indobert-large-p1 + CNN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "351ad27c",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2024-04-11T17:09:25.805522Z",
     "iopub.status.busy": "2024-04-11T17:09:25.805141Z",
     "iopub.status.idle": "2024-04-11T17:09:26.716254Z",
     "shell.execute_reply": "2024-04-11T17:09:26.715343Z"
    },
    "papermill": {
     "duration": 0.923259,
     "end_time": "2024-04-11T17:09:26.718740",
     "exception": false,
     "start_time": "2024-04-11T17:09:25.795481",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Suicidal dataframe:          label                                        twitterText\n",
      "0     SUICIDAL  \"pengen bunuh diri aja\"\\nyang paling sering te...\n",
      "1     SUICIDAL  Secara tidak langsung punya Fikiran pengen bun...\n",
      "2     SUICIDAL          sejam lalu haha hihi sekarang pengen mati\n",
      "3     SUICIDAL  Baru kelar 1 masalah datang lagi malah lainnya...\n",
      "4     SUICIDAL       fuck bundir aku lagi pgn mabok smpk ga sadar\n",
      "...        ...                                                ...\n",
      "8727  SUICIDAL  jujur cape hidup kaya gini terus, aku nganggur...\n",
      "8728  SUICIDAL  Siapa yang dari sd udah kepikiran mau mati ? A...\n",
      "8729  SUICIDAL   Apa aku doang kalo stres suka pengen bunuh diri?\n",
      "8730  SUICIDAL  Mood ga karuan. Ovt ga karuan. Selalu kek miki...\n",
      "8731  SUICIDAL   pengen bundir tapi takut gabisa dengerin mineral\n",
      "\n",
      "[8732 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import re\n",
    "import os\n",
    "\n",
    "df_suicidal = pd.read_csv(\"/kaggle/input/manual-annotation-indo-twitter-suicide/final-annotated-suicide-dataset.csv\", usecols=['label', 'twitterText'])\n",
    "df_suicidal.dropna(inplace=True)\n",
    "df_suicidal['label'] = \"SUICIDAL\"\n",
    "df_suicidal = df_suicidal.sample(frac=1, random_state=42).reset_index(drop=True)\n",
    "print(\"Suicidal dataframe:\", df_suicidal)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "00bd0d2d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:09:26.737476Z",
     "iopub.status.busy": "2024-04-11T17:09:26.737157Z",
     "iopub.status.idle": "2024-04-11T17:09:27.186294Z",
     "shell.execute_reply": "2024-04-11T17:09:27.185014Z"
    },
    "papermill": {
     "duration": 0.460519,
     "end_time": "2024-04-11T17:09:27.188349",
     "exception": false,
     "start_time": "2024-04-11T17:09:26.727830",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Non-suicidal dataframe:               label                                        twitterText\n",
      "0      NOT_SUICIDAL  yall the feminists i know on twitter are happi...\n",
      "1      NOT_SUICIDAL  Imagine bini sedap masak and his brain thinks ...\n",
      "2      NOT_SUICIDAL  Kita masak macam katering dah raya ni dia kalu...\n",
      "3      NOT_SUICIDAL  Masak ni survival skills, bukan gender roles. ...\n",
      "4      NOT_SUICIDAL  Istg if I were a man, I'd surely date her. \\n\\...\n",
      "...             ...                                                ...\n",
      "35307  NOT_SUICIDAL  @leo_edw W slalu punya feel kalo dia cuma ghos...\n",
      "35308  NOT_SUICIDAL  Jog! Lagi ngrasake bersyukur tapi yo kesel ki,...\n",
      "35309  NOT_SUICIDAL  @mks_fess itulah pentingnya tidak berharap ket...\n",
      "35310  NOT_SUICIDAL  Moots pdkt yuk tapi enih harus delapan kata ad...\n",
      "35311  NOT_SUICIDAL  Pengertian Ghosting dan Cara Menghindarinya: H...\n",
      "\n",
      "[35312 rows x 2 columns]\n",
      "Non-suicidal dataframe:               label                                        twitterText\n",
      "0      NOT_SUICIDAL  [cm] sekarang aku paham kenapa anak semester a...\n",
      "1      NOT_SUICIDAL  Semoga kita kuat kuat aja ya ngejalanin hidup ...\n",
      "2      NOT_SUICIDAL                              pengen bundir lmaoooo\n",
      "3      NOT_SUICIDAL  sumpah pengen gua katain anjing tolol bego ban...\n",
      "4      NOT_SUICIDAL                   pengen beli immortal trus bundir\n",
      "...             ...                                                ...\n",
      "35392  NOT_SUICIDAL         @kadyasandya coba bundir di jembatan suhat\n",
      "35393  NOT_SUICIDAL  @BluueBleGum tapi gua jadi mikir nih, kan pres...\n",
      "35394  NOT_SUICIDAL  @amordevaleri @WeningAA @CNNIndonesia iya aku ...\n",
      "35395  NOT_SUICIDAL  Gua pas itu gini\\n\\n\"Capek bgt pikiran bundir ...\n",
      "35396  NOT_SUICIDAL  Dulu pernah ampe udh cari cara di google tapi ...\n",
      "\n",
      "[35397 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "df_not_suicidal_a = pd.read_csv(\"/kaggle/input/manual-annotation-indo-twitter-suicide/final-non-suicide-dataset.csv\", usecols=['label', 'twitterText'])\n",
    "df_not_suicidal_a.dropna(inplace=True)\n",
    "df_not_suicidal_a['label'] = \"NOT_SUICIDAL\"\n",
    "print(\"Non-suicidal dataframe:\", df_not_suicidal_a)\n",
    "\n",
    "df_not_suicidal_b = pd.read_csv(\"/kaggle/input/manual-annotation-indo-twitter-suicide/final-annotated-non-suicide-dataset.csv\", usecols=['label', 'twitterText'])\n",
    "df_not_suicidal_b.dropna(inplace=True)\n",
    "df_not_suicidal_b['label'] = \"NOT_SUICIDAL\"\n",
    "print(\"Non-suicidal dataframe:\", df_not_suicidal_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3d808fb2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:09:27.206653Z",
     "iopub.status.busy": "2024-04-11T17:09:27.206393Z",
     "iopub.status.idle": "2024-04-11T17:09:27.221019Z",
     "shell.execute_reply": "2024-04-11T17:09:27.220335Z"
    },
    "papermill": {
     "duration": 0.025799,
     "end_time": "2024-04-11T17:09:27.223094",
     "exception": false,
     "start_time": "2024-04-11T17:09:27.197295",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Non-suicidal dataframe:              label                                        twitterText\n",
      "0     NOT_SUICIDAL  Gak mudah mengakhiri hubungan di umur 26thn di...\n",
      "1     NOT_SUICIDAL                   Canda. Sesempetnya aja ya bous 😎\n",
      "2     NOT_SUICIDAL  @thekuxell Cara Israel mengurangi populasi ten...\n",
      "3     NOT_SUICIDAL  Aku hanya lelah hidup seperti ini, tak tau har...\n",
      "4     NOT_SUICIDAL  Ada orang yg pintar karena tekun belajar dan a...\n",
      "...            ...                                                ...\n",
      "8727  NOT_SUICIDAL  @kkmanti @kangadang1 @Yanuardychandra @floriel...\n",
      "8728  NOT_SUICIDAL  @Don_gindo Kok dek akibat singo di ajar makan ...\n",
      "8729  NOT_SUICIDAL  Mande :\"Jadi mo cium gak\"\\nSurya :\"ntar aja ka...\n",
      "8730  NOT_SUICIDAL  Tuhan...knp z harus kenal,cyg,n nikah dg org y...\n",
      "8731  NOT_SUICIDAL  Demi tuhan, ini narasi paling berkesan. Gemes,...\n",
      "\n",
      "[8732 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "#Combine non-suicidal datasets from non-suicide keyword query (a) and suicide keyword but annotated non-suicide (b)\n",
    "#Randomly sample 8732 data to match number of suicidal data\n",
