSkQWX5z85Qi5STIUdnsvSonp3vJdNyLHbEuP5Wbg5zyfu5NRQrwC8X6+bV0D02wrLmEm5c3kk4lMTTJliNRVs2p5s+/+wYNZRF+6fo5lBWGeHl3J1cvauL3v/U68YwkFAmxtFTje9uHec+aBgp9imBJJcsbiuiYlFSXhBgbH6GkvJx/ur2Jwe5Bhn0GH37XcjYMjfI/r3YjleTx1w5RXbKMJfNqiXb3MFRei5Ge4PtbermjMUDfwSEMXzWaZWLZJnWlNfzObQuIj03iq6ygRcX5/I/3oRk6RT6BJXWXW2tcM6uQcamzrDJAUcigtbSEGxbZ1AYE3REfmkhTUxJgIiNJxJLs74uyZUjy4QUBHtk/Br7AGfW9kBJlZhwd6YxXS59Gffnf8oeMZSJtC3QfIhBAGL7z2o7TxQklamVbYFkoM41KJVHJBDI+iYxNYMUmsOOTyEQMO5VEmiknHEbKKcl0srubvO0QNFeGmVtbTNBvcPOyBnYeGUITglnVJVi2xLQky5pKWN5SxlA0Tixl8fiOASZSJgHDabOUCs0waCwNcfhIHz/YNkAqk6GqsowbFlSiY7OjbYiy6jIiykIYPnxC8uqubsoqS7lyQS3XtJZx2/p52KPD7BtJUVgQoiyoUVxazAeXlfHozgGkJggF/JRFfAgliAQNrESK1poQW9tGeevoCHELykpDJGJJjECQ8ohOcVGEtc3FNFcUcMuiCoRts6kjzifW1/Dq4fFcbwho6xphx7ji/SuqkVIhpcKy7DOe3AQ4yZYuhlhDpVBWBpmIIeOTkEk7k3/ef0z5/vbhhJzMik1MPaByxoR8A/QxduS8Z7iQOySGgwb/+PB29vXFuGV5DQMTCarKguzsGMOy4T+e3M1tq5tIpkweePEoGxbVkLEkm/b1kjKdh0jHEzy0c4gVVX4e2RfllrXN/PTFfTy+ZxgbwY1JyYHDw8TaE1QGWoiPTnIoOsGKphJeePMID+0YAL+fT9xQihGO4G/v5VCqGjnQSV/aYDRhsWZWOe0pne6xBHOrCmiPjhOLJRmKJTjQPsH2zhTVSyvw69A3GOVwtACzd4CeuEVhqcZ/P7mXrUMZbphTwqQ9QtdgjLFYnNe6k1TpJpZt88rBUUTQT11Q8b23+hlP67zelcltEXWG70hJN1+HPzCj6+BtwbSNRpRlYdsWaBqa348w/KAZU1eLv53NO5Ezuvkz31GeqHepwXHWZi0v5LZOhfzRoNyJQwmBplTWoau48EviFXk6rK6jh8JgOJwuu23uGbQy+8aVAl1HKyjEMRycHpFknfqajl5QkI3KmK4WXAjk950SmpPqwOdzRUkvQMC15+ZsblNw8kc4J87oS8nlNxUib5m+F6nuWQxzBEj2mEBNWQeWv2nShUMeMdk2diLuElrgJPedrFTvi+cvlAjOZCMIt3+lRKZSaKHwBeATx4ezr4e7q6mZxjbTTr4Tw3AskrrhpAyfsh/cuX/nJyayCz3GzgZTojM8X1CeYXeGbZRmMpdfSEwxQwtQUmInE2ghEL7AORHFlXI2oz8jw4f7r1IKmcmg+fwIw3eBnOvT2zbdmOZ+kTYyY0Mm7YS86YazU45uIFwrqWLaPWeJiy6s6lwjK9IokduxCKaJWxcD1zo5vGQ9MplwslYZzg4pZ91yqVDGmfeAcDminUqiR9wdQ1Ven15MXZsXgOykkTed/PwIJwLHDYFDN3JiZXaCVrkIo9N4pnc8keXCrhxxUWoCYUDQZ5BKufn6ubjGwYngpcq2UgmMcASE7xw03tug7ywKEsLJipVKZsXGi7lP84MnskcsC2nZgLMfgNB0J12driM0A3TN0TuFyE7ap+KIf8cTWW6TDIHmF7RWF3DVwjKK0fj2a51MTFiXpOYpLBuZSqGHnMj0M4bC2aH0HI0Elck4KcEvwuzA+ZKjmOlM3nEFYFtgK5TpRSAJl8g0NE13EmdqWja063h4RxGZmvZLKHdDBwMaKwu4Z3U965qLGRuP8uTuIZIpJ4r8Yp91jwdlmkgjg/A7YUVn9gzu7qXur7PpB080l6kUmqaBz5+N5LkY9LTTg7dll8iZHd0gDJBIaeYFup/4yd5RROZBuA9v64rCQoPbljdz97JqIqR57VAHP9oRpX04iW15M9AlRGb5U61S2OkkPsPIijFnAnUGe44dv3FuBGgiCWFQfr8TC3qJUdlM0SVTjzs/HCP1iW2q7zgiyyq1AZ3ls4r46Lom5haHiUUH6E3ESNuSgaTEtjxF9vSj0C8GZB2r0nYse8HQGU0VDqdR58RSmXP1KiQSkknHMeBzt8rIW1508Xf5yVqYbwD5BSIyj2A0n8Zdq6u5b0k1B0diPNHWweKaMPNnz2JNYSkTtPHAK23IlJ11Sl78L31mCIUbceF3AnVPuwAc4484N7H0WbM+gJLYySQ6IkdocEkJDucC7ygi82TosF+nzK/44bYuXmyLoWk2daWFBESIsuICPn5VC0eGYjy/pw/tRJuRX8TI+tAETlC2ZSH8p09k3tIeTXGO9KZ8755yCS2BoST4A5dgT589Lhoiy67PzWrg0zJx5Et1QpEX0JGFl40iFs/w7c0DmLZE2BJl6LzVF2P1rEkSyRB1lSX81k0LGZiIsa9jEmHnCG1K6FV++0R2fr7IVivhiGFmBs3vP4lJOZebJGd0deP9pLNh+rmN2BDZ8u1kCiFBCwby1n29A3AKwdEXBZEp5a189ijJSSwqhALcLXNV3hIaabuTpHStPU7yMikl3sKCNCBwtmiVCHa1W4wtr6NwIkphpIDFzeX8+nXz+ZMHNzM+mkITrg9EEyhNIFxTrZPEFBwSm0bVFw1Ebt3eKYqMCoGPDFUiQ3XAojSUQRgG5+35XIeuED6EL8BFNEWdFZxwPAUsOO4154XIVN6/U2ZN7497wOvmLEdSXnIcO5uByptlhZqp5BycbIMCx7Gaq8sJgJb0jqQ4OjhJdXEhiUSSSCTEhsX13LtmFl9/djcy42St9TIhOe3UnNg2TTjOSE1HaJ5D0q1D5WXSOp4V6rzDTaNgWQjfDFbGKQGwAl2ZrI3E+eDqGlYvaqGotBifP+BMLOd1Ejl5r5xMeM8/fzEI+qeybObcEdm0SOYplmY8ruSKXAo3/YAbcWHbCGVPMSVnk9F4EofIlZpnKJ5y/ZSGTIEgmTB5vX2Y1S31jEfHCIdDFAQNPrhhEa8d7mPnoWG0vOh210sLto2yQWG6RYlczJtuOMSXl33qgsBLYWdZrs8sD8oLIXM6M6LSfHIefOyujRRXVnHSYTrFOzvt+Kkcm+G8ApL9E9ixjNs2QaimEN2vE+8ax85YhOtKMCI+4j3jKNPJnS6EINxQigDi3WMYxUFCZQVYyQzJgQnCDSXohnHsJHO+cIrr6M4xJ8vnMcLhTFmuJJ2kOdJyVlMrNz+tEtlOOcYLL479mnf2ODh2FlcCsOHN9jG6xmK0+sMkkkkikTCNVYV85NpFHO55jVQ8MyXz74xPKKWTm8S2UaSdRam6D81wsk+hace997zBmcXc7FMzVy+UwIfJZxcKPvH+G/GHIqe3hGmmS0/12JSmKpCKnhcPMPZmB5mRFMGGQlruXcXwWx2MvtGFZmgYET9zP3EVnU/sIn5wBCybQHMpC3/1WpIDk+z9v89SvLyWxZ/bSLxjnL3//hwr//Iu9LLTyZ51djg2D8vMOPHKaE9Rdh2fWfaR/eqxL093ssHKQDqJSk4iExMQn0QlY5BOoKyMk+bNDR5VXjq06WzpHMMTVweHU7x0sJu0BaNjo0gp8etw48pm1i9uRGk5UdGzuk1ZWatwk5i604gCIRVYaWQqhkpMQioBlpnVFb1+yqXyPI9QTjqIvNeUdRwqodhQHOf+u67GHwqf33acBEITtL5vNbPetwYjorP48zeiaRqjm9tZ+OsbWPGFO9EjfvpeOMTCX72eijVNFLZWsuy3byJQHGbwtSMEqwuIHR4mORR3Jmwz17fK2zz+IlGfT0hkzlocJwuQcqORHfZuI2wLYWZQqQQyGUMmJlDxSVQygcqkUKaJsF2jRP74Oka0Os/zTtZgqbBNxcv7h+gbnSCdsonFkmjCoDQS4v4NCygrDpG7WmEbOuHSMA0NJQhf/kJJjnmBApx8kZm02x9xMDMIZZNL43menjWrHypn93a8BuYaGVImv3RVCwVl5VxI0dZ9/Wg+Dc3QUAh0v05qPAEhHwVN5QRKw4SbSkkNTaIZwgnM1QS6z8CMpRnf0UvV9XMxwn5GtnU5vkJxLFFdLJtUnCRrng3KAisNmSSk4hCPodzcHjKVQJlpJ5AyL5pd4M347i/3+/Shlv1+HsdftnjXutI5lOCp/YNkJIxHJ7FtiU/TWDengjuvmIUwNNAF5VUR7r62mb+6dykfXNtAMKjPUKj7aHk/hLsAFGkiUwlnzy0z5eZ0zL3288PXlLvtUK6JHmYHkixbNNs5nn0356A+lf9xBR51vOOuEKMgmxvdKYailnJUxqL7mb0MvnGU6NZuShbVIvI3bRcwvr8PazzFwIuHseNphrccxTZthFSkh+Okoskpe19PqX+m9r4NhHhCnUwmYrmZ0Y2m8KTQnDFiukx6oe09x4cAbEvy3L4BbphfTWt1AeOTY5SXVhAJBfjANYvZ2zNES1UJty6tpyQsONI3wiO7ekgmbE712XI9ohDSdpIQ6Rk0v5O27LxBedZURX5UpgLmFEFBUdHx7z3D+qb+zB8L0+2A062BCi3oI1hTCJogUldC60fX0/3YLgZfbaNsVSMN71oIAvTiAD7bQkpJ9PAQVTe20nzPKuK94xz57hbS8QT+4hCHvrWJkmV1zL5vjRsr6dWYb9qe1i6X0M5nio0T5vhouv/fLw5+e46QFQR9GnesbebXN8wjHNKpr6/HMlNMxFKMxtJIO0M0OkoiFmPXkUG+vmOU+Bg4joIzfRluGgS/H+ELojTP2HMOX64C4fehhQuy5Xra4IdbLL74qVudTTDO0YDyxo6SYMZTaAEDw28gbRszkcEXDiA0gRlPoxk6etBAIJBSYsXSyLSNHvFhBP2gKeyURWY4gRYy8JeEULbCzlgIDeyUjb84iG3a2EkTf2EQoWlkJpMoCXpQBwlWykTzOb5N4bpflGWjAN1vOCnfbcd/aoQNZ8nKGWZU84KDHWns+AVcFM7ofEwxy880ARzPpHwyiLz8Wqbkxf39XDO7ilVNRfT29RGJhEmZGdITo8RTKfzhII0NDUwmM2yICX62Yxhhnvng9HLkq0wKZZmIUAg0nzOvnsuchXLmYF8zmXJcJdq5f+XSttn79Zcoaixn1ntXMrSlnY4ndrL8927BimXY+Y9PUzyvivmf2oBA0fGznfQ/ewAsm+KlDbTev5Zk/wRtD7xGejiGryxE6y+txxxP0vvsXmrftYi272xh/seupKCpnB3/+BQLPnkNBS0VtH33DSYOD7L0925GAnv+43nkeBp8OpFZJRjhANGdfYiwTuU1rUzuHiDRPYqmaRTPq2H2h68gWHLsDqPnEhfhTps525hjaHGNLuT9nel7/mf6sbxSPd4xkbTY0juMFtQJBgL09Q0yPDhCZWU5c2a1UFJYRCgcob6+kutbyykv9lbCnrkU7xCUcFwaWcOIcsOMzp7AFAqpbKZNVYCgfUKSTJxZpuDjwm2ybmiUzq5m9M0urITJ0BsdFNaW4SsIMry1C5GWTOzuJz0cY7J9hN5Hd9P47mXM/eTVpAcmGNvZRccP3sRfHmbRb99AUWsVfS8eQCYy2BMZZMZGDaU4+sO3yIwlHCKyFOnhGNHdfZC0GdnWRaSykOWfv5lgaZjadbNZ+EvXQNKisLGMxZ/aSP3aedjRNJVXzGb+J68h0T5KzxN7zm2fzICLgsimE4ESTmiT0oXTQg3Qnb9KB6Ur0BS4f5WmHPP79Gu9jy6QmlOTEgp/iZ/3XjOb+6+cQ8rMcLizk4xpsmBuKw21dZSVFCPc5J81ZRVUlgS5Y2ktwpdvezw9eGSUy4agUKmEQ2j5x88GnrM3TwIQgIbicMLgyJFOr/JzUJlXOiAE5SsaSY8nGN/TR+zQABVrm1GmzejrR6m+YR5GUYixHd2kByYxAj6q1s6ifEUTS37/Zorn1ZDqm6Rm7SyK5lYz5/61zLv/SoSeG556aRAj6KP7sd2enMbI9k4CpUGqNs5l+PV2pK0IlITRdB095MdXFAQ0JvcPsv/rLxPvGkYB/rIwpUsaKFlWx2T7CMo+v1rRRSEuZjeEExq2gHQ6g7SsLNEdo6zmRxZ4jlicqAcNGynyh5EjpmkaGIEQjXXF/PL1c7l6VhnxiXGSCZNwQZjSopCjXAjw+XyUFhcyMhEj7A9QXVbFumaT19tG2d+dREimtel0npW85wKZSSCEQhjBqRedCTzF3rWcZXtMwDhhfvjKIeYvmIPP9ZOdS2U/XFtMQX0pXY/sBJ9B0dxqJjtGiHWNYdo21niSodfaaXrPcuyUxfi+PgLlBbT/dDtVV83CVxFhaGcPkVkV9L18GDORpKipgqzO49douW8N+//tBayEiZ2xGH69neRYAnNbB2bvJPHOUQpnV0zrEEXhwmpm3bsao8BZqW0nM8Q6RpjYN0hkfiVCP7/GuouCyECgNEiYaRKjKSyjAOUPkWfOzL80j/V5w1VzyDSTQkxMItOT7gU6jp9OgeajvsbP5+5fQ12hYP++durqSliwaA7dPT1kLMHYWJRwOOzI68VFTEzGySiLgsIQJdEQdy2tpW34KFZCoZ3FupApt0lQqRSENDB8eefPgICzu58cu9JZCHi412Dt81u49V1Xo7uZrs4GwisYx+9Vvq6ZzgffonLDHHyRAD07eyiYX0nTPStIj8bo+tE2jAI/VRvncuSbm5FAZHY5ZQvqCBSGOPyNTez44hOIgMac+9djJU20kIEwNPSwj0hzGfXvXkL3IztJDk9iRpO0fmQtvtIwnT/ezsiuLgpnVaCFdITPcbloQYPJvYPs/erzVFw7GyNsMPRiG0MvtRGqL6bp9iVn1Qen1E8X0rro8QJbQBrJeN8EqnYeKhBx9JT8mfZYi3DuoCe1KIWWTqDad6MSYwgCKGG7HEqjNKTxriuaOTAap76qiA+un8V9d1zP8NgY7Z19VBQXU1ZWTFlZKVJKxqITDI2O4Q8GSKdSdHb18rVXj/DG4Tiauzwml0T1DHvAexZNQ4QKULozOM4mM7AejiCMQG5+ErnXWE2MP99Qzo0b1+MLhqYWcNwqRZZL5vus8ic/Bdhpk8xECiMSwAj5yESTCF3DVxhE2ZLMWAJfYQBhCFIjCZRpE6woRAvooARmLEV6PE6gKIyvKIRMmZjJDEbYhxU3CZSFUJYkE02hB31YKZNgaQQ0yEymQCn8RSEy0SS630AP+zGjKey06faLH2laYCkQCl9REM1n5DZDmeG5TnQsn3TECcSCC27CV4BtCMYHR0kHq5HlDVMvmOkBTwiBNt6HbN/tyEnCRigDheXQo9Lwhf00zKrh5vmF/O6Hb6C5vpa9+9tQCAojQRrqa/EZPixb0tXdiw0Yfh/9fb3s6x3jH548SDLm5eA7mxXF+U5TwPChhSIgzkxV9vRFLRhC9yQB77hwJiEpoFileF+T5P0b5tHU3Ig/HEHL23LoeDjZU3r3nq3wdWaC+IWrSwGablycJnzvpUjASmRQZYUIqTmGjdNJjjllSgEVLETomuMPAZRwslIJZQASaZokkpJd/TYvbt7O/fdU09BQx979+wn4dcZGo1RXV6JrGgF/gGgihu4zqKqoIh5LcOOCMh7dPoowc+vXzgxTiUDZJiqTRviDZ2fO9xQyl8vm62YaMCFCfLPT5pFvH2BR4S4WleoU+XUco9MxhTmMUNcQhnFuDDTvEHiZrJSAX/ulO4973QXWyRxCUjgDTGTiqEAx2f2+Z+RiaobjuTcvFIh0EiltV1fzTI6gsNxjIG1F93ia147EWHbgMCuWLKCirJTJeAKf4acklQKhkU6nSCVTSEAX0N3dy7IyeKNIMDCqoUkvsuDsoQHKTCN8fnex6JnNs9klQzOkIlfuyjsDSV3AZn5ViNlVhRT7dbQZqnKFYkco0PW3jcNcCsi+9ZN0ygU3fDj7QWvofh052AuhEkQghLLdaXfKbuzCtdPnD2rPCuIIycJKwXBPTleS3i4j3iyt0DQd3RBIG7b3xnlpx0Hmz2qisaGB7Tt30zc0SDKdYHZzM+OTUQaHR4knE+zctYtnnv05N27YwK1LlvDN13pQqbNPQCPyvynlxIMGQie448SlCelaGMWxLdMUNOiT/NrqIm6+ZgPF5RXZqIcTib7nShR8J+FU++SCEpkbA4Fu24TKSjC7BxBtu6CsGnxBjl0D41nPyHsy23GeAZgpVHQY4uNOxLyS3sYluRKEIFJUhGH4UEIwGDV5vS3Osu072Xj1ehob6tm1bz/9/f3EJ2L0DA2ya+9etmx5g57ObpKJBL19vdyx/ipePTLEoR4TTapcrv2z7hPl7sXsbDp+OuV5C2KVa12caidyxL6FgSh/f/cCFixfgtB1prZYzMD9smcuYzpU9p8T4oKLiwLQJITCIcyKYpJDo9A36S5fcHSebECUR1x5U4jjYfMeQyKQTroA6XMIzR0dmgBN1wgXFRApLsztLaxgb2+CV/Z2smT+HKorK9i1V9E/OMSe3XvZs38fHZ2d2EjKqytJJJO88eZbNM1r4c758/m3kW7sBIgpDTsbCCejr50B7TRzYWQZusrlTck6zBSVWoy/vG0OC1cuc1MNHFPzOWj/ZUzHBSKynC7mQKDZUFRWhi8SITUaxcpk0DMxdE1QWBBBad4+1DkvUjZ6eoqRRIEUSM1HKj5BypIEAgUYAR/+cBBfMJj1JwlACcFEyubNDptXtuzi6jXLicbjbNmxnbaDhxBS0dzYjOY36Onrxu9XGJXl/PTJF9j47girmiNsORDPS51wdgPV6xllmmiG/zQNIDlTuxOnqJFdvasUH5kXYNmapdkyL8XNHS9FnDWRHSuXejrUdEuVcm0WruVQSpAKlO0s51cSTSpCmiBQGkEGiyk0TWqrSvngB+/NJmHy7CEoUMJbXCFzOhvSWcCnaUwcOsiPn36D/ngRmtARwjF6CPdeILtr5aGBOK8cHOaFF/+FXbu3MzAwQEVFJS2ts9CA+GSMkD+E1A3a6uoYvO5K+sai/MaSOvZ0HSAR085tjJptOX2kazP08YnhJPmyAR/C7aNKkeCOK5egGwGmhIO8XbAy0NMLuh9qa8BMQv8g+EJQWw3pOPQPOdcGIlBb6axh7BuAcDFUlgIKBvshaUJdPfh0SMZgYBiKy6C0CKQFvX0gBdTVgaFfcFn3rInME5KcwNncAguUAttJ2aZsmc0+JaTK5vgAmO6nEwi0jAQtQEtDNf6ATTigZs505kXW5xlDBDoK8OmKe++4gfHRON97+hDScre8EcLZhUPXUIaz8ZvQBCkTtvWkaDF16moaGB0aImNLxsfHwAJTCHoLInStWEly3RWIeQsYfvN1tiUmuLK1hGf3xFDW1C2IzuTdTrnTtkEz8qyXJykx/7Rt5f1QzCuQ1DXUZfvtbUUmDl/4PPz4FShvge98A/719+HpNyBmw789AGV9cN9vQlkJrLoV/v1P4P98Gl7YA74C+I8HQB2EX/sTSKXg9k/An38CPvdJePMQlNTDNx+Al74OX3rQ6bsP/zb8xWcds/AFxImJbIrukzsgppx3uZJtgy2ReVmohFLHzMLHOM5nTF8mkDo0VFcwkRgBoRDMkKDGi0JwzIcu3bpkLiVC05jbXIXPaCNteUYB5cx2FpB28oygC9B1unszLFm1iBXl8zh4YC9WxsZMWIwWF9DROovJK67EKow4aeDGoohFC3j2hRf4gwXNvNW+n7EJHV2deZT+9IdTtpnNPnW6w0TaNpqSKKGDgroiA1/Q2zz97Rx0Cva8DD/bDf/1gMOh6srgs38Bf+iHu2+AlzbDnQ1Q1gSf+jBcfyt074Kft8GjT8GXPg8/ehxGnoH3fA5ub4SP/DGsKIZ9cXjyGfitX4JHnoOtz8L9vwOhQ/D6/qmqxAXCCYlsSvM87iMdziSl5XAq6X6ydglXCxD5BDXV43O8edmhEYcfGsokHAySsULZ1AHZFomc7pFDzrHrLDMXxE2TkkiQoF8jlcpxmRy/dVMF2Apsm0zG5I19/TReUc3iRUs4MjLA7uUrGL/jdjLFhajBUacMIVDjIzC7mXjTPJ4d7ufWpdU8+Nog0hJ52ubZvFyBknY2ffbpwskO5mRPFig3fEib9lLfJhw+CL0d8LlPgAzBt34EK+bAlidg3AfvvhWqYnDFfHjkO/DAj+Ab/wj+Sfj1z0D/QVjUCAOjcPNcmNMIWhz2HYLiemish+Y6GBuDD38Mfv0vwF8A//z1C87F4GThCnYGlUkhE3FULIacnETGYqhEAtJpJytT1vHpSiHZfB75/x1jKJ6ZwNwRYAU1mkrCJJNxZrU0ownPoexyM89Wnp9HJK9sBNgoRpJpin0+5rWUoYl8s7Zzz9Q2OPpc/0iS1w8NsWVshO23vIuhz/0GxUvmUVJZBSE/4HBtZWVgYAwWzmZL2sTfWE5LlZ5rwFlCAMJL8HpGUEjLdAheCCajKWzTPPlt5wMBP9TNg4cehUYfbN4Bbz4Dv/tX8IUvwzWLYPZy+J9vwFf+AaJd4K+B7z0In/0kLG2F1lkQCcBkDKJjoEJQVQ6pGKQzMDYOAQ2+/l344tfgDz4E/+//Qcw6afPON07IyexYDMgfji6vEp7x/dxNille5dMprtBZ2FSCpdm0zmvN5b0/hlQ9TBXQJKAj6J+Is7S1hRVzjzA0nOBoXzynC87Yfg2lFEcGTRbe+h7awxpqdBQjWMvCggCbyquxuruRehqUhoyOo5UVkZk3m4cPHuU9y+rpGGlHpZ3dL8+OabiJiZSNs5rg9CDAWYGtgigh2BuD6OgY5XVhx7wPb59utnQtmP8Mv/l56E7A/Er47V+BdhN+8t+wZz+s1OC7L8BQJ8y/FkoU/M1XoHcY2sfhj94DNQPwlS/AD0Kw5ga4/Vb4xrfgA/fBkX74tath03fhf78K9ghUrQbfhedkJySyfCHN+X2+GuyuJzN0Csr9rJldRLDAx6q169EN3b0ib8DO2AxPUHT8ajoaI2mTjlic+2+9Hr/2Oj/4+UGGRtM55+yMBQlGJ1M0dqSY3eLnYHsbA8XFNPmLaK4qpS0aRUyaKCVAWajBEURLM4eOHGWgUGdpnZ/tnTIbpX/mcJ0U0ib3Jk4dChyOK200XdCbCbF52wHuqK0BtKzR6PxDQMsKePA7sOMQLF7tiH5/+EWIppxLKlpg3VwQVUAArr8RioPwgQ9AdxTWXg1zGmD+n8K8KyEm4KYboDQCP/ieU+6iVbB8Hvzv9+Dl18EIw3U3QOCCBzWdOAq//s4vnhsd/iRQKNAEwRIf6+aXUdVQxbJli9Cn7X9wfL+OS2DZTeYEuhJkBOjKoi7sZ25xMd95ZDMPPruL0Un7pHpTKKizauNsHg3FMa9YTXHTbNaHg/x8aAjz6BGUpTlhS0oimuogNk7d61v5tbIGvv7kAdIJDU1xVtxCoRA+PyLo5aA4jbJcnVXzBxHhMELB8tAE//GJK6lqbMiVJXIukdOt4hcd2dQscEKn4wVNP+DZLpSm4SvQWT27lMqmclYsm48yjtX4TzQhONCccEW3bF0pEAY9iQx7olE+ds81fPqedZQXBrLi4kw5OxSQTFoM7x9ihfKhDh8hOjlBv2mxoKQUSqpB5HaUUUPDUNVIf0U1b8k4G+eUgH5uujbr7jiT6U6AtDJOanQh2JEs4Es/3kx0eCj31Nn8g+5HXf6c6if7jk7yGi4okQmFE+8b0VnSUkT17AqWL10EhoYu3RCpUy/NMRYonHVTKCcGXzmbBI3ETfZGx7n/jvX8xn3riIT0vM45tpuUgLaucRon/ZQMjaC6u9mXSVBtGJRUl0PAj8AGZaDSadTwOGrJIp6JDlE3v5KKAonUxAwkfFo9lNe002QxWbeGQqbTbsIenZ/0R/iLbz5H+74D2FbGffq3RWD5hcUFExc9vcgqCtBaq7FqXhMrr1iM3zBwVH3XGTBlbJ1ooE2dXmzhEJtAYAnwKQ1TWCwqLWZBaTFf/J+n+PEzu7CkNsWN7pTkGveFpKmqkIqrangy5EOtW0lrdQ21PoPNXT2Y3d3OdkqA0n3os+pR+9pY1d/BNZMGP3h9CM30/HvKfYJTJxaFAqGhRYpQ4vSChaf0ixDokUJnJxoESkmatBh3zg5w3aIaauqrCfkDiIuF2I4reV2Msqyje5fX1hy3cRd20abPIFRqsKChnHmLZuMzjKwny11uOd04f2LkhdxrbjmOG1ughAVK0D4Spy4U5GN3rGP7/l72d4zO3DYAJegenKShv4za6jS9nV0cLSqh3ldAXVkJnePjkIij0ME2kSMT6HNns7PnCNfUlTG/fIxDgzIvP/2ZDBJn88OzGl9KIVNJ9HBB1hndpQr56mHFNw/1UqR1Eha25+E8i4rOHsKdWIRhODlPdN1dsqO8s+QmrIsBTiue+ePbj3vFBSEyp7vALvCxsNJPbW0lJSVFruEi3/h9Ot3oDpD8bZjcTQgcxV7gwyAjBHuGxriqqYYb17ZypG8MM6OmlSSy36SC/XsGuKp+Dj9p70JW13Ag6GNRJEJvZRVm51EnBlNoqLFxZHEhcv4yntq7jXtW1NL2fBd2WnMS72QbdhpPpbKPcYZwF8VaFiKTRvMHc0tyhCAhgiQcxTjH0VX21qlQp3HsLO8XaSCjULqO0g00Q0foBmQ3YLw4+K6TSPbEL/Vt18m8PpW6IFLgo6bAT8PsJqQ7CL32imzq5PzPyTDztZpy1nt5hDdo2gwkU6xdVE9FaTBr/J+xRCUYm0wRb4s6G5Z2HmUwlmFCShorStAKSt1lbwKwUUPDiKYmDoWD9BQYLK8NgO7sOnK6kRvZ2BF1pg5pD04aWDudznMJ5HRYb4LSpobp5FlDyB0/1WNncf8Ub6hto2XSqEQCOTmJiseQySTSTIE03b45lYq8tzzz8TP9OEHqJyb3t53IvG1plU+nqlCjtLKcAn/AEQ+dlMG52fRsP+Du8OkMcBvpmNyVTvvYJHObaqirCGR3AZ2RXwgnROvg3j6WmiHCPT2ovn7a0mnm+gMEqqvBpwFOzkYVm3B2clmykp8NDbJiRR2BQMYd0Gcy9wpQ8iw9bi5RKRs7GXciSbyy8ycy9+eF/nhNy580vcghZVmQSUMiiYrFULE4Kp5wgoa9HYamZVHOTbt5UUjuWPPG25n8l9fQE+ICWBdduvcJigMalbW12JpwsiVpzkboUtOQQpzlR0PifJw9nnUQmps+RDKcTmMEA9RXFGAYXsuOFUG8wRhL2wztGWaNFkK17yc6kaAnYzK/pABKikFpLkeTMDQC5RWM1VWyzY5zzZwypJGbSzktYlPuUqCz7XdXw7Vs7FTC3crJexsn3y3yYsEUd5TCWdJjmah0CpVIOkQ3GUNNxiAeg2Tc2VUnk0ZYGYRlOZvYK9cFk3WRnO5HIqRyInLkiUO3LozhQwB+QUHIT11pEREhMIXMzV7ZfnRH1pQdz2c4dqLjWeQpCkqhK4mUijkNVfh8vThhfTON5Fx5BzuG2DivgnJznJHeTg4V+FlfUEhxZRUT0TjSMkEEUJk4cnwcfcFSXnnxGT43r5GqjnGGJnzZ3PenNKhdeV8p6XLbc0MKysxgawI9EL5YrAenianmMDXlmzMhCSmdzBTZC6dNoNmgc5HdYQdEdr+5fFOVY09TWT+tM5nm9uM7mf/27Scy4cybuqFRXhhkbkUJPmHgLb88H8E+KvsanLKVAInCJwSzGmvx+bZzvNk8R+8C04JDO7pZe10jT7UfJlNRT4c/yNziQrZWFUFvKuv7UyMjyNmzmWxZwDM9HWxYUMGPto6d0c4wwvMWn2XXZCUwQKXTjqskEJ6Rg18qmNpukXdcnOjCPGliqmRxPLXymKLy5uyTdd4F4mQCTSjKQgECIiezKk9fO9fVZb+5hOxu0WtKm4riCCG/xgT5cvwMo1k493b0xWjpSdJSnuZoxyG6i1Ywt8RPaWUVIyNxZxMJGUDYJmp4EDG3ia09R7myvoKWrhRHB5IYljg9gpGn45Q/AbKP5xJtKk2ENOURg7AnSeR3vpjWI+o4x8gJEPnH8suYcnyGYzMJK8ct9wRlnlK7pt0/k+F3qvNIMcNjzXjtTLgwRKbh7K7iN9DwMiYpxOkMvDNEfhW2lISDAUIBA3D8aMfPoei0USnFjm3dXH1bK91d7ZjV1WwPNbE8GOal2gqszk6QGWdF83gCVVSIvXgZj2zfyl3Laul8sdNNuppjTcd9bOGJI9J50VN2jzx9KJGbZOYGYrxveTnXLG+lprYCn8/LjT/t+b2qZhphJzt2Jve/nXXNdP+pzPDT7j/ZLadFZPnblU5ts5h23YlVJIXAzupfAkd4djeNEOdHZASyfjiv1ZYCQwgMN/wp5/Kc8e7smbFJk+jBSZbPC/Hm0aOMVlaQ8PlpKS3h8Og4Kjru3mLB4DCiqZ72tlIGVIaF9QF2t6fQrdPg10ohlESeZoo4r93O+3AsYn4s7m2R/Ma9N1LT0oSmG3kT/JnKENMpZCaKmT5yTnTPqR7L/30CKeSkOBvZ6eS9doacbOqDTX/UrOyvVNZk7yiKbnMkSKVh25Yzu7rXnO/sSbkV1k59SrpDUDkNPmlX5xkt9uzr4+rWRewfHCLW08/eoI8rQmG6K6tIxhNg2SAkKpFGRuNoSxby/Kub+eUFVbQNdpOZyKnrpwJlSyc/iTh9S6CXLEhD8pE58HufuJNISXnufN630ynb6y/btGl/+C2K5lRTsbKZifZhBl4+QMvdq0GDrsd2UdBQSvVVrUhNEG8fYeDFA1gZi4rVzZQvb8RKmfS/eIBYxyjhhmLqrl9AvG+c0R1dlC1vpP/VwzTfsRw96KPj4a3U37yYcE0hva8eJtkzRuOdy7CTFh0Pb0PYzkjzlxfgKwkQbxtBKUHJ8lpSfROkh2JoAYOy5Q2ULq1HN84sM/KpkvVpmfBzcq1nZVGuFcdCmRmUmYR0ApWMIROTzicexY5FseNRVCzq+pEmkZkM0t187UyDzE8X3o73HqR0fkt5ijKCZxVUgkRa0bu9mysDBYij3UxOJOkzLeaVFaCVliCkK+tpCkbGUCUlDNVW86YZZ1lDGOnlxzmFqoVSIK0z4jSKnDRxRVGaz33oRsLFZXiT2vTP6Tm0nBZpBpjRDH1P70NZFkOvHiFxdBw9qDO+t4/eJ/bQ+fAOzESazEicvf/2HPG+CaSp2P9fLzOyp4cjP3mL3if3oYV99D13mEM/eINkT5ThNzpJDMYYeuoQ7T96EyuWZujVdjLRJJlYhp6fbKfv8X2M7+1DDxgUzq4kun8QmTApaColum+ARO8ExXMqCZRGGN7aiTmeRGVsDnztFYZeP5rtp9N36J1E3HdxEiJzLS9KOsslLBNlplDJOCoxgYznfZIxZDKBTKeQZsZZnm95eUAcKvKsWE6RCls5G517HGw6EUwdKblPtozjfNzCsh/l/vW6RLmcS0qJkhI7a689xUEsHA64/+gELWOCyvFBVNcAR5Im1YZBcU0FIuhD2CBsHWWlYXQCMWcuw/YYQZlEID1f6ClB2fYZmYS8SJegyvArG1ooranJi6Q5W3grrDUq17YQ6xgj3h1lbGcP5WubEbrO0OYjlF/RgjRtooeGmGwbQiYt5n/qGuZ/+loW/+5N+IuCjG/tofHupcz76JUs/aObqb1uLohcalsKdCb2DDK8tdOZuICJgwNIpShd08TQ5nZ8YT+1183FVxqiYE4lVVe0IHQNO54m3j2GpjtJhYqW1DLvk1dTsbqRoU1HnKRP51GKOiGRyWQcmZjEdglJJSYdQsqkUabpZKiSkplp2s276BG9B+VksSoP+qgsCODTFIYAXThLW5zlI7mPkhKlJAobhY3ERinnu1LymA/e/UiUsFHC8TFpwsmp4tMVAUMQ9umEfBpFkRB1lcVZbex0utqyJdu3dXJNIIzoOko6OsLBjM3iSAStqtRxgmvKcVpGo6waH+UeK8zOzjjCmp7T/8RQUjp642nHZjl1NAYtrli5YMbMwWcM7/0iKG6twhf20/fz/ZiTCcqXNZAemWRi7yC+wiC+UICh19qcZ1ESadkoy8YaS2JnbDDATltIaWNPZrCjqbw5T+ErCFF/1xJ6Ht8FSQslFUObjuALBwgUBpnY109qNO42yHmXyjUjGgUBCpvL8EV82f5TlkSmbYTv/MdjnFgny6SzdJOVP09A8Sd6/VO0OCnZuHwRH7l5AwWhALpmoDTwdolVrv7mcaWsaV/NYErNBTtmid0RfZxZMH82FMLhoEo6L0G6YVYbr1jC67sHsU9JbPSexzGWdA/FWdhv0liSoau9k96iUlqLItRX1tA1HkPFYhQjuL8wQtNbb7H7QAemXZY1tJzA0pKFsw2adKIbtNMbFJ50v7BUp7isbGqfnSWyU5IAI+KnZEU9/U/up2hZDYGqQvqeO4DwgdAUwcZixvf0UX/LYvzlYfZ++Xn0cIBU9ygLfusGqtfPpvvRXYzv7yN2aJiKDa0U1pTgOX0lULluNuMH+xnd0kl6NM74/n6KF9WCu1/GyI4u6jcuOGbuMhMmk/3jqCI/CMXwa+1Ed/UQ7xhl7q9ey4zb2ZxDnDglXH5E+9kYb2aATwhCfh8aEl0oAv7AcSo4E20tr4z8TMEoksmEG7unO5soCQj4NRAaAtsb+qdcjVKwdUcvN986h//t6cCqrWV7sJFan4W/uopri0K8L5VmqRXnYG09b745QdLSEN767ayV6GTPo5zEOAHjFK4/ti8KCyNoxvn12NRtnEcgEqRoYQ2aLvCXhmn9+JVUrm7BjKfpe/EgRoGfxZ+/ieE32pFpm5IPrKawtZKiWZWE6ktIdo1Tsb6VqtVNJPqj1N66iMKmchpvX4xR4Kf1g1dQ1FKJvzxC/R1LqLt+PnrQR9GCatcwBDUb5xGqKQGhqFjXQkFtCQqFz29Qd/MCzNEUwq/T8v41FM6qOOlznS1Omkhn5h9nB0VWTZuik808w55txbnZ1nE1aeTWnbkbWiiPFs+sruHxNPFDoyxuDrCz/TBJw0dhQOfLoSKuEIqCUIjH45JvZoL0m22EpOnK6adJLJaF8Ht5TE6nhQIrk3bTfp/OfaeHcF0JqdY4saNDxHvGKWwpp6ClzJEm/BrhykKsaJrCuZUULaohdnSEcGMpyf4Jxvf2ULW+FW1lIwObj9D7/AGEz6B0YY3DvaWi99l9GIUBSpbUMHFgECE0Rt7qoGL9LEI1xYzt6qH/hYOUr2rCiATof+kQdjyDvzxEyaI6pGmR6p5AD/nwlQQpbC4/7xZtuICxi9JWCOmIEm+LadGFppw1YjkuQjarE5DTeU7B6ZuV71G8sXuQNYW1lMT7WTlhc2V5iIbySjqKavn2uOBpzSBR5idcVY7qGHAdzNmSTqHlCqSNsu3Tyh/iRf73jpukUymCvkAuBu+cDzCNwZePEN0/QKCmgI4fvUXz+1dTd/18ooeG2PeVFwg3lbPiT29BExrtD76JLxxgZGsn6YkENdfOw06ZdP5wK6GGEqxEhoHn9lF7y2I6f/gW5VfOJlgRwRpL0v34XsrWNWCXRogeGODgf75MoLLQ4ZgvH2Lhp6+l4yfb0Qr8aFLR9/P91Fw/n/Yfb6Vy/WwCqgBHnjn/uDBE5ubgQJDdwfJtq1pId7MJ1/EtAdvGTkyA7nO3kp0hJfhMyJOj4ynJG5u68Qt4Ru+lZ2EL/fFudhkHiS5fDE2VqOYyUosaCQ6O4YunPOZ6SiSWJWczjdCN0xBrHRHqQFTR1d7N3CXF51Qqmd5GgaJsRQNzPrqOjoe20/fMPqqvbGXotaMUzq8m2R9l8ugIRfOrqVw3m/YfbsOKp5j/2esQPh1SFkIIQg0l2AmT1GgcoWvgNwg1llDQUEqqaxw97KNoTjWFsyvpfGQHkZYyFn1uI5nxJNv/6meM7u0Hoai6qhVdE3Q/vguZsdB9BsGmYgpbKtBOU789U1yQRDrCU2bOdln9mWAK18xt3CeURKaT2IlJsDMc4zc4LrzVRYpkGqIpwUhc8PIbXRzY34/W1onvwCGYSEJSIRfMIllVhtS0U2bgWc1YgbJMd2nF6UgAiv+/vTMPs6O6Dvzv3qp6S++t3tTautVSC+0rMpvZZIwRZjMeGwYDsSGGL8ZrZrJ8mWRmPDPYDniI48QBxxMCDrZjsImNHRsjm8UIIRaBBEhCe2tfWq1e31517/xxq+rVa7Va3Y02Ozrf97rfq+UuVffcs59zWCd58oV1uIVcOJ9AwXTCWQkBwhY41Ql0XpPvzdDz1l5qzmkiXpPk0CsdCCGYeOVs3L4MFdMbqZ05PtQ/aCB3KEWhN2PU7p6GgiK9p4dsl9EguqkcR97ZS7ZrAJV1scvjSMfCTsaQlkTlTenifU9vYNeP1zL5I4twKpPogiKzu4d8d/qUrb3TgGRmZspT4as9hdxixMhmxhLVWAotEJ6Ll06hC9lRDcxwoIGs53tYaIWTzlK+aSdyy3Y40ouqq6AwYyJeMkZo4xgp+LkuVD438pQEvmJFCXh8Q5bXXnoNpSLRxFHb4nsG01B6Zzcd//YWe37xDjULJ9C/9RBuKk/v5oOgoGfdbgr9WZKNlSSaqqhua0A6ER9WDWXN1ZRPqCG1/RCFVBYRE9TOHE9iXDkoRaKpkll3Xkz9vEnUn9fKkXV72PbDN9jy3ZfRaGram9Bomi+fgVXhYNkSIQQyIamZ1UysMuE7IZz81XfaEul4nlec3qlkF31PhzAaKCJ76UDTpxUqk0bEFTJWNsjQNzSIUBkRtRgapi7WP0D5uq30149DV1agJzSjnY0g8ozWi0NrTA0CzzU5L0bw8IIRHdYJ/uqnm/iqbbP4gsVI2wnHeEJAQHJaHYWsS2r3ESZeNYfxl7azf+VmJl43hynL55PvzbDtidfJ9qaoKI9TNbuJ5IQqc7sQSMeiauEE0gd6EJag9ealJJurqJ7eSOcrHTj1ZdTObKZydhPCEiAFje+bivY8ul7biZV0mPm5yyibVE31vAlUndNEoq6cvj1HaJg3ifK2eg6t3kFyQhXVMxqNu8pJhmFTwjVf8scnHs21ye9x9fKl3HfXcpIxiW3HiPslgk42ZLNZ4+mBBiHRWvHEr97kL//uJ6GbV3R3E04cmSiPrOWxLUhlSdJTJxJf2MYsx2PTbzfjDOQjyDlCMCTXJJdJVkQCQI/fSLCxNMs0nz53HNcuW0hd8wQsJzai+0cyOBNobCIqhPQ9fJTvGSKFKYSqiy5dWilDoSMbmVJeMUFNcLwk5svPQ+bLVNo/H/QjpPD3Sl16vxAQ2EKFr/gZkQnl+CCG0SKdpmxVOozaCLI4aa1Zu3YtP/rRj3Ach09/+tM89dRTbNmyhUsvvZS2tjZWrlzJnXfeyYoVK8hkMnR2dnLJJZfw2GOPAbBgwQJuuOEGEokEAD09PaxYsYJzzjmHWbNm8fTTT7Njxw6uvfYampoaTdlbETh7mXjYIHQ0cBnShZxR9Jcg2qgmC4D0PMq278Has4fWxdPoqS/jcLqAjISKjap5z0PnsxBPjlh7EnhQ7FdlfOWVAb7/5m84b2KCaU0VxBKnNu+iRvsFGE3Kt5LxvwdzyimF0JdAc9uNHzzmZacFyQQCz5fJgue7e/duvvSlL3HzzTcza9YsHnnkEd555x1uuukm7rvvPu655x4effRRLrnkEr797W9zww038OSTT9LU1MSzzz7LX/zFX/Dggw+SSqW44447EEKwevVqHnjgAW699VYymQzf/va3mT17Fvfddz/33//XSFlE8NLxRf8ajZ6SwmcdRz3ZEKRSqCxs2HaIeQum8HxXCj1QGB27JoInqNGFrAlVsUdWKFBHNlsPi8N5yZbDWfKexknmGbHv5nuBojjs20ilYXutYqq3MRVkO41wvEzXp0mF77s3hUoHzbvvvktlZSW33347juPwta99jZtvvplrrrmGJ554gq6uLlpaWvjFL35BNpvlfe97H9///vcBqKysZNmyZWzZsoUXXniBlpYWEokEy5Yt47nnngOMDJjJpOnt7WPr1i24rkss8DIZ5p0GrInO59DSAicRylujwbhogMrB/gItXSlaJ1SxfdsRrDFkfAtMfV4ugyUlCDuysw57F2U6z/UtcMuy+UyfMY1EeXnIep0KKHGLO2W9nhwYybZ02hQf0SRBQkNTUxP79+9n1apVtLW1MW3aNNasWcPixYvZu3cvM2fOxHEc/vZv/5arrrqKcePGhXx8oVBg69atrF69mosuuoienh4SiQRSyvCaJUuWcNddd/PII4+wYMFCEwk8AhNCUX2uUbkMluUYZGN0CyS4VmuNUor1u7u5YO4EdlYmoCczYrnqqJEpD51NI5LlgDWsDV0AdSLHXy5r4uqrLiZWHmQUlqckKv33DczyOX5832lCMo0XVJD0Rzh79mzuueceHnroIZqamrj77rt56KGHuPfee7nuuuu4+OKLaW1t5dlnn+WjH/0oyWSShQsX0tjYSHV1NQ888ABLly7ljjvuoLq62vSiNe3t7TQ0NJDNZnnlldXMmzePz3zmM0bw9hF8pMZgrYwtTSbKQ9ZrdPQMXyBXDORh975u5kyq4q1UDqugo5eMtCnzij0Xnc0gEuVH5asITBMAZeT5b8uauO76KxCOE5U8RzuLszAKOOXaRY2RDS77wAL+9p7rqEo62NLGVR47duxg0qRJaK3Zu3cvkydPpqOjg1gsRnt7O0IIDh06xPbt2xFCMHv2bLLZLOvXryeRSDBt2jT6+vqoq6tj3759TJ8+nb6+PlKpFPF4nI6ODjSayZMmU1dXi9AOHi4/+tUb/OXfP4Xyju0gHFrTBFjJylAOglEuT63BsZEVFcSk5MI543n1UC+pg5mwndEqG0Om1okj48liGu7wvEBozcenFPhfX/goibLy0l58+egsjA6C6BAjSp5h2kVDyUqFxZ6eHu644w6+/vWvk8lkuPfee/nOd77D3Xffjeu6PPXUU2SzWT73uc9hWRbjx4/npptu4t5776W6uprJkyezZMkSHn/8cW699Va++MUvct999+E4DqtWraK5uZknnniC9vZ2br/9di5fdlmJE/zIlpjPNuZzWJaNFiP32igB3zs6p2Dbrk5mttTzWm8Oyy8ePxrLVUiNtEAUTBVREU8YR2iKuowqkeOWy+cQL6sgRL1TySNqDal+8IDKShNeks9BKg3lFRBzjP2vfwDsGFSUmfsG+sFV5h7bMhmC+/shloRk3Myltw+QUFV50sNWxgKjlnYHOxsN5Xw09LlSFx6tIkIZUN9QT3t7OytWrOCZZ55h6dKl7Nmzh+rqahobG3n99dd5/PHHsW2bRx99lG984xusWLGCyZMn893vfpf777+fiooKstksrutSV1fHww8/zP79+ykUChQKBRobGzn//PNpbW0tGa8YNP6hIFDsC9+1SSt3bEyWIPQ4EWj29hUQ6RyNTRUof4GMXskX5GQHXciichk/HblpTAvN9ApF+zltobnilMORPfCRK+D626E/b2LjHvyfsGgJ/OoNSHXDn38aLrwQbvsCpAvw9GOw7FK4eBm8tNE8txefhHMXwX2PGvey//dVuPj9cOUN0HH41M9rBDBqSlbk4HXJr2Mt0WOxXga/igH1tm2zfPlyHn74YbTWfPnLX+a5556jr68Py7L41a9+RTKZZNy4cZSXlyOEoLu7m/r6esrKyo7alWfOnMm4ceN48sknmTNnDgADAwN0Hu4kl8uVTmakLkrhPRpVKCAsZ5hZDg1m3WsTdiItlBZs3dXNObMm0NmVgYG8nyxn7EY5Ucgbw2wigRbGXamtxiFZXhmO9ZSyh9qDR78FThN095qXv20NPLkaGssgm4fnnoCXDsKj34Pqaih0wv/5Otz1V3DhHJg4GQY64Rv/BG2ToW8Adq6Dv3scvvr3MHU8NFedujmNAo6f4yNwbtO6GOKvPJNP3Cug3ALazaPdgp+WIA+FPDqfMznKXT83iOfnIPc847bkKgJDmRYm18cFF1zArl27yGazTJkyhRdffJHly5dz5ZVXsnr1ahYtWsTLL7/M/fffz7/+679y/vnn8/TTT/M3f/M3PPHEE2Sz2XDktm1z1113sXPnTjzPKFmqq6uZNHEiqVTKLDENjDocPyh8UEBqk1l/tKDRZhf282/05QV9hweY2lyFsooKlZGivhiEMhqN9vKoTArhuWg0sfIkwi9yLwYbf08maA2bX4NfrIV7bjchF/k0/M3X4aY/gonVgII334TOHfCHt8Gf/A/YtAl27IPv/DXcfAusXAc/eBAmXADvN5sm2zfCoUPwP78En7obNu0fxW556mBYSqZyGfCC/BkBskFJercICH00K3JUynoE2pJ42azfZvGCSZMm8Wd/9mfU1tZSUVHB1VdfzW233YZSirKyMi655BK+/vWv8/zzz5PL5bj77ruprKxk1apVuK7L5Zdfzs0338ycOXOQUrJgwQK+9rWvEY/Hqa2tJZfLsWvXbiZPnoIMwtr10Yv0uCAApUxym1GmEwtj0DyFiBFS0Y79fSye1cTeqgSF7gxBBpDRKfUHFcnzPINo8SSZ/rypURY7DWL4T/4FutPwvR/Azi3wTw/BM69Ctgbe3gFP/RgmFmDGeXDv3fCJT8H6K8Apg//9TXj22/Dwd6FzJTQthnXrwN0Gs6+GqvHw4D/DV74Av1wJC1pP/fyOA8MjWTZTRKWIkuCYzMYQgvRRRwJWSXmhQTUAx3G48847w99f/OIXw+9//Md/DMDEiRNZvnx5yB7eeOON3HjjjeF1S5cuBWDhwoUA3HLLLeG5yy+/nFwu66eCiypeRrf9BWpv5RZ8J9tRsIvBpZ5nEN2fR1bBrj09TJ1QxcZUDpnzhnyex4MSe5uxO6CzabYeUKT6+qmsT4y6zfcMF30YrKmwfwvE4jBlBnzh85DOwWtxqB4H5y+AHz0AD1ogK2DBYmiphYe/DbvfhOWfgSmLoHMAnt0GuWqYvRDsf4CHHoRth+D2tlM/txHA8Nta1IZ0gjVRxq1qBIs7Iu6FKu5BY9GDrjvq/vCrKFG7FimFiHxGAZ5L6Hg6KtB+NLYCywoHvrcvx8IGRVVtnIFDOSxdms57bGDu3dpbYMNbGzlvWcNJjIweqnsB77/afA51wLxL4aM3QMI2z2/6JJh9GUyrB5WAzfvg7j+FRTPgO/8EP/01XHs93PhhKPPNJhfOhXQtnH8u/NM/wPNr4MZb4Mrzzkhz33F4h5MwYr9JrUz6PBGSx8G6Ph8i7OSxkVIMQrDIwgxZ2EC7ocNmRaAsDzb+Uu712FPwrws1pGNRMWqF9hTSsghSl2lg175e2lrreKe3C501Mt97QjGfWqaI8cgz65i3ZB5lVTUnfNMcETS2widbi78tGz56W/H3jZ8ovf6cJfCnS45u58Krit8v+KD5nMEwPJKdhPcQ4EJXKs1re44weVyBqmSMigpN3LaxhEAKs8sa/ItgzxBG06IxXYcL1YiNxgHZUxpXKQquIpvL05dKkc579OdcujOCvt4Mv317r7l+1ARJG80ZhhoZdB6+kaLBWYCbB8cm9JJG0JX2aOzPUNNQRveefoTiKC+O0YEIN4UVezQ//OGvuPW267CTZaH5pHSLO4VKkd8TOB6vMazHR8OCT554XY0GLQRuwwRUwzQSMZt4WZKymhoSjkXMkjiWwJYSS2iklEhpRHopMLFCQVMa42iML+YphVLgaVNMwtUeBVeT9zzyniCfK5Dr6SRfcMkrl4JKoJQilj5AZf/BYk7+EayyQMEgE2UmLwiMCMlKGrAkVkWlX4SueGfcgRnT6nlrZzdkit7DY1r7uhRJK0SeL17YyC03fYjymnEUi+GNuYf/8KABeeZ5fIDWHgN5l35XgxfzsxEX/LNR/s3/HWwG4ZooLg6BGIIKDZKzBMaZNqP9Zm2DmQIQQYTxyGsz++aysdcOE5gMWW4BYrGwX40mWxD0dPYzvrGSvXt6sbyxmQqCfqJt96sYf72yi5e3PcYnls1l0fxzqKitwbLs4sTMxUP/Hs2x6LnBxwcfK7VBHL/dY7V5Ku6PHvMlFRlPciw4LZU2izyTBSIWahvBHjSZgPkTECq1delxXUwHHkIJddYE+RVD2Udpk8tPer45witRfo9qOb+HAn0C8PI5LCdW4m8IsK87x8zaSg4nbQoD+TGp9I/qz39WeSx+fUCy8nvvMPHH62iujlEXDxqXCNtBWNbpkdt+R+Gb//2OY547TUGbgJ/3XssCQsSNr532QMiIHCYiLzoaQhxRagy1DkIy4398214Q+SyQYdEJU5HQjrB6o1TnK8/sG8MWEBzmftczGaicwOHYjMH14OChPiY2VbEjewTtvgdqFvTlx8EEMloOh205h20HoThvDaKAtjTScRC2A8hSTuHECxG/s2BEacE3h7nmNKUfAJQwxSqUBSLvs3C+Olv7UbLa5H8oJsVn0DqL0O2SeqiidBcWwuRX1AAWGr/qPRIhNEKoYjujnkwQfBrqKkd+q3+9yuWwAnubrwARQFdfjpqaJMmaBNmujNEUUjw/Wojmrg8UHH53lD5Ybcrxei5a5ExYjB0Dy+cGRLHF4l8d+fv77dMfkoARLpfTJJMpHFsxvjrPuLpaBlxJzrHxCgrlGeWFVto3Gpt0ZsXcG5FJ+oUl0JjKLVobpJESaZlARGkJLEcgbUGZZRHPSapiMWrKK2iqKUd7Lpvf7WPrO6bl0SwPk/UKY1R+L4HFrosq5JFOPDxksssJ9h3oZeKkWrb25yDnnZLlWyKOaQX5HCqfA8syxnfbAWGF+9p7ZWN/Z2GEjM/pYRe1IGYpLmyx+PhVs2mor8VTDq5n1O2uUrieChHN0+AJkwmplEU0S04KiSUw6n+pkVJgScv8tyS2FNhCYEkF2sOREtuykAJWrnuLnl2CrX6ttFGDX79NMMKsw9HnEG0mm0XbNgjLZ+fMPNM5QX4gQ/W4BH0HUwhviJtPMAz1HASgXRfleZDPIaQFtmPyc0gZhtb8R+AlzeZTfEfHm/OpRzKfsytks8Sw6dh9gGnNtWjtEhY/EhLts3XG108av0hhslzpSFtB9t7iIV0qs+HLMkL436RhF1Hkcoru3gF6ewuEaqJR8XumLx2xN43tmWi08tCDoq6DFju7MkyePI6BRB6dLmpgTzX1CPNVaoX2NHgFw2FIC2FZCNtGSNsv6uFPLTrWQTJ09KmJEvJ5cmBoVIiMIpTTi2MJmOASXNK+uOG5Rc+dYeA0UTJQmQL9/Sm27NnJefPaSSSNeksq5dcTE2ik2S20Qvn6kCAptmkIXztYfKkK4SMRSKxA6kGgkKH3hymV1LFnJwLYe/Cwj8yj3IXDLW3sGkYIxEmNyhXAyiOiOSiFJucK+o8M0FBTxoFcP5arT3NCp4gspnUxwqLg55GUEiwbYdkmQY8IPkGA1HvclMY66pJNNFCMhSjk57AUBA7wYUyeMgVJtOcZLXikiqvg+ErY01TVRSNyLls7DrFozhRWvv42H7xwEZaUKGmjMZ7/ylcpW/gK/KCIe3RSgWLExw8L/HrNwg9kDHwkBQgbrRUSxcHuI2za0UF3V56+Xr/4QwR3RwIhSr4HNX5kImY7yGRN9qmwlphRNHT25WmpSmKVWej+QfbEUwil9DOiPAGzOE32UrMwCzlCDbGUxpFA2obVFD7HEqzSkzyVotNFJBmsxnBI2vNxzjWO454f1qUCBZlByJBn0gwKFxp+8CcVyYakC4H2TCsOH+jh4N4+ps0az6o3N7B03hzi8aA2sjS0LCwxFFCwYphNsaZYkQppoUEKvJKq58H5HFpIdh7s5pXX19LQVM9Pf7PahLvIyI5GRKM5eOhHHTGLqvgCiixRNPREDNdKtFyT9lCZDFZ5OVpa4QwU0HV4gMb6CvZn+pCuPvM0Dn5yIihS51A+8BTaBUTeL39okK9YDN7ybZf+8QA5QywWYR8BBxN2G7y36DuLlmkNqY7PlvtytPZLK4cRGUftstH1I4rrbpT2w1NIyQbvJOBlC/z61Q1MbhtHdXUZv1j5EpObGpnVNpWyMgdLG695LUyAp+H2xNDYG9Hya+UihET6ygiNIue5HOjs5u13N5MpuEyf2sqKtRvYkhSotnp/ZEE7fjbhgEIqkMrfobUpNC+0RipMnTFpYcWEoaDar7umCW3mobwROQZFtqkUBNrz8DIpZLKipHxtf9ajMueSqIqRP5IrtVqM6l2cfDDKm0GjEsHbD5DPV2YJMLW+ixvUUQxlBMeKF4niP98uGjgcRDe2YoZuXdJW0JEo6XPQgIlu4WODk4dk/vMKF4KUYFkoJwbxGCppoWNxDlTG+JfXN/Gp8+cyZ9pUdh/o5ImnVxCPxWhqqKO5qYHaynLK43Ec2zYq+SCPeqQz7durPKVwPY9s3qMvlaazp4f9h7oY6B+goiLJlIkTkEh+smYNzx3Kk26oDwdblHsjDzf6wvw+RIiI/kSD/xqECqhs8Z4QuTyDrFJjEFVphDJ8v/A00tMIJYtILHJIJ47UtpEPEPSksrRMqWVjViPyLlIZpA6o6GAXs1OOfCPqO8J0DmL9A8rPoKcbUkgdPapL/g3+EX4LEXSIEUXHO8zDikiho4aT4yDsbx3KttHxGF5FAq8yiapI4CVsVCyGjlkoS6KlMQ83FtIsHV/HsulTmVRXRw5Ff28fR3p76EtnyRUKKA1SaCxpYduGp7eERArwUITaRumQiNmUx2NUliUpqygnbtv09g/w6o49PL99F9uxKDhJ0NLXSIaPZKinVJwXx7puqHPBzlmkvkFB96IhM8ri+PcG2spgPn5JJgnEhOLaxjJe6Hbpy0qkJ3BcjZVX5uOqoIdjzeaMhqFp+7HOBGejyHaMGZ+UB1HkznY9ds8xezjhlExrDZbEq64k11SDV5PELYuhYpap6hHKPuZjmDnBAbuKn3dlef7QGqbELebW1jK1ro6GcbWMb05QGbNNjSlpI9GmUEUolJo6cUoLXM8j52YZyLv0pnLsOniE/Rt30NE7wPZ0hm7h4FoVFM1aEcHmmI/Jf5FHbXWDkAkodaQctK2HYpQoUs6htn4fKYfYkykIeK0vS1Ml7FcKPBuR0EiliHkWZWlJbCDQpA47qTMShqN+x73rDJ3qsEimowsw8l2UnCmVkQQCbdvk66vJj69GJSTaT+CCFmgl/PQ9RYWFRiFUASWhT1ayXsOGrj7szm4coEJIKoVFmS1MGIyQ2MIgmacVCo3rKVwF2YIiJSCjFFkFeSF9iumgnUqfhc0jhMQ4DY8UjvcGI0LDManUUV8ZGlGj56ICvpFjOlyXpoyFsBVKpEEkcKWLwkU6MWJSILyRz+wsnFwYnpJl8yU/g1iuUB7y5RcRVcUiEK5LfH8XMpPDq07ilcfwkg7aEWhbAhbGLaeoEvWV7wBo7aGxyAtBXkhSWBxEIFyNLhhDcigDCcBHGABtCfNd6kD6xmTULJj9QAgEltkmjkr8M8zCjC78IfCi+JCGonhHXUQo/Jeo4wJmcvC4AobJCHaekOzP24YpVBZS21jYJHKCREYhwwKLv98+hL8rMDwly2aOWoJRJiRYCkoEtZdFqIaV6TTx7l6IOahEHJWMoZJxVMJGxWOouIOK2WjHQtsWWkqEpcHSIKSJ7Aod8hUIFUHsiBYqICAl7JUkVJ8LAAu0gwjsZlpDiRvQYGyJYJEefDw4NpTOafD1/nVCg1B+Ni9zrCjQBwgUVSP7PpSB6jlQtihtlCBKYtsKy0tgexJZUNh5DyvvRuqdnUWvMwWOm35gMFIFoAkIhb/z+mtVa4Vwizoi0sbmYUmBbdtg2wjHQTsO2jF1qZTtI5ptoR0LZVsoR4It0ZaFtgRayvC/kNJEE1uGgmqfmhVtLjrQGkTmInwiFBwMCv7pIsGTBqtDvwQVziL6SCIaLl0U14he62shAUNhFULbEcKlwu8igkRCC6Q2EQNCKYTSRhvpaoNgrtFAojRSCCwnjhCWqXVN0X54FsHOLBgWyY4l5pceKFXHmkODNVsa7SmTqzCXLyKFZTwbbNsxWZuCSikB6+l/tPTdpqTw/eJ8imkJtBQoIUzFRin8a4R/TharOPpIFD0mfPaRAFEtC+E4IKTvVS/CfJMlqoxAUxjmoTRnLdsGLIOkgYoeYf5rn5IF7amiPS2gVsVzFPuMeLMU7W0+VsscIp6EWPyMtpn9R4dTZowWAVkMZRAFBQ8KLkpmQVphkKCQkbglrZGYsrMQlL/VJYZOK+yDEnkoYGdLyqUOTicXoW5aCOzySrRlhRtIUTrSJf6tQhOhVqYvK5k0RQKjyX38f8H8fRoZoTnFFkrp0DCoEsT3KI3KpMEtYMWTJvsT+MGZI2jnLJwSOG05PgKDbmCtF66Lcl2EyKFtGxmLIWzb918sGoDNzaULJ2RldXGpRrlFEZHhBstMItIsgLDzfsJSjvZY0JF7GHxKQy6PlPHwQj34/DHGHBwY3O6QepXotwCZ8y6uO4CMxxGxOIFb0mD1yVk4PXDqkCzAj+KXyD9RXGRaIwomxz6WhXRipqyOKKUqg3dpUWyGQaeGHMeQpzRopULXnaHZ42PdK9Cua3LcS2vo+4dqbpgBHxdBAvZag8kUnEHn88h4HBWL+WYKA0W0P130bThKbUi8DnwfhysXOvpuKckMcVImPoxWmtNIyY4FJT6anotyPSjkkbG4QbaoZu5k9O+nDx9T21qj3QIiZh3/2hMIJf58yjgYk88j4glwbJ8ih0zvCND/xEOUgR36+Rbl8RM7Oh22W+zpRM9/+PZOXTXuEUJxOZhfQmDC8zNpRCoFngnujCogThQYxd8YqqRTfIkqnw8VGKcH/IF4Ll56AHeg31TYUSZLqhykBz0VECLYMZ6L0e0ITKLYESVvH3nfYlD/JxyOXzX6jEOyocD474EuuOhUCnLZiFZjaEvX2PohdM4dc4PKzz4VRBGf0uU8BHgeKp3BG+hH59Lgeb5iRocyoRAecTuNLfNoNDErQ8zKGAWN1ihddMA2HwiLOgZZwXRpmUfTdvFXMib4o2vmc8eH5hC3hN+OCj3kr1g4gb+4+QKm1FcQcCsBug1ut7SmUOnYwpEFvzUUJtfRf8tFZGdNMIl1HQtlRRzMdbGXaNvF+QUdR+cYUeB5BYaDM45dHBJEcakKrVDZLHgKmUhgjMqhGuG99xPEGY2Zo9CofBbLjhW1gKcFBk1AeeiMh5fNg+MgYw7YNp6A98/YxJ/c+Bsef3EJ//76HB6488ekckn+9JHrmdPawhULW83CUpqOA938Zt0uelJ5NCJMMxdlLQIHZR3OXzCpvoo/uvZc8q7imTUdjK8tZ+r4GtZs2c+uzhSf/NAiLpw9kY6DfXz/uY2+48DR2Saj9QqC3tyJtWQvnIHd1U/8hQ0IpfHqKsheNgeZylGoTZL50HzKymKIWIzUR5YSf2075b94g0CfJkra9n1u/K4DFjsMqdHaF2UiNfeGgd8NJCOqHPSpWj5vDN/JJCJQ4r9HVts8WH+/HGVbURZXex46n0XEEidBxhjLmIJffoBiIYdXyDNtQifXXrqNpdO3s2jKNjrn2sxo3s0l52zgYG8lH7/oVZKJ+Xz2+qXkCgVcT5OISX7zZgef/dYKMgVNbZlDTXmc/kyerpSL0AqNRX2lRXkiQe9Ahp5MgV2H+vjrH7xIrlCgN5Xlv/6npdxw0Uy+8v3f8tizm/nWT15h1fpGnlu7g6baBGDT1TeAp6GxMoZtSbr6cuQ8TV2FQ3kyQX8qS3e6gHYLZC6bbcw9b+0kfniA7KJW+q9fQnL1FmI/eRWRKeC8tRNv/DiybY3IHZ0mOqTMwT6SQfrRC8qWeDVloEH2pbHyHuDhxWN45TY6ncM63IfwVFGDfZzccGNCsiJVIUwJYHwCi6Q1uosFOeYJd6CjXz+RXaRImcLAkMi1xfwW2i1AWkGy3GS8Dch+RGMXqv793IqhXjIYe9CTlgTVHYoaRh1BkpGjigBUPoPl2AjhHPf6UwZCRJ6sZmpTJ5+/9lkcjBz5gQVbkHhIrZhU53HLpa/zszW3A5qfrnqXx37zNt/6/Ie5YM4UWhurmNdaz2euO5f66nJ6U1kefeYt/nnFem68sI0vfPQ8yuMxDvel+NoPXmJ9RyefvmYpfaksWlgsf187IPnCRy7gYE+Gi+dO4vKF0+g4MMAnP7SAlqZq7vrGLzjQ1c9jf349tmVx+31Pcfn8SXz6miVUlyc53Jvi7//tVX78WgfOlv1kFk+l0N6M072V3IIW0BBbsx3Z1kT2g/ORUiIO9JhETtMa6PmrG/GqksTf3k35Iy+gkw6pWy8mf84EQJN4dx/Jf1yBl3BIf/YavOYaRMEj8Zu3iP3wBezcyFRwY5bJpBBMaWmgZXKD8aRA+yx4IAhqTKyWKvKxWqBDFyBTdihI2GlirAJM8xkBZe4p8v0hHhmc1RrtKnQ6bcLbgwXk889CgxLaj6wW/j1m/Epoo3bXEMSUWVj8wccuY+rkejyhUAHVHOWzCSiiymZBq4B5YvQtnRwIXLDe2j6RP/2HZRxJlbH7cC2f+8aHeOntKbja4YEfnceXH/0A+ZwDaMZVlDFnSgNlcYd8wWNKYyX/7ROX0FhTwUvrd1GecPgvHzufi2Y3cesH51OVdPh/P3+FVCbPf7p0DmUJm9qKBDXlCQ71pNnX2Qto3t11gANdfVQkYtRVluF6Hmu3HaSptpILZ01g9tQG2ppr2XGwl/pKhz+56f1opfjBr9+kpjzOn99yMS21SZw3dgCQX9hCvq4Cb2ojVvcA9qb9KMeiUJVEJYp5U7zmWuSBbkTeI33edLJLW0l/eBG5Ra3Yb2zDfruD9OI2MtcvpnDhLPLtzdgrN2Bv2oO3eBpMaii69B0HxsYuas3smZP5u/s/x579h/nq/d9j4sRGKisSvLpmIzPap7Bt+15apjQSj8XxPI99+49w7uIZrHlzEwOpPJdcNIc31m5j89Y9IKCmuor5c1rI5Qrkcjnq6msoi8fYtHUvLZOb6DzczZRJDWzevp/2tgms37STutoqXn1jC1dePp9EeRl9OZfK8iR79h9hfGMNb23oYN6sFjp2d7JgdgsF12Pthg6mTR5PRUWSte9sZ3rreLbtOsCEpnG8tGYLlWUWUxqqWXjOFGxH8OIb2zl4eGBUPF+4TbgFdD4L8bISUf1MgLJ4jts/9AYTx3VTEc+TzTucP3s/reP7QEFrcz8/X5VkxiSTafnyRVO5ZMEUUpkcD/9yLcmYTU1Fgp+v3sznv7WCv7r1Qj511WLmtzXz8vrdTJ8wj48tW8jW3Z08+cJ68nkjtwhg3fbDrHp3H1Mn1PHD327ilc2H+VRQckzAM2/s4A+uXMBFcyYxoa4CS1o8/doWpk9uorIsRlfvAM31lWTyLlXlSSbXV7Ftw16svgxuezO5xVPxKpMkV23C6kvjhrKG9rXHmvgrW6n45lNkrluK+8kPkG+ph9Z6tAa7LAZS4mYLeFOaiD21GrtnAO/CWci9XVgvbUAfPEIJ4zUMjA3JBPT0ZTh0uI98Ls/8eW3cetOVdPf0c9655zBp0nj+9fFfc+2H309jQw179x2krraWXXsPMmH8OGa0txCzBR+9/lLWv7uT2upKfvnMy3zq9g/j2BbpVIZcvoCnFFe7HvPmtrPh3e0cPHSE5Veex/vOncNLq9fy7pZ9rHlzK5+560Y2v9vBJZcuYcVv3+STH29h/abdXHHRAqZOGc+GzTtpqK9heusEXnzlLRbNbae7r5+2yQ1cdN5ctmzbw7adh1j52mbOXTiTisp9XHHpIjZt30vLxCb+7z+vGNtDQqNyWVPoz46HR083aAQx2+MjF69nUt0RLFyaauATV7yNJQpoJB8+bxtPPj/Pvx5+uWo93/3ZqxzsTrPjUD8fWNKG0ooJdRVMaypnSkMVaE1/Ks0bm3pIpTM4luTjyxYwq7WJe775c6NECM15hn1vHldOVTLC6iPYuPMI2/YdYWH7BGa2NNDVn2HVhr3Ma2tCKdh5qIefvrSVgVQGISU7O3uwj2RxNu0lt2QauasWgdbYL29EpNLoXM60nc+js1nTf1Ml7sRa1MR60GClc6hMHuEp9GtbEI5FsmcAe9sBxP5u4j9aZWS4+S1kb7kU50A3zsqNvpg0PIyNXdRgCcnhrl7apk6krXU8UkgS8RjdvWmax9dzzfILQAiy2QL/8I//zoHObqrKy7Bsm2wuj7QkW7bt5e//8af8n/u/xwur3qGQLzAwkKa6ppL9+4/wyusbWTBvBq+9sZGWyc389Ocv8b5z5/LMr1dz0fmLOLCvi8/efR3KU/zk5y9x4MARfvTUi1i25N+efpkPLVvCc6ve4mBXL8mEg21L8q6HlJCIOew90E3H7oMsu3gh6zZu54t/uJy4Y6GVYue+w7z05ibKE3719FFCVI7TmbRv3zszQKBJZeJ8/htX89yaNlxt87+/eylf+Zf342Gx4o3p3PaVj7Bux0TfbqnZe3iAlev3sW1/L8rTvL5hL29u2sui6c08+eWbuXzRVLbs6WTFyre5atEUPvuRC7hgbgsxR6I8D53LEuY7cQts6DiM5yk+c915XHPedF89DkIpUpkcz7+5ncpEjMbqMl7ftIf9h/p57Z2dbOw4yAWzW/nqnR/gpg8sZnxNFYcPdENvL7GV69FCkK+vwD7ch/N2h3FI97xIYIYRIwpt4xn4yh+QvXw+ViqL/comnOfXA5C95TIyt1xG9oKZqJ40XvM4Mp+4DPeSueiKpJmDN3I5YtgcH/WtNwx5UqOxpGTxgnbKkwlefWMjs2e2UlYW5/U3NjF/XhtSSjLpHI5ts3HzLuJxh8UL29m4aSfZbJ5F89vZsGkne/Z2ESgm5sxqRXke5eVJjnT30t+fY8b0iezbf5jGhnGsfXsL5y+dzdatezinfQrrN+9kWksTCsGWrXtonz6Zd7fsZs6CmXTsPcg/P/B57vqzB5k0vo6PffgC5syYzNce+jd6e9OUlyV4+c0tNDfW0jKhnrUbOpg7cwqe59Ldn6Omvo5DR/qoKE/y7vYDY6qvrEMBEZM2oazChOqcAeimAQuXTy5fw5Smbr715Hm0jO/jY5e+xS9fPYfn17aj0Vw0v4UbLpnDb9ft4GcvbgyVJ2hBc0MZ//mKhbROrGf/4R5+uGIt2/b1UlMe4+MfWMDc6eNJZ/M8+fzbbN7VxZdufj/pnMs3H1+FQHHjpbOYO20SP3tpA9Mm1DKnrZnv/vJN1nccYsakcdxxzblYUvCzFzfw27d2I1C0jq/h5isXMKGhhq27DvH9Z9ZxuCcDQuBWl5H9+EWouIOz/QCJX65BKEFh7mSyl83DWbsd2dlL7spFyI6D6JoKVH0VzupNJF7ZjLYkufPPobBkGkJpnFc2EX99q0Hc97WTP3c6wrGx3+kg9tw7WAWPwH9m78//8pgvdUxIRqDkKDZD0Wox2n0/ouM+ERZ5ASSTLJjbzrnzp/Loj39Lc0Mtn7p5GQPpDN/53q8ZyOT8Wl3R3cgfuQBh2VjllYT06D3U6SqRxSwbK1nh+zaePihqh4eRFAPtcPSC0OVN+KeOvtvkmj16FYxmXUQL+UQPlow3ajALtKb66D4DNc9RfRBRWmhK0sKFYpaOXBfazEzQvYYwQ8BJQrIzE8LBSsskBxVWqKY3b83P6SGOUfDcV+tLIbEqqktyHp6YsekiogVG9FORBP4kQ5jO8ChfNzEixcDY4EQZ+o89wOiZ4P0dy5wzHJL9TrhVjQZEaI03WqQwYlj4djERFLHQQ+5wYHbEMB3ACRuX/8VzUekBtHYjlOF3GwaXgys9ebJ6fe8IZqjW0SlmhzK2GNlUMJYJ/V4hWcQG7Rc/8B1idPERBexpSeGKaAMBh+SdWCQr9ompBJJKod08vih+Yvs6rSAinzMbjjXSEz2DYZFMKw9TJqbUeTJ0nPSNviVnT7fNNZAlPNcvBDFKCTHYkpUqsngndEJ+IQztoTMDkM9E2tec/gd4FkYPw7+vYe1k3kA/YVEAafJomFpU0vfykL5YUUwNPcbSyScQfGdOrdGeC1bMPzq6NpTnhozEyeN4NF42g/AUMpHEeO5Hqe5ZOPPh+Ctr+JRwfmxVMX9TPvwWVuGwLITtlzn1kS50fTqtoFGFAsKJjU1r6Rpn12IC1BM/obDFfBZPBXk6jF3utD++szAKGP5tDatdPAtn4Sy8d/i9UnychbNwJsJZJDsLZ+Ekw1kkOwtn4STDWSQ7C2fhJMNZJDsLZ+Ekw1kkOwtn4STD/weDd1MJHkVhuAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==========================================================================================\n",