    "df_not_suicidal = pd.concat([df_not_suicidal_a, df_not_suicidal_b])\n",
    "df_not_suicidal = df_not_suicidal.sample(n=8732, random_state=42).reset_index(drop=True)\n",
    "print(\"Non-suicidal dataframe:\", df_not_suicidal)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "177400f4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:09:27.240836Z",
     "iopub.status.busy": "2024-04-11T17:09:27.240586Z",
     "iopub.status.idle": "2024-04-11T17:09:27.249320Z",
     "shell.execute_reply": "2024-04-11T17:09:27.248296Z"
    },
    "papermill": {
     "duration": 0.01974,
     "end_time": "2024-04-11T17:09:27.251256",
     "exception": false,
     "start_time": "2024-04-11T17:09:27.231516",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              label                                        twitterText\n",
      "0          SUICIDAL  \"pengen bunuh diri aja\"\\nyang paling sering te...\n",
      "1          SUICIDAL  Secara tidak langsung punya Fikiran pengen bun...\n",
      "2          SUICIDAL          sejam lalu haha hihi sekarang pengen mati\n",
      "3          SUICIDAL  Baru kelar 1 masalah datang lagi malah lainnya...\n",
      "4          SUICIDAL       fuck bundir aku lagi pgn mabok smpk ga sadar\n",
      "...             ...                                                ...\n",
      "17459  NOT_SUICIDAL  @kkmanti @kangadang1 @Yanuardychandra @floriel...\n",
      "17460  NOT_SUICIDAL  @Don_gindo Kok dek akibat singo di ajar makan ...\n",
      "17461  NOT_SUICIDAL  Mande :\"Jadi mo cium gak\"\\nSurya :\"ntar aja ka...\n",
      "17462  NOT_SUICIDAL  Tuhan...knp z harus kenal,cyg,n nikah dg org y...\n",
      "17463  NOT_SUICIDAL  Demi tuhan, ini narasi paling berkesan. Gemes,...\n",
      "\n",
      "[17464 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "df = pd.concat([df_suicidal, df_not_suicidal])\n",
    "df = df.reset_index(drop=True)\n",
    "print(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c6f97162",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:09:27.269616Z",
     "iopub.status.busy": "2024-04-11T17:09:27.268987Z",
     "iopub.status.idle": "2024-04-11T17:09:27.293430Z",
     "shell.execute_reply": "2024-04-11T17:09:27.292486Z"
    },
    "papermill": {
     "duration": 0.035899,
     "end_time": "2024-04-11T17:09:27.295643",
     "exception": false,
     "start_time": "2024-04-11T17:09:27.259744",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       label                                        twitterText\n",
      "0          1  \"pengen bunuh diri aja\"\\nyang paling sering te...\n",
      "1          1  Secara tidak langsung punya Fikiran pengen bun...\n",
      "2          1          sejam lalu haha hihi sekarang pengen mati\n",
      "3          1  Baru kelar 1 masalah datang lagi malah lainnya...\n",
      "4          1       fuck bundir aku lagi pgn mabok smpk ga sadar\n",
      "...      ...                                                ...\n",
      "17459      0  @kkmanti @kangadang1 @Yanuardychandra @floriel...\n",
      "17460      0  @Don_gindo Kok dek akibat singo di ajar makan ...\n",
      "17461      0  Mande :\"Jadi mo cium gak\"\\nSurya :\"ntar aja ka...\n",
      "17462      0  Tuhan...knp z harus kenal,cyg,n nikah dg org y...\n",
      "17463      0  Demi tuhan, ini narasi paling berkesan. Gemes,...\n",
      "\n",
      "[17464 rows x 2 columns]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_24/3044191695.py:1: FutureWarning: Downcasting behavior in `replace` is deprecated and will be removed in a future version. To retain the old behavior, explicitly call `result.infer_objects(copy=False)`. To opt-in to the future behavior, set `pd.set_option('future.no_silent_downcasting', True)`\n",
      "  df['label'] = df['label'].replace({'SUICIDAL': 1, 'NOT_SUICIDAL': 0})\n"
     ]
    }
   ],
   "source": [
    "df['label'] = df['label'].replace({'SUICIDAL': 1, 'NOT_SUICIDAL': 0})\n",
    "print(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c65fe3f",
   "metadata": {
    "papermill": {
     "duration": 0.008659,
     "end_time": "2024-04-11T17:09:27.314256",
     "exception": false,
     "start_time": "2024-04-11T17:09:27.305597",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e284faba",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:09:27.333014Z",
     "iopub.status.busy": "2024-04-11T17:09:27.332771Z",
     "iopub.status.idle": "2024-04-11T17:09:53.945314Z",
     "shell.execute_reply": "2024-04-11T17:09:53.944414Z"
    },
    "papermill": {
     "duration": 26.624617,
     "end_time": "2024-04-11T17:09:53.947726",
     "exception": false,
     "start_time": "2024-04-11T17:09:27.323109",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting indonlp\r\n",
      "  Downloading indoNLP-0.3.4-py3-none-any.whl.metadata (3.4 kB)\r\n",
      "Downloading indoNLP-0.3.4-py3-none-any.whl (121 kB)\r\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m121.9/121.9 kB\u001b[0m \u001b[31m1.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\r\n",
      "\u001b[?25hInstalling collected packages: indonlp\r\n",
      "Successfully installed indonlp-0.3.4\r\n",
      "Collecting unidecode\r\n",
      "  Downloading Unidecode-1.3.8-py3-none-any.whl.metadata (13 kB)\r\n",
      "Downloading Unidecode-1.3.8-py3-none-any.whl (235 kB)\r\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m235.5/235.5 kB\u001b[0m \u001b[31m2.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\r\n",
      "\u001b[?25hInstalling collected packages: unidecode\r\n",
      "Successfully installed unidecode-1.3.8\r\n"
     ]
    }
   ],
   "source": [
    "!pip install indonlp\n",
    "!pip install unidecode"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71e8f990",
   "metadata": {
    "papermill": {
     "duration": 0.009751,
     "end_time": "2024-04-11T17:09:53.967970",
     "exception": false,
     "start_time": "2024-04-11T17:09:53.958219",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Data Pre Processing"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "293247cf",
   "metadata": {
    "papermill": {
     "duration": 0.009597,
     "end_time": "2024-04-11T17:09:53.987346",
     "exception": false,
     "start_time": "2024-04-11T17:09:53.977749",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "1. Transliteration (there's cases makası, whose letter 'i' is a Turkish lowercase, making it seem like a 'new' word. We try to make it uniform to English letters, thus becoming 'makasi')\n",
    "2. Remove newlines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5aa86928",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:09:54.008843Z",
     "iopub.status.busy": "2024-04-11T17:09:54.008544Z",
     "iopub.status.idle": "2024-04-11T17:09:54.447011Z",
     "shell.execute_reply": "2024-04-11T17:09:54.446214Z"
    },
    "papermill": {
     "duration": 0.452181,
     "end_time": "2024-04-11T17:09:54.449423",
     "exception": false,
     "start_time": "2024-04-11T17:09:53.997242",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from unidecode import unidecode\n",
    "df['twitterText'] = df['twitterText'].apply(unidecode)\n",
    "df['twitterText'] = df['twitterText'].str.replace(\"\\n\", \"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "200a97f6",
   "metadata": {
    "papermill": {
     "duration": 0.009899,
     "end_time": "2024-04-11T17:09:54.469879",
     "exception": false,
     "start_time": "2024-04-11T17:09:54.459980",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "3. Remove Elongation\n",
    "4. Convert Emoji To Indonesian Words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "cecb86cb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:09:54.491709Z",
     "iopub.status.busy": "2024-04-11T17:09:54.491129Z",
     "iopub.status.idle": "2024-04-11T17:10:08.824721Z",
     "shell.execute_reply": "2024-04-11T17:10:08.823714Z"
    },
    "papermill": {
     "duration": 14.347179,
     "end_time": "2024-04-11T17:10:08.827105",
     "exception": false,
     "start_time": "2024-04-11T17:09:54.479926",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "       label                                        twitterText\n",
      "0          1  \"pengen bunuh diri aja\"yang paling sering teru...\n",
      "1          1  Secara tidak langsung punya Fikiran pengen bun...\n",
      "2          1          sejam lalu haha hihi sekarang pengen mati\n",
      "3          1  Baru kelar 1 masalah datang lagi malah lainnya...\n",
      "4          1       fuck bundir aku lagi pgn mabok smpk ga sadar\n",
      "...      ...                                                ...\n",
      "17459      0  @kkmanti @kangadang1 @Yanuardychandra @floriel...\n",
      "17460      0  @Don_gindo Kok dek akibat singo di ajar makan ...\n",
      "17461      0  Mande :\"Jadi mo cium gak\"Surya :\"ntar aja kalo...\n",
      "17462      0  Tuhan...knp z harus kenal,cyg,n nikah dg org y...\n",
      "17463      0  Demi tuhan, ini narasi paling berkesan. Gemes,...\n",
      "\n",
      "[17464 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "from indoNLP.preprocessing import emoji_to_words, replace_word_elongation, remove_url, pipeline\n",
    "\n",
    "def preprocessing(twitterText):\n",
    "    pipe = pipeline([emoji_to_words, replace_word_elongation, remove_url])\n",
    "    return pipe(twitterText)\n",
    "df['twitterText'] = df['twitterText'].apply(preprocessing)\n",
    "print(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0ca9fbf",
   "metadata": {
    "papermill": {
     "duration": 0.010956,
     "end_time": "2024-04-11T17:10:08.850067",
     "exception": false,
     "start_time": "2024-04-11T17:10:08.839111",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "5. Formalize Slang Words (Lemmatization)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f742fd91",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:10:08.871781Z",
     "iopub.status.busy": "2024-04-11T17:10:08.871477Z",
     "iopub.status.idle": "2024-04-11T17:10:08.951011Z",
     "shell.execute_reply": "2024-04-11T17:10:08.950022Z"
    },
    "papermill": {
     "duration": 0.092827,
     "end_time": "2024-04-11T17:10:08.953135",
     "exception": false,
     "start_time": "2024-04-11T17:10:08.860308",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          slang         formal\n",
      "0          woww            wow\n",
      "1         aminn           amin\n",
      "2           met        selamat\n",
      "3        netaas        menetas\n",
      "4       keberpa       keberapa\n",
      "...         ...            ...\n",
      "15001  gataunya  enggak taunya\n",
      "15002      gtau     enggak tau\n",
      "15003     gatau     enggak tau\n",
      "15004     fans2        fan-fan\n",
      "15005   gaharus   enggak harus\n",
      "\n",
      "[15006 rows x 2 columns]\n"
     ]
    }
   ],
   "source": [
    "# Step 1: Read dataset of slang vs formal words\n",
    "# Taken from Colloquial Indonesian Lexicon by Salsabila, Winatmoko, Septriandi, and Jamal\n",
    "# https://ieeexplore.ieee.org/abstract/document/8629151\n",
    "slang_df = pd.read_csv('/kaggle/input/manual-annotation-indo-twitter-suicide/colloquial-indonesian-lexicon.csv', usecols=['slang', 'formal'])\n",
    "print(slang_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "5b9f1035",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:10:08.976930Z",
     "iopub.status.busy": "2024-04-11T17:10:08.976662Z",
     "iopub.status.idle": "2024-04-11T17:10:08.985542Z",
     "shell.execute_reply": "2024-04-11T17:10:08.984743Z"
    },
    "papermill": {
     "duration": 0.021692,
     "end_time": "2024-04-11T17:10:08.987242",
     "exception": false,
     "start_time": "2024-04-11T17:10:08.965550",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Step 2: Create a dictionary mapping slang words to their normalized versions\n",
    "# Note: Is equivalent to IndoNLP's replace_slang(), but we also added specific suicidal slangs here, which don't exist in the dataset\n",
    "slang_mapping = dict(zip(slang_df['slang'], slang_df['formal']))\n",
    "slang_mapping.update({\n",
    "    'm*ti': 'mati',\n",
    "    'm4ti': 'mati',\n",
    "    'bundir': 'bunuh diri',\n",
    "    'bund1r': 'bunuh diri'\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "51809b2e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:10:09.008628Z",
     "iopub.status.busy": "2024-04-11T17:10:09.008385Z",
     "iopub.status.idle": "2024-04-11T17:10:09.223669Z",
     "shell.execute_reply": "2024-04-11T17:10:09.222978Z"
    },
    "papermill": {
     "duration": 0.228054,
     "end_time": "2024-04-11T17:10:09.225595",
     "exception": false,
     "start_time": "2024-04-11T17:10:08.997541",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import re\n",
    "# Step 3: Compile regular expressions for each slang word\n",
    "slang_patterns = {re.compile(r'\\b' + re.escape(slang) + r'\\b'): normalized for slang, normalized in slang_mapping.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "dbac9aea",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:10:09.247491Z",
     "iopub.status.busy": "2024-04-11T17:10:09.247194Z",
     "iopub.status.idle": "2024-04-11T17:15:33.613924Z",
     "shell.execute_reply": "2024-04-11T17:15:33.613040Z"
    },
    "papermill": {
     "duration": 324.391158,
     "end_time": "2024-04-11T17:15:33.627237",
     "exception": false,
     "start_time": "2024-04-11T17:10:09.236079",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<bound method NDFrame.head of        index  label                                        twitterText\n",
      "0          0      1  \"pengin bunuh diri saja\"yang paling sering ter...\n",
      "1          1      1  Secara tidak langsung punya Fikiran pengin bun...\n",
      "2          2      1          sejam lalu haha hihi sekarang pengin mati\n",
      "3          3      1  Baru kelar 1 masalah datang lagi malah lainnya...\n",
      "4          4      1  fuck bunuh diri aku lagi pengin mabok smpk eng...\n",
      "...      ...    ...                                                ...\n",
      "17459  17459      0  @kkmanti @kangadang1 @Yanuardychandra @floriel...\n",