      "Text1: #COVID19 Vaccination Tracker\n",
      "(upto 09th February, 2021 till 08:00 AM)\n",
      "#LargestVaccineDrive\n",
      "#StayAlert\n",
      "#StaySafe https://t.co/JmMMayb8hn\n",
      "==========================================================================================\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==========================================================================================\n",
      "Text2: #COVID19 Vaccination Tracker\n",
      "#LargestVaccineDrive\n",
      "#StayAlert\n",
      "#StaySafe https://t.co/eo8wP3Fcdy\n",
      "==========================================================================================\n",
      ">>>>> Label: Contradictory\n",
      "==========================================================================================\n"
     ]
    }
   ],
   "source": [
    "visualize(1395) #NoEntailment(Neutural)\n",
    "visualize(12) #Implies\n",
    "visualize(13) #Contradictory"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a28cb44a",
   "metadata": {},
   "source": [
    "### 1.3. Data Split"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c9754f39",
   "metadata": {},
   "source": [
    "The dataset suffers from class imbalance problem."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7700a8f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NoEntailment     1182\n",
       "Implies           109\n",
       "Contradictory     109\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"label\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "105891a9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total training examples: 1197\n",
      "Total validation examples: 63\n",
      "Total test examples: 140\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# stratified split (consider number of labels in train/val/test set)\n",
    "\n",
    "# 10% for test\n",
    "train_df, test_df = train_test_split(df, test_size=0.1, stratify=df[\"label\"].values, random_state=42)\n",
    "# 4.5% for validation, 85.5% for train\n",
    "train_df, val_df = train_test_split(train_df, test_size=0.05, stratify=train_df[\"label\"].values, random_state=42)\n",
    "\n",
    "print(f\"Total training examples: {len(train_df)}\")\n",
    "print(f\"Total validation examples: {len(val_df)}\")\n",
    "print(f\"Total test examples: {len(test_df)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "451ba0c1-253b-4019-b30a-3d73b4520b6c",
   "metadata": {},
   "source": [
    "### 1.4. Multi-modal Dataset, Dataloader"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74304c67-df08-48a6-afaa-dc8ee5bdcf87",
   "metadata": {},
   "source": [
    "**Input Batch**: text_inputs, image_inputs, labels\n",
    "\n",
    "**1. Text Inputs**\n",
    "```\n",
    "{'input_ids': tensor(batch_size, 128),  'token_type_ids': tensor(batch_size, 128),  'attention_mask': tensor(batch_size, 128)}\n",
    "```\n",
    "- 'input ids': [CLS] text1 [SEP] text2 [SEP]\n",
    "    - [CLS]: 101, [SEP]: 102 \n",
    "    - max_length = 128, padding = 0\n",
    "- 'token_type_ids': text1(0), text2(1)\n",
    "- 'attention_mask': all 1 in text positions\n",
    "\n",
    "**2. Image Inputs**\n",
    "```\n",
    "{'image1': tensor(batch_size, 3, 128, 128),  'image2': tensor(batch_size, 3, 128, 128)}\n",
    "```\n",
    "- color image -> resize(128) -> to tensor\n",
    "\n",
    "**3. Labels**\n",
    "```\n",
    "tensor(batch_size,)\n",
    "```\n",
    "             "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "73f5c992-cc7a-48c1-8690-d85040a0691f",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Dataset\n",
    "\n",
    "from transformers import AutoTokenizer\n",
    "import torchvision\n",
    "from PIL import Image\n",
    "\n",
    "class MultiModal_Dataset(torch.utils.data.Dataset):\n",
    "    def __init__(self, df, tokenizer, img_size=128):\n",
    "        \n",
    "        # text\n",
    "        self.text1 = df[\"text_1\"].tolist()\n",
    "        self.text2 = df[\"text_2\"].tolist()\n",
    "        self.tokenizer = tokenizer\n",
    "        \n",
    "        # image\n",
    "        self.image1 = df[\"image_1_path\"].tolist()\n",
    "        self.image2 = df[\"image_2_path\"].tolist()\n",
    "        self.img_transform = torchvision.transforms.Compose([torchvision.transforms.Resize(size=(img_size, img_size)),\n",
    "                                                             torchvision.transforms.ToTensor()])\n",
    "        #label\n",
    "        self.label = df[\"label_idx\"].tolist()\n",
    "    \n",
    "    def __len__(self):\n",
    "        return len(self.text1)\n",
    "    \n",
    "    def __getitem__(self, idx):\n",
    "        \n",
    "        # text inputs\n",
    "        text1 = self.text1[idx]\n",
    "        text2 = self.text2[idx]\n",
    "        text_inputs = self.tokenizer(text1, text2, padding=\"max_length\", truncation=True, max_length=128, return_tensors=\"pt\") #batch별 max_len 자동 커스텀화\n",
    "        \n",
    "        # image inputs\n",
    "        ## number of channels are different depending on the image mode\n",
    "        ### If the input has 4 channels, the ``mode`` is assumed to be ``RGBA``.\n",
    "        ### If the input has 3 channels, the ``mode`` is assumed to be ``RGB``.\n",
    "        ### If the input has 2 channels, the ``mode`` is assumed to be ``LA``.\n",
    "        ##### https://pytorch.org/vision/main/_modules/torchvision/transforms/transforms.html#ToPILImage\n",
    "        ##### https://stackoverflow.com/questions/9166400/convert-rgba-png-to-rgb-with-pil\n",
    "        image1 = Image.open(self.image1[idx]).convert(\"RGB\")\n",
    "        image1 = self.img_transform(image1)\n",
    "        image2 = Image.open(self.image2[idx]).convert(\"RGB\")\n",
    "        image2 = self.img_transform(image2)\n",
    "        image_inputs = {'image1': image1, 'image2': image2}\n",
    "        \n",
    "        # labels\n",
    "        labels = self.label[idx]\n",
    "        \n",
    "        return text_inputs, image_inputs, labels\n",
    "        "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ccfd507-cc05-4d69-9924-3117e320f2eb",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00947e05",
   "metadata": {
    "tags": []
   },
   "source": [
    "## 2. Modeling"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3df34f1e",
   "metadata": {},
   "source": [
    "The model consists of the following elements:\n",
    "\n",
    "![image](multimodal.png)\n",
    "\n",
    "\n",
    "- A standalone encoder for the images. We will use a **ResNet50** pre-trained on the ImageNet-1k dataset for this.(torchvision)\n",
    "\n",
    "- A standalone encoder for the texts. A pre-trained **BERT** will be used for this.(huggingface transformers)\n",
    "\n",
    "After extracting the individual embeddings, they will be projected in an identical space. Finally, their projections will be concatenated and be fed to the final classification layer."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b0460e1",
   "metadata": {},
   "source": [
    "Model Architecture\n",
    "\n",
    "- text encoder\n",
    "- vision encoder\n",
    "- projection(text, vision)\n",
    "- Dense(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b495cf84-3bdf-4977-b3a4-5f0209eef882",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "True\n",
      "NVIDIA GeForce RTX 2080 Ti\n",
      "1\n"
     ]
    }
   ],
   "source": [
    "## GPU check\n",
    "print(torch.cuda.is_available())\n",
    "print(torch.cuda.get_device_name(0))\n",
    "print(torch.cuda.device_count())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "3ea78935-fb71-43cb-b1e6-f54bd8eb5b19",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "device(type='cuda')"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "device"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2b31c618-a63c-4236-a8a6-b762238ca5b6",
   "metadata": {},
   "source": [
    "### 2.0. projection embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "26f33ea0-46d1-46aa-8301-50acb52e9495",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Project_Embeddings_one_layer(nn.Module):\n",
    "    def __init__(self, projection_dim=256, dropout_rate=0.2):\n",
    "        super().__init__()\n",
    "        self.gelu_linear_dropout = nn.Sequential(nn.GELU(),\n",
    "                                                 nn.Linear(projection_dim, projection_dim),\n",
    "                                                 nn.Dropout(dropout_rate))\n",
    "        self.layer_norm = nn.LayerNorm(projection_dim)\n",
    "        \n",
    "    def forward(self, input_x):\n",
    "        x = self.gelu_linear_dropout(input_x)\n",
    "        x = x + input_x #residual connection\n",
    "        x = self.layer_norm(x)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "fdc3ce60-e6d9-4a5d-9d3a-b075cc68b2bd",
   "metadata": {},
   "outputs": [],
   "source": [
    "class Project_Embeddings(nn.Module):\n",
    "    def __init__(self, num_projection_layers=3, hidden_dim=768, projection_dim=256, dropout_rate=0.2):\n",
    "        super().__init__()\n",
    "        self.first_layer = nn.Linear(hidden_dim, projection_dim)\n",
    "        self.embedding_layers = nn.ModuleList([Project_Embeddings_one_layer() for i in range(num_projection_layers)])\n",
    "        \n",
    "    def forward(self, embeddings):\n",
    "        output = self.first_layer(embeddings)\n",
    "        for layer in self.embedding_layers:\n",
    "            output = layer(output)\n",
    "        return output  "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a7a5b09",
   "metadata": {},
   "source": [
    "### 2.1. Text Encoder"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7e81091-9f0a-4b80-9b5a-b94921b211e7",
   "metadata": {},
   "source": [
    "[CLS]: 101, [SEP]: 102"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "07498706",
   "metadata": {},
   "source": [
    "- input: type_ids (batch_size, 128), mask (batch_size, 128), word_ids (batch_size, 128)\n",
    "- BERT (trainable?)\n",
    "- projection\n",
    "    - Dense\n",