      "17460  17460      0  @Don_gindo Kok dek akibat singo di ajar makan ...\n",
      "17461  17461      0  Mande :\"Jadi mau cium enggak\"Surya :\"entar saj...\n",
      "17462  17462      0  Tuhan...kenapa saja harus kenal,cyg,n nikah de...\n",
      "17463  17463      0  Demi tuhan, ini narasi paling berkesan. Gemes,...\n",
      "\n",
      "[17464 rows x 3 columns]>\n"
     ]
    }
   ],
   "source": [
    "# Step 4: Iterate through the Twitter dataset and replace slang words with normalized versions\n",
    "def replace_slang(text):\n",
    "    for pattern, normalized in slang_patterns.items():\n",
    "        text = pattern.sub(normalized, text)\n",
    "    return text\n",
    "\n",
    "df['twitterText'] = df['twitterText'].apply(replace_slang)\n",
    "df = df.reset_index()\n",
    "print(df.head)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "160f2e0d",
   "metadata": {
    "papermill": {
     "duration": 0.010667,
     "end_time": "2024-04-11T17:15:33.648591",
     "exception": false,
     "start_time": "2024-04-11T17:15:33.637924",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "6. Remove twitter tags (@user)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "6494bd82",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:15:33.671046Z",
     "iopub.status.busy": "2024-04-11T17:15:33.670732Z",
     "iopub.status.idle": "2024-04-11T17:15:33.710196Z",
     "shell.execute_reply": "2024-04-11T17:15:33.709194Z"
    },
    "papermill": {
     "duration": 0.053178,
     "end_time": "2024-04-11T17:15:33.712093",
     "exception": false,
     "start_time": "2024-04-11T17:15:33.658915",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<bound method NDFrame.head of        index  label                                        twitterText\n",
      "0          0      1  \"pengin bunuh diri saja\"yang paling sering ter...\n",
      "1          1      1  Secara tidak langsung punya Fikiran pengin bun...\n",
      "2          2      1          sejam lalu haha hihi sekarang pengin mati\n",
      "3          3      1  Baru kelar 1 masalah datang lagi malah lainnya...\n",
      "4          4      1  fuck bunuh diri aku lagi pengin mabok smpk eng...\n",
      "...      ...    ...                                                ...\n",
      "17459  17459      0      Habis TL adanya Ribut Terus.Gak ada yang B...\n",
      "17460  17460      0   Kok dek akibat singo di ajar makan 2 buah pis...\n",
      "17461  17461      0  Mande :\"Jadi mau cium enggak\"Surya :\"entar saj...\n",
      "17462  17462      0  Tuhan...kenapa saja harus kenal,cyg,n nikah de...\n",
      "17463  17463      0  Demi tuhan, ini narasi paling berkesan. Gemes,...\n",
      "\n",
      "[17464 rows x 3 columns]>\n"
     ]
    }
   ],
   "source": [
    "def remove_twitter_tag(text):\n",
    "    return re.sub(r'@\\w+', '', text)\n",
    "\n",
    "df['twitterText'] = df['twitterText'].apply(remove_twitter_tag)\n",
    "print(df.head)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "70826cc5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:15:33.734775Z",
     "iopub.status.busy": "2024-04-11T17:15:33.734520Z",
     "iopub.status.idle": "2024-04-11T17:15:33.745468Z",
     "shell.execute_reply": "2024-04-11T17:15:33.744329Z"
    },
    "papermill": {
     "duration": 0.024382,
     "end_time": "2024-04-11T17:15:33.747444",
     "exception": false,
     "start_time": "2024-04-11T17:15:33.723062",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1 0]\n"
     ]
    }
   ],
   "source": [
    "print(df['label'].unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0a379459",
   "metadata": {
    "papermill": {
     "duration": 0.010905,
     "end_time": "2024-04-11T17:15:33.769064",
     "exception": false,
     "start_time": "2024-04-11T17:15:33.758159",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "# Classify Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "47c1ca1e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:15:33.791887Z",
     "iopub.status.busy": "2024-04-11T17:15:33.791629Z",
     "iopub.status.idle": "2024-04-11T17:16:32.409708Z",
     "shell.execute_reply": "2024-04-11T17:16:32.408679Z"
    },
    "papermill": {
     "duration": 58.632186,
     "end_time": "2024-04-11T17:16:32.411999",
     "exception": false,
     "start_time": "2024-04-11T17:15:33.779813",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "6307d75b0704430d9d47399ddabecc67",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer_config.json:   0%|          | 0.00/2.00 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "222a3cbb9a954b9683c76dd7ed0630e1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "vocab.txt:   0%|          | 0.00/229k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a7138ffc64894f03b5bdaf7e2d096dea",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "special_tokens_map.json:   0%|          | 0.00/112 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "64b9af66d54b400bb62727647077d7ca",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "config.json:   0%|          | 0.00/1.53k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "22da459c521045539fab0d5483186d5f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "pytorch_model.bin:   0%|          | 0.00/1.34G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()\n",
      "  return self.fget.__get__(instance, owner)()\n"
     ]
    }
   ],
   "source": [
    "# Load the IndobertModel and tokenizer\n",
    "from transformers import BertTokenizer, BertModel\n",
    "model_name = 'indobenchmark/indobert-large-p1'\n",
    "tokenizer = BertTokenizer.from_pretrained(model_name)\n",
    "bert_model = BertModel.from_pretrained(model_name)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1f632704",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:32.439245Z",
     "iopub.status.busy": "2024-04-11T17:16:32.438468Z",
     "iopub.status.idle": "2024-04-11T17:16:32.444318Z",
     "shell.execute_reply": "2024-04-11T17:16:32.443554Z"
    },
    "papermill": {
     "duration": 0.020463,
     "end_time": "2024-04-11T17:16:32.446146",
     "exception": false,
     "start_time": "2024-04-11T17:16:32.425683",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Prepare the data\n",
    "texts = df['twitterText'].tolist()\n",
    "labels = df['label'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d214e5e0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:32.471920Z",
     "iopub.status.busy": "2024-04-11T17:16:32.471674Z",
     "iopub.status.idle": "2024-04-11T17:16:34.191734Z",
     "shell.execute_reply": "2024-04-11T17:16:34.190943Z"
    },
    "papermill": {
     "duration": 1.735157,
     "end_time": "2024-04-11T17:16:34.194141",
     "exception": false,
     "start_time": "2024-04-11T17:16:32.458984",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "# Split the data into train and test sets 8:2\n",
    "train_texts, test_texts, train_labels, test_labels = train_test_split(texts, labels, test_size=0.2, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "fda3f076",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:34.221943Z",
     "iopub.status.busy": "2024-04-11T17:16:34.221465Z",
     "iopub.status.idle": "2024-04-11T17:16:34.376306Z",
     "shell.execute_reply": "2024-04-11T17:16:34.375340Z"
    },
    "papermill": {
     "duration": 0.171407,
     "end_time": "2024-04-11T17:16:34.378389",
     "exception": false,
     "start_time": "2024-04-11T17:16:34.206982",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "\n",
    "if torch.cuda.is_available():\n",
    "    device = torch.device(\"cuda\")\n",
    "else:\n",
    "    print('No GPU available, using the CPU instead.')\n",
    "    device = torch.device(\"cpu\")\n",
    "\n",
    "def tokenize_and_create_input_tensors(texts):\n",
    "    input_ids = []\n",
    "    attention_masks = []\n",
    "\n",
    "    for text in texts:\n",
    "        encoded_dict = tokenizer.encode_plus(\n",
    "            text,\n",
    "            add_special_tokens=True,\n",
    "            max_length=128,  # You can adjust the max_length as per your needs\n",
    "            padding='max_length',\n",
    "            truncation=True,\n",
    "            return_attention_mask=True,\n",
    "            return_tensors='pt'\n",
    "        )\n",
    "\n",
    "        input_ids.append(encoded_dict['input_ids'])\n",
    "        attention_masks.append(encoded_dict['attention_mask'])\n",
    "\n",
    "    # Convert the lists to tensors\n",
    "    input_ids = torch.cat(input_ids, dim=0)\n",
    "    attention_masks = torch.cat(attention_masks, dim=0)\n",
    "\n",
    "    return input_ids, attention_masks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "60762367",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:34.404948Z",
     "iopub.status.busy": "2024-04-11T17:16:34.404648Z",
     "iopub.status.idle": "2024-04-11T17:16:51.046396Z",
     "shell.execute_reply": "2024-04-11T17:16:51.045401Z"
    },
    "papermill": {
     "duration": 16.656673,
     "end_time": "2024-04-11T17:16:51.048442",
     "exception": false,
     "start_time": "2024-04-11T17:16:34.391769",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Before:\n",
      "torch.Size([13971, 128])\n",
      "torch.Size([13971, 128])\n",
      "torch.Size([13971])\n",
      "After:\n",
      "torch.Size([13971, 128])\n",
      "torch.Size([13971, 128])\n",
      "torch.Size([13971])\n"
     ]
    }
   ],
   "source": [
    "train_input_ids, train_attention_masks = tokenize_and_create_input_tensors(train_texts)\n",
    "train_labels = torch.tensor(train_labels).to(device)\n",
    "\n",
    "print(\"Before:\")\n",
    "print(train_input_ids.shape)\n",
    "print(train_attention_masks.shape)\n",
    "print(train_labels.shape)\n",
    "\n",
    "train_input_ids = train_input_ids.squeeze(dim=1).to(device)\n",
    "train_attention_masks = train_attention_masks.squeeze(dim=1).to(device)\n",
    "\n",
    "print(\"After:\")\n",
    "print(train_input_ids.shape)\n",
    "print(train_attention_masks.shape)\n",
    "print(train_labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "00e10226",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:51.074001Z",
     "iopub.status.busy": "2024-04-11T17:16:51.073701Z",
     "iopub.status.idle": "2024-04-11T17:16:55.110117Z",
     "shell.execute_reply": "2024-04-11T17:16:55.109101Z"
    },
    "papermill": {
     "duration": 4.051576,
     "end_time": "2024-04-11T17:16:55.112423",
     "exception": false,
     "start_time": "2024-04-11T17:16:51.060847",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "After:\n",
      "torch.Size([3493, 128])\n",
      "torch.Size([3493, 128])\n",
      "torch.Size([3493])\n"
     ]
    }
   ],
   "source": [
    "test_input_ids, test_attention_masks = tokenize_and_create_input_tensors(test_texts)\n",
    "test_labels = torch.tensor(test_labels).to(device)\n",
    "\n",
    "test_input_ids = test_input_ids.squeeze(dim=1).to(device)\n",
    "test_attention_masks = test_attention_masks.squeeze(dim=1).to(device)\n",
    "\n",
    "print(\"After:\")\n",
    "print(test_input_ids.shape)\n",
    "print(test_attention_masks.shape)\n",
    "print(test_labels.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "963ebb24",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:55.139239Z",
     "iopub.status.busy": "2024-04-11T17:16:55.138450Z",
     "iopub.status.idle": "2024-04-11T17:16:55.197434Z",
     "shell.execute_reply": "2024-04-11T17:16:55.196693Z"
    },
    "papermill": {
     "duration": 0.07461,
     "end_time": "2024-04-11T17:16:55.199710",
     "exception": false,
     "start_time": "2024-04-11T17:16:55.125100",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "torch.save(train_input_ids, '/kaggle/working/train_input_ids.pt')\n",
    "torch.save(train_attention_masks, '/kaggle/working/train_attention_masks.pt')\n",
    "torch.save(train_labels, '/kaggle/working/train_labels.pt')\n",
    "\n",
    "torch.save(test_input_ids, '/kaggle/working/test_input_ids.pt')\n",
    "torch.save(test_attention_masks, '/kaggle/working/test_attention_masks.pt')\n",
    "torch.save(test_labels, '/kaggle/working/test_labels.pt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "680ecd09",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:55.226010Z",
     "iopub.status.busy": "2024-04-11T17:16:55.225739Z",
     "iopub.status.idle": "2024-04-11T17:16:55.234881Z",
     "shell.execute_reply": "2024-04-11T17:16:55.234137Z"
    },
    "papermill": {
     "duration": 0.023999,
     "end_time": "2024-04-11T17:16:55.236599",
     "exception": false,
     "start_time": "2024-04-11T17:16:55.212600",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class BertCNNClassifier(nn.Module):\n",
    "    def __init__(self, bert_model, num_filters, kernel_sizes, num_classes):\n",
    "        super(BertCNNClassifier, self).__init__()\n",
    "        self.bert = bert_model\n",
    "        self.conv1d = nn.ModuleList([nn.Conv1d(bert_model.config.hidden_size, num_filters, kernel_size, bias=True) for kernel_size in kernel_sizes])\n",
    "        self.dropout = nn.Dropout(0.45)\n",
    "        self.fc = nn.Linear(len(kernel_sizes) * num_filters, num_classes, bias=True)\n",
    "\n",
    "    def forward(self, input_ids, attention_mask):\n",
    "        token_type_ids = torch.zeros_like(input_ids)  # Create token_type_ids tensor\n",
    "        outputs = self.bert(input_ids=input_ids, attention_mask=attention_mask, token_type_ids=token_type_ids)\n",
    "        last_hidden_state = outputs[0].permute(0, 2, 1)  # (batch_size, hidden_size, seq_len)\n",
    "\n",
    "        conved = [torch.relu(conv(last_hidden_state)) for conv in self.conv1d]  # Apply convolution to each kernel\n",
    "        pooled = [F.max_pool1d(conv, conv.shape[2]).squeeze(2) for conv in conved]  # Max-over-time pooling\n",
    "        cat = self.dropout(torch.cat(pooled, dim=1))  # Concatenate pooled outputs and apply dropout\n",
    "\n",
    "        logits = self.fc(cat)  # Apply linear layer for classification\n",
    "        return logits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "a80ad996",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:55.262512Z",