    "    - gelu -> Dense -> Dropout -> Add(residual connection) -> Layer Normalization (num_layers 만큼 반복)\n",
    "    - output: (batch_size, 256)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "fd8bc411-1d1f-4fc9-9b6a-492f3e09463e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "c51201ae91194305ae6237df853dc4f8",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading:   0%|          | 0.00/570 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "BertConfig {\n",
       "  \"architectures\": [\n",
       "    \"BertForMaskedLM\"\n",
       "  ],\n",
       "  \"attention_probs_dropout_prob\": 0.1,\n",
       "  \"gradient_checkpointing\": false,\n",
       "  \"hidden_act\": \"gelu\",\n",
       "  \"hidden_dropout_prob\": 0.1,\n",
       "  \"hidden_size\": 768,\n",
       "  \"initializer_range\": 0.02,\n",
       "  \"intermediate_size\": 3072,\n",
       "  \"layer_norm_eps\": 1e-12,\n",
       "  \"max_position_embeddings\": 512,\n",
       "  \"model_type\": \"bert\",\n",
       "  \"num_attention_heads\": 12,\n",
       "  \"num_hidden_layers\": 12,\n",
       "  \"pad_token_id\": 0,\n",
       "  \"position_embedding_type\": \"absolute\",\n",
       "  \"transformers_version\": \"4.5.0\",\n",
       "  \"type_vocab_size\": 2,\n",
       "  \"use_cache\": true,\n",
       "  \"vocab_size\": 30522\n",
       "}"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from transformers import AutoConfig\n",
    "\n",
    "# pretrained 모델과 동일한 config\n",
    "model_config = AutoConfig.from_pretrained('bert-base-uncased')\n",
    "model_config"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "5fef5aba-2355-472e-9ac2-58c5d3281005",
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import (AutoTokenizer, AutoModel)\n",
    "\n",
    "class Text_Encoder_BERT(nn.Module):\n",
    "    def __init__(self, model_name, trainable=False, num_projection_layers=2, hidden_dim=768, projection_dim=256, dropout_rate=0.2, device=device):\n",
    "        super().__init__()\n",
    "        self.model_name = model_name\n",
    "        \n",
    "        self.text_encoder = AutoModel.from_pretrained(model_name) # fine tuning 여부 체크!\n",
    "        \n",
    "        for param in self.text_encoder.parameters():\n",
    "            param.requires_grad = trainable\n",
    "        \n",
    "        self.projection = Project_Embeddings(num_projection_layers=2, hidden_dim=768, projection_dim=256, dropout_rate=0.2)\n",
    "        \n",
    "    def forward(self, input_ids, token_type_ids, attention_mask):\n",
    "\n",
    "        text_inputs = {'input_ids': input_ids, 'token_type_ids': token_type_ids, 'attention_mask':attention_mask}\n",
    "        \n",
    "        bert_embedding = self.text_encoder(**text_inputs)\n",
    "        \n",
    "        cls_output = bert_embedding.pooler_output\n",
    "        \n",
    "        output = self.projection(cls_output)\n",
    "        \n",
    "        return output #torch.Size([batch_size, 256])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "1f694e73-291d-42a5-b9c1-2b6eb495cbd1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "================================================================================\n",
       "Layer (type:depth-idx)                                  Param #\n",
       "================================================================================\n",
       "Text_Encoder_BERT                                       --\n",
       "├─BertModel: 1-1                                        --\n",
       "│    └─BertEmbeddings: 2-1                              --\n",
       "│    │    └─Embedding: 3-1                              (23,440,896)\n",
       "│    │    └─Embedding: 3-2                              (393,216)\n",
       "│    │    └─Embedding: 3-3                              (1,536)\n",
       "│    │    └─LayerNorm: 3-4                              (1,536)\n",
       "│    │    └─Dropout: 3-5                                --\n",
       "│    └─BertEncoder: 2-2                                 --\n",
       "│    │    └─ModuleList: 3-6                             (85,054,464)\n",
       "│    └─BertPooler: 2-3                                  --\n",
       "│    │    └─Linear: 3-7                                 (590,592)\n",
       "│    │    └─Tanh: 3-8                                   --\n",
       "├─Project_Embeddings: 1-2                               --\n",
       "│    └─Linear: 2-4                                      196,864\n",
       "│    └─ModuleList: 2-5                                  --\n",
       "│    │    └─Project_Embeddings_one_layer: 3-9           66,304\n",
       "│    │    └─Project_Embeddings_one_layer: 3-10          66,304\n",
       "================================================================================\n",
       "Total params: 109,811,712\n",
       "Trainable params: 329,472\n",
       "Non-trainable params: 109,482,240\n",
       "================================================================================"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from torchinfo import summary\n",
    "\n",
    "text_encoder_bert = Text_Encoder_BERT(model_name=model_name, num_projection_layers=2, hidden_dim=768, projection_dim=256, dropout_rate=0.2)\n",
    "summary(text_encoder_bert)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f5536331-c4e1-4785-85db-c11c500d8958",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2ac6cd86",
   "metadata": {},
   "source": [
    "### 2.2. Vision Encoder"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac8296d0",
   "metadata": {},
   "source": [
    "- input: image1 (batch_size, 128, 128, 3), image2 (batch_size, 128, 128, 3)\n",
    "- embedding 1 (resnet for embedding image 1) -> avg pooling\n",
    "- embedding 2 (resnet for embedding image 2) -> avg pooling\n",
    "    - avg means that global average pooling will be applied to the output of the last convolutional block, and thus the output of the model will be a 2D tensor.\n",
    "- concatenate embedding 1 & embedding 2\n",
    "- projection\n",
    "    - Dense\n",
    "    - gelu -> Dense -> Dropout -> Add(residual connection) -> LayerNormalization (num_layers 만큼 반복)\n",
    "    - output: (batch_size, 256)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c5f379c3",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torchvision\n",
    "\n",
    "class Image_Encoder_ResNet(nn.Module):\n",
    "    def __init__(self, trainable=False):\n",
    "        super().__init__()\n",
    "        \n",
    "        resnet_image1 = torchvision.models.resnet50(pretrained=True)\n",
    "        self.resnet_image1 = nn.Sequential(*list(resnet_image1.children())[:-1]) # remove linear classifier\n",
    "        \n",
    "        resnet_image2 = torchvision.models.resnet50(pretrained=True)\n",
    "        self.resnet_image2 = nn.Sequential(*list(resnet_image2.children())[:-1])\n",
    "        \n",
    "        for param1, param2 in zip(self.resnet_image1.parameters(), self.resnet_image2.parameters()):\n",
    "            param1.requires_grad = trainable\n",
    "            param2.requires_grad = trainable\n",
    "        \n",
    "        self.projection = Project_Embeddings(num_projection_layers=1, hidden_dim=4096, projection_dim=256, dropout_rate=0.2)\n",
    "        \n",
    "    def forward(self, image1, image2):\n",
    "        # image_inputs = {'image1': tensor(B,3,128,128), 'image2': tensor(B,3,128,128)}\n",
    "        \n",
    "        output1 = self.resnet_image1(image1) # (B,C,1,1)\n",
    "        output2 = self.resnet_image2(image2) # (B,C,1,1)\n",
    "        \n",
    "        concat = torch.cat((output1, output2), dim=1) # (B,2C,1,1)\n",
    "                        \n",
    "        batch_size = concat.size(0)\n",
    "        output = concat.view((batch_size, 4096)) # (B,2C)\n",
    "        \n",
    "        output = self.projection(output)\n",
    "        return output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "ed73407b-a007-4cb8-acfb-8c444ae7d6e0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "====================================================================================================\n",
       "Layer (type:depth-idx)                             Output Shape              Param #\n",
       "====================================================================================================\n",
       "Image_Encoder_ResNet                               --                        --\n",
       "├─Project_Embeddings: 1                            --                        --\n",
       "│    └─ModuleList: 2-1                             --                        --\n",
       "├─Sequential: 1-1                                  [1, 2048, 1, 1]           --\n",
       "│    └─Conv2d: 2-2                                 [1, 64, 64, 64]           (9,408)\n",
       "│    └─BatchNorm2d: 2-3                            [1, 64, 64, 64]           (128)\n",
       "│    └─ReLU: 2-4                                   [1, 64, 64, 64]           --\n",
       "│    └─MaxPool2d: 2-5                              [1, 64, 32, 32]           --\n",
       "│    └─Sequential: 2-6                             [1, 256, 32, 32]          --\n",
       "│    │    └─Bottleneck: 3-1                        [1, 256, 32, 32]          (75,008)\n",
       "│    │    └─Bottleneck: 3-2                        [1, 256, 32, 32]          (70,400)\n",
       "│    │    └─Bottleneck: 3-3                        [1, 256, 32, 32]          (70,400)\n",
       "│    └─Sequential: 2-7                             [1, 512, 16, 16]          --\n",
       "│    │    └─Bottleneck: 3-4                        [1, 512, 16, 16]          (379,392)\n",
       "│    │    └─Bottleneck: 3-5                        [1, 512, 16, 16]          (280,064)\n",
       "│    │    └─Bottleneck: 3-6                        [1, 512, 16, 16]          (280,064)\n",
       "│    │    └─Bottleneck: 3-7                        [1, 512, 16, 16]          (280,064)\n",
       "│    └─Sequential: 2-8                             [1, 1024, 8, 8]           --\n",
       "│    │    └─Bottleneck: 3-8                        [1, 1024, 8, 8]           (1,512,448)\n",
       "│    │    └─Bottleneck: 3-9                        [1, 1024, 8, 8]           (1,117,184)\n",
       "│    │    └─Bottleneck: 3-10                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    │    └─Bottleneck: 3-11                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    │    └─Bottleneck: 3-12                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    │    └─Bottleneck: 3-13                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    └─Sequential: 2-9                             [1, 2048, 4, 4]           --\n",
       "│    │    └─Bottleneck: 3-14                       [1, 2048, 4, 4]           (6,039,552)\n",
       "│    │    └─Bottleneck: 3-15                       [1, 2048, 4, 4]           (4,462,592)\n",
       "│    │    └─Bottleneck: 3-16                       [1, 2048, 4, 4]           (4,462,592)\n",
       "│    └─AdaptiveAvgPool2d: 2-10                     [1, 2048, 1, 1]           --\n",
       "├─Sequential: 1-2                                  [1, 2048, 1, 1]           --\n",
       "│    └─Conv2d: 2-11                                [1, 64, 64, 64]           (9,408)\n",
       "│    └─BatchNorm2d: 2-12                           [1, 64, 64, 64]           (128)\n",
       "│    └─ReLU: 2-13                                  [1, 64, 64, 64]           --\n",
       "│    └─MaxPool2d: 2-14                             [1, 64, 32, 32]           --\n",
       "│    └─Sequential: 2-15                            [1, 256, 32, 32]          --\n",
       "│    │    └─Bottleneck: 3-17                       [1, 256, 32, 32]          (75,008)\n",
       "│    │    └─Bottleneck: 3-18                       [1, 256, 32, 32]          (70,400)\n",
       "│    │    └─Bottleneck: 3-19                       [1, 256, 32, 32]          (70,400)\n",
       "│    └─Sequential: 2-16                            [1, 512, 16, 16]          --\n",
       "│    │    └─Bottleneck: 3-20                       [1, 512, 16, 16]          (379,392)\n",
       "│    │    └─Bottleneck: 3-21                       [1, 512, 16, 16]          (280,064)\n",
       "│    │    └─Bottleneck: 3-22                       [1, 512, 16, 16]          (280,064)\n",
       "│    │    └─Bottleneck: 3-23                       [1, 512, 16, 16]          (280,064)\n",
       "│    └─Sequential: 2-17                            [1, 1024, 8, 8]           --\n",
       "│    │    └─Bottleneck: 3-24                       [1, 1024, 8, 8]           (1,512,448)\n",
       "│    │    └─Bottleneck: 3-25                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    │    └─Bottleneck: 3-26                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    │    └─Bottleneck: 3-27                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    │    └─Bottleneck: 3-28                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    │    └─Bottleneck: 3-29                       [1, 1024, 8, 8]           (1,117,184)\n",
       "│    └─Sequential: 2-18                            [1, 2048, 4, 4]           --\n",
       "│    │    └─Bottleneck: 3-30                       [1, 2048, 4, 4]           (6,039,552)\n",
       "│    │    └─Bottleneck: 3-31                       [1, 2048, 4, 4]           (4,462,592)\n",
       "│    │    └─Bottleneck: 3-32                       [1, 2048, 4, 4]           (4,462,592)\n",
       "│    └─AdaptiveAvgPool2d: 2-19                     [1, 2048, 1, 1]           --\n",
       "├─Project_Embeddings: 1-3                          [1, 256]                  --\n",
       "│    └─Linear: 2-20                                [1, 256]                  1,048,832\n",
       "│    └─ModuleList: 2-1                             --                        --\n",
       "│    │    └─Project_Embeddings_one_layer: 3-33     [1, 256]                  66,304\n",
       "====================================================================================================\n",
       "Total params: 48,131,200\n",
       "Trainable params: 1,115,136\n",
       "Non-trainable params: 47,016,064\n",
       "Total mult-adds (G): 2.67\n",
       "====================================================================================================\n",
       "Input size (MB): 0.39\n",
       "Forward/backward pass size (MB): 116.14\n",
       "Params size (MB): 192.52\n",
       "Estimated Total Size (MB): 309.05\n",
       "===================================================================================================="
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#\n",
    "res = Image_Encoder_ResNet()\n",
    "summary(res, ((1,3,128,128),(1,3,128,128)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42fe853c",
   "metadata": {
    "tags": []
   },
   "source": [
    "### 2.3. Multi-modal Classification Layer"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f27cd74",
   "metadata": {},
   "source": [
    "- concatenate text projection (batch_size, 256) & image projection (batch_size, 256) -> output: (batch_size, 512)\n",
    "- 3 classes: `NoEntailment`, `Implies`, `Contradictory`\n",
    "- Dense layer: input (batch_size, 512) -> output (batch_size, 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "4d4af843",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MultiModal_Classification(nn.Module):\n",
    "    def __init__(self, text_model_name = 'bert-base-uncased', text_trainable=False, image_trainable=False, projection_dim=256, num_class=3):\n",
    "        super().__init__()\n",
    "        self.text_encoder = Text_Encoder_BERT(model_name=text_model_name, trainable=text_trainable, num_projection_layers=2, hidden_dim=768, projection_dim=256, dropout_rate=0.2)\n",
    "        self.image_encoder = Image_Encoder_ResNet(trainable=image_trainable)\n",
    "        \n",
    "        self.classifier = nn.Linear(2*projection_dim, num_class)\n",
    "        \n",
    "    def forward(self, input_ids, token_type_ids, attention_mask, image1, image2):\n",
    "        \n",
    "        # text encoder: BERT\n",
    "        text_projection = self.text_encoder(input_ids, token_type_ids, attention_mask) # (B,256)\n",
    "        \n",
    "        # image_encoder: ResNet\n",
    "        image_projection = self.image_encoder(image1, image2) # (B,256)\n",
    "        \n",
    "        # concat\n",
    "        text_image = torch.cat((text_projection, image_projection), dim=1) # (B,512)\n",
    "        \n",
    "        # classifier\n",
    "        output = self.classifier(text_image)\n",
    "        \n",
    "        return output        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "804a677e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "=====================================================================================\n",
       "Layer (type:depth-idx)                                       Param #\n",
       "=====================================================================================\n",
       "MultiModal_Classification                                    --\n",
       "├─Text_Encoder_BERT: 1-1                                     --\n",
       "│    └─BertModel: 2-1                                        --\n",
       "│    │    └─BertEmbeddings: 3-1                              (23,837,184)\n",
       "│    │    └─BertEncoder: 3-2                                 (85,054,464)\n",
       "│    │    └─BertPooler: 3-3                                  (590,592)\n",
       "│    └─Project_Embeddings: 2-2                               --\n",
       "│    │    └─Linear: 3-4                                      196,864\n",
       "│    │    └─ModuleList: 3-5                                  132,608\n",
       "├─Image_Encoder_ResNet: 1-2                                  --\n",
       "│    └─Sequential: 2-3                                       --\n",
       "│    │    └─Conv2d: 3-6                                      (9,408)\n",
       "│    │    └─BatchNorm2d: 3-7                                 (128)\n",
       "│    │    └─ReLU: 3-8                                        --\n",
       "│    │    └─MaxPool2d: 3-9                                   --\n",
       "│    │    └─Sequential: 3-10                                 (215,808)\n",
       "│    │    └─Sequential: 3-11                                 (1,219,584)\n",
       "│    │    └─Sequential: 3-12                                 (7,098,368)\n",
       "│    │    └─Sequential: 3-13                                 (14,964,736)\n",
       "│    │    └─AdaptiveAvgPool2d: 3-14                          --\n",
       "│    └─Sequential: 2-4                                       --\n",
       "│    │    └─Conv2d: 3-15                                     (9,408)\n",
       "│    │    └─BatchNorm2d: 3-16                                (128)\n",
       "│    │    └─ReLU: 3-17                                       --\n",
       "│    │    └─MaxPool2d: 3-18                                  --\n",
       "│    │    └─Sequential: 3-19                                 (215,808)\n",
       "│    │    └─Sequential: 3-20                                 (1,219,584)\n",
       "│    │    └─Sequential: 3-21                                 (7,098,368)\n",
       "│    │    └─Sequential: 3-22                                 (14,964,736)\n",
       "│    │    └─AdaptiveAvgPool2d: 3-23                          --\n",
       "│    └─Project_Embeddings: 2-5                               --\n",
       "│    │    └─Linear: 3-24                                     1,048,832\n",
       "│    │    └─ModuleList: 3-25                                 66,304\n",
       "├─Linear: 1-3                                                1,539\n",
       "=====================================================================================\n",
       "Total params: 157,944,451\n",
       "Trainable params: 1,446,147\n",
       "Non-trainable params: 156,498,304\n",
       "====================================================================================="
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "multi = MultiModal_Classification(text_model_name = 'bert-base-uncased', text_trainable=False, image_trainable=False, projection_dim=256, num_class=3)\n",
    "summary(multi)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf941711",
   "metadata": {},
   "source": [
    "## 3. Training / Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8dcd913d-6053-4419-b899-66a18f02a49a",
   "metadata": {},
   "source": [
    "### wandb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "acbd4ca1-a5d7-4773-b7d1-4613231ae793",
   "metadata": {},
   "outputs": [],
   "source": [
    "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "515489d3-1649-4afe-b0fe-736bb942ce88",
   "metadata": {},
   "outputs": [],
   "source": [
    "import wandb"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4279be40-ba97-4da5-b2b7-eb77051ad392",
   "metadata": {},
   "source": [
    "This is happening whenever you use multiprocessing (Often used by data loaders). The way to disable this warning is to set the TOKENIZERS_PARALLELISM environment variable to the value that makes more sense for you. By default, we disable the parallelism to avoid any hidden deadlock that would be hard to debug, but you might be totally fine while keeping it enabled in your specific use-case.\n",
    "\n",
    "You can try to set it to true, and if your process seems to be stuck, doing nothing, then you should use false.\n",
    "\n",
    "We'll improve this message to help avoid any confusion (Cf huggingface/tokenizers#328)\n",
    "\n",
    "https://github.com/huggingface/transformers/issues/5486"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "c80ebf2a-53b9-4a78-8b5d-c222a193c1d0",
   "metadata": {},
   "outputs": [],
   "source": [
    "EPOCHS = 10\n",
    "BATCH_SIZE = 32\n",
    "LEARNING_RATE = 0.001\n",
    "\n",
    "args={\"epochs\": EPOCHS, \"batch_size\": BATCH_SIZE, \"learning_rate\" : LEARNING_RATE}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "c6d83256-4b4d-4942-8569-9d10e671dcb4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "Finishing last run (ID:260xbpga) before initializing another..."