     "iopub.status.busy": "2024-04-11T17:16:55.262200Z",
     "iopub.status.idle": "2024-04-11T17:16:55.709022Z",
     "shell.execute_reply": "2024-04-11T17:16:55.708056Z"
    },
    "papermill": {
     "duration": 0.461993,
     "end_time": "2024-04-11T17:16:55.711288",
     "exception": false,
     "start_time": "2024-04-11T17:16:55.249295",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "num_filters = 700  # Number of filters for each kernel size\n",
    "kernel_sizes = [2, 3, 4]  # List of kernel sizes\n",
    "num_classes = 2\n",
    "\n",
    "# Initialize the CNN model\n",
    "model = BertCNNClassifier(bert_model, num_filters, kernel_sizes, num_classes).to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f277d075",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:55.740148Z",
     "iopub.status.busy": "2024-04-11T17:16:55.739855Z",
     "iopub.status.idle": "2024-04-11T17:16:55.747147Z",
     "shell.execute_reply": "2024-04-11T17:16:55.746239Z"
    },
    "papermill": {
     "duration": 0.023766,
     "end_time": "2024-04-11T17:16:55.749096",
     "exception": false,
     "start_time": "2024-04-11T17:16:55.725330",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Define the loss function and optimizer\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=2.45e-5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "e571700e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:55.776423Z",
     "iopub.status.busy": "2024-04-11T17:16:55.775909Z",
     "iopub.status.idle": "2024-04-11T17:16:55.781462Z",
     "shell.execute_reply": "2024-04-11T17:16:55.780578Z"
    },
    "papermill": {
     "duration": 0.021144,
     "end_time": "2024-04-11T17:16:55.783178",
     "exception": false,
     "start_time": "2024-04-11T17:16:55.762034",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from torch.utils.data import TensorDataset, DataLoader, RandomSampler, SequentialSampler\n",
    "\n",
    "# Create TensorDatasets and DataLoaders\n",
    "batch_size = 32\n",
    "\n",
    "train_data = TensorDataset(train_input_ids, train_attention_masks, train_labels)\n",
    "train_sampler = RandomSampler(train_data)\n",
    "train_dataloader = DataLoader(train_data, sampler=train_sampler, batch_size=batch_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ff1f8466",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:55.809708Z",
     "iopub.status.busy": "2024-04-11T17:16:55.808982Z",
     "iopub.status.idle": "2024-04-11T17:16:55.813643Z",
     "shell.execute_reply": "2024-04-11T17:16:55.812776Z"
    },
    "papermill": {
     "duration": 0.019767,
     "end_time": "2024-04-11T17:16:55.815553",
     "exception": false,
     "start_time": "2024-04-11T17:16:55.795786",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "437\n"
     ]
    }
   ],
   "source": [
    "print(len(train_dataloader)) # number of batches"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e4450e37",
   "metadata": {
    "papermill": {
     "duration": 0.012382,
     "end_time": "2024-04-11T17:16:55.840448",
     "exception": false,
     "start_time": "2024-04-11T17:16:55.828066",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "Training for 4 epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "810fb0ac",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T17:16:55.867855Z",
     "iopub.status.busy": "2024-04-11T17:16:55.867050Z",
     "iopub.status.idle": "2024-04-11T18:22:52.478128Z",
     "shell.execute_reply": "2024-04-11T18:22:52.477377Z"
    },
    "papermill": {
     "duration": 3956.627357,
     "end_time": "2024-04-11T18:22:52.480564",
     "exception": false,
     "start_time": "2024-04-11T17:16:55.853207",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch  0  batch  0   2024-04-11 17:16:56.074022\n",
      "epoch  0  batch  100   2024-04-11 17:20:38.508751\n",
      "epoch  0  batch  200   2024-04-11 17:24:24.889390\n",
      "epoch  0  batch  300   2024-04-11 17:28:11.303922\n",
      "epoch  0  batch  400   2024-04-11 17:31:57.684879\n",
      "epoch  1  batch  0   2024-04-11 17:33:20.671131\n",
      "epoch  1  batch  100   2024-04-11 17:37:06.917701\n",
      "epoch  1  batch  200   2024-04-11 17:40:53.545118\n",
      "epoch  1  batch  300   2024-04-11 17:44:39.882826\n",
      "epoch  1  batch  400   2024-04-11 17:48:26.369094\n",
      "epoch  2  batch  0   2024-04-11 17:49:49.400635\n",
      "epoch  2  batch  100   2024-04-11 17:53:35.742332\n",
      "epoch  2  batch  200   2024-04-11 17:57:22.104398\n",
      "epoch  2  batch  300   2024-04-11 18:01:07.969647\n",
      "epoch  2  batch  400   2024-04-11 18:04:54.562447\n",
      "epoch  3  batch  0   2024-04-11 18:06:17.656988\n",
      "epoch  3  batch  100   2024-04-11 18:10:04.181919\n",
      "epoch  3  batch  200   2024-04-11 18:13:50.641016\n",
      "epoch  3  batch  300   2024-04-11 18:17:37.167376\n",
      "epoch  3  batch  400   2024-04-11 18:21:23.718253\n"
     ]
    }
   ],
   "source": [
    "import datetime\n",
    "import gc\n",
    "# Train the model\n",
    "\n",
    "def save_ckp(state):\n",
    "    f_path = '/kaggle/working/' + 'checkpoint_epoch' + str(state['epoch']) + '.pt'\n",
    "    torch.save(state, f_path)\n",
    "\n",
    "num_epoch = 4\n",
    "gc.collect()\n",
    "torch.cuda.empty_cache()\n",
    "\n",
    "for i in range(num_epoch):\n",
    "  num_batch = 0\n",
    "  for batch in train_dataloader:\n",
    "    if (num_batch % 100 == 0):\n",
    "      print('epoch ', i, ' batch ', num_batch, ' ', datetime.datetime.now())\n",
    "    input_ids, attention_masks, labels = batch\n",
    "    outputs = model(input_ids, attention_masks)\n",
    "    loss = criterion(outputs, labels)\n",
    "\n",
    "    optimizer.zero_grad()\n",
    "    loss.backward()\n",
    "    optimizer.step()\n",
    "    num_batch+=1\n",
    "    del input_ids\n",
    "    del attention_masks\n",
    "    del labels\n",
    "    del outputs\n",
    "    gc.collect()\n",
    "    torch.cuda.empty_cache()\n",
    "\n",
    "checkpoint = {\n",
    "    'epoch': num_epoch - 1,\n",
    "    'state_dict': model.state_dict(),\n",
    "    'optimizer': optimizer.state_dict()\n",
    "}\n",
    "save_ckp(checkpoint)\n",
    "\n",
    "del train_data\n",
    "del train_sampler\n",
    "del train_dataloader\n",
    "gc.collect()\n",
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "a35466ad",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T18:22:56.630222Z",
     "iopub.status.busy": "2024-04-11T18:22:56.628953Z",
     "iopub.status.idle": "2024-04-11T18:24:13.079947Z",
     "shell.execute_reply": "2024-04-11T18:24:13.078905Z"
    },
    "papermill": {
     "duration": 76.517452,
     "end_time": "2024-04-11T18:24:13.099042",
     "exception": false,
     "start_time": "2024-04-11T18:22:56.581590",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9107\n",
      "Recall: 0.9349\n",
      "Precision: 0.8934\n",
      "F1-Score: 0.9137\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score, recall_score, precision_score, f1_score\n",