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<br/>Waiting for W&B process to finish, PID 11895... <strong style=\"color:green\">(success).</strong>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "VBox(children=(Label(value=' 0.00MB of 0.00MB uploaded (0.00MB deduped)\\r'), FloatProgress(value=1.0, max=1.0)…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<style>\n",
       "    table.wandb td:nth-child(1) { padding: 0 10px; text-align: right }\n",
       "    .wandb-row { display: flex; flex-direction: row; flex-wrap: wrap; width: 100% }\n",
       "    .wandb-col { display: flex; flex-direction: column; flex-basis: 100%; flex: 1; padding: 10px; }\n",
       "    </style>\n",
       "<div class=\"wandb-row\"><div class=\"wandb-col\">\n",
       "<h3>Run history:</h3><br/><table class=\"wandb\"><tr><td>Train Accuracy</td><td>▁▁▃▄▄▄▅▆██</td></tr><tr><td>Train F1 Macro</td><td>▁▁▃▄▄▄▅▆██</td></tr><tr><td>Train F1 Weighted</td><td>▁▃▄▅▅▅▅▆██</td></tr><tr><td>Train Loss</td><td>█▅▅▄▄▄▃▂▁▁</td></tr><tr><td>Valid Accuracy</td><td>▇▇████▄▁▁▃</td></tr><tr><td>Valid F1 Macro</td><td>▁▁████▅▄▄▅</td></tr><tr><td>Valid F1 Weighted</td><td>██████▄▁▁▃</td></tr><tr><td>Valid Loss</td><td>▁▁▁▁▁▂▄▆█▆</td></tr></table><br/></div><div class=\"wandb-col\">\n",
       "<h3>Run summary:</h3><br/><table class=\"wandb\"><tr><td>Train Accuracy</td><td>0.92231</td></tr><tr><td>Train F1 Macro</td><td>0.73782</td></tr><tr><td>Train F1 Weighted</td><td>0.93163</td></tr><tr><td>Train Loss</td><td>0.22059</td></tr><tr><td>Valid Accuracy</td><td>0.77778</td></tr><tr><td>Valid F1 Macro</td><td>0.36418</td></tr><tr><td>Valid F1 Weighted</td><td>0.80247</td></tr><tr><td>Valid Loss</td><td>0.95399</td></tr></table>\n",
       "</div></div>\n",
       "Synced 5 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)\n",
       "<br/>Synced <strong style=\"color:#cdcd00\">robust-flower-12</strong>: <a href=\"https://wandb.ai/yookyungkho/multimodal_entailment/runs/260xbpga\" target=\"_blank\">https://wandb.ai/yookyungkho/multimodal_entailment/runs/260xbpga</a><br/>\n",
       "Find logs at: <code>./wandb/run-20220117_094927-260xbpga/logs</code><br/>\n"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Successfully finished last run (ID:260xbpga). Initializing new run:<br/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "                    Syncing run <strong><a href=\"https://wandb.ai/yookyungkho/multimodal_entailment/runs/322w6zci\" target=\"_blank\">easy-aardvark-13</a></strong> to <a href=\"https://wandb.ai/yookyungkho/multimodal_entailment\" target=\"_blank\">Weights & Biases</a> (<a href=\"https://docs.wandb.com/integrations/jupyter.html\" target=\"_blank\">docs</a>).<br/>\n",
       "\n",
       "                "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<button onClick=\"this.nextSibling.style.display='block';this.style.display='none';\">Display W&B run</button><iframe src=\"https://wandb.ai/yookyungkho/multimodal_entailment/runs/322w6zci?jupyter=true\" style=\"border:none;width:100%;height:420px;display:none;\"></iframe>"
      ],
      "text/plain": [
       "<wandb.sdk.wandb_run.Run at 0x7f4aa5e27fd0>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wandb.init(project=\"multimodal_entailment\", entity=\"yookyungkho\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "3864a27c-d153-4f45-bd25-cd0d677ddd44",
   "metadata": {},
   "outputs": [],
   "source": [
    "wandb.config.update(args)\n",
    "\n",
    "torch.manual_seed(0) #랜덤 시드 고정\n",
    "\n",
    "\n",
    "# Load the tokenizer for the BERT model\n",
    "tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')\n",
    "\n",
    "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
    "\n",
    "# Create Dataset objects for train/validation sets.\n",
    "train_dataset = MultiModal_Dataset(df=train_df, tokenizer=tokenizer, img_size=128)\n",
    "valid_dataset = MultiModal_Dataset(df=val_df, tokenizer=tokenizer, img_size=128)\n",
    "\n",
    "train_dataloader = torch.utils.data.DataLoader(train_dataset, batch_size = 32)\n",
    "valid_dataloader = torch.utils.data.DataLoader(valid_dataset, batch_size = 32)\n",
    "\n",
    "# model\n",
    "#model = MultiModal_Classification(text_model_name = 'bert-base-uncased', text_trainable=False, image_trainable=False, projection_dim=256, num_class=3) # freeze\n",
    "model = MultiModal_Classification(text_model_name = 'bert-base-uncased', text_trainable=True, image_trainable=True, projection_dim=256, num_class=3) # fine-tuning\n",
    "model = model.to(device)\n",
    "\n",
    "wandb.watch(model, log=\"all\")\n",
    "\n",
    "# CE Loss\n",
    "criterion = torch.nn.CrossEntropyLoss()\n",
    "\n",
    "# Optimizer : Adam\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=LEARNING_RATE)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "cd8f4058-c4d0-4abc-a12b-fec35c43bd77",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_epoch(model, optimizer, dataloader):\n",
    "    model.train()\n",
    "    \n",
    "    losses = 0\n",
    "    acc = 0\n",
    "    total = 0\n",
    "    f1_macro = 0\n",
    "    f1_weighted = 0\n",
    "\n",
    "    #for batch_idx, (text_inputs, image_inputs, labels) in enumerate(tqdm(dataloader)):\n",
    "    for batch_idx, (text_inputs, image_inputs, labels) in enumerate(dataloader):\n",
    "        \n",
    "        input_ids = text_inputs['input_ids'].squeeze(1).to(device) # (B,1,128) -> (B,128)\n",
    "        token_type_ids = text_inputs['token_type_ids'].squeeze(1).to(device)\n",
    "        attention_mask = text_inputs['attention_mask'].squeeze(1).to(device)\n",
    "        \n",
    "        image1 = image_inputs['image1'].to(device)\n",
    "        image2 = image_inputs['image2'].to(device)\n",
    "        labels = labels.to(device)\n",
    "        \n",
    "        output = model(input_ids,token_type_ids, attention_mask, image1, image2).to(device)\n",
    "        predicted = torch.max(output, 1)[1]\n",
    "        \n",
    "        # loss, Update parameters\n",
    "        optimizer.zero_grad()\n",
    "        loss = criterion(output, labels)\n",
    "        loss.backward()\n",
    "        optimizer.step()\n",
    "        \n",
    "        losses += loss.item()\n",
    "        \n",
    "        # accuracy\n",
    "        total += labels.size(0)\n",
    "        acc += (predicted == labels).sum().item()\n",
    "        # f1 score\n",
    "        f1_macro += f1_score(predicted.cpu().detach().numpy(), labels.cpu().detach().numpy(), average='macro')\n",
    "        f1_weighted += f1_score(predicted.cpu().detach().numpy(), labels.cpu().detach().numpy(), average='weighted')\n",
    "             \n",
    "    train_loss = losses / len(dataloader)\n",
    "    train_acc = acc / total\n",
    "    train_f1_macro = f1_macro / len(dataloader)\n",
    "    train_f1_weighted = f1_weighted / len(dataloader)\n",
    "    \n",
    "    #return train_loss, train_acc\n",
    "    return train_loss, train_acc, train_f1_macro, train_f1_weighted\n",
    "\n",
    "def evaluate_epoch(model, dataloader, test = False):\n",
    "    model.eval()\n",
    "    \n",
    "    losses = 0\n",
    "    acc = 0\n",
    "    total = 0\n",
    "    f1_macro = 0\n",
    "    f1_weighted = 0\n",
    "    \n",
    "    with torch.no_grad():\n",
    "        #for batch_idx, (text_inputs, image_inputs, labels) in enumerate(tqdm(dataloader)):\n",
    "        for batch_idx, (text_inputs, image_inputs, labels) in enumerate(dataloader):\n",
    "            \n",
    "            input_ids = text_inputs['input_ids'].squeeze(1).to(device) # (B,1,128) -> (B,128)\n",
    "            token_type_ids = text_inputs['token_type_ids'].squeeze(1).to(device)\n",
    "            attention_mask = text_inputs['attention_mask'].squeeze(1).to(device)\n",
    "        \n",
    "            image1 = image_inputs['image1'].to(device)\n",
    "            image2 = image_inputs['image2'].to(device)\n",
    "            labels = labels.to(device)\n",
    "        \n",
    "            output = model(input_ids, token_type_ids, attention_mask, image1, image2).to(device)\n",
    "            predicted = torch.max(output, 1)[1]\n",
    "            \n",
    "            if test == True:\n",
    "                if batch_idx == 0:\n",
    "                    pred_labels = predicted\n",
    "                    real_labels = labels\n",
    "                else:\n",
    "                    pred_labels = torch.cat((pred_labels, predicted))\n",
    "                    real_labels = torch.cat((real_labels, labels))\n",
    "            \n",
    "            # loss\n",
    "            loss = criterion(output, labels)\n",
    "            losses += loss.item()\n",
    "            # accuracy\n",
    "            total += labels.size(0)\n",
    "            acc += (predicted == labels).sum().item()\n",
    "            # f1 score\n",
    "            f1_macro += f1_score(predicted.cpu().detach().numpy(), labels.cpu().detach().numpy(), average='macro')\n",
    "            f1_weighted += f1_score(predicted.cpu().detach().numpy(), labels.cpu().detach().numpy(), average='weighted')\n",
    "            \n",
    "    valid_loss = losses / len(dataloader)\n",
    "    valid_acc = acc / total\n",
    "    valid_f1_macro = f1_macro / len(dataloader)\n",
    "    valid_f1_weighted = f1_weighted / len(dataloader)\n",
    "    \n",
    "    if test == True:\n",
    "        return valid_loss, valid_acc, valid_f1_macro, valid_f1_weighted, pred_labels, real_labels\n",
    "    else:\n",
    "        return valid_loss, valid_acc, valid_f1_macro, valid_f1_weighted"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "20bff1a8-a907-4205-b111-0835ca7d192e",
   "metadata": {},
   "outputs": [],
   "source": [
    "class CheckPoint:\n",
    "    \n",
    "    def __init__(self, dir_checkpoint, file_name):\n",
    "        self.best_loss = 9999\n",
    "        if not os.path.exists(dir_checkpoint):\n",
    "            os.mkdir(dir_checkpoint)\n",
    "        self.dir_checkpoint = dir_checkpoint\n",
    "        self.file_name = file_name\n",
    "    \n",
    "    def step(self, model, loss, acc, f1_macro, f1_weighted, epoch):\n",
    "        if loss < self.best_loss:\n",
    "            print(f'Model Save : Best Loss {self.best_loss:.4f} -> {loss:.4f}')\n",
    "            \n",
    "            state = {'model_state_dict': model.state_dict(),\n",
    "                     'loss': loss,\n",
    "                     'accuracy': acc,\n",
    "                     'f1_macro': f1_macro,\n",
    "                     'f1_weighted': f1_weighted,\n",
    "                     'epoch': epoch}\n",
    "\n",
    "            torch.save(state, f'{self.dir_checkpoint}/{self.file_name}_multi_modal.pth')\n",
    "        \n",
    "            self.best_loss = loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "46fa46cc-23d3-4419-bd4b-82a179bb5472",
   "metadata": {},
   "outputs": [],
   "source": [
    "def save_history(epoch, dir_checkpoint, file_name, log_dict):\n",
    "    \n",
    "    file_path = f\"{dir_checkpoint}/{file_name}_history_multi_modal.json\"\n",
    "    \n",
    "    if epoch == 1:\n",
    "        history = {'Train Loss': [log_dict['Train Loss']], 'Train Accuracy': [log_dict['Train Accuracy']],\n",
    "                   'Train F1 Macro': [log_dict['Train F1 Macro']], 'Train F1 Weighted': [log_dict['Train F1 Weighted']],\n",
    "                   'Valid Loss': [log_dict['Valid Loss']], 'Valid Accuracy': [log_dict['Valid Accuracy']],\n",
    "                   'Valid F1 Macro': [log_dict['Valid F1 Macro']], 'Valid F1 Weighted': [log_dict['Valid F1 Weighted']]}\n",
    "        \n",
    "        json.dump(history, open(file_path,'w'), indent=4)\n",
    "    else:\n",
    "        with open(file_path, \"r\") as json_file:\n",
    "            json_data = json.load(json_file)\n",
    "\n",
    "        json_data['Train Loss'].append(log_dict['Train Loss'])\n",
    "        json_data['Train Accuracy'].append(log_dict['Train Accuracy'])\n",
    "        json_data['Train F1 Macro'].append(log_dict['Train F1 Macro'])\n",
    "        json_data['Train F1 Weighted'].append(log_dict['Train F1 Weighted'])\n",
    "        json_data['Valid Loss'].append(log_dict['Valid Loss'])\n",
    "        json_data['Valid Accuracy'].append(log_dict['Valid Accuracy'])\n",
    "        json_data['Valid F1 Macro'].append(log_dict['Valid F1 Macro'])\n",
    "        json_data['Valid F1 Weighted'].append(log_dict['Valid F1 Weighted'])\n",
    "\n",
    "        with open(file_path, 'w') as outfile:\n",
    "            json.dump(json_data, outfile, indent=4)\n",
    "            \n",
    "def visualize_history(dir_checkpoint, file_name):\n",
    "    file_path = f\"{dir_checkpoint}/{file_name}_history_multi_modal.json\"\n",
    "    history = json.load(open(file_path,'r'))\n",
    "\n",
    "    fig, ax = s\n",
    "\n",
    "    ax[0, 0].plot(range(len(history['Train Loss'])), history['Train Loss'])\n",
    "    ax[0, 0].plot(range(len(history['Valid Loss'])), history['Valid Loss'])\n",
    "    ax[0, 0].legend(['Train','Valid'], loc='upper right')\n",
    "    ax[0, 0].set_title(\"Loss\")\n",
    "\n",
    "    ax[0, 1].plot(range(len(history['Train Accuracy'])), history['Train Accuracy'])\n",
    "    ax[0, 1].plot(range(len(history['Valid Accuracy'])), history['Valid Accuracy'])\n",
    "    ax[0, 1].legend(['Train','Valid'], loc='upper right')\n",
    "    ax[0, 1].set_title(\"Accuracy\")\n",
    "\n",
    "    ax[1, 0].plot(range(len(history['Train F1 Macro'])), history['Train F1 Macro'])\n",
    "    ax[1, 0].plot(range(len(history['Valid F1 Macro'])), history['Valid F1 Macro'])\n",
    "    ax[1, 0].legend(['Train','Valid'], loc='upper right')\n",
    "    ax[1, 0].set_title(\"F1 Macro\")\n",
    "\n",
    "    ax[1, 1].plot(range(len(history['Train F1 Weighted'])), history['Train F1 Weighted'])\n",
    "    ax[1, 1].plot(range(len(history['Valid F1 Weighted'])), history['Valid F1 Weighted'])\n",
    "    ax[1, 1].legend(['Train','Valid'], loc='upper right')\n",
    "    ax[1, 1].set_title(\"F1 Weighted\")\n",
    "    \n",
    "    plt.suptitle(f'History Plot for {file_name}_multi_modal', fontweight =\"bold\", fontsize=15)\n",
    "    fig.tight_layout()\n",
    "    plt.show()\n",
    "    plt.savefig(f'{dir_checkpoint}/{file_name}_plot_multi_modal.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "db334f48-361b-4ee6-ab9c-0089338b9e0e",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Out of Memory 방지\n",
    "import gc\n",
    "gc.collect()\n",
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "f17aa44a-a61a-47b4-99a5-6647a8ac7880",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ignore warnings\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(action='ignore') "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e7f5910-9ccb-4629-b4a4-a70fc22ea6a0",
   "metadata": {},
   "source": [
    "### freeze weights\n",
    "\n",
    " Freeze pre-trained BERT, ResNet weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "0e5691e9-ccc3-4ab6-8275-ee76bedf6a08",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                                                                                                                            | 0/10 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 1 (time = 39.413s)\n",
      "[Train] Loss: 0.7061, Accuracy: 0.840, F1_macro: 0.325, F1_weighted: 0.908\n",
      "[Valid] Loss: 0.5389, Accuracy: 0.841, F1_macro: 0.305, F1_weighted: 0.914\n",
      "Model Save : Best Loss 9999.0000 -> 0.5389\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 20%|████████████████████████████████████                                                                                                                                                | 2/10 [01:22<05:28, 41.05s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 2 (time = 37.962s)\n",
      "[Train] Loss: 0.5207, Accuracy: 0.845, F1_macro: 0.322, F1_weighted: 0.915\n",
      "[Valid] Loss: 0.5509, Accuracy: 0.841, F1_macro: 0.305, F1_weighted: 0.914\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 30%|██████████████████████████████████████████████████████                                                                                                                              | 3/10 [02:03<04:46, 40.98s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 3 (time = 38.465s)\n",
      "[Train] Loss: 0.4723, Accuracy: 0.865, F1_macro: 0.441, F1_weighted: 0.918\n",
      "[Valid] Loss: 0.5490, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 40%|████████████████████████████████████████████████████████████████████████                                                                                                            | 4/10 [02:43<04:03, 40.61s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 4 (time = 38.025s)\n",
      "[Train] Loss: 0.4378, Accuracy: 0.880, F1_macro: 0.510, F1_weighted: 0.922\n",
      "[Valid] Loss: 0.5428, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 50%|██████████████████████████████████████████████████████████████████████████████████████████                                                                                          | 5/10 [03:23<03:21, 40.28s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 5 (time = 37.684s)\n",
      "[Train] Loss: 0.4145, Accuracy: 0.880, F1_macro: 0.510, F1_weighted: 0.921\n",
      "[Valid] Loss: 0.5611, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 60%|████████████████████████████████████████████████████████████████████████████████████████████████████████████                                                                        | 6/10 [04:03<02:40, 40.16s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 6 (time = 37.855s)\n",