    "import torch.nn.functional as F\n",
    "\n",
    "# Evaluate the model on the test set\n",
    "model.eval()\n",
    "test_preds = []\n",
    "test_proba = []\n",
    "test_labels_list = []\n",
    "\n",
    "test_data = TensorDataset(test_input_ids, test_attention_masks, test_labels)\n",
    "test_sampler = SequentialSampler(test_data)\n",
    "test_dataloader = DataLoader(test_data, sampler=test_sampler, batch_size=batch_size)\n",
    "\n",
    "with torch.no_grad():\n",
    "    for batch in test_dataloader:\n",
    "        input_ids, attention_masks, labels = batch\n",
    "        outputs = model(input_ids, attention_masks)\n",
    "        preds = torch.argmax(outputs, dim=1)\n",
    "        test_preds.extend(preds.tolist())\n",
    "        test_proba.extend(F.softmax(outputs, dim=1)[:, 1])\n",
    "        test_labels_list.extend(labels.tolist())\n",
    "\n",
    "accuracy = accuracy_score(test_labels_list, test_preds)\n",
    "recall = recall_score(test_labels_list, test_preds)\n",
    "precision = precision_score(test_labels_list, test_preds)\n",
    "f1 = f1_score(test_labels_list, test_preds)\n",
    "test_proba = [i.cpu().tolist() for i in test_proba]\n",
    "\n",
    "print(f\"Accuracy: {accuracy:.4f}\")\n",
    "print(f\"Recall: {recall:.4f}\")\n",
    "print(f\"Precision: {precision:.4f}\")\n",
    "print(f\"F1-Score: {f1:.4f}\")\n",
    "\n",
    "del test_data\n",
    "del test_sampler\n",
    "del test_dataloader\n",
    "gc.collect()\n",
    "torch.cuda.empty_cache()\n",
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "98c3479e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T18:24:13.134067Z",
     "iopub.status.busy": "2024-04-11T18:24:13.133732Z",
     "iopub.status.idle": "2024-04-11T18:24:13.728647Z",
     "shell.execute_reply": "2024-04-11T18:24:13.727652Z"
    },
    "papermill": {
     "duration": 0.614504,
     "end_time": "2024-04-11T18:24:13.730698",
     "exception": false,
     "start_time": "2024-04-11T18:24:13.116194",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "cf_matrix = confusion_matrix(test_labels_list, test_preds)\n",
    "\n",
    "cf = sns.heatmap(cf_matrix, annot=True, fmt='d')\n",
    "cf.xaxis.set_ticklabels(['Negative', 'Positive'])\n",
    "cf.set_xlabel('Predicted Label')\n",
    "cf.yaxis.set_ticklabels(['Negative', 'Positive'])\n",
    "cf.set_ylabel('True Label')\n",
    "cf.set_title(\"Confusion Matrix for Indonesian Language Suicidal Posts Classification\", pad=20)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "bc880e37",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-11T18:24:13.764493Z",
     "iopub.status.busy": "2024-04-11T18:24:13.763978Z",
     "iopub.status.idle": "2024-04-11T18:24:14.103318Z",
     "shell.execute_reply": "2024-04-11T18:24:14.102405Z"
    },
    "papermill": {
     "duration": 0.358527,
     "end_time": "2024-04-11T18:24:14.105341",
     "exception": false,
     "start_time": "2024-04-11T18:24:13.746814",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ROC AUC Score:\n",
      "0.9633421883207285\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_auc_score, roc_curve, RocCurveDisplay\n",
    "\n",
    "def plot_sklearn_roc_curve(y_real, y_pred, roc_score):\n",
    "    fpr, tpr, _ = roc_curve(y_real, y_pred)\n",
    "    df_fpr = pd.DataFrame(fpr)\n",
    "    df_fpr.to_csv(\"/kaggle/working/fpr.csv\")\n",
    "    df_tpr = pd.DataFrame(tpr)\n",
    "    df_tpr.to_csv(\"/kaggle/working/tpr.csv\")\n",
    "    print(\"ROC AUC Score:\")\n",
    "    print(roc_score)\n",
    "    roc_display = RocCurveDisplay(fpr=fpr, tpr=tpr, roc_auc=roc_score, estimator_name='ROC Curve').plot()\n",
    "    roc_display.figure_.set_size_inches(5,5)\n",
    "    plt.plot([0, 1], [0, 1], color='green', linestyle='dashed')\n",
    "    plt.title(\"ROC Curve for Indonesian Language Suicidal Posts Classification\")\n",
    "\n",
    "roc_score = roc_auc_score(test_labels_list, test_proba)\n",
    "plot_sklearn_roc_curve(test_labels_list, test_proba, roc_score)"
   ]
  }
 ],
 "metadata": {
  "kaggle": {
   "accelerator": "nvidiaTeslaT4",
   "dataSources": [
    {
     "datasetId": 4750496,
     "sourceId": 8083396,
     "sourceType": "datasetVersion"
    }
   ],
   "isGpuEnabled": true,
   "isInternetEnabled": true,
   "language": "python",
   "sourceType": "notebook"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 4493.41198,
   "end_time": "2024-04-11T18:24:16.255749",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2024-04-11T17:09:22.843769",
   "version": "2.5.0"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "009afb300a0247b497874ed2439e9c89": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_cfa94f1ef8814f4b8659739f7102d2e1",
       "placeholder": "​",
       "style": "IPY_MODEL_c635ab672b3b42d2a56e6fee82163c62",
       "value": "special_tokens_map.json: 100%"
      }
     },
     "0b73866e982545dd961bc9ecb6d2c42c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "0bab75b657bc41c385360bd310f23b9a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "0be2bef83453458ca466909ddd94f7a4": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_5cb2035b89fa4733920278830602000c",
       "placeholder": "​",
       "style": "IPY_MODEL_966631709b5f450fa1c780d471c4ee5e",
       "value": "vocab.txt: 100%"
      }
     },
     "0d1ad5abeeee438fa560363512b939c6": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "19375d16905e4a4cab001d20752b5869": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "1999773e354f40c5a7a40ab5af52cda2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_b3654128e908454fa35c5559a7bb11da",
       "placeholder": "​",
       "style": "IPY_MODEL_e94b5aaa7ff14a27a8d4930596f4ba0a",
       "value": "tokenizer_config.json: 100%"
      }
     },
     "1a32f9a855ec432f86770bf0922016df": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_4dc1e4797e894194842e6554f48f651f",
       "max": 112.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_d58189f54b964543bf3f27bc0c8e0516",
       "value": 112.0
      }
     },
     "1bcb4071b6b14b19b28ac058779418af": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_ded867d45fe141aca62a6383e4c4c9cd",
       "max": 2.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_f496506177da47038492b6601be6bc1d",
       "value": 2.0
      }
     },
     "1bdb1d314c2c48a9b270590990bb4348": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "212d424172174d58af9476fbf81e9630": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "222a3cbb9a954b9683c76dd7ed0630e1": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_0be2bef83453458ca466909ddd94f7a4",
        "IPY_MODEL_821ac45ab2f3443abe63ffea083a7602",
        "IPY_MODEL_c385b36e2f4f4345b4c44986e49f6aa7"
       ],
       "layout": "IPY_MODEL_be8af13661fc47adaf19378f01130d2a"
      }
     },
     "22da459c521045539fab0d5483186d5f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_874a740d989847f88234cbb5ca2b8369",
        "IPY_MODEL_530b66f95a3141389adb8f90e7fba1b2",
        "IPY_MODEL_7e569e4150cc4b57b7ac2fb0fa6c7c01"
       ],
       "layout": "IPY_MODEL_212d424172174d58af9476fbf81e9630"
      }
     },
     "2dfeac5cfe3f482cac9be06e54f60a02": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "3529dff98da64c168a34ebac301cb5ba": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "417a8f42e43142338434a8bc1a5b49fa": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "48b80663b75e48d58516a8e548576b8e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "4dc1e4797e894194842e6554f48f651f": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "530b66f95a3141389adb8f90e7fba1b2": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_48b80663b75e48d58516a8e548576b8e",
       "max": 1340657204.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_c76c4763ad424a59a8615b8a9d6e2b19",
       "value": 1340657204.0
      }
     },
     "5cb2035b89fa4733920278830602000c": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "5f2d4f0a34e64fd193073d5e5095870a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_b2c7a3c42ac845f9b335aaab372d4727",
       "placeholder": "​",
       "style": "IPY_MODEL_1bdb1d314c2c48a9b270590990bb4348",
       "value": " 1.53k/1.53k [00:00&lt;00:00, 140kB/s]"
      }
     },
     "6307d75b0704430d9d47399ddabecc67": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_1999773e354f40c5a7a40ab5af52cda2",
        "IPY_MODEL_1bcb4071b6b14b19b28ac058779418af",
        "IPY_MODEL_9cfa470786474e6ea7ea2da7a9181c0a"
       ],
       "layout": "IPY_MODEL_0bab75b657bc41c385360bd310f23b9a"
      }
     },
     "64af69877b9744508f245d04e77330fd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "64b9af66d54b400bb62727647077d7ca": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_bf9eccec37254c6691b00874d4cc0f1b",
        "IPY_MODEL_c3e613834e774f4b88a0ccf492aa0663",
        "IPY_MODEL_5f2d4f0a34e64fd193073d5e5095870a"
       ],
       "layout": "IPY_MODEL_f2b0efa8eac049128f9240886f3a83be"
      }
     },
     "668750b890ac4c77a090b640f38c2a93": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "6acb4733d20647fc97b8225cc01a0f73": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "6e907bf0c6024c96a79aa902d84ad946": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "78fb150651644908b60c6fadebce4bee": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "7d241eaf7f1c4e108d8aef18d9734932": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "7e569e4150cc4b57b7ac2fb0fa6c7c01": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_6e907bf0c6024c96a79aa902d84ad946",
       "placeholder": "​",
       "style": "IPY_MODEL_6acb4733d20647fc97b8225cc01a0f73",
       "value": " 1.34G/1.34G [00:46&lt;00:00, 29.7MB/s]"
      }
     },
     "821ac45ab2f3443abe63ffea083a7602": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_78fb150651644908b60c6fadebce4bee",
       "max": 229167.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_be53c9aaf21841d794b0e6f51e1f0be6",
       "value": 229167.0
      }
     },
     "85b21e643c624841b839683e1acd6e7d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_0d1ad5abeeee438fa560363512b939c6",
       "placeholder": "​",
       "style": "IPY_MODEL_417a8f42e43142338434a8bc1a5b49fa",
       "value": " 112/112 [00:00&lt;00:00, 9.75kB/s]"
      }
     },
     "874a740d989847f88234cbb5ca2b8369": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_64af69877b9744508f245d04e77330fd",
       "placeholder": "​",
       "style": "IPY_MODEL_87bbb796b93c49f3b7c4089e315fc296",
       "value": "pytorch_model.bin: 100%"
      }
     },
     "87bbb796b93c49f3b7c4089e315fc296": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "8ed96eb820f7493796e13883cf733563": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "966631709b5f450fa1c780d471c4ee5e": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "9cfa470786474e6ea7ea2da7a9181c0a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_2dfeac5cfe3f482cac9be06e54f60a02",
       "placeholder": "​",
       "style": "IPY_MODEL_7d241eaf7f1c4e108d8aef18d9734932",
       "value": " 2.00/2.00 [00:00&lt;00:00, 157B/s]"
      }
     },
     "a7138ffc64894f03b5bdaf7e2d096dea": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_009afb300a0247b497874ed2439e9c89",
        "IPY_MODEL_1a32f9a855ec432f86770bf0922016df",
        "IPY_MODEL_85b21e643c624841b839683e1acd6e7d"
       ],
       "layout": "IPY_MODEL_0b73866e982545dd961bc9ecb6d2c42c"
      }
     },
     "b2c7a3c42ac845f9b335aaab372d4727": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "b3654128e908454fa35c5559a7bb11da": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "be53c9aaf21841d794b0e6f51e1f0be6": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "be8af13661fc47adaf19378f01130d2a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "bee51f36dfee4d5995f45d1e42693895": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "bf9eccec37254c6691b00874d4cc0f1b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_3529dff98da64c168a34ebac301cb5ba",
       "placeholder": "​",
       "style": "IPY_MODEL_668750b890ac4c77a090b640f38c2a93",
       "value": "config.json: 100%"
      }
     },
     "c2aa0af0632c43cc929508cedf110268": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "c385b36e2f4f4345b4c44986e49f6aa7": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_c2aa0af0632c43cc929508cedf110268",
       "placeholder": "​",
       "style": "IPY_MODEL_8ed96eb820f7493796e13883cf733563",
       "value": " 229k/229k [00:00&lt;00:00, 5.08MB/s]"
      }
     },
     "c3e613834e774f4b88a0ccf492aa0663": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "1.5.0",
       "_view_name": "ProgressView",
       "bar_style": "success",
       "description": "",
       "description_tooltip": null,
       "layout": "IPY_MODEL_bee51f36dfee4d5995f45d1e42693895",
       "max": 1535.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_19375d16905e4a4cab001d20752b5869",
       "value": 1535.0
      }
     },
     "c635ab672b3b42d2a56e6fee82163c62": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "c76c4763ad424a59a8615b8a9d6e2b19": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "cfa94f1ef8814f4b8659739f7102d2e1": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "d58189f54b964543bf3f27bc0c8e0516": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "ded867d45fe141aca62a6383e4c4c9cd": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "e94b5aaa7ff14a27a8d4930596f4ba0a": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "DescriptionStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "DescriptionStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "description_width": ""
      }
     },
     "f2b0efa8eac049128f9240886f3a83be": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "1.2.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "1.2.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "overflow_x": null,
       "overflow_y": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "f496506177da47038492b6601be6bc1d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "1.5.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "1.5.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "1.2.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}