      "[Train] Loss: 0.3911, Accuracy: 0.880, F1_macro: 0.525, F1_weighted: 0.921\n",
      "[Valid] Loss: 0.6277, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 70%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▉                                                      | 7/10 [04:42<02:00, 40.08s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 7 (time = 37.896s)\n",
      "[Train] Loss: 0.3406, Accuracy: 0.886, F1_macro: 0.557, F1_weighted: 0.921\n",
      "[Valid] Loss: 0.7603, Accuracy: 0.794, F1_macro: 0.368, F1_weighted: 0.826\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 80%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████                                    | 8/10 [05:23<01:20, 40.16s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 8 (time = 38.323s)\n",
      "[Train] Loss: 0.2646, Accuracy: 0.902, F1_macro: 0.636, F1_weighted: 0.928\n",
      "[Valid] Loss: 0.9155, Accuracy: 0.746, F1_macro: 0.349, F1_weighted: 0.761\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 90%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████                  | 9/10 [06:03<00:40, 40.08s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 9 (time = 37.863s)\n",
      "[Train] Loss: 0.2107, Accuracy: 0.921, F1_macro: 0.743, F1_weighted: 0.933\n",
      "[Valid] Loss: 1.0859, Accuracy: 0.746, F1_macro: 0.346, F1_weighted: 0.756\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 10/10 [06:40<00:00, 40.10s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 10 (time = 35.779s)\n",
      "[Train] Loss: 0.2206, Accuracy: 0.922, F1_macro: 0.738, F1_weighted: 0.932\n",
      "[Valid] Loss: 0.9540, Accuracy: 0.778, F1_macro: 0.364, F1_weighted: 0.802\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# Start training! (10 epochs)\n",
    "\n",
    "dir_checkpoint='./checkpoint'\n",
    "file_name='exp1_freeze'\n",
    "checkpoint = CheckPoint(dir_checkpoint='./checkpoint', file_name=file_name)\n",
    "\n",
    "\n",
    "for epoch in tqdm(range(1, EPOCHS+1)):\n",
    "    start_time = timer()\n",
    "    train_loss, train_acc, train_f1_macro, train_f1_weighted = train_epoch(model, optimizer, train_dataloader)\n",
    "    #train_loss, train_acc = train_epoch(model, optimizer, train_dataloader)\n",
    "    end_time = timer()\n",
    "    \n",
    "    valid_loss, valid_acc, valid_f1_macro, valid_f1_weighted = evaluate_epoch(model, valid_dataloader, test=False)\n",
    "    \n",
    "    print(f\">>>>> Epoch {epoch} (time = {(end_time - start_time):.3f}s)\")\n",
    "    print(f\"[Train] Loss: {train_loss:.4f}, Accuracy: {train_acc:.3f}, F1_macro: {train_f1_macro:.3f}, F1_weighted: {train_f1_weighted:.3f}\")\n",
    "    #print(f\"[Train] Loss: {train_loss:.4f}, Accuracy: {train_acc:.3f}\")\n",
    "    print(f\"[Valid] Loss: {valid_loss:.4f}, Accuracy: {valid_acc:.3f}, F1_macro: {valid_f1_macro:.3f}, F1_weighted: {valid_f1_weighted:.3f}\")\n",
    "\n",
    "    # checkpoint\n",
    "    checkpoint.step(model, valid_loss, valid_acc, valid_f1_macro, valid_f1_weighted, epoch)\n",
    "    \n",
    "    log_dict = {'Train Loss': train_loss, 'Train Accuracy': train_acc, 'Train F1 Macro': train_f1_macro, 'Train F1 Weighted': train_f1_weighted,\n",
    "                'Valid Loss': valid_loss, 'Valid Accuracy': valid_acc, 'Valid F1 Macro': valid_f1_macro, 'Valid F1 Weighted': valid_f1_weighted}\n",
    "    \n",
    "    #log_dict = {'Train Loss': train_loss, 'Train Accuracy': train_acc,\n",
    "    #            'Valid Loss': valid_loss, 'Valid Accuracy': valid_acc, 'Valid F1 Macro': valid_f1_macro, 'Valid F1 Weighted': valid_f1_weighted}\n",
    "    \n",
    "    # history\n",
    "    save_history(epoch, dir_checkpoint, file_name, log_dict)\n",
    "    \n",
    "    wandb.log(log_dict) ### wandb 페이지에 log 기록"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "7787f76e-98ce-4200-b892-5e4cf5a50ad7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## visualize and save manual plot image to png in checkpoint folder\n",
    "visualize_history(dir_checkpoint, file_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3de6f2df-4983-42b5-b541-684fc6c74ad9",
   "metadata": {},
   "source": [
    "### fine-tuning\n",
    "\n",
    " Fine-tune pre-trained BERT, ResNet weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "734159d3-fbd7-4f50-9956-3408ce193f30",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|                                                                                                                                                                                            | 0/10 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 1 (time = 47.330s)\n",
      "[Train] Loss: 0.6835, Accuracy: 0.816, F1_macro: 0.390, F1_weighted: 0.865\n",
      "[Valid] Loss: 0.8630, Accuracy: 0.667, F1_macro: 0.310, F1_weighted: 0.654\n",
      "Model Save : Best Loss 9999.0000 -> 0.8630\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 10%|██████████████████                                                                                                                                                                  | 1/10 [00:50<07:30, 50.10s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 2 (time = 47.058s)\n",
      "[Train] Loss: 0.4717, Accuracy: 0.867, F1_macro: 0.486, F1_weighted: 0.911\n",
      "[Valid] Loss: 0.5712, Accuracy: 0.841, F1_macro: 0.371, F1_weighted: 0.896\n",
      "Model Save : Best Loss 0.8630 -> 0.5712\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 20%|████████████████████████████████████                                                                                                                                                | 2/10 [01:41<06:46, 50.76s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 3 (time = 47.196s)\n",
      "[Train] Loss: 0.4588, Accuracy: 0.878, F1_macro: 0.503, F1_weighted: 0.918\n",
      "[Valid] Loss: 0.5467, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n",
      "Model Save : Best Loss 0.5712 -> 0.5467\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 30%|██████████████████████████████████████████████████████                                                                                                                              | 3/10 [02:32<05:57, 51.10s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 4 (time = 47.294s)\n",
      "[Train] Loss: 0.4669, Accuracy: 0.876, F1_macro: 0.504, F1_weighted: 0.918\n",
      "[Valid] Loss: 0.5330, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n",
      "Model Save : Best Loss 0.5467 -> 0.5330\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 50%|██████████████████████████████████████████████████████████████████████████████████████████                                                                                          | 5/10 [04:14<04:13, 50.78s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 5 (time = 47.406s)\n",
      "[Train] Loss: 0.4578, Accuracy: 0.877, F1_macro: 0.502, F1_weighted: 0.919\n",
      "[Valid] Loss: 0.5939, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 60%|████████████████████████████████████████████████████████████████████████████████████████████████████████████                                                                        | 6/10 [05:03<03:21, 50.32s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 6 (time = 47.260s)\n",
      "[Train] Loss: 0.4691, Accuracy: 0.869, F1_macro: 0.489, F1_weighted: 0.911\n",
      "[Valid] Loss: 0.5829, Accuracy: 0.841, F1_macro: 0.415, F1_weighted: 0.884\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 70%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▉                                                      | 7/10 [05:52<02:29, 49.95s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 7 (time = 47.167s)\n",
      "[Train] Loss: 0.4580, Accuracy: 0.876, F1_macro: 0.495, F1_weighted: 0.915\n",
      "[Valid] Loss: 0.5424, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 80%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████                                    | 8/10 [06:42<01:39, 49.87s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 8 (time = 47.660s)\n",
      "[Train] Loss: 0.4609, Accuracy: 0.875, F1_macro: 0.497, F1_weighted: 0.915\n",
      "[Valid] Loss: 0.5636, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 90%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████                  | 9/10 [07:32<00:49, 49.91s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 9 (time = 47.852s)\n",
      "[Train] Loss: 0.4441, Accuracy: 0.874, F1_macro: 0.499, F1_weighted: 0.915\n",
      "[Valid] Loss: 0.5936, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 10/10 [08:22<00:00, 50.23s/it]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      ">>>>> Epoch 10 (time = 47.700s)\n",
      "[Train] Loss: 0.4492, Accuracy: 0.876, F1_macro: 0.502, F1_weighted: 0.913\n",
      "[Valid] Loss: 0.5997, Accuracy: 0.857, F1_macro: 0.418, F1_weighted: 0.908\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# Start training! (10 epochs)\n",
    "\n",
    "dir_checkpoint='./checkpoint'\n",
    "file_name='exp1_tuning'\n",
    "checkpoint = CheckPoint(dir_checkpoint='./checkpoint', file_name=file_name)\n",
    "\n",
    "\n",
    "for epoch in tqdm(range(1, EPOCHS+1)):\n",
    "    start_time = timer()\n",
    "    train_loss, train_acc, train_f1_macro, train_f1_weighted = train_epoch(model, optimizer, train_dataloader)\n",
    "    #train_loss, train_acc = train_epoch(model, optimizer, train_dataloader)\n",
    "    end_time = timer()\n",
    "    \n",
    "    valid_loss, valid_acc, valid_f1_macro, valid_f1_weighted = evaluate_epoch(model, valid_dataloader, test=False)\n",
    "    \n",
    "    print(f\">>>>> Epoch {epoch} (time = {(end_time - start_time):.3f}s)\")\n",
    "    print(f\"[Train] Loss: {train_loss:.4f}, Accuracy: {train_acc:.3f}, F1_macro: {train_f1_macro:.3f}, F1_weighted: {train_f1_weighted:.3f}\")\n",
    "    #print(f\"[Train] Loss: {train_loss:.4f}, Accuracy: {train_acc:.3f}\")\n",
    "    print(f\"[Valid] Loss: {valid_loss:.4f}, Accuracy: {valid_acc:.3f}, F1_macro: {valid_f1_macro:.3f}, F1_weighted: {valid_f1_weighted:.3f}\")\n",
    "\n",
    "    # checkpoint\n",
    "    checkpoint.step(model, valid_loss, valid_acc, valid_f1_macro, valid_f1_weighted, epoch)\n",
    "    \n",
    "    log_dict = {'Train Loss': train_loss, 'Train Accuracy': train_acc, 'Train F1 Macro': train_f1_macro, 'Train F1 Weighted': train_f1_weighted,\n",
    "                'Valid Loss': valid_loss, 'Valid Accuracy': valid_acc, 'Valid F1 Macro': valid_f1_macro, 'Valid F1 Weighted': valid_f1_weighted}\n",
    "    \n",
    "    #log_dict = {'Train Loss': train_loss, 'Train Accuracy': train_acc,\n",
    "    #            'Valid Loss': valid_loss, 'Valid Accuracy': valid_acc, 'Valid F1 Macro': valid_f1_macro, 'Valid F1 Weighted': valid_f1_weighted}\n",
    "    \n",
    "    # history\n",
    "    save_history(epoch, dir_checkpoint, file_name, log_dict)\n",
    "    \n",
    "    wandb.log(log_dict) ### wandb 페이지에 log 기록"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "37ae07f5-edcb-4f6e-98f9-7ee73f2ef6b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## visualize and save manual plot image to png in checkpoint folder\n",
    "visualize_history(dir_checkpoint, file_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2e20aa2-7ffc-4ac7-80f5-0641acf5d55c",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf4f81ee-4894-4bce-bb9c-5f3c58bbc0f4",
   "metadata": {},
   "source": [
    "## 4. Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "3ab5674b-3491-46f2-b4f9-dd9e4f84878f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Load pretrained model\n"
     ]
    }
   ],
   "source": [
    "torch.manual_seed(0) #랜덤 시드 고정\n",
    "\n",
    "\n",
    "# Load the tokenizer for the BERT model.\n",
    "tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')\n",
    "\n",
    "# Create Dataset objects for train/validation sets.\n",
    "test_dataset = MultiModal_Dataset(df=test_df, tokenizer=tokenizer, img_size=128)\n",
    "\n",
    "test_dataloader = torch.utils.data.DataLoader(test_dataset, batch_size = 32)\n",
    "\n",
    "weights = torch.load(f'{dir_checkpoint}/{file_name}_multi_modal.pth')\n",
    "\n",
    "model = MultiModal_Classification(text_model_name = 'bert-base-uncased', text_trainable=False, image_trainable=False, projection_dim=256, num_class=3)\n",
    "model = model.to(device)\n",
    "model.load_state_dict(weights['model_state_dict'])\n",
    "\n",
    "print('Load pretrained model')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "22b59df3-59d5-4eab-b6ea-c03511168b14",
   "metadata": {},
   "source": [
    "### fine-tuning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "cd9aee9e-2031-48d8-be16-cdc6bae1f42f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Test] Loss: 0.4976, Accuracy: 0.886, F1_macro: 0.513, F1_weighted: 0.916\n"
     ]
    }
   ],
   "source": [
    "test_loss, test_acc, test_f1_macro, test_f1_weighted, pred_labels, real_labels = evaluate_epoch(model, test_dataloader, test = True)\n",
    "print(f\"[Test] Loss: {test_loss:.4f}, Accuracy: {test_acc:.3f}, F1_macro: {test_f1_macro:.3f}, F1_weighted: {test_f1_weighted:.3f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "b5025cdd-c7f7-4ac2-be38-86241fc47429",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "               precision    recall  f1-score   support\n",
      "\n",
      "Contradictory       0.70      0.64      0.67        11\n",
      "      Implies       0.00      0.00      0.00        11\n",
      " NoEntailment       0.90      0.99      0.94       118\n",
      "\n",
      "     accuracy                           0.89       140\n",
      "    macro avg       0.53      0.54      0.54       140\n",
      " weighted avg       0.81      0.89      0.85       140\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n",
      "/opt/conda/lib/python3.7/site-packages/sklearn/metrics/_classification.py:1318: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    },
    {
     "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>Contradictory</th>\n",
       "      <th>Implies</th>\n",
       "      <th>NoEntailment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Contradictory</th>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Implies</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NoEntailment</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>117</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Contradictory  Implies  NoEntailment\n",
       "Contradictory              7        0             4\n",
       "Implies                    2        0             9\n",
       "NoEntailment               1        0           117"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "\n",
    "class_names = list(label_map.keys())\n",
    "\n",
    "# classification report\n",
    "report = classification_report(real_labels.cpu().detach().numpy(), pred_labels.cpu().detach().numpy(), target_names=class_names)\n",
    "print(report)\n",
    "\n",
    "# confusion matrix\n",
    "pd.DataFrame(confusion_matrix(real_labels.cpu().detach().numpy(), pred_labels.cpu().detach().numpy()),index=class_names, columns=class_names)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae7c65e3-d1e6-4266-a96b-87a45ecdc60d",
   "metadata": {},
   "source": [
    "### freeze"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "94ed331d-cfaf-401d-9822-c126cc6afa4d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[Test] Loss: 0.5966, Accuracy: 0.843, F1_macro: 0.329, F1_weighted: 0.890\n"
     ]
    }
   ],
   "source": [
    "test_loss, test_acc, test_f1_macro, test_f1_weighted, pred_labels, real_labels = evaluate_epoch(model, test_dataloader, test = True)\n",
    "print(f\"[Test] Loss: {test_loss:.4f}, Accuracy: {test_acc:.3f}, F1_macro: {test_f1_macro:.3f}, F1_weighted: {test_f1_weighted:.3f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "8a492be3-9fb6-4efa-8e11-aa57d922f6a4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "               precision    recall  f1-score   support\n",
      "\n",
      "Contradictory       0.00      0.00      0.00        11\n",
      "      Implies       0.00      0.00      0.00        11\n",
      " NoEntailment       0.84      1.00      0.91       118\n",
      "\n",
      "     accuracy                           0.84       140\n",
      "    macro avg       0.28      0.33      0.30       140\n",
      " weighted avg       0.71      0.84      0.77       140\n",
      "\n"
     ]
    },
    {
     "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>Contradictory</th>\n",
       "      <th>Implies</th>\n",
       "      <th>NoEntailment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Contradictory</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Implies</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NoEntailment</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>118</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Contradictory  Implies  NoEntailment\n",
       "Contradictory              0        0            11\n",
       "Implies                    0        0            11\n",
       "NoEntailment               0        0           118"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix, classification_report\n",
    "\n",
    "class_names = list(label_map.keys())\n",
    "\n",
    "# classification report\n",
    "report = classification_report(real_labels.cpu().detach().numpy(), pred_labels.cpu().detach().numpy(), target_names=class_names)\n",
    "print(report)\n",
    "\n",
    "# confusion matrix\n",
    "pd.DataFrame(confusion_matrix(real_labels.cpu().detach().numpy(), pred_labels.cpu().detach().numpy()),index=class_names, columns=class_names)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